Restitution of Overpaid Tax 9781472561336, 9781849461733

Since the decision of the House of Lords in Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] AC

236 119 2MB

English Pages [352] Year 2013

Recommend Papers

Flat Tax

This new and updated edition of The Flat Tax sets forth the flat-tax plan developed by Robert Hall and Alvin Rabushka, s

177 79 5MB Read more

The Foundations of Restitution for Wrongs 9781472560032, 9781841136479

Restitution for wrongs’, or ‘restitutionary damages’, is the judicial award which compels the wrongdoer to give up to th

166 14 2MB Read more

Landmark Cases in the Law of Restitution 9781472559876, 9781841135885

It is now well established that the law of unjust enrichment forms an important and distinctive part of the English law

161 49 2MB Read more

Restitution and the Politics of Repair: Tropes, Imaginaries, Theory 9781474453110

Analyses the social imaginary of undoing, repair and return underpinning the international norm of restitution-making Ap

114 20 1MB Read more

Post-Communist Restitution and the Rule of Law 9786155211713

Eastern European societies underwent large-scale deprivations of property by the authoritarian regimes, beginning after

128 13 2MB Read more

Post-communist restitution and the rule of law 9786155211713, 9789639776401

Eastern European societies underwent large-scale deprivations of property by the authoritarian regimes, beginning after

100 99 67MB Read more

OECD Tax Policy Studies No.16 Fundamental Reform of Corporate Income Tax (Oecd Tax Policy Studies) 9264038116, 9789264038110

Policy makers in OECD countries are concerned about whether they can maintain their current levels of corporate income t

414 8 2MB Read more

Restitution in Private International Law 9781472562180, 9781841131429

This important new book fills a large gap in legal literature by examining restitution in private international law,incl

171 95 1MB Read more

The Ecology of Tax Systems 9781788116862, 9781788116879

This groundbreaking book analyzes how the ecology of taxation is fundamental for the success or failure of tax systems.

473 29 348KB Read more

Tax Policy Reform and Economic Growth, OECD Tax Policy Studies

479 82 2MB Read more

Restitution of Overpaid Tax
9781472561336, 9781849461733

Author / Uploaded
Steven Elliott
Birke Häcker
Charles Mitchell (editors)

0 0 0
Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up

File loading please wait...

Citation preview

Preface The essays in this collection were, with one exception, first presented at a conference held at Merton College, Oxford on 9 and 10 July 2010. Publication has been delayed to allow the authors to take account of the Supreme Court’s decision in Test Claimants in the FII Group Litigation v HMRC [2012] UKSC 19, [2012] 2 AC 337. In the meantime, three papers given at the conference have been published elsewhere: Charles Mitchell, ‘Restitution of Ultra Vires Payments by Public Bodies’ [2010] Public Law 747; Elise Bant, ‘Change of Position as a Defence to Restitution of Unlawfully Exacted Tax’ [2012] Lloyd’s Maritime and Commercial Law Quarterly 122; James Lee, ‘Defences to Claims for Restitution of Overpaid Tax’ [2012] Restitution Law Review 13. Thanks are due to One Essex Court and to Pump Court Tax Chambers, including Richard Vallat, for sponsoring and hosting the conference at Merton College. Thanks are further due to Eoin O’Dell, who promoted the conference and collection and who solicited the contributions. Finally, the editors thank Hart Publishing for their dedication and efforts. Steven Elliott made no editorial contribution to essays concerning his current cases. Steven Elliott Birke Häcker Charles Mitchell

Contributors Catherine Barnard is a Fellow of Trinity College, Cambridge, and a Professor of European Union Law at Cambridge University. Monica Bhandari is a Senior Lecturer in Tax Law at University College London. Robert Chambers is a Professor of Property Law at University College London. Niamh Cleary is a Visiting Lecturer in Law at King’s College London. Niamh Connolly is an Assistant Professor of Law at Trinity College Dublin. Simone Degeling is a Professor of Law at the University of New South Wales. Steven Elliott is a barrister at One Essex Court. Nelson Enonchong is Barber Professor of Law in the University of Birmingham, and a barrister at No 5 Chambers, Birmingham. Julian Ghosh QC is a member of Pump Court Tax Chambers. He is also a Judge of the FirstTier and Deputy Judge of the Upper-Tier Tax Tribunals, and a Visiting Professor at the International Tax Centre, University of Leiden. Birke Häcker is a Senior Research Fellow at the Max Planck Institute for Tax Law and Public Finance in Munich, and a Fellow of All Souls College, Oxford. Charles Mitchell is a Professor of Law at University College London. Anne Sanders is a Research Fellow and Lecturer in Law at the University of Cologne. Maximilian Schlote is a pupil barrister at One Essex Court. Duncan Sheehan is a Professor of Commercial Law at the University of East Anglia. Charlie Webb is a Senior Lecturer in Law at the London School of Economics. Rebecca Williams is a Fellow and Tutor in Law at Pembroke College, Oxford, and a CUF Lecturer in Law at Oxford University.

Table of Cases Australia ACI Operations Pty Ltd v Commonwealth of Australia [1995] FCA 1770............................314 AID/WATCH Inc v Commissioner of Taxation of the Commonwealth Bank of Australia (2010) 241 CLR 539...................................................................................................316, 321 Antill Ranger & Co Pty Ltd v Commissioner of Motor Transport (1955) 93 CLR 83............320 Attorney General v Gray [1977] NSWLR 406.......................................................................133 Avon Products Pty Ltd v Commissioner of Taxation [2006] HCA 29....................................323 Bank of New South Wales v The Commonwealth (1948) 76 CLR 1......................................321 Bell Bros Pty Ltd v Shire of Serpentine-Jarrahdale [1969] WAR 104; [1969] WAR 155 (SCWA); (1969) 121 CLR 137 (HCA)................................................64–65, 69, 75–76, 325 British American Tobacco Australia Ltd v State of Western Australia (2003) 217 CLR 30.............................................................................................................19, 316–21 Chippendale Printing Co Pty Ltd v The Commonwealth of Australia and the Commissioner of Taxation [1996] FCA 1259............................................................................................314 Commissioner of State Revenue (Victoria) v Royal Insurance Australia Ltd (1994) 182 CLR 51..............................................................................133, 136–39, 318–20, 323–24, 326 Commissioner of Stamps (SA) v Telegraph Investment Co Pty Ltd (1995) 184 CLR 453......................................................................................................................319, 326 Commonwealth of Australia v Burns [1971] VR 825.............................................................133 Comptroller General of Customs v Kawasaki Motors Pty Ltd (No 2) (1991) 32 FCR 243....................................................................................................................... 324–25 Cook v Lumbers (2008) 232 CLR 635....................................................................................325 David Securities Pty Ltd v Commonwealth Bank of Australia (1992) 175 CLR 353 (HC).......................................................................................................128–29, 278, 325–26 Director of Housing v Sudi [2011] VSCA 266........................................................................322 Eastman v Chief Executive, Dept of Justice and Community Safety [2011] ACTSC 33 .......322 Federal Airports Corporation v Aerolineas Argentinas (1997) 147 ALR 649 ........................325 Ha v New South Wales (1997)189 CLR 465..........................................................................317 Hollis v Atherton Shire Council [2003] QSC .........................................................................314 ICM Agriculture Pty Ltd v Commonwealth of Australia [2009] HCA 51 ............................321 James v The Commonwealth (1939) 62 CLR 339 .................................................................320 Jeffrey v Fitzroy Collingwood Rental Housing Association Ltd [1999] VSC 335....................314 Kruger v The Commonwealth (1997) 190 CLR 1..................................................................320 Lamesa Holding BV v Commissioner of Taxation [1999] FCA 612.......................................314 Mallinson v Scottish Australian Investment Co Ltd (1920) CLR 66......................................324 Mason v New South Wales (1959) 102 CLR 108 (HC) ..................................... 74, 76, 283, 325 Momcilovic v R [2011] HCA 34.............................................................................................322 Pape v Commissioner of Taxation of the Commonwealth of Australia (2009) 238 CLR 1..................................................................................................................................324

xii Table of Cases Park Oh Ho v Minister for Immigration and Ethnic Affairs (1988) 14 ALD 787, (1988) 20 FCR 104, (1989) 167 CLR 637..................................................................320, 324 R v War Pensions Entitlement Appeal Tribunal, ex parte Bott (1933) 50 CLR 228...............323 Roxborough v Rothmans of Pall Mall Australia Ltd (1999) 161 ALR 253, [1999] FCA 107, [1999] FCA 1535; (2001) 208 CLR 516 .........................19, 314, 317–20, 325–26 Sabet v Medical Practitioners Board of Victoria (2008) 20 VR 414 ......................................322 Sargood Brothers v The Commonwealth (1910) 11 CLR 258 (HC)........................... 71, 73, 78, 86, 88, 283, 325 State Bank of New South Wales Ltd v Commissioner of Taxation of Commonwealth of Australia [1995] FCA 1652 ...............................................................................................314 The Commonwealth v SCI Operations Pty Ltd (1998) 192 CLR 285 ...................................324 Werrin v The Commonwealth (1938) 59 CLR 150....................................................76–77, 320 Williams v Commonwealth of Australia [2012] HCA 23..............................................314, 319

Canada Air Canada v British Columbia (Attorney General) (1986) 4 BCLR 356 (BCCA); [1989] 1 SCR 1161, 59 DLR (4th) 161....................................... 79, 81, 86, 89, 106, 136–37, 139, 264, 273, 275, 278, 289–90, 296, 299, 308 Alberta v Elder Advocates of Alberta Society 2011 SCC 24, [2011] 2 SCR 261....................................................................................... 87, 256, 305, 309–10, 312 Amax Potash Ltd v Government of Saskatchewan [1977] 2 SCR 576............................ 289–90 Baumgartner v Baumgartner (1987) 164 CLR 137...............................................................228 Bell Mobility Inc v Anderson 2012 NWTCA 4.......................................................................305 BMP Global Distribution Inc v Bank of Nova Scotia 2009 SCC 15, [2009] 1 SCR 504, 304 DLR (4th) 292..........................................................................222, 236, 305 Canadian Pacific Airlines Ltd v British Columbia (Attorney General) [1989] 1 SCR 1133 ........................................................................................................................296 Chaulk v R [1990] 3 SCR 1303 .............................................................................................142 Chiasson v Canada (Attorney-General) (2008) 295 DLR (4th) 744 (FCTD) .....................296 Eadie v Township of Bradford [1967] SCR 573........................................................................76 Eurig Estate, Re [1998] 2 SCR 565 ..........................................................................................79 Garland v Consumers’ Gas Co 2004 SCC 25, [2004] 1 SCR 629, 237 DLR (4th) 385 ....................................................................19, 129, 222–23, 227, 236, 287, 304–08, 310 Garland v Enbridge Gas Distribution Inc 2006 CanLII 41291 (Ont SCJ)............................306 Kerr v Baranow 2011 SCC 10, [2011] 1 SCR 269..................................................................222 Kingstreet Investments Ltd v New Brunswick (Department of Finance) 2007 SCC 1, [2007] 1 SCR 3, 276 DLR (4th) 342..........................19–20, 23, 51, 73, 79, 87, 90, 136, 215, 222, 236, 241, 245, 256, 273–76, 287–90, 293–95, 298, 301, 303, 305, 307–10, 312, 315 Lange v Australian Broadcasting Corporation (1997) 189 CLR 520.......................315–16, 319 Marcotte v Longueuil (Ville) 2009 SCC 43, [2009] 3 SCR 65...............................................137 Merchant Law Group v Canada (Revenue Agency) 2009 FC 755, (2009) 311 DLR (4th) 124 (FCTD); 2010 FCA 184.............................................................................296, 309 Momi v Canada (Minister of Citizenship and Immigration) 2006 FC 738, 55 Imm LR (3d) 133, [2007] 2 FCR 291 (Fed Ct)................................................................................311

Table of Cases xiii

Muschinski v Dodds (1985) 160 CLR 583..............................................................................228 Pape v Commissioner of Taxation of the Commonwealth of Australia (2009) 238 CLR 1..................................................................................................................................315 Peter v Beblow [1993] 1 SCR 980, 101 DLR (4th) 621..................................................228, 306 Pettkus v Becker [1980] 2 SCR 834, 117 DLR (3d) 257.................................................228, 306 Port Alberni (City) v Catalyst Paper Corp 2010 BCSC 402...................................................312 R v Heywood [1994] 3 SCR 761.............................................................................................142 Regional Municipality of Peel v Canada [1992] 3 SCR 762..................................................118 Seversen (AJ) Inc v Qualicum Beach [1980] 3 WWR 375 (BC SC)......................................135 Sivia v British Columbia (Superintendent of Motor Vehicles) 2012 BCSC 1030...................310 Sorbara v Canada (Attorney General) (2008) 304 DLR (4th) 470 (Ont SCJ); 2009 ONCA 506, 309 DLR (4th) 162; 2009 CanLII 61389 (SCC)...................................136, 309 Sorochan v Sorochan [1986] 2 SCR 35, 29 DLR (4th) 1.......................................................306 Soulos v Korkontzilas [1997] 2 SCR 217, 146 DLR (4th) 214...............................................305 Storthoaks (Rural Municipality) v Mobil Oil Canada Ltd [1976] 2 SCR 147.......................129 TimberWest Forest Corp v Campbell River (City) 2009 BCSC 1862.....................................312

Cayman Islands A v Rothschild Trust Cayman Ltd [2006] WTLR 1129..........................................................150 Barclays Private Bank & Trust (Cayman) Ltd v Chamberlain [2007] WTLR 1697.....150, 155

European Court of Human Rights Bäck v Finland (2004) 40 EHRR 1184...................................................................................142 James v United Kingdom (1986) 8 EHRR 123................................................................ 141–42 National & Provincial Building Society v United Kingdom (1997) 25 EHRR 127 ......51, 141–43 Pressos Compania Naviera v Belgium (1996) 21 EHRR 301.................................................141

European Court of Justice/Court of Justice of the European Union Alphabetical Abbey National plc v HM Revenue & Customs (Case C–169/04) [2006] ECR I–4027............5 Accession by the Community Courts to the ECHR (Opinion 2/94) [1996] ECR I–1759.....147 Amministrazione delle Finanze dello Stato v Salumi (Joined Cases 66/79, 127/79 and 128/79) [1980] ECR 1237 ...................................................................................................38 Amministrazione delle Finanze dello Stato v SpA San Giorgio (Case 199/82) [1983] ECR 3595..........................................................3, 5, 14–16, 18, 24, 29, 35–36, 43, 56–57, 60, 71, 106, 112, 126, 188–90, 192, 196–99, 203–10, 289 Amminstrazione delle Finanze dello Stato v Simmenthal SpA (Case 106/77) [1978] ECR 629................................................................................................................................27 Bartsch v Bosch und Siemens Hausgeräte (BSH) Altersfürsorge GmbH (Case C–427/06) [2008] ECR I–7245............................................................................................................185

xiv Table of Cases Belgische Staat v Direct Parcel Distribution Belgium (Case C–264/08) [2010] ECR I–731............................................................................................................................71 BLP Group plc v Customs & Excise Commissioners (Case C–4/94) [1996] 1 WLR 174..........................................................................................................................118 BP Supergas Anonimos Etairia Geniki Emporiki-Viomichaniki kai Antiprossopeion v Greece (Case C–62/93) [1995] STC 805............................................................................188 Brasserie du Pêcheur SA v Bundesrepublik Deutschland and R v Secretary of State for Transport, ex parte Factortame Ltd (Joined Cases C–46/93 and C–48/93) [1996] ECR I–1029..........................................................................................................205, 289–90 Cadbury Schweppes plc v Inland Revenue Commissioners (Case C–196/04) [2006] ECR I–7995................................................................................................................170, 180 Comateb v Directeur Général des Douanes et Droits Indirects (Case C–192/95) [1997] ECR I–165..........................................................................................................................193 Commission v Greece (Case 68/88) [1989] ECR 2965 ..........................................................185 Commission v United Kingdom (Case C–466/98) [2002] ECR I–9427................................186 Cucchi v Avez SpA (Case 77/76) [1977] ECR 987 ................................................................289 Danfoss A/S and Sauer-Danfoss ApS v Skatteministeriet (Case C–94/10), 20 October 2011................................................................. 9, 111, 188, 191–93, 195–96, 199 Defrenne v Sabena (Case 149/77) [1978] ECR 1365...............................................................68 Edilizia Industriale Siderurgica Srl (Edis) v Ministero delle Finanze (Case C–231/96) [1998] ECR I–4951.................................................................................... 34, 38, 42, 55, 197 Elida Gibbs Ltd v Customs & Excise Commissioners (Case C–317/94) [1996] ECR I–5339, [1997] QB 499..............................................................................................112 Faccini Dori v Recreb Srl (Case C–91/92) [1994] ECR I–3325 ............................................169 Fantask A/S and others v Industriministeriet (Erhvervsministeriet) (Case C–188/95) [1997] ECR 1–6783....................................................................................6, 68, 191–92, 205 Francovich and Bonifaci v Italian Republic (Joined Cases C–6/90 and C–9/90) [1991] ECR I–5357...................................................................... 14–15, 17, 31, 43, 205, 289 Haahr Petroleum Ltd v Åbenrå Havn (Case C–90/94) [1997] ECR I–4085.........................197 Halifax plc v Commissioners of Customs & Excise (Case C–255/02) [2006] ECR I–1609........................................................................................................................164 HM Revenue & Customs v Weald Leasing Ltd (Case C–103/09) [2011] STC 596...............164 HSBC Holdings plc v HM Revenue & Customs (Case C–569/07) [2009] ECR I–9047 ...........................................................................................................................5 ICI Chemical Industries plc v Colmer (HM Inspector of Taxes) (Case C–264/96) [1998] ECR I–4695................................................................................................................5 JP Morgan Fleming Claverhouse Investment Trust plc v HM Revenue & Customs (Case C–363/05) [2007] ECR I–5517.......................................................5, 114–15, 117–18 Just I/S v Danish Ministry for Fiscal Affairs (Case 68/79) [1980] ECR 501......................5, 289 Kolpinghuis Nijmegen BV (Case 80/86) [1987] ECR 3969...................................................182 Kücükdeveci v Swedex GmbH & Co KG (Case C–555/07) [2010] ECR I–365.........14, 184–86 Kühne & Heitz NV v Produktschap voor Pluimvee en Eieren (Case C–453/00) [2004] ECR I–837................................................................................................................17 Kyrian v Celní úřad Tabor (Case C–233/08) [2010] ECR I–177.............................................55 Lady & Kid A/S v Skatteministeriet (Case C–398/09), 6 September 2011, [2012] STC 854, [2012] All ER (EC) 410.................................................. 5, 43, 46, 71, 191–92, 269 Levez v Jennings (Harlow Pools) Ltd Case (C–326/96) [1998] ECR I–7835..........................55

Table of Cases xv

Littlewoods Retail Ltd v HM Revenue & Customs (Case C–591/10) [2012] ECR I–000, [2012] STC 1714................. 6, 15, 35–37, 39, 42–44, 47, 55, 112, 189, 200–02, 208–10, 258 Mangold v Helm (Case C–144/04) [2005] ECR I–9981...........................................14, 184–86 Marks & Spencer plc v Customs & Excise Commissioners (Case C–62/00) [2002] ECR I–6325, [2003] QB 866.................................... 6, 15–16, 28, 38–39, 42–43, 57, 197–98 Marks & Spencer plc v Halsey (HM Inspector of Taxes) (Case C–446/03) [2006] STC 237................................................................................................................................16 Marleasing SA v La Comercial Internacional de Alimentación SA (Case C–106/89) [1991] ECR I–4135..................................................................................14, 61, 170, 178–79 Metallgesellschaft Ltd v Inland Revenue Commissioners and Hoechst AG v Inland Revenue Commissioners (Joined Cases C–397/98 and C–410/98) [2001] ECR I–1727, [2001] Ch 620....................... 5, 26, 40–43, 53, 59, 138, 188–90, 200–01, 203, 205, 239, 289 Ministero delle Finanze v IN.CO.GE.’90 Srl and Others (Joined Cases C–10/97 to C–22/97) [1998] ECR I–6307......................................................................................27, 209 Ministre du Budget, des Comptes publics et de la Fonction publique v Accor SA (Case C–310/09) 15 September 2011.......................................................................... 5, 43, 46, 269 Pfeiffer v Deutsches Rotes Kreuz, Kreisverband Waldshut eV (Joined Cases C–397/01 to C–403/01) [2004] ECR I–8835............................................ 170, 172, 176, 178–79, 181, 183 Preston v Wolverhampton Healthcare NHS Trust (Case C–78/98) [2000] ECR I–3201........55 Pupino (Case C–105/03) [2005] ECR I–5285.................................................170–71, 176, 179 Reemstma Cigarettenfabriken GmbH v Ministero delle Finanze (Case C–35/05) [2007] ECR I–2425............................................................................ 9, 29–30, 111, 193, 231 Rewe-Handelsgesellschaft Nord mbH v Hauptzollamt Kiel (Case 158/80) [1981] ECR 1805............................................................................................................................208 Rewe-Zentralfinanz eG and Rewe-Zentral AG v Landwirtschaftskammer für das Saarland (Case 33/76) [1976] ECR 1989............................................................34, 189, 197 Schul Douane Expediteur BV v Inspecteur der Invoerrechten en Accijnzen, Roosendaal (Case 15/81) [1982] ECR 1409..........................................................................................113 Scotch Whisky Association v Compagnie financière européenne de Financière Européenne de Prises de Participation (Cofepp) and Centrale d’Achats et de Services Alimentaires SARL (Casal) (Case C–136/96) [1998] ECR I–4571........................................................171 Test Claimants in Class IV of the ACT Group Litigation (Case C–374/04) [2006] ECR I–11673......................................................................................................................186 Test Claimants in the Franked Investment Income Group Litigation v Inland Revenue Commissioners (Case C–446/04) [2006] ECR I–11753, [2007] 1 CMLR 35.......... 5, 27, 31, 43–44, 118, 175, 177–78, 183, 186, 188–89, 201, 203–05, 210 Test Claimants in the Thin Cap Group Litigation v Inland Revenue Commissioners (Case C–524/04) [2007] ECR I–2107.................................................................................27 Unibet (London) Ltd v Justitiekanslern (Case C–432/05) [2007] ECR I–2271......................39 Weber’s Wine World Handels-GmbH v Abgabenberufungskommission Wien (Case C–147/01) [2003] ECR I–11365........................................................................................191 Wheels Common Investment Fund Trustees Ltd, National Association of Pension Funds Ltd, Ford Pension Fund Trustees Ltd, Ford Salaried Pension Fund Trustees Ltd, Ford Pension Scheme for Senior Staff Trustee Ltd v Commissioners for Her Majesty‘s Revenue and Customs (Case C–424/11) [2011] OJ C311/23...........................................114

xvi Table of Cases Chronological 33/76 Rewe-Zentralfinanz eG and Rewe-Zentral AG v Landwirtschaftskammer für das Saarland [1976] ECR 1989 ...........................................................................34, 189, 197 77/76 Cucchi v Avez SpA [1977] ECR 987.............................................................................289 106/77 Amminstrazione delle Finanze dello Stato v Simmenthal SpA [1978] ECR 629 ........27 149/77 Defrenne v Sabena [1978] ECR 1365 .........................................................................68 66/79, 127/79 and 128/79 Amministrazione Delle Finanze dello Stato v Salumi [1980] ECR 1237..............................................................................................................................38 68/79 Just I/S v Danish Ministry for Fiscal Affairs [1980] ECR 501 ................................5, 289 158/80 Rewe-Handelsgesellschaft Nord mbH v Hauptzollamt Kiel [1981] ECR 1805............................................................................................................................208 15/81 Schul Douane Expediteur BV v Inspecteur der Invoerrechten en Accijnzen, Roosendaal [1982] ECR 1409 ...........................................................................................113 199/82 Amministrazione delle Finanze dello Stato v SpA San Giorgio [1983] ECR 3595..........................................................3, 5, 14–16, 18, 24, 29, 35–36, 43, 56–57, 60, 71, 106, 112, 126, 188–90, 192, 196–99, 203–10, 289 80/86 Kolpinghuis Nijmegen BV [1987] ECR 3969...............................................................182 68/88 Commission v Greece [1989] ECR 2965......................................................................185 C–106/89 Marleasing SA v La Comercial Internacional de Alimentación SA [1991] ECR I–4135..............................................................................................14, 61, 170, 178–79 C–6/90 and C–9/90 Francovich and Bonifaci v Italian Republic [1991] ECR I–5357.................................................................................. 14–15, 17, 31, 43, 205, 289 C–91/92 Faccini Dori v Recreb Srl [1994] ECR I–3325.........................................................169 C–46/93 and C–48/93 Brasserie du Pêcheur SA v Bundesrepublik Deutschland and R v Secretary of State for Transport, ex parte Factortame Ltd [1996] ECR I–1029..........................................................................................................205, 289–90 C–62/93 BP Supergas Anonimos Etairia Geniki Emporiki-Viomichaniki kai Antiprossopeion v Greece [1995] STC 805.........................................................................188 Opinion 2/94 on Accession by the Community Courts to the ECHR [1996] ECR I–1759........................................................................................................................147 C–4/94 BLP Group plc v Customs & Excise Commissioners [1996] 1 WLR 174..................118 C–90/94 Haahr Petroleum Ltd v Åbenrå Havn [1997] ECR I–4085....................................197 C–317/94 Elida Gibbs Ltd v Customs & Excise Commissioners [1996] ECR I–5339, [1997] QB 499....................................................................................................................112 C–188/95 Fantask A/S and others v Industriministeriet (Erhvervsministeriet) [1997] ECR 1–6783................................................................................................6, 68, 191–92, 205 C–192/95 Comateb v Directeur Général des Douanes et Droits Indirects [1997] ECR I–165..........................................................................................................................193 C–136/96 Scotch Whisky Association v Compagnie financière européenne de Financière Européenne de Prises de Participation (Cofepp) and Centrale d’Achats et de Services Alimentaires SARL (Casal) [1998] ECR I–4571...............................................................171 C–231/96 Edilizia Industriale Siderurgica Srl (Edis) v Ministero delle Finanze [1998] ECR I–4951................................................................................................ 34, 38, 42, 55, 197 C–264/96 ICI Chemical Industries plc v Colmer (HM Inspector of Taxes) [1998] ECR I–4695............................................................................................................................5 C–326/96 Levez v Jennings (Harlow Pools) Ltd [1998] ECR I–7835......................................55

Table of Cases xvii

C–10/97 to C–22/97 Ministero delle Finanze v IN.CO.GE.’90 Srl and Others [1998] ECR I–6307..................................................................................................................27, 209 C–78/98 Preston v Wolverhampton Healthcare NHS Trust [2000] ECR I–3201....................55 C–397/98 and C–410/98 Metallgesellschaft Ltd v Inland Revenue Commissioners and Hoechst AG v Inland Revenue Commissioners [2001] ECR I–1727, [2001] Ch 620............................................................................................. 5, 26, 40–43, 53, 59, 138, 188–190, 200–01, 203, 205, 239, 289 C–466/98 Commission v United Kingdom [2002] ECR I–9427............................................186 C–62/00 Marks & Spencer plc v Customs & Excise Commissioners [2002] ECR I–6325, [2003] QB 866.......................................................... 6, 15–16, 28, 38–39, 42–43, 57, 197–98 C–453/00 Kühne & Heitz NV v Produktschap voor Pluimvee en Eieren [2004] ECR I–837............................................................................................................................17 C–147/01 Weber’s Wine World Handels-GmbH v Abgabenberufungskommission Wien [2003] ECR I–11365.................................................................................................191 C–397/01 to C–403/01 Pfeiffer v Deutsches Rotes Kreuz, Kreisverband Waldshut eV [2004] ECR I–8835.............................................................. 170, 172, 176, 178–79, 181, 183 C–255/02 Halifax plc v Commissioners of Customs & Excise [2006] ECR I–1609...............164 C–105/03 Pupino [2005] ECR I–5285.............................................................170–71, 176, 179 C–446/03 Marks & Spencer plc v Halsey (HM Inspector of Taxes) [2006] STC 237..............16 C–144/04 Mangold v Helm [2005] ECR I–9981.......................................................14, 184–86 C–169/04 Abbey National plc v HM Revenue & Customs [2006] ECR I–4027........................5 C–196/04 Cadbury Schweppes plc v Inland Revenue Commissioners [2006] ECR I–7995................................................................................................................170, 180 C–524/04 Test Claimants in the Thin Cap Group Litigation v Inland Revenue Commissioners [2007] ECR I–2107.....................................................................................27 C–374/04 Test Claimants in Class IV of the ACT Group Litigation [2006] ECR I–11673......................................................................................................................186 C–446/04 Test Claimants in the Franked Investment Income Group Litigation v Inland Revenue Commissioners [2006] ECR I–11753, [2007] 1 CMLR 35...................................................................... 5, 27, 31, 43–44, 118, 175, 177–78, 183, 186, 188–89, 201, 203–05, 210 C–35/05 Reemstma Cigarettenfabriken GmbH v Ministero delle Finanze [2007] ECR I–2425........................................................................................ 9, 29–30, 111, 193, 231 C–363/05 JP Morgan Fleming Claverhouse Investment Trust plc v HM Revenue & Customs [2007] ECR I–5517.....................................................................5, 114–15, 117–18 C–432/05 Unibet (London) Ltd v Justitiekanslern [2007] ECR I–2271..................................39 C–427/06 Bartsch v Bosch und Siemens Hausgeräte (BSH) Altersfürsorge GmbH [2008] ECR I–7245........................................................................................................................185 C–555/07 Kücükdeveci v Swedex GmbH & Co KG [2010] ECR I–365.....................14, 184–86 C–569/07 HSBC Holdings plc v HM Revenue & Customs [2009] ECR I–9047.......................5 C–233/08 Kyrian v Celní úřad Tabor [2010] ECR I–177.........................................................55 C–264/08 Belgische Staat v Direct Parcel Distribution Belgium [2010] ECR I–731...............71 C–103/09 HM Revenue & Customs v Weald Leasing Ltd [2011] STC 596...........................164 C–310/09 Ministre du Budget, des Comptes publics et de la Fonction publique v Accor SA, 15 September 2011............................................................................ 5, 43, 46, 269 C–398/09 Lady & Kid A/S v Skatteministeriet, 6 September 2011, [2012] STC 854, [2012] All ER (EC) 410 ............................................... 5, 43, 46, 71, 191–192, 269

xviii Table of Cases C–94/10 Danfoss A/S and Sauer-Danfoss ApS v Skatteministeriet, 20 October 2011................................................................. 9, 111, 188, 191–93, 195–96, 199 C–591/10 Littlewoods Retail Ltd v HM Revenue & Customs [2012] ECR I–000, [2012] STC 1714................. 6, 15, 35–37, 39, 42–44, 47, 55, 112, 189, 200–02, 208–10, 258 C–424/11 Wheels Common Investment Fund Trustees Ltd, National Association of Pension Funds Ltd, Ford Pension Fund Trustees Ltd, Ford Salaried Pension Fund Trustees Ltd, Ford Pension Scheme for Senior Staff Trustee Ltd v Commissioners for Her Majesty’s Revenue and Customs [2011] OJ C311/23............................................114

Germany BayObLG (9.12.1998), NJW 1999, 1194.......................................................................244, 262 BFH (19.10.1976), BFHE 120, 424 .......................................................................................246 BFH (16.9.2010), case refs V R 46/09, 48/09, 49/09, 51/09, 52/09, 57/09......................17, 252 BGH (22.6.1956), BGHZ 21, 102..........................................................................................226 BGH (19.4.1961), BGHZ 35, 103..........................................................................................232 BGH (21.12.1961), BGHZ 36, 232........................................................................................220 BGH (29.4.1968), BGHZ 50, 90............................................................................................232 BGH (9.6.1969), Wertpapier-Mitteilungen 1969, 1083 ........................................................232 BGH (11.6.1970), NJW 1970, 1637 ......................................................................................260 BGH (7.1.1971), BGHZ 55, 128....................................................................................220, 266 BGH (26.10.1978), BGHZ 72, 252 .......................................................................................266 BGH (17.2.1982), BGHZ 83, 278..........................................................................................231 BGH (8.7.1982), BGHZ 84, 361............................................................................................228 BGH (6.12.1990), NJW-RR 1991, 574 ..................................................................................235 BGH (2.10.1991), BGHZ 115, 261........................................................................................228 BGH (14.7.1993), NJW-RR 1993, 1457 ................................................................................232 BGH (13.7.1994), BGHZ 127, 48..........................................................................................228 BGH (14.12.1994), NJW 1995, 662 ......................................................................................235 BGH (6.3.1998), BGHZ 138, 160 = NJW 1998, 2354 ..........................................................258 BGH (12.5.1998), NJW 1998, 2529 ......................................................................................258 BGH (18.5.1999), NJW 1999, 2887 ............................................................................. 235–236 BGH (30.6.1999), BGHZ 142, 137 .......................................................................................228 BGH (5.2.2003), NJW 2003, 1449.........................................................................................236 BGH (14.7.2003), NJW-RR 2004, 556 ..................................................................................236 BGH (3.2.2004), BGHZ 158, 1, 9 = NJW 2004, 1315 ................................................... 261–62 BGH (30.3.2004), case ref XI ZR 145/03....................................................................... 261–62 BGH (21.3.2005), ZIP 2005, 753...........................................................................................230 BGH (23.1.2007), NJW 2007, 1584.......................................................................................236 BGH (24.4.2007), BGHZ 172, 147 = NJW 2007, 2401.........................................................258 BGH (19.3.2008), NJW-RR 2008, 1237.................................................................................236 BGH (24.5.2012), Wertpapier-Mitteilungen 2012, 1208................................................ 261–62 BSG (25.11.1977), case ref 2 RU 127/75...............................................................................260 BVerfG (29.5.2012 and 6.6.2012), case refs 1 BvR 1234/11, 1390/11, 1395/11, 1403/11 .......................................................................................................................17, 252 BVerwG (7.6.1962), BVerwGE 14, 222 .................................................................................261

Table of Cases xix

BVerwG (17.9.1970), BVerwGE 36, 108................................................................................266 BVerwG (18.3.1973), NJW 1973, 1854 .................................................................................261 BVerwG (18.5.1973), NJW 1973, 2122 .........................................................................258, 261 BVerwG (24.11.1977), Deutsches Verwaltungsblatt 1978, 608 .............................................258 BVerwG (1.2.1980), NJW 1980, 2538....................................................................................244 BVerwG (22.4.1982), case ref 3 C 71.81................................................................................258 BVerwG (7.2.1985), BVerwGE 71, 48 = NJW 1985, 2208 ....................................................260 BVerwG (12.3.1985), BVerwGE 71, 85 = NJW 1985, 2436...............................245–46, 266–67 BVerwG (13.12.1991), Neue Zeitschrift für Verwaltungsrecht 1992, 131..............................259 BVerwG (4.5.1994), NJW 1994, 3116....................................................................................260 BVerwG (27.10.1998), BVerwGE 107, 304 = NJW 1999, 1201...................................... 261–62 BVerwG (18.1.2001), BVerwGE 112, 351..............................................................................266 BVerwG (28.2.2002), BVerwGE 116, 74= Neue Zeitschrift für Verwaltungsrecht 2002, 854............................................................................................................................252 BVerwG (30.4.2003), Neue Zeitschrift für Verwaltungsrecht 2003, 1385....................... 260–62 BVerwG (7.9.2004), case ref 3 B 35/04 .......................................................................... 261–62 BVerwG (15.6.2006), NJW 2006, 3225 .................................................................................245 LG Göttingen (13.11.2003), case ref 6 O 16/03 and 25/03 ..................................................262 LG Hannover (28.9.2004), case ref 18 O 379/03 .................................................................262 LG Potsdam (22.2.2008), Neue Zeitschrift für Verwaltungsrecht: RechtsprechungsReport 2008, 513 ................................................................................................................262 OLG Hamm (19.10.2000), NJW-RR 2001, 1140, 1440................................................244, 262 OLG Karlsruhe (30.12.1987), NJW 1988, 1920....................................................................254 OLG Karlsruhe (6.3.2003), case ref 11 U 21/01....................................................................262 OLG Karlsruhe (30.9.2004), Wertpapier-Mitteilungen 2005, 645........................................259 OLG Köln (22.12.2000), case ref 2 Wx 32/00, Rechtspfleger 2001, 203 ...............................262 OVG Berlin–Brandenburg (19.5.2011), case ref OVG 10 B 2.08.........................................259 OVG Hamburg (9.7.1930), Deutsche Juristen-Zeitung 1931, 174........................................261 OVG Koblenz (28.11.1991), Neue Zeitschrift für Verwaltungsrecht 1992, 796.....................253 OVG Lüneburg (23.1.2004), case ref 11 LB 257/03..............................................................261 OVG Münster (20.4.1966), Die Öffentliche Verwaltung 1967, 271 .....................................244 OVG Münster (27.2.1992), NJW 1992, 2245 ...............................................................244, 254 OVG Weimar (22.10.2002), Neue Zeitschrift für Verwaltungsrecht: RechtsprechungsReport 2003, 830.................................................................................................................254 OVG Weimar (18.11.2009), case ref 1 KO 693/07................................................................254 RG (20.1.1890), RGZ 25, 302 ....................................................................................... 242–243 RG (17.2.1903), RGZ 54, 24 ..................................................................................................242 RG (19.12.1905), RGZ 62, 238 ..............................................................................................242 RG (8.10.1909), RGZ 71, 432 ................................................................................................232 RG (29.10.1909), RGZ 72, 152 ......................................................................................242, 260 RG (14.12.1909), RGZ 83, 159 ..............................................................................................242 RG (12.5.1911), RGZ 76, 270 ................................................................................................242 RG (21.2.1912), RGZ 78, 303 ................................................................................................230 RG (11.2.1913), RGZ 81, 303 ................................................................................................230 RG (3.10.1913), RGZ 83, 161 ................................................................................................242 RG (12.6.1917), RGZ 90, 314 ................................................................................................242 RG (12.3.1918), RGZ 92, 310 ................................................................................................243

xx Table of Cases RG (30.4.1920), RGZ 99, 41 ........................................................................................... 242–43 RG (11.5.1920), RGZ 99, 161 ................................................................................................266 RG (12.6.1923), RGZ 107, 189 ....................................................................................... 243–44 RG (25.6.1925), RGZ 111, 151 ......................................................................................226, 228 RG (27.4.1936), RGZ 151, 123 ..............................................................................................258 RG (13.11.1940), RGZ 165, 193.............................................................................................230 RG (30.10.1942), RGZ 170, 65 ..............................................................................................266 VG Berlin (21.10.2003), case ref 8 A 218.00, Zeitschrift für das Fürsorgewesen 2005, 162............................................................................................................................254 VG Berlin (11.2.2010), case ref 16 K 117.09 ........................................................................258 VG Gera (11.3.2004), case ref 4 K 277.02.GE, ThürVGRspr 2005, 196 ..............................254 VG Lüneburg (24.6.2003), case ref 4 A 78/02.......................................................................254 VGH Kassel (17.7.1990), NJW 1991, 510..............................................................................254 VGH Mannheim (16.3.1977), NJW 1978, 2050....................................................................258 VGH Mannheim (18.10.1990), Neue Zeitschrift für Verwaltungsrecht 1991, 583 ...............254

Hong Kong Greatworth v Sun Fook Kong Construction Ltd [2006] HKCFI 356......................................121 Yew Sang Hong Ltd v Hong Kong Housing Authority [2008] HKCA 109.............................121

Ireland A v Governor of Arbour Hill Prison [2006] IESC 45, [2006] 4 IR 88 (SC)...........297–300, 302 Article 26 and the Health (Amendment) (No 2) Bill 2004, Re [2005] IESC 7, [2005] 1 IR 105 (SC).......................................52, 138, 146, 278–82, 286, 291, 294, 296, 301 Bank of Ireland Trust Services Ltd v Revenue Commissioners [2002] 4 IR 178 (HC).............................................................................................................................49, 280 Blehein v Minister for Health and Children [2008] IESC 40; [2010] IEHC 329.............................................................................................................. 274, 293, 296, 298 Boliden Tara Mines Ltd v Cosgrove [2007] IEHC 60.............................................................149 Byrne v Ireland [1972] IR 241 (SC) ......................................................................285, 289, 294 Casey v Irish Sailors and Soldiers’ Land Trust [1937] 1 IR 208...............................................63 CC v Ireland [2006] IESC 33, [2006] 4 IR 1 (SC).................................................................300 Cox v Ireland [1992] 2 IR 503 (Irish High Ct)......................................................................142 Crotty v An Taoiseach [1987] IR 713 (SC).............................................................................285 Damache v Director of Public Prosecutions [2012] IESC 11..................................................300 De Búrca v Attorney General [1976] IR 38 (SC)...................................................292, 297, 299 Director of Public Prosecutions v Cunningham [2012] IECCA 64................274, 296–300, 302 Dolan v Neligan [1967] IR 247 (SC)................................................. 52, 276–81, 283, 286, 301 Dublin Corporation v Building and Allied Trade Union [1996] 2 ILRM 547 (SC)......................................................................................................................278, 281 Duff v Minister for Agriculture (No 2) [1997] 2 IR 22 (SC)..................................................278 Harris v Quigley [2005] IEHC 81, [2005] IESC 79, [2006] 1 IR 165.........18, 278–80, 283–84 Heaney v Ireland [1994] 3 IR 593 (SC).................................................................................142

Table of Cases xxi

Irish Pensions Trust Ltd v Central Remedial Clinic [2005] IEHC 87, [2006] 2 IR 126 ........149 Keating v Crowley [2010] IESC 29.........................................................................................292 Kennedy v Law Society of Ireland [2002] 2 IR 458 (SC)........................................................289 McDonnell v Ireland [1998] 1 IR 134 (SC)....................................................................293, 299 McMahon v Attorney General [1972] IR 69 (SC) .........................................................292, 297 Meade v Cork County Council (SC) ......................................................................................277 Meskell v Córas Iompair Éireann [1973] IR 121 (SC) ............................................289, 291–92 Muckley v Ireland [1985] IR 472 (SC)...................................................................281, 297–299 Murphy v Attorney General [1982] 1 IR 241 (SC)....................... 18–19, 68–69, 132, 138, 146, 161, 273–77, 279–86, 288, 290–302 O’Rourke v Revenue Commissioners [1996] 2 IR 1 (HC)..................... 18, 49, 69, 136, 275–81, 283–84, 297, 299, 301 Redmond v Minister for the Environment (No 2) [2004] IEHC 24, [2006] 3 IR 1....................................................................................................................289, 291–92 Rogers v Louth County Council [1981] IR 265 (SC).......................... 52, 276–80, 283, 286, 301 Ryan v Attorney General [1965] IR 294 (SC)................................................................285, 298 State (Burke) v Lennon [1940] IR 136 (SC)..........................................................................285 The State (Byrne) v Frawley [1978] IR 326 (SC)...........................................................298, 300

Jersey Friedman and Asiatrust Ltd [2006] JRC 187.................................................................. 149–50 Green GLG Trust, Re 2002 JLR 571........................................................................................149 Howe Family Number 1 Trust, Re [2007] JRC 248, 2007 JLR 660........................................150 Seaton Trustees Ltd, Re [2009] JRC 050 ................................................................................150 V Settlement, Re [2011] JRC 046...........................................................................................150 Vistra Trust Co (Jersey) Ltd, Re [2008] JRC 111....................................................................150

New Zealand Julian v Mayor of Auckland [1927] NZLR 453........................................................................77 National Bank of New Zealand Ltd v Waitaki International Processing (NI) Ltd [1999] 2 NZLR 211............................................................................................................269 Waikato Regional Airport Ltd v Attorney General [2003] UKPC 50, [2004] 3 NZLR 1..........................................................................................................84–86, 88, 146

South Africa Willis Faber Enthoven (Pty) Ltd v Receiver of Revenue (1992) (4) SA 202 (SC)..................278

United Kingdom Abacus Trust Co (Isle of Man) Ltd v Barr [2003] EWHC 114 (Ch)..............................152, 157

xxii Table of Cases Abacus Trust Co (Isle of Man) Ltd v National Society for the Prevention of Cruelty to Children [2001] STC 1344 ................................................................................152, 157, 161 Aberdeen Estates Ltd, VTD 13622, 28 September 1995 (Edinburgh VAT & Duties Tribunal)............................................................................................................................116 Abrahams’ Will Trust, Re; Caplan v Abrahams [1969] 1 Ch 463..........................................151 Allnutt v Wilding [2007] EWCA Civ 412, [2007] WTLR 941...............................................159 AMP (UK) plc v Barker [2001] Pens LR 77.............................................................150–51, 155 Armstrong DLW GmbH v Winnington Networks Ltd [2012] EWHC 10 (Ch), [2012] Bus LR 1199...........................................................................................................122 Ashcroft v Barnsdale [2010] EWHC 1948 (Ch), [2010] STC 2544......................................153 Auckland Harbour Board v R [1924] AC 318 (PC)..................30, 133, 240, 247, 254, 295, 298 Bank of Credit and Commerce International SA (in liq) v Ali (No 1) [2001] UKHL 8, [2002] 1 AC 251...................................................................................................66 Banque Financière de la Cité v Parc (Battersea) Ltd [1999] 1 AC 221 (HL).................................................................................................. 9, 116, 123–24, 189, 207 Barclays Bank Ltd v WJ Simms, Son and Cooke (Southern) Ltd [1980] QB 677 (QB)................................................................................ 53, 101, 123, 222, 252, 305 Barclays Bank plc v O’Brien [1994] 1 AC 180 (HL)..............................................................101 Barclays Mercantile Business Finance Ltd v Mawson [2004] UKHL 51 ...............................164 Barons Finance Ltd v Kensington Mortgage Co Ltd [2011] EWCA Civ 1592...............129, 131 Barros Mattos Junior v MacDaniels Ltd [2004] EWHC 1188 (Ch), [2005] 1 WLR 247..................................................................................................................131, 134 Barton v Armstrong [1976] AC 104 (PC)........................................................................73, 101 Beaulane Properties Ltd v Palmer [2005] EWHC 817 (Ch), [2006] Ch 79..........................142 Benedetti v Sawiris [2010] EWCA Civ 1427 .........................................................................304 Birmingham Hippodrome Theatre Trust Ltd v HM Revenue & Customs [2011] UKFTT 117 (TC)...............................................................................................................118 Black-Clawson International Ltd v Papierwerke Waldhof–Aschaffenburg AG [1975] AC 591 (HL)...........................................................................................................126 Blackpool and Fleetwood Tramroad Co v Bispham with Norbreck Urban District Council [1910] 1 KB 592......................................................................................................71 Bloomsbury International Ltd v Sea Fish Industry Authority (Bloomsbury) [2009] EWHC 1721 (QB), [2010] 1 CMLR 12......................... 13, 127–128, 130, 132, 134–35, 144 Boake Allen Ltd v HM Revenue & Customs [2006] EWCA Civ 25, [2006] STC 626.......12, 53 Boddington v British Transport Police [1999] 2 AC 143 (HL).................................................25 Breadner v Granville-Grossman [2001] Ch 523...................................... 150–51, 153, 155, 162 British Eventing Ltd v HM Revenue & Customs [2010] UKFTT 382 (TC)..........................119 British Steel plc v Customs and Excise Commissioners [1997] 2 All ER 366 (CA).......................................................................................................... 4, 32–34, 58, 61, 82 Brocklebank (T and J) Ltd v The King [1924] 1 KB 647; [1925] 1 KB 52.......................74, 277 Brook’s Wharf and Bull Wharf Ltd v Goodman Bros [1937] 1 KB 534 (CA)........................118 Burmah Oil Co (Burma Trading) Ltd v Lord Advocate [1965] AC 75 (HL).................. 144–45 Burrell v Burrell [2005] EWHC 245 (Ch), [2005] STC 569.............................150–52, 155–56 Cannon Brewery Company Ltd v Central Control Board (Liquor Traffic) [1918] 2 Ch 101 (CA)....................................................................................................................145 Cantiere San Rocco SA v Clyde Shipbuilding and Engineering Co [1924] AC 226 (HL)....................................................................................................................................215

Table of Cases xxiii

Carter Commercial Developments v Bedford Borough Council [2001] EWHC 669 (QB) (Admin) (2001) 34 EG 99....................................................................................................33 Chalke (FJ) Ltd v HM Revenue & Customs [2009] EWHC 952(Ch), [2009] STC 2027, [2009] 3 CMLR 14........................................................................... 55, 59, 69, 191, 200, 257 Chalke (FJ) Ltd v HM Revenue & Customs [2010] EWCA Civ 313, [2010] STC 1640...................................................................................... 53–55, 59, 68–69, 236, 257 Charles Terence Estates Ltd v Cornwall Council [2011] EWHC 2542 (QB), [2012] 1 P & CR 2................................................................................................129–33, 247 Chase Manhattan Bank NA v Israel-British Bank (London) Ltd [1981] Ch 102 (CA)..........61 Chief Constable of Greater Manchester v Wigan Athletic AFC Ltd [2008] EWCA Civ 1449, [2009] 1 WLR 1580...........................................................................................116 Clark v University of Lincolnshire and Humberside [2000] 1 WLR 1988 (CA)......................12 Clark v University of Lincolnshire and Humberside [2000] 1 WLR 1988 (HL)..........32–34, 46 Commerzbank AG v Price-Jones [2003] EWCA Civ 1663 ....................................................129 Costello v MacDonald [2011] EWCA Civ 930.......................................................................121 Cressman v Coys of Kensington (Sales) Ltd [2004] EWCA Civ 47, [2004] 1 WLR 2775................................................................................................................116, 129 CTN Cash & Carry Ltd v Gallaher Ltd [1994] 4 All ER 714 (CA)...................................3, 231 de Freitas v Permanent Secretary of Ministry of Agriculture, Fisheries, Lands and Housing [1999] 1 AC 69 (PC)................................................................................... 141–142 De Keyser’s Royal Hotel Ltd, Re [1919] 2 Ch 197 (CA).........................................................144 Derrick v Williams [1939] 2 All ER 559 (CA).............................................................63, 65–66 Deutsche Morgan Grenfell [2003] EWHC 1779, [2003] STC 1017........................................67 Deutsche Morgan Grenfell Group plc v Inland Revenue Commissioners [2005] EWCA Civ 78, [2006] Ch 243........................................................ 31, 58–59, 61, 63–64, 136 Deutsche Morgan Grenfell Group plc v Inland Revenue Commissioners [2006] UKHL 49, [2007] 1 AC 558................. 4, 7, 10–12, 15, 17, 19, 23, 26–27, 31–32, 34, 37–38, 40–47, 52–53, 58–60, 63, 71–72, 83–85, 96, 115, 123, 138, 189, 206–08, 213, 215–16, 236, 239, 241, 255–56, 264–65, 269, 275, 293, 303 Dew v Parsons (1819) 2 B & Ald 562, 106 ER 471....................................................73–74, 276 Dextra Bank & Trust Co Ltd v Bank of Jamaica [2001] UKPC 50, [2002] 1 All ER (Comm) 193 (PC)............................................................................. 123, 129, 132, 135, 269 Dimes v The Proprietors of the Grand Junction Canal (1852) 3 HLC 759, 10 ER 301 ..........26 Dimskal Shipping Co SA v International Transport Workers’ Federation [1992] 2 AC 152 (HL)......................................................................................................................73 Donaldson v Smith [2006] EWHC B9 (Ch), [2007] WTLR 421..........................................162 Eagerpath Ltd v Edwards [2001] STC 26.................................................................................60 Earl Cadogan v Sportelli [2008] UKHL 71, [2010] 1 AC 226...............................................142 Ensign Tankers v Stokes [1992] STC 617........................................................................160, 163 Equal Opportunities Commission v Secretary of State for Employment [1995] 1 AC 1 (HL)..........................................................................................................................27 F (a minor) (publication of information), Re [1977] Fam 58 (CA).....................................126 Factortame Ltd v Secretary of State for Transport (No 2) [1991] 1 AC 603 (HL).............27, 59 Farmer v Sloan [2004] EWHC 606 (Ch), [2005] WTLR 521........................................ 152–53 Fernley v Branson (1851) 20 LJQB 178 ..................................................................................90 Fibrosa Spolka Akcyjna v Fairbairn Lawson Combe Barbour Ltd [1943] AC 32 (HL).........106 Filby v Mortgage Express (No 2) Ltd [2004] EWCA Civ 759........................................118, 121

xxiv Table of Cases Fleming (trading as Bodycraft) and Condé Nast Publications Ltd v HM Revenue and Customs [2008] UKHL 2, [2008] 1 WLR 195, [2008] STC 324 .......................6, 16, 26–27, 56, 115, 176, 183 Foskett v McKeown [2000] UKHL 29, [2001] 1 AC 102.......................................................307 Furniss v Dawson [1984] AC 474...........................................................................................164 Futter v Futter [2010] EWHC 449 (Ch), [2010] STC 982....................................150, 153, 155 Gallaher Ltd v Gallaher Pensions Ltd [2005] EWHC 42 (Ch), [2005] Pens LR 103.........................................................................................................................150, 155 Ghaidan v Godin-Mendoza [2004] UKHL 30, [2004] 2 AC 557...........................171–74, 176, 178–79, 181–82 Gibb v Maidstone and Tunbridge Wells NHS Trust [2010] EWCA Civ 678, [2010] IRLR 786.....................................................................................................................3, 7, 118 Gibbon v Mitchell [1990] 1 WLR 1304..................................................................................159 Goss v Chilcott [1996] AC 788 (PC)......................................................................................265 Grantham Cricket Club v Customs & Excise Commissioners [1998] BVC 2272.......................9 Great Western Railway Co v Sutton (1869) LR 4 HL 226........................................................90 Greater Manchester Police v Wigan Athletic AFC Ltd [2008] EWCA Civ 1449, [2009] 1 WLR 1580............................................................................................................304 Green v Associated Newspapers Ltd [2005] QB 972 (CA).....................................................126 Green v Cobham [2002] STC 820................................................................................... 153–54 Greenway v Hurd (1792) 4 Term Rep 553, 100 ER 1171........................................................74 Gresh v RBC Trust Co (Guernsey) Ltd 2009–2010 GLR 216.................................................153 Griffiths (deceased), Re [2008] EWHC 118 (Ch), [2009] Ch 162..........................152–53, 163 Guinness Mahon & Co Ltd v Kensington and Chelsea Royal London Borough Council [1999] QB 215 (CA).............................................................................................247 Hastings–Bass (deceased), Re [1975] Ch 25 (CA).....................................10, 149–162, 164–65 Haugesund Kommune v Depfa ACS Bank [2009] EWHC 2227 (Comm); [2010] EWCA Civ 579, [2012] Bus LR 1...................................................................129, 265 Hazell v Hammersmith & Fulham London Borough Council [1992] 2 AC 1 (HL) .......................................................................................................................32, 53, 247 Hemming (trading as Simply Pleasure Ltd) v Westminster City Council [2012] EWHC 1260 (Admin) .....................................................................................33–34, 50, 247 Henderson v Folkestone Waterworks Co (1885) 1 TLR 329 (CA).....................................63, 88 Henderson v Merrett Syndicates Ltd [1995] 2 AC 145 (HL)...........................32–34, 46, 56, 84 Hills v Street (1825) 5 Bing 36, 130 ER 973.................................................................73, 77, 90 HM Revenue & Customs v IDT Card Services Ireland Ltd [2006] STC 1252 (CA)..............................................................................................................170–74, 179–182 Hooper v Mayor and Corporation of Exeter (1887) 56 LJQB 457.............................77, 276–77 Inco Europe Ltd v First Choice Distribution (a firm) [2000] 1 WLR 586 (HL)................................................................................................................173–75, 177–78 Inland Revenue Commissioners v Burmah Oil Co Ltd [1982] STC 30..................................164 Inland Revenue Commissioners v Deutsche Morgan Grenfell Group plc [2005] EWCA Civ 78, [2006] Ch 242..............................................................................................54 Inland Revenue Commissioners v Duke of Westminster [1936] AC 1............................160, 163 Investment Trust Companies (in liquidation) v HM Revenue & Customs [2012] EWHC 458 (Ch), [2012] STC 1150 ...........................6–9, 15, 29, 38–39, 42, 46, 51, 56, 59, 61, 111–17, 119, 121–26, 189, 193–200, 207–08, 304

Table of Cases xxv

Irving v Wilson (1791) 4 Term Rep 485, 100 ER 1132............................................................74 James, Ex parte (1874) LR 9 Ch App 609................................................................................71 Jenks v Dickinson [1997] STC 853.........................................................................................174 Jones v Kernott [2011] UKSC 53, [2011] 3 WLR 1121..........................................................306 Jones v Powys Local Health Board [2008] EWHC 2562 (Admin).......................33–34, 50, 247 Jons v Perchard (1796) 2 Esp 507, 170 ER 432........................................................................73 Kerr v British Leyland (Staff) Trustees Ltd [2001] WTLR 1071....................................152, 163 Kiriri Cotton Company Limited v Dewani [1960] AC 192 (PC) ..........................................278 Kleinwort Benson Ltd v Birmingham City Council [1997] QB 380 (CA) ....................121, 207 Kleinwort Benson Ltd v Lincoln City Council [1999] 2 AC 349 (HL)........ 3–4, 7, 9–10, 31, 45, 52–53, 62–67, 72, 85, 124, 242, 247, 278, 295 Legal Services Commission v Henthorn [2011] EWHC 258 (QB), [2011] EWCA Civ 1415, [2012] 1 WLR 1173...........................................................................................126 Legal Services Commission v Loomba [2012] EWHC 29 (QB), [2012] 1 WLR 2461...........126 Lewis v Maningham (1550) 1 Plowden 60, 75 ER 96..............................................................74 Lipkin Gorman (a firm) v Karpnale Ltd [1988] UKHL 12, [1991] 2 AC 548 (HL).........................................................127–29, 132, 135, 139, 190–91, 264, 269, 281, 307 Litster v Forth Dry Dock and Engineering Company Ltd [1989] 1 All ER 1134 (HL)....................................................................................................................................174 Littlewoods Retail Ltd v HM Revenue & Customs [2010] EWHC 1071 (Ch), [2010] STC 2072......................................13, 35–36, 38–39, 42–43, 46, 55–56, 61, 69–70, 118, 128, 130, 132, 134–36, 139–40, 190–92, 200, 202, 207–10, 257–58, 264–65 Littlewoods Retail Ltd v HM Revenue & Customs [2010] EWHC 2771 (Ch), [2011] STC 171..............................................................................................................................112 Lloyds Bank plc v Independent Insurance Co [2000] QB 110 (CA)......................................123 London & North Western Railway Co v Evans [1893] 1 Ch 16 (CA)....................................144 Longdill v Jones (1816) 1 Stark 345, 171 ER 492 ....................................................................73 Lovell v Simpson (1800) 3 Esp 153, 170 ER 570......................................................................73 Malik v Bank of Credit and Commerce International SA (in liq) [1997] AC 20 (HL)...........66 Malkin v The King [1906] 2 KB 886........................................................................................74 Mallusk Cold Storage Ltd v Dept of Finance and Personnel [2003] NIQB 58, [2003] RA 370................................................................................................................................136 Martin v Tomkinson [1893] 2 QB 121.....................................................................................74 Maskell v Horner [1915] 3 KB 106...............................................................................71, 73, 77 Meadows v Grand Junction Waterworks Co (1905) 21 TLR 538.............................................71 Menelaou v Bank of Cyprus plc [2012] EWHC 1991 (Ch) ..............................................9, 122 Mettoy Pension Trustees Ltd v Evans [1990] 1 WLR 1587 ......................................150–52, 155 Miliangos v George Frank Textiles Ltd [1976] AC 443...................................................... 44–45 Monro v HM Revenue & Customs [2008] EWCA Civ 306, [2009] Ch 69.............................................................................................7, 61, 64, 68, 125, 141, 143 Morgan v Palmer (1824) 2 B & C 729, 107 ER 554.............................................74–75, 276–77 Morgan Guaranty Trust Corporation of New York v Lothian Regional Council 1995 SLT 299 (Inner House)........................................................................................................65 Moses v Macferlan (1760) 2 Burr 1005, 97 ER 676...............................................................318 Moule v Garrett (1872) LR 7 Ex 101......................................................................................118 NEC Semi-Conductors Ltd v Inland Revenue Commissioners [2006] EWCA Civ 25, [2006] STC 606; [2007] UKHL 25, [2007] 1 WLR 1386..............................................12, 82

xxvi Table of Cases Niru Battery Manufacturing Co v Milestone Trading Ltd [2002] EWHC 1425 (Comm), [2002] 2 All ER (Comm) 705; [2003] EWCA Civ 1446, [2004] QB 985 (CA)...............269 Norwich City Council v Stringer (2001) 33 HLR 15 (CA)......................................................86 Nurdin & Peacock plc v DB Ramsden & Co Ltd [1999] 1 WLR 1249 (Ch)..................108, 231 O’Reilly v Mackman [1983] 2 AC 237 (HL)................................................................12, 32–33 O’Rourke (HM Inspector of Taxes) v Binks [1992] STC 710.................................................173 Pao On v Lau Yiu Long [1980] AC 614 (PC)...........................................................................73 Parker v Great Western Railway Co (1844) 7 Man & G 253, 135 ER 107...............................90 Parsons v Blandy (1810) Wight 22, 145 ER 1160 ...................................................................73 Pascoe v First Secretary of State [2006] EWHC 2356 (Admin), [2007] 1 WLR 885............142 Philip Collins Ltd v Davis [2000] 3 All ER 808 (Ch).............................................129, 131, 135 Phillips-Higgins v Harper [1954] 1 QB 411 ......................................................................53, 57 Pickstone v Freemans plc [1988] 2 All ER 803 (HL)..............................................................174 Pilkington’s Will Trust, Re; Pilkington v Inland Revenue Commissioners [1964] AC 612 (HL), reversing [1964] AC 612 (CA)...................................................................151 Pirelli Cable Holding NV v Inland Revenue Commissioners [2006] UKHL 4........................27 Pirelli Cable Holding NV v HM Revenue and Customs [2008] EWCA Civ 70.......................27 Pitt v Holt [2010] EWHC 45 (Ch), [2010] 1 WLR 1199............150, 153–55, 158–59, 161–62 Pitt v Holt [2011] EWCA Civ 197, [2012] Ch 132................ 10, 150, 153–55, 158–59, 161–62 Practice Statement (Judicial Precedent) House of Lords [1966] 1 WLR 1234 ..............31, 44 Queens of the River Steamship Co Ltd v Conservators of the River Thames (1899) 15 TLR 474...........................................................................................................................73 R v A (No 2) [2002] 1 AC 45..................................................................................................173 R v Abbott [1897] 2 IR 362.....................................................................................................145 R v Attorney-General of England and Wales [2003] UKPC 22, [2003] EMLR 24................100 R v Customs & Excise Commissioners, ex parte Building Societies Ombudsman Co Ltd [2000] STC 892......................................................................................................................9 R v Director of Public Prosecutions, ex parte Kibelene [2000] 2 AC 326 (HL) 380...............143 R v Inland Revenue Commissioners, ex parte Woolwich Equitable Building Society [1987] STC 654 (QB); [1990] 1 WLR 1400 (HL).........................................................32, 51 R v Panel on Take–overs and Mergers, ex parte Datafin Plc [1987] QB 815 (CA)..................24 R v Secretary of State for the Home Department, ex parte Fire Brigades Union [1995] 2 AC 513 (HL)......................................................................................................................32 R v Secretary of State for the Home Department, ex parte Pierson [1998] AC 539 (HL).....289 R v Secretary of State for Transport, ex parte Factortame Ltd [1990] 2 AC 85 (HL).............172 R v Somerset County Council, ex parte Fewings [1995] 1 All ER 513 (HL)............................24 R (Burnley Borough Council) v First Secretary of State [2006] EWHC 798 (Admin).........141 R (Child Poverty Action Group) v Secretary of State for Work and Pensions [2009] EWCA Civ 1058, [2010] 1 WLR 1886 ..............................................................................141 R (Child Poverty Action Group) v Secretary of State for Work and Pensions [2010] UKSC 54, [2011] 2 AC 15..............................................................................61, 125–26, 247 R (Clays Lane Housing Co–Operative Ltd) v The Housing Corporation [2004] EWCA Civ 1658, [2005] 1 WLR 2229........................................................................ 142–43 R (Daly) v Secretary of State for the Home Department [2001] 2 AC 532..................... 141–42 R (Fisher) v English Nature [2003] EWHC 1599 (Admin), [2004] 1 WLR 503..................142 R (Rottman) v Metropolitan Police Commissioner [2002] UKHL 20, [2002] 2 AC 692.............................................................................................................................126

Table of Cases xxvii

R (SRM Global Master Fund LP) v Treasury Commissioner [2009] EWCA Civ 788, [2009] UKHRR 1219.................................................................................................. 141–42 R (Trailer and Marina (Leven) Ltd) v Secretary of State for the Environment, Food and Rural Affairs [2004] EWCA Civ 1580, [2005] 1 WLR 126........................................142 Racal Group Services Ltd v Ashmore [1995] STC 1151.................................................153, 159 Ramsay (WT) Ltd v IRC [1982] AC 300...............................................................................164 Redgrave v Hurd (1881) LR 20 Ch D 1 (CA)........................................................................101 Reg v Tower Hamlet London Borough Council, ex parte Chetnik Ltd [1988] 1 AC 858 (HL)......................................................................................................................................71 Relfo Ltd (in liq) v Varsani [2012] EWHC 2168 (Ch) ......................................................9, 122 Rose v Ford [1937] AC 826 (HL).............................................................................................66 Rowe v Vale of White Horse District Council [2003] EWHC 388 (Admin), [2003] 1 Lloyd’s Rep 418 ..............................................................................................................304 Satchwell v McIntosh 2006 SLT (Sh Ct) 117 .........................................................................228 Scottish Equitable plc v Derby [2001] EWCA 369, [2001] All ER 818..........................129, 131 Sebel Products Ltd v Customs & Excise Commissioners [1949] Ch 409 (Ch) ............3, 71, 108 Secretary of State for the Home Department v JJ [2007] UKHL 45, [2008] 1 AC 385 (HL)...............................................................................................................................25 Sempra Metals Ltd (formerly Metallgesellschaft Ltd) v HM Revenue & Customs [2007] UKHL 34, [2008] 1 AC 561......................................... 4, 6, 8, 18, 26, 35, 53, 59, 140, 189, 200, 209, 257–58, 262–63, 270 Shepherd v Hills (1855) 11 Ex 55, 156 ER 743......................................................................324 Sieff v Fox [2005] EWHC 1312 (Ch), [2005] 1 WLR 3811 ......................150–53, 155–57, 163 Simmonds, Ex parte (1885) 16 QBD 308.................................................................................71 Slater v Mayor of Burnley (1888) 59 LT 636 ...........................................................................76 Slocock, Re [1979] 1 All ER 358.............................................................................................163 Smith v Hughes (1871) LR 6 QB 597 (CA)...........................................................................101 Smith v Secretary of State for Trade and Industry [2007] EWHC 1013 (Admin) [2008] 1 WLR 394..........................................................................................................................142 SmithKline Beecham plc v Apotex Europe Ltd [2006] EWCA Civ 658, [2007] Ch 71..........121 Smithson v Hamilton [2007] EWHC 2900 (Ch), [2008] 1 WLR 1453................................156 Soteriou (Andreas) v Ultrachem Ltd [2004] EWHC 983 (QB).............................................141 South of Scotland Electricity Board v British Oxygen Co Ltd (No 2) [1959] 1 WLR 587 (HL).........................................................................................................................86, 90 Stannard v Fisons Pension Trust Ltd [1991] PLR 224 ...................................................152, 163 Steele v Williams (1853) 8 Ex 625, 155 ER 1502 .......................................................74–76, 276 Stephenson v The Queen (1865) 2 WW & A’BL 143..............................................................321 Test Claimants in the ACT Litigation v HM Revenue & Customs [2010] EWHC 359 (Ch), [2010] STC 1678........................................................................................................58 Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2008] EWHC 2893 (Ch), [2009] STC 254......... 10, 12–13, 54, 56, 59, 61, 69–70, 127, 130, 132–35, 137, 140, 144–45, 149, 161, 190–91, 264–65, 270, 297 Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2010] EWCA Civ 103, [2010] STC 1251.......... 7–8, 12–14, 35, 44, 54–56, 59–61, 69, 82, 118, 123, 130, 132, 141, 172–78, 181–86, 190, 203–07, 209–10, 213, 256, 264–65

xxviii Table of Cases Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2012] UKSC 19, [2012] 2 AC 337, [2012] Bus LR 1033.............4–18, 28–31, 34, 37–46, 49, 53–54, 56–57, 61, 63–64, 69, 82, 85, 96, 100, 102, 105, 112, 124, 130, 132, 171–72, 175–83, 189, 191, 200, 207–10, 213–16, 223, 231, 239, 241, 255–56, 270, 313 Test Claimants in the Thin Cap Group Litigation v HM Revenue & Customs [2009] EWHC 2908 (Ch), [2010] STC 301 .............................................................................27, 59 Thorpe v HM Revenue & Customs [2009] EWHC 611 (Ch), [2009] STC 2107..................143 Times Right Marketing Limited (in liq) v HM Revenue & Customs, 13 March 2008 (London VAT and Duties Tribunal) .........................................................................117, 195 Traherne v Gardner (1856) 5 El & Bl 914, 119 ER 721...........................................................75 Trustor AB v Smallbone (CA), 9 May 2000............................................................................121 Twyford v Manchester Corporation [1946] Ch 236.................................................................75 Umphelby v M’Lean (1817) 1 B & Ald 42, 106 ER 16.............................................................77 United Railways of Havana and Regla Warehouses Ltd, In re [1961] AC 1007 (HL).............44 Universe Tankships Inc of Monrovia v International Transport Workers’ Federation (The Universe Sentinel) [1983] AC 366 (HL).......................................................72–73, 100 Uren v First National Home Finance Ltd [2005] EWHC 2529 (Ch) .......................................3 Vestey v Inland Revenue Commissioners (Nos 1 and 2) [1980] AC 1148 (HL)......................45 Vodafone 2 v HM Revenue & Customs (No 2) [2009] STC 1480 (CA)........171–174, 180, 183 Waikato Regional Airport Ltd v Attorney General [2003] UKPC 50, [2004] 3 NZLR 1..........................................................................................................84–86, 88, 146 Waste Recycling Group Ltd v HM Revenue & Customs [2008] EWCA Civ 849, [2009] STC 200......................................................................................................................5 Waterhouse v Keen (1825) 4 B & C 200, 107 ER 1033............................................................74 Watkins v Secretary of State for the Home Department [2006] AC 395 (HL) ......................293 Westdeutsche Landesbank Girozentrale v Islington London Borough Council [1994] 1 WLR 938 (CA); [1996] AC 669 (HL)......................................................... 3, 214, 247, 257 Western Bank Ltd v Schindler [1977] Ch 1 (CA)..................................................................174 WHA Ltd v HM Revenue & Customs [2007] EWCA Civ 728, [2007] STC 1695.................164 Wheels Common Investment Fund Trustees Ltd v HM Revenue & Customs [2011] UKFTT 534 (TC), [2011] Pens LR 387.............................................................................114 Whiteside v Whiteside [1950] Ch 56......................................................................................159 Wilkins (John) (Motor Engineers) Ltd v Revenue and Customs Commissioners [2009] UKUT 175 (TCC), [2009] STC 2485; [2010] EWCA Civ 923..........................171–72, 177, 179–81, 183–85, 258 William Whiteley Ltd v The King (1909) 101 LT 741......................................................76, 277 Wilson v First County Trust Ltd (No 2) [2003] UKHL 40, [2004] 1 AC 816................. 142–43 Woolwich Equitable Building Society v Inland Revenue Commissioners [1989] 1 WLR 137 (QB).................................................................................................. 3, 71, 88, 93 Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] AC 70 (HL).................................................. 3–5, 7, 10–12, 15–19, 24–25, 28–33, 35–39, 42, 44–47, 49–58, 60, 63, 66, 69–76, 80–91, 93, 96–97, 100–02, 104–09, 127, 132–36, 139–40, 142, 145, 178, 190, 207–10, 213–17, 223–25, 231, 237, 239–42, 245–48, 252–53, 255–57, 264–65, 270, 273–74, 276–77, 279–81, 283–91, 293–94, 299, 301–03, 312–16, 318–19, 322, 325

Table of Cases xxix

United States of America Atchison, Topeka & Santa Fe Railway Co v O’Connor 223 US 280 (1911).......................78, 88 Bivens v Six Unknown Named Agents of Federal Bureau of Narcotics 403 US 388 (1971).................................................................................................................................321 Gaar, Scott & Co v Shannon 223 US 468 (1911).....................................................................78 United States v Butler 297 US 1 (1936)...................................................................................86

Table of Legislation Australia Administrative Decisions (Judicial Review) Act 1977 (Cth)...................................... 324–325 s 5........................................................................................................................................324 s 16......................................................................................................................................324 s 16(1).................................................................................................................................324 s 16(1)(d)................................................................................................................... 324–325 Business Franchise Licenses (Tobacco) Act 1987 (NSW)....................................................317 Charter of Human Rights and Responsibilities Act 2006 (Vict)................................ 321–322 s 20......................................................................................................................................321 s 31......................................................................................................................................322 s 36......................................................................................................................................322 s 39......................................................................................................................................322 s 39(1).................................................................................................................................322 s 39(3)–(4)..........................................................................................................................322 Constitution.............................................................................................20, 315–317, 319–321 s 7........................................................................................................................................316 s 24......................................................................................................................................316 s 51(xxxi)............................................................................................................................321 s 61......................................................................................................................................319 s 64......................................................................................................................................316 s 75(v).................................................................................................................................323 s 76 (i).................................................................................................................................317 s 77 (iii)...............................................................................................................................317 s 90............................................................................................................................. 317–320 s 92......................................................................................................................................320 s 128....................................................................................................................................316 Crown Suits Act 1947 (WA)..................................................................................................317 Federal Airports Corporation Act 1986 (Cth) s 56......................................................................................................................................325 Franchise Fees Windfall Tax (Collection) Act 1997 (Cth)...................................................318 s 4............................................................................................................................... 317–318 s 6........................................................................................................................................318 Human Rights Act 2004 (ACT).................................................................................... 321–322 Pt 3......................................................................................................................................322 s 8........................................................................................................................................322 s 40B...................................................................................................................................322 s 40C...................................................................................................................................322 Judiciary Act s 39B...................................................................................................................................323

xxxii Table of Legislation Law Reform (Property, Perpetuities and Succession) Act 1962 s 23........................................................................................................................................64 s 23(2)...................................................................................................................................64 Migration Act.........................................................................................................................324 Property Law Act 1969 s 124(2).................................................................................................................................64 Sales Tax Assessment Act 1992 (Cth)....................................................................................323 Pt 4......................................................................................................................................323 Stamps Act 1958 (Vict) s 111....................................................................................................................................323 s 111(1)...................................................................................................................... 323–324 Stamps Act 1978.....................................................................................................................323

Canada Charter of Rights and Freedoms s 15......................................................................................................................................310 Customs Act...........................................................................................................................309 Excise Tax Act.........................................................................................................................309 s 312....................................................................................................................................309 Financial Administration Act........................................................................................309, 311 s 19(2).................................................................................................................................311 Immigration and Refugee Protection Act.............................................................................311

European Union EC Treaty................................................................................................................................170 Art 10 (ex Art 5 EC, now Art 4(3) TEU)....................................................................40, 170 Art 52 (ex Art 43 EC, now Art 49 TFEU)...........................................................................26 EU Charter on Fundamental Rights.....................................................................................147 Art 17..................................................................................................................................147 Art 52..................................................................................................................................147 Treaty on European Union (TEU) Art 6....................................................................................................................................147 Treaty on the Functioning of the European Union (TFEU)........................................175, 180 Art 4......................................................................................................................................40 Art 28..................................................................................................................................188 Art 49..........................................................................................................188, 201, 203–204 Treaty of Lisbon.....................................................................................................................239 Directives Council Directive 92/81/EEC of 19 October 1992 on the harmonization of the structures of excise duties on mineral oils...................................................................192

Table of Legislation xxxiii

Council Directive 1999/70/EC of 28 June 1999 concerning the framework agreement on fixed-term work concluded by ETUC, UNICE and CEEP.....................................185 Council Directive 2000/78/EC of 27 November 2000 establishing a general framework for equal treatment in employment and occupation.......................................... 184–185 Council Directive 2006/112/EC of 28 November 2006 on the common system of value added tax........................................................................................................................112 Sixth Council Directive 77/388 of 17 May 1977 on the harmonisation of the laws of the Member States relating to turnover taxes.......................................112, 181–182, 193 Art 13B(d)(6).....................................................................................................................114 Art 17..................................................................................................................................118 Art 17(1)–(2)......................................................................................................................117 Art 18(2).............................................................................................................................117 Art 22(5).............................................................................................................................117

Germany Abgabenordnung/AO (General Tax Code)......................................................................17, 259 § 37 I...................................................................................................................................249 § 37 II..........................................................................................................249–250, 262, 269 § 118...................................................................................................................................248 § 125...................................................................................................................................248 § 165 I 2 Nr 4.......................................................................................................................17 § 172.....................................................................................................................................17 § 233...................................................................................................................................259 § 233a..........................................................................................................................259, 262 §§ 234–238.........................................................................................................................259 § 347...................................................................................................................................248 § 361 I.................................................................................................................................249 Baugesetzbuch/BauGB (Federal Building Code) § 133 III 4...........................................................................................................................259 Beamtenrechtsrahmengesetz/BRRG (Federal Civil Service Framework Act) § 126...................................................................................................................................244 Bundesbesoldungsgesetz/BBesG (Federal Salary Law) § 2 I.....................................................................................................................................251 § 12 II..................................................................................................................249, 250, 252 Bürgerliches Gesetzbuch/BGB (Civil Code )........................... 17, 219–220, 223, 228, 230–231, 233–234, 237, 242, 246, 249, 252, 259, 266, 268–269 § 119...................................................................................................................................251 § 134...................................................................................................................................230 § 142...........................................................................................................................233, 251 § 199 I.................................................................................................................................236 § 288...................................................................................................................................259 § 289...................................................................................................................................259 §§ 291–292.........................................................................................................................259 § 313...................................................................................................................................228 § 313b.................................................................................................................................230

xxxiv Table of Legislation § 518 II................................................................................................................................230 § 766...................................................................................................................................230 § 812...........................................................................................219–220, 242–246, 258, 260 § 812 I.................................................................................................................................247 § 812 I 1.......................................................219–221, 224, 226, 229, 232–233, 237, 243, 252 § 812 I 2...................................................................................................... 219, 224, 237, 252 § 813.......................................................................................................................... 219–220 § 814...........................................................................219–220, 227, 230–232, 242, 252–255 §§ 815–816................................................................................................................ 219–220 § 817...................................................................................................219–220, 226, 232, 237 § 818.......................................................................................................................... 219–220 § 818 I.........................................................................................................258–261, 263, 266 § 818 II................................................................................................................258, 263, 266 § 818 III......................................................................................................246, 258, 266–269 § 818 IV..............................................................................................246, 258–260, 266, 269 § 819.......................................................................................................................... 219–220 § 819 I................................................................................................. 246, 258, 260, 266, 269 §§ 820–822................................................................................................................ 219–220 § 987 II................................................................................................................................259 Finanzgerichtsordnung/FGO (Court of Procedure of the Fiscal Courts) § 33.....................................................................................................................................244 § 69 I...................................................................................................................................249 Gerichtsverfassungsgesetz/GVG (Judicature Act) § 13............................................................................................................................ 242–243 § 71 II Nr 1.........................................................................................................................244 Gesetz über die Rechte des Fiskus hinsichtlich der Zinsen (Prussian Law concerning the rights of the fisc as regards interest) 7 July 1833 §§ 2–3.................................................................................................................................260 Grundgesetz/GG (Basic Law) Art 20 III.....................................................................................................................245, 266 Insolvenzordnung/InsO (Insolvency Code) § 143 I 2..............................................................................................................................262 Reichsabgabenordnung/RAO (General Tax Code of the German Reich).............................243 § 127...................................................................................................................................243 § 227...................................................................................................................................243 Sozialgerichtsgesetz/SGG (Code of Procedure of the Social Courts) § 51.....................................................................................................................................244 Sozialgesetzbuch/SGB (Social Code) SGB IV § 26 II..................................................................................................................................249 SGB X § 50 ....................................................................................................................249–250, 259 § 102...................................................................................................................................249 Verwaltungsgerichtsordnung/VwGO (Code of Procedure of the Administrative Courts) § 40.............................................................................................................................242, 244 § 68.....................................................................................................................................248 § 70 I...................................................................................................................................248

Table of Legislation xxxv

§ 74 I...................................................................................................................................248 § 80 II 1...............................................................................................................................249 § 113...........................................................................................................................249, 251 Verwaltungsverfahrensgesetz/VwVfG (Administrative Procedure Act ).........................17, 249 § 35 I...................................................................................................................................248 § 44.....................................................................................................................................248 § 48.......................................................................................................17, 250–251, 267–268 § 49.............................................................................................................250–251, 267, 268 § 49a............................................................................................ 250–251, 259, 266, 267, 268 § 50.....................................................................................................................................268 § 54.....................................................................................................................................251 § 59.....................................................................................................................................251 § 62.....................................................................................................................................251

Ireland Constitution (Bunreacht na hÉireann) 1937.................................................................274, 298 Art 34.3.2............................................................................................................................285 Art 34.5.1............................................................................................................................289 Art 40.3...............................................................................................................................298 Art 40.3.1............................................................................................................................289 Art 40.3.2............................................................................................................................291 Art 43.1...............................................................................................................................291 Customs Consolidation Act 1876 ss 30–31..............................................................................................................................276

New Zealand Biosecurity Act 1933................................................................................................................85 Judicature Act 1908 s 94A.....................................................................................................................................62 s 94A(2)................................................................................................................................64 Judicature Amendment Act 1958 s 2..........................................................................................................................................64

South Africa Constitution of South Africa Art 38..................................................................................................................................293

United Kingdom Arbitration Act 1980..............................................................................................................175

xxxvi Table of Legislation Bill of Rights 1688............................................................................ 98, 215, 217, 253, 313, 320 Art 4..............................................................................................................................81, 313 Canada Act 1982 s 24(1).................................................................................................................................293 Crown Proceedings Act 1949 s 21........................................................................................................................................50 European Communities Act 1972...................................................................................27, 172 s 2(1)–(2)............................................................................................................................172 s 2(4)...................................................................................................172–173, 175–176, 178 Finance Act 1923 s 24..........................................................................................................................................3 Finance Act 1972 Pt V.....................................................................................................................................132 Finance Act 1986 s 102....................................................................................................................................156 Finance Act 1991 s 53........................................................................................................................................51 Finance Act 1996 s 197....................................................................................................................................179 Finance Act 1997 s 47......................................................................................................................................198 Finance Act 2004 s 320.......................................................................... 4, 16, 35, 37–43, 56–57, 62, 67–68, 208 Finance Act 2006....................................................................................................................155 s 88......................................................................................................................................154 Sch 12.................................................................................................................................154 Finance Act 2007 s 107...................................................................................... 4, 16, 35, 37–43, 56–57, 68, 208 Finance Act 2008 s 121................................................................................................................................16, 56 Sch 39...................................................................................................................................30 Sch 39, para 36...................................................................................................................115 Finance Act 2009 s 100................................................................................................................................58, 64 Sch 52.............................................................................................................................58, 64 Human Rights Act 1998.................................................................................................141, 144 s 2(1)...................................................................................................................................141 s 3........................................................................................................................172–173, 176 s 3(1)...................................................................................................................................141 Income and Corporation Taxes Act 1988.............................................................................180 s 18........................................................................................................................................61 s 231....................................................................................................174, 176–178, 181, 183 s 231(1)...............................................................................................................174–175, 177 Income Tax Act 2007..............................................................................................................165 Income Tax (Earnings and Pensions) Act 2003....................................................................165 Income (Trading and Other Income) Act 2005....................................................................165 Inheritance Tax Act 1984

Table of Legislation xxxvii

ss 2–3..................................................................................................................................149 s 3A.....................................................................................................................................149 s 7........................................................................................................................................149 s 18......................................................................................................................................152 s 49......................................................................................................................................155 s 64......................................................................................................................................156 s 71......................................................................................................................................155 ss 105(1)(bb)......................................................................................................................155 s 106....................................................................................................................................155 s 122....................................................................................................................................155 s 142..............................................................................................................................10, 149 Inheritance Tax Act 1994 s 65......................................................................................................................................156 Law Reform (Miscellaneous Provisions) Act 1934.................................................................65 Limitation Act 1980.................................................................................................................61 s 32(1)(c)................................................ 4, 10, 15, 35, 37, 53, 56–59, 62, 115, 207–208, 255 Prescription and Limitation (Scotland) Act 1973 s 6..........................................................................................................................................65 Sch 1, para 1.........................................................................................................................65 Senior Courts Act 1981............................................................................................................32 Social Security Administration Act 1992 s 71........................................................................................................................61, 125, 247 Supreme Court Act 1981 s 31(4)...................................................................................................................................32 s 35A.............................................................................................................................93, 257 Taxation of Chargeable Gains Act 1992 ss 17–18..............................................................................................................................149 s 69(2)........................................................................................................................ 153–154 s 71(1).................................................................................................................................156 s 260....................................................................................................................................156 s 286....................................................................................................................................149 Taxes Management Act 1970...............................................................................58, 61–62, 178 s 33............................................................... 3, 7, 58, 60–61, 68, 150, 155, 174, 178, 182, 207 s 33(2A)......................................................................................174–176, 178–179, 181–183 s 33(2A)(a).........................................................................................................................178 s 42..............................................................................................................................150, 155 Value Added Tax Act 1994............................................................. 8–9, 111–112, 114, 119, 124 s 4(2)...................................................................................................................................118 s 25......................................................................................................................................118 s 25(1).................................................................................................................................195 s 25(2)........................................................................................................................ 194–195 s 25(3).........................................................................................................................117, 195 s 26..................................................................................................................................9, 118 s 26(1)........................................................................................................................ 194–195 s 26(2).................................................................................................................................195 s 26(2)(a)............................................................................................................................194 ss 25–26..............................................................................................................................118

xxxviii Table of Legislation s 36......................................................................................................................................195 s 78.................................................................. 35–36, 39, 54–55, 61, 179–181, 200, 208–209 s 78(1).................................................................................................................................179 s 78(3).................................................................................................................................179 s 80.......................7, 35–36, 39, 54–55, 61, 111, 114–116, 124–125, 150, 155, 200, 208–209 s 80(2A)..........................................................................................................8, 115–116, 119 s 80(3)...................................................................................................................61, 114, 122 s 80(4).................................................................................................................................115 s 80(7)..................................................................................112, 116, 124, 126, 150, 155, 194 Statutory Instruments Civil Procedure (Modification of Supreme Court Act 1981) Order 2004 (SI 2004/1033)......32 Civil Procedure Rules 1998 (SI 1998/3132) CPR Part 54........................................................................................................................247 rule 54.3................................................................................................................................33 Income Tax (Building Societies) Regulations 1986 (SI 1986/482)............................51, 80, 93 Value Added Tax Regulations 1995 (SI 1995/2518) reg 29(1A)....................................................................................................................16, 183 reg 101(2)(b)–(c)...............................................................................................................118

Table of Conventions, Treaties etc European Convention for the Protection of Human Rights and Fundamental Freedoms 1952................................................................................................141–143, 146–147, 176 Section I.............................................................................................................................291 First Protocol, Art 1.............................................................................13, 140–143, 147, 291 UN International Covenant on Civil and Political Rights 1966 Art 2.3.................................................................................................................................289

1 Introduction STEVEN ELLIOTT, BIRKE HÄCKER AND CHARLES MITCHELL

A. BACKGROUND

Time was when money paid to the government was surprisingly difficult to recover when it transpired that the payment was not due. Tax statutes gave claimants some recovery rights,1 but these were limited in scope. So too were their rights at common law, which were essentially no different from their rights against private individuals.2 Money could be recovered if it was paid by mistake, but only if the mistake related to a fact and not if it related to a proposition of law. Money could also be recovered if it was paid under duress, including duress colore officii, but such cases were (and are) uncommon. Where money was paid pending the outcome of a dispute with a tax authority, it was sometimes possible to imply an agreement that the money would be repaid if the taxpayer were successful;3 resort to this artifice highlighted the inadequacy of the remedies that were generally available to taxpayers. Times have changed. In 1992 the House of Lords made a new start in Woolwich Equitable Building Society v IRC.4 Encouraged by academic writings5 and developments in European law,6 the court held that money paid as tax pursuant to ultra vires legislation could be recovered, although there was no illegitimate compulsion, mistake, or contract to repay. This was a bold step. Unjust enrichment claims generally lie only when the case falls ‘within or close to some established category or factual recovery situation’,7 but Woolwich shows us that the courts have the power to recognise new grounds for recovery.8 1 Statutory recovery rights were first introduced by the Finance Act 1923, s 24, which was the statutory precursor of the present-day Taxes Management Act 1970, s 33. 2 See the survey in Law Commission, Restitution: Mistakes of Law and Ultra Vires Public Authority Receipts and Payments (Law Com No 227, 1994) Part C. 3 Sebel Products Ltd v Customs & Excise Commissioners [1949] Ch 409; Woolwich Equitable Building Society v Inland Revenue Commissioners [1989] 1 WLR 137 (QB). 4 Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] AC 70 (HL) (hereafter ‘Woolwich’). 5 P Birks, ‘Restitution from the Executive: a Tercentenary Footnote to the Bill of Rights’ in P Finn (ed), Essays on Restitution (Sydney, Law Book Co, 1990) 164; W Cornish, ‘“Colour of Office”: Restitutionary Redress Against Public Authority’ (1987) 14 Journal of Malaysian and Comparative Law 41. 6 See Woolwich (n 4) 177, where Lord Goff cited Case 199/82 Amministrazione delle Finanze dello Stato v SpA San Giorgio [1983] ECR 3595 (hereafter ‘San Giorgio’). 7 Uren v First National Home Finance Ltd [2005] EWHC 2529 (Ch) [16]–[18]. 8 See especially Woolwich (n 4) 172 (Lord Goff), but contrast 197 (Lord Browne-Wilkinson). See too CTN Cash & Carry Ltd v Gallaher Ltd [1994] 4 All ER 714 (CA) 720; Westdeutsche Landesbank Girozentrale v Islington London Borough Council [1996] AC 669, 691 and 722, and Kleinwort Benson Ltd v Lincoln City Council [1999] AC 349 (hereafter ‘Kleinwort Benson’) 372 and 393. For discussion of the limits to the courts’ power see Gibb v Maidstone and Tunbridge Wells NHS Trust [2010] EWCA Civ 678, [2010] IRLR 786 (hereafter ‘Gibb’) [26]–[27].

4 Steven Elliott, Birke Häcker and Charles Mitchell In 1999 the House of Lords took a further significant step in Kleinwort Benson Ltd v Lincoln City Council,9 when it abolished the mistake of law bar. In obiter dicta, the court also held that restitution could be awarded on the ground of mistake where payments were made pursuant to an understanding of the law which was correct at the time but which was retrospectively falsified by a later court decision. In the first decade of the present century, claimants who paid money in the mistaken belief that they owed a tax liability therefore had two new avenues to recovery – Woolwich and mistake of law – and they made the most of them. The Woolwich case left some important questions unanswered. Does the principle established by the case apply where taxes are demanded on the basis of a misapplication of valid legislation? According to the Court of Appeal in British Steel plc v Customs & Excise Commissioners (No 1),10 the answer is yes. Does the principle only operate where the amount was demanded by the state? In Test Claimants in the FII Group Litigation v HMRC,11 the Supreme Court has now told us that demand is inessential, with the result that the Woolwich principle also applies to taxes collected by self-assessment. More fundamentally, does Woolwich cover the field of overpaid tax? In Deutsche Morgan Grenfell Group plc v IRC,12 the House of Lords held that it does not, and that Woolwich claims sit alongside, and are merely alternatives to, any other restitution claims that are also available. More fundamentally still, how does the Woolwich principle fit within the structure of the English law of unjust enrichment, which generally requires claimants to establish a positive ground for restitution, and does not generally treat the absence of a legal ground for a transfer as a sufficient reason to award recovery?13 In FII,14 members of the Supreme Court touched inconclusively on this question, but it awaits a definitive judicial answer. In Kleinwort Benson, Lord Browne-Wilkinson dissented from the court’s findings that ‘retrospective’ mistakes could ground recovery, and would be subject to the same limitation rules as other mistakes, because he foresaw that these findings would enable claimants to recover payments stretching back for decades.15 In DMG,16 the House of Lords confirmed that claimants relying on ‘retrospective’ mistake to recover money paid as tax were entitled to invoke the Limitation Act 1980, section 32(1)(c). This postpones the inception of the six-year period applicable to mistake claims to the date when the claimant knew of, or could reasonably be expected to have discovered, his mistake. Also, the date when the claimant’s mistake became reasonably discoverable in such a case was usually no earlier than the date of the court decision which retrospectively changed the law. In response to these findings, legislation was enacted to disapply section 32(1)(c) in relation to mistakes of law relating to taxation matters under the care and management of the Revenue.17 Kleinwort Benson (n 8). British Steel plc v Customs & Excise Commissioners (No 1) [1997] 2 All ER 366. 11 Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2012] UKSC 19, [2012] 2 AC 337 (hereafter, ‘FII (SC)’) [64]–[81] and [171]–[174]. 12 Deutsche Morgan Grenfell Group plc v Inland Revenue Commissioners [2006] UKHL 49, [2007] 1 AC 558 (hereafter ‘DMG’). 13 Kleinwort Benson (n 8) 405; DMG (n 12) [21] and [155]; Sempra Metals Ltd v Inland Revenue Commissioners [2007] UKHL 34, [2008] 1 AC 561 (hereafter ‘Sempra’) [23]–[25]. 14 FII (SC) (n 11) [81] and [162]. 15 Kleinwort Benson (n 8) 364. 16 DMG (n 12). 17 The Finance Act 2004, s 320, which applied prospectively to actions brought on or after 8 September 2003, and the Finance Act 2007, s 107, which extended the disapplication retrospectively to any action brought before 8 September 2003 for relief from the consequences of such a mistake of law. 9

10

Introduction 5

However, a majority of the Supreme Court has now held in FII 18 that these statutory measures have not achieved their purpose with regard to claims by taxpayers with EU lawsourced rights of recovery, as the government failed to implement transitional measures to protect taxpayers with accrued restitutionary rights. A reference has been made to the CJEU to determine whether this is indeed the position. While Woolwich and mistake of law are now the grounds of recovery generally relied on by English claimants bringing actions at common law,19 the recognition of these grounds does not alone account for the explosion of overpaid tax litigation that has occurred over the past decade. This may also be attributed to other factors on the European and the domestic planes, factors which have combined to create a perfect storm for HMRC. The first factor is the new assertiveness of the Court of Justice of the European Union20 in tax matters. Value Added Tax (VAT) is rooted in European directives, so that the CJEU has always had the last word. In contrast, direct taxes, such as corporation tax and income tax, find their source in national law. The CJEU has nonetheless increasingly insisted that these taxes should not be structured in such a way as to infringe basic principles of European law. A series of references beginning in 1998 with ICI plc v Colmer,21 and including most prominently Metallgesellschaft Ltd v IRC 22 and Test Claimants in the FII Group Litigation v IRC,23 ‘turned into a root and branch onslaught on many established features of the system for taxing companies’.24 At the same time the European and English courts have also held that other types of tax have been wrongly collected.25 All said, a great deal of tax has been paid that should not have been paid. A second factor is the CJEU’s uncompromising requirement that taxes collected contrary to European law should be repaid.26 This European right to repayment must be given effect by the national courts applying the procedures available to them, but European law regulates that process closely. The CJEU has taken a very strict line in relation to the defences which national legal systems may legitimately allow to such claims, apparently restricting this to reasonable time-bars and literal passing on.27 So, for example, the FII (SC) (n 11). Woolwich has had a mixed reception internationally. It has been followed in Ireland, but it has neither been received into Australian law nor rejected there. Meanwhile, Canadian law has taken an altogether different path. For discussion see chapters 13, 14, and 15 of the present volume. 20 Hereafter ‘CJEU’. In this chapter, references to the CJEU include the former European Court of Justice (ECJ). 21 Case C-264/96 ICI Chemical Industries plc v Colmer (HM Inspector of Taxes) [1998] ECR I-4695. 22 Joined Cases C-397/98 and C-410/98 Metallgesellschaft Ltd v Inland Revenue Commissioners and Hoechst AG v Inland Revenue Commissioners [2001] ECR I-1727. 23 Case C-446/04 Test Claimants in the Franked Investment Income Group Litigation v Inland Revenue Commissioners [2006] ECR I-11753. 24 A Park, ‘A Judge’s Tale: Corporation Tax and Community Law’ [2006] British Tax Review 322, 332. 25 For a few examples that have prompted restitution claims currently under consideration see Case C-169/04 Abbey National plc v HM Revenue & Customs [2006] ECR I-4027 and Case C-363/05 JP Morgan Fleming Claverhouse Investment Trust plc v HM Revenue & Customs [2007] ECR I-5517 (VAT); Case C-569/07 HSBC Holdings plc v HM Revenue & Customs [2009] ECR I-9047 (stamp duty reserve tax); and Waste Recycling Group Ltd v HM Revenue & Customs [2008] EWCA Civ 849, [2009] STC 200 (landfill tax). 26 In England, this requirement is commonly associated with the CJEU’s decision in San Giorgio (n 6), and claims made under it are commonly called ‘San Giorgio claims’. This is perplexing, and may reflect only the fact that Lord Goff cited that case in Woolwich (n 4). The issue in San Giorgio concerned the defence of passing on; the EU law principle requiring repayment of taxes collected contrary to EU law had already been established, for example in Case 68/79 Hans Just I/S v Danish Ministry for Fiscal Affairs [1980] ECR 501. The CJEU does not accord the San Giorgio case the seminal importance that the English courts appear to. 27 See most recently Case C-398/09 Lady & Kid A/S v Skatteministeriet, CJEU, 6 September 2011, [2012] STC 854, and Case C-310/09 Ministre du Budget, des Comptes Publics et de la Fonction Publique v Accor SA, 15 September 2011. For discussion see R Williams, ‘Unjust Enrichment and the Court of Justice of the European Communities: A Loss of National Competence and Principle?’ [2011] British Tax Review 631. 18 19

6 Steven Elliott, Birke Häcker and Charles Mitchell ‘prevailing practice’ defence, which has historically provided an important protection to tax authorities, is not allowed.28 Moreover, while reasonable time-bars are allowed, the courts have struck down, as contrary to European law, attempts by Parliament to stem the tide of claims by curtailing limitation periods without advance notice.29 A third factor is the absence of a comprehensive and satisfactory statutory scheme regulating the repayment of unlawfully collected tax. In England there is a patchwork of taxspecific schemes, each with its own peculiar features. Some taxes are not subject to any such scheme, so that common law claims can be freely brought. Other statutory schemes have proven to be unduly restrictive and offensive to European law, and have therefore been restrictively construed or disapplied,30 again allowing claimants to bring common law claims. The conjunction of these factors has been an irresistible enticement for some taxpayers. It has also caught the attention of accountancy firms which, with the same enthusiasm that they once dedicated to the creation and marketing of tax avoidance schemes (now less saleable following the enactment of anti-avoidance legislation), have focused their attention on the creation and marketing of schemes to challenge aspects of UK tax legislation with a view to obtaining restitution of money paid to HMRC. If tax can be shown to have been collected unlawfully, and especially in breach of European law, then the taxpayer need only find a way to escape from any restrictive statutory scheme that may cover the case in order to recover payments going back for many years, along with interest that may well outstrip the principal sums claimed, particularly if compound interest is awarded.31 This volume takes a detailed look at all these developments and their ramifications for the recovery of money paid as tax and for the law of unjust enrichment in England, the European Union, and other jurisdictions. In the remainder of this introductory chapter, we set the contributors’ work in context and offer some critical observations of our own. Our discussion tracks the division made in the book between chapters on English law, European law, and comparative law.

28 Case C-188/95 Fantask A/S v Industriministeriet (Erhvervsministeriet) [1997] ECR I-6783; FII (SC) (n 11) [116]–[120] and [204]–[205]. 29 In addition to FII (SC) (n 11) see Case C-62/00 Marks & Spencer plc v Customs & Excise Commissioners [2002] ECR I-6325 (hereafter ‘Marks & Spencer’); Fleming (trading as Bodycraft) v HM Revenue & Customs [2008] UKHL 2, [2008] 1 WLR 195 (hereafter ‘Fleming’). 30 eg FII (SC) (n 11) (statutory scheme construed as being non-exclusive of common law claims where an exclusive scheme would have offended European law by reason of the statutory prevailing practice defence); Investment Trust Companies (in liquidation) v HM Revenue & Customs [2012] EWHC 458 (Ch), [2012] STC 1150 (hereafter ‘ITC ’) (statutory scheme disapplied where it provided no remedy for end consumers). 31 As the House of Lords effectively held that it should be in Sempra (n 13). In Case C-591/10 Littlewoods Retail Ltd v HM Revenue & Customs, 19 July 2012, [2012] STC 1714 (hereafter ‘Littlewoods (CJEU)’), the compound interest on a principal amount of some £205 million was said to be some £1.25 billion. The CJEU declined in that case to rule whether Littlewoods’ European law right to repayment also entitled it to compound interest. Media reports suggest that HMRC are facing claims whose total value runs to billions of pounds: eg N Huber, ‘Retailers Hope to Bag VAT Windfall from European Ruling’ Accountancy Age, 22 February 2012; published online at: www.accountancyage.com/aa/analysis/2154026/retailers-hope-bag-vat-windfall-european-ruling. This is borne out by the provision made for repayment claims in HMRC’s 2008/09 accounts, which was increased from £2.22 billion to £8.545 billion: HM Revenue & Customs, 2008–09 Accounts pp 101–2, paras 9 and 11; published online at www.hmrc.gov.uk/about/hmrc-accs-0809.pdf.

Introduction 7

B. ENGLISH LAW

The chapters collected in Section II of this book concern the English law governing the recovery of money paid as tax which is not due. In practice, most claims of this kind are brought under statutory recovery schemes, such as those established by the Taxes Management Act 1970, section 33, and the Value Added Tax Act 1994, section 80. These schemes are generally exclusive, ie claimants who fall within their scope are forbidden to rely on their common law rights, and may only rely on their statutory rights. Nevertheless, a series of very high-value claims have been made in recent years by claimants who have sought to rely on their common law rights, either because they have no statutory rights or (more commonly) because their statutory rights are more limited than their common law rights, for example because they are subject to shorter limitation periods or because the statutory schemes make less generous provision for the award of interest than would otherwise be available. To overcome these disadvantages, some claimants have argued that they should be permitted to rely on their common law rights because their case falls outside the scope of any statutory regime,32 while others have argued that although their case falls within the scope of a statutory regime, it would be contrary to EU law to confine them to their statutory rights because these are insufficiently generous to give them an effective remedy for their EU law-sourced rights of recovery.33 This litigation has caused the English courts to focus their attention on the content and scope of common law claims in unjust enrichment to recover money paid as tax. Their findings have an obvious significance for tax lawyers, and they also have important ramifications for the law of unjust enrichment. To make out a claim in unjust enrichment, the claimant must show that the defendant was enriched, that this enrichment was gained at the claimant’s expense, and that it was ‘unjust’, meaning that the circumstances in which it occurred coincided with, or were analogous to, those of a previous case in which restitution was awarded.34 Where all these elements are present, a remedy will be awarded unless the defendant can raise a defence. Questions relating to all of these matters have arisen in the overpaid tax cases, and we will say something here about each of them.

(1) Enrichment Money is obviously beneficial, and so one might have thought that enrichment would not be a difficult issue in the context of claims to recover money paid as tax. However, there are some complexities. These derive from the fact that money not only has an exchange value, but also has a use value. The exchange value of money is always simply its face value, reflecting the fact that money is not only a store of value, but also a measure of value. However the use value of money varies with the recipient’s identity, as it is measured by As in eg DMG (n 12) esp [19], [55], and [135]. As in eg Monro v HM Revenue & Customs [2008] EWCA Civ 306, [2009] Ch 69 esp [34]; Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2010] EWCA Civ 103, [2010] STC 1251 (hereafter, ‘FII (CA)’) esp [261], but compare FII (SC) (n 11) [116]–[119] and [204]–[205]; ITC (n 30) [111]–[171]. 34 Woolwich (n 4) 196–97; Kleinwort Benson (n 8) 737; Gibb (n 8) [26]–[27]. 32 33

8 Steven Elliott, Birke Häcker and Charles Mitchell asking what it would have cost the recipient to borrow the same amount, and some recipients (for example, the government) can borrow money more cheaply than others. Thus, in Sempra,35 the majority of the House of Lords held that the Revenue were enriched by the use value of money received from the claimants, measured as compound interest because this was the type of interest that would have been charged had the government borrowed the same sum on the market. However, the rate charged would have been lower than other lenders would have had to pay and so this lower rate was the measure of the government’s enrichment.36 More difficult enrichment issues have arisen in cases where claimants have not paid money as tax, but have foregone a valuable right. One such is the FII case,37 where the claimants used tax reliefs to offset invalid tax liabilities that they would otherwise have used to offset valid tax liabilities. According to the Court of Appeal, ‘[u]tilisation of the reliefs may have been a detriment to the claimants, but did not represent a gain to the Revenue for the purpose of a restitutionary cause of action’.38 In reaching this conclusion, Arden LJ reasoned that recovery could not be permitted because the claimants’ valid tax payments were ex hypothesi payments that HMRC would have been entitled to receive. Unfortunately this missed the point of the claimants’ argument. As Maximilian Schlote observes in chapter ten, the enrichment in respect of which the claim was made was not the value of the money paid to HMRC in respect of later, lawful tax liabilities, but the value of HMRC’s discharged obligation to allow the claimants a credit against lawfully due tax. The value of that discharged obligation could not be calculated until later, but even so it was an enrichment in HMRC’s hands at the time when the reliefs were used. As Charles Mitchell discusses in chapter six, a similar confusion was avoided by Henderson J in the ITC case.39 There the claimants paid money as VAT to fund managers for supplies which were later discovered to be exempt. To pay their supposed output tax liability, the managers used a combination of cash and input tax credits to which it was assumed they were entitled. When the parties’ mistake was discovered, the managers recovered the amount of their cash payments to HMRC and passed this money back to the claimants. However the Value Added Tax Act 1994, section 80(2A) did not allow the managers to recover the value of the credits. Hence the claimants brought their own common law claims against HMRC in respect of these. HMRC contended that they were not enriched by receipt of this value because it represented money which they had been legally entitled to receive as output tax from the managers’ suppliers. Henderson J held this to be irrelevant because the enrichment claimed was not the value of the money paid to HMRC by the suppliers, but the value of HMRC’s discharged obligation to allow the managers to set off their input tax credits against valid output tax liabilities. Following Schlote, Mitchell argues that Henderson J was right to reject HMRC’s argument, but that he should have held that HMRC were not enriched for a different reason. This was that the Value Added Sempra (n 13). Many legal scholars consider that this decision illustrates the principle that a defendant need not make restitution of the objective market value of a benefit where he can show that subjectively he would only have been willing to pay a lower price for the benefit if he had been given a choice: eg A Burrows, The Law of Restitution, 3rd edn (Oxford, OUP, 2011) 51; C Mitchell, P Mitchell, and S Watterson (eds), Goff and Jones: The Law of Unjust Enrichment, 8th edn (London, Sweet & Maxwell, 2011) paras 4.06–4.07 and 5.05–5.10. But for a different view see A Lodder, Enrichment in the Law of Unjust Enrichment and Restitution (Oxford, Hart Publishing, 2012) 76–77. 37 FII (CA) (n 33). This aspect of the CA’s decision was not considered on appeal: FII (SC) (n 11). 38 ibid [179]. 39 ITC (n 30). 35 36

Introduction 9

Tax Act 1994 never gave the managers a legal right to an input tax credit. Section 26 provides that such a credit can only be claimed for ‘input tax . . . attributable to . . . taxable supplies’, and this condition was not satisfied as the managers had only provided the claimants with exempt supplies.

(2) At the Claimant’s Expense The ITC case also contains an interesting discussion of the requirement that a defendant’s enrichment must have been gained at the claimant’s expense. This requirement is barely mentioned in most overpaid tax cases, as these concern direct payments by a taxpayer to HMRC, and it is uncontroversial that HMRC’s enrichment was received from the taxpayer. In the ITC case, however, the claimants had to show that the value of the money which they had paid the managers relevantly corresponded to the value of the benefits received from the managers by HMRC. That is, the claimants had to establish that HMRC were enriched at their expense although HMRC were only the ‘remote’ rather than the ‘immediate’ recipients of benefits emanating from the claimants. As Mitchell discusses, it is a controversial topic whether, and if so, when, such claims are possible under the law of unjust enrichment. Henderson J considered that as a general rule they are not permitted, but that this rule is subject to exceptions, and that the circumstances of the ITC case were sufficiently exceptional for him to hold that HMRC were enriched at the claimants’ expense. His reasons were specific to the scheme of the Value Added Tax Act 1994, under which ‘in economic terms the person at whose expense unlawful VAT is paid to HMRC is indubitably the consumer’.40 His conclusion is consistent with cases where the CJEU has held that a claim to recover money paid as indirect tax can sometimes be brought by the customer who was the ultimate source of the money.41 It is also consistent with earlier English cases where businesses were ordered to account for the fruits of their actions to recover overpaid VAT to the customers who had borne the cost of paying for this.42 However, there are other cases concerning different situations in which a negative answer has been given to the same general question, and this whole area of the law is in need of restatement and rationalisation by an appellate court.43

(3) Unjust Factors An ‘unjust factor’ is a reason established by the cases why a defendant’s enrichment at a claimant’s expense should be reversed.44 At least three unjust factors could be in play in ibid [81]. Case C-35/05 Reemstma Cigarettenfabriken GmbH v Ministero delle Finanze [2007] ECR I-2425 and Case C-94/10 Danfoss A/S and Sauer-Danfoss ApS v Skatteministeriet, CJEU, 20 October 2011. 42 Grantham Cricket Club v Customs & Excise Commissioners [1998] BVC 2272; R v Customs & Excise Commissioners, ex parte Building Societies Ombudsman Co Ltd [2000] STC 892. The significance of these cases is well explained in M Chowdry, ‘Unjust Enrichment and Section 80(3) of the Value Added Tax Act 1994’ [2004] British Tax Review 620, 629. 43 For the case law to date see Goff & Jones (n 36) paras 6.12–6.62, and also Menelaou v Bank of Cyprus plc [2012] EWHC 1991 (Ch) [22] and Relfo Ltd (in liq) v Varsani [2012] EWHC 2168 (Ch) [86]–[87]. 44 For judicial use of ‘unjust factor’ terminology see eg Banque Financière de la Cité v Parc (Battersea) Ltd [1999] 1 AC 221, 227; Kleinwort Benson (n 8) 363, 386, 395 and 409; FII (SC) (n 11) [81]. 40 41

10 Steven Elliott, Birke Häcker and Charles Mitchell cases where a claimant pays money to HMRC as tax that is not due: mistake, duress colore officii, and the Woolwich principle. These are discussed by several contributors: Rebecca Williams, Duncan Sheehan, Nelson Enonchong, and Charlie Webb. They examine the definition and scope of each unjust factor, and the interplay between them, asking whether the law should operate a hierarchy of claims, ie whether a claimant should be required to bring an action founded on Woolwich where the facts would support such an action, even though another unjust factor is also present on the facts, and although the claimant would prefer to rely on this other ground. A special situation is also considered by Monica Bhandari in chapter eight, which concerns a line of cases in which trustees have invoked the rule in Re Hastings-Bass (deceased) 45 to undo the tax consequences of transactions they have entered, on the ground that they failed to take these consequences into account. Bhandari notes that the scope for trustees to invoke this rule has been curtailed by the Court of Appeal’s decision in Pitt v Holt,46 which confines its operation to cases where the trustees’ failure to take all relevant considerations into account constitutes a breach of duty. However, she argues that the rule still places trustees (and by extension, their beneficiaries) in too favourable a position by comparison with other taxpayers who cannot recover money paid as tax on the ground that they would have arranged their affairs in a more tax-efficient way if they had known that they could do this.47 Note that a trustee who seeks to recover money paid as tax by invoking the HastingsBass principle might rely on mistake as the ground of recovery, but he might also rely on his own want of authority and/or the beneficiaries’ lack of consent to the transaction.48 There are obvious reasons why a trustee might not feel much enthusiasm for bringing a claim based on his own breach of duty,49 but he has a power and a duty to do so.50 Mistake is discussed in chapter three by Duncan Sheehan, who notes that mistakes are easy to prove on the facts of many overpaid tax cases, and who believes that ‘retrospective mistakes’ of the kind recognised in Kleinwort Benson51 are ‘real’ mistakes, notwithstanding Lord Hoffmann’s later concession in DMG that they are ‘deemed’ mistakes that ground recovery because of ‘practical considerations of fairness’.52 Given the court’s decision in the latter case, that the claimant could take advantage of the limitation rule established by the Limitation Act 1980, section 32(1)(c), it seems that Lord Hoffmann thought that this rule could be invoked even by claimants relying on a ‘deemed mistake’. This liberal interpretation of the subsection sits uneasily alongside the Supreme Court’s restrictive finding in the FII case that the subsection may only be invoked by claimants who rely on mistake as an element of their cause of action.53

Re Hastings-Bass (deceased) [1975] Ch 25 (CA). Pitt v Holt [2011] EWCA Civ 197, [2012] Ch 132 (hereafter ‘Pitt’). 47 cf Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2008] EWHC 2893 (Ch), [2009] STC 254 (hereafter, ‘FII (Ch)’) [257], where Henderson J accepted counsel’s supposition that HMRC owes no liability to repay money paid as inheritance tax following the taxpayer’s mistaken failure to make an election in a deed of variation pursuant to the Inheritance Tax Act 1984, s 142. 48 Goff & Jones (n 36) paras 8.94–8.101. 49 cf Pitt (n 46) [130]. 50 Goff & Jones (n 36) paras 8.109–8.119. 51 Kleinwort Benson (n 8). 52 DMG (n 12) [23]. For the view that retrospective mistakes are not real mistakes see eg A Nair, ‘“Mistakes of Law” and Legal Reasoning: Interpreting Kleinwort Benson v Lincoln City Council’ in R Chambers, C Mitchell, and J Penner (eds), Philosophical Foundations of the Law of Unjust Enrichment (Oxford, OUP, 2009) 373. 53 FII (SC) (n 11) [42]–[63] and [177]–[185]. 45 46

Introduction 11

In chapter four, Nelson Enonchong makes the point that the courts have not clearly defined the scope of the Woolwich rule; that this makes it uncertain when claims can be brought in reliance on the rule; and that it also makes it uncertain when claims cannot be brought in reliance on the rule, so that claimants must rely on some other unjust factor. Thus he argues that duress colore officii may still have some practical relevance as an unjust factor because the Woolwich rule probably does not apply to every type of payment received by a public authority acting beyond its powers, and cases to which it does not apply might be pleaded as duress cases. Enonchong quotes from Lord Walker’s speech in Deutsche Morgan Grenfell where his Lordship considered the different types of payment that might conceivably fall within the scope of the Woolwich principle, and emphasised the difficulty of drawing a clear dividing line within a spectrum which stretches from central government taxes and duties through rates, community charge, drainage rates and charges, special levies and licence fees imposed by statute on different industrial and commercial activities, and charges made by statutory undertakers.54

In chapter five, Charlie Webb also stresses how little work the courts have done to define the Woolwich principle, observing that their Lordships’ speeches in the case neither specify the circumstances in which recovery should be allowed nor spell out the principle of justice that actuated their decision. Webb notes textbook treatments of the case which identify the constitutional principle of ‘no taxation without Parliament’ as the reason for recovery. However he prefers the different, claimant-focused, explanation that the claimant building society did not truly consent to the Revenue’s enrichment in the case, albeit that it made no mistake and was not subject to illegitimate pressure, because its intention to benefit the Revenue was conditional upon the money being due. If this is correct, then Webb’s analysis suggests that ‘vitiation of a claimant’s intention to benefit the defendant’ may be a larger category of unjust factors than has been generally appreciated. However, his claimantfocused analysis puts him at odds with many other legal scholars who consider that the rule is defendant-focused, in the sense that its point is to control the behaviour of public authority defendants because this lies in the public interest. One writer who espouses this view is Rebecca Williams, who sets out her conception of the Woolwich rule in chapter two.55 She argues that the reason for restitution in the Woolwich cases is that an ultra vires public law event has taken place, and that if this were fully recognised then the law would become simpler and more certain because public lawyers have a clear and well-defined idea of what the ultra vires doctrine entails. Williams also argues that in cases where a Woolwich-based hybrid public–private law action would lie, claimants should be confined to bringing such an action, and should not be allowed to bring a purely private law action based on mistake or duress, even if this would also be available on the facts. She considers that restricting claimants to hybrid claims is in the public interest because it will enable greater understanding of public law issues in play in such cases and reduce the risk that these will be overlooked. In contrast, Sheehan and Enonchong believe that concurrency of actions should be the norm, and Sheehan objects to Williams’ proposed hierarchy of claims that it would deprive 54 DMG (n 12) [140]. See too Lord Walker’s comment in FII (SC) (n 11) [80]: ‘there are bound to be borderline cases’. 55 See also her monograph on the subject: R Williams, Unjust Enrichment and Public Law (Oxford, Hart Publishing, 2010).

12 Steven Elliott, Birke Häcker and Charles Mitchell claimants of causes of action that fulfil their needs, and would force them to engage in more complex and more expensive enquiries; in Williams’ rendering, it would also expose them to much tighter time limits. Williams would not deny that her rule might have such effects, but considers this justifiable in light of the special nature of public body defendants; she also stresses that it is a commonplace of public law, exemplified by O’Reilly v Mackman 56 and the Clark case,57 that litigants may be treated less generously when they sue the government than they would when suing other private individuals. Whichever of these positions is preferable in principle, the trend of recent cases is against Williams, since the House of Lords refused to limit the claimants to a Woolwich claim in DMG, and the Supreme Court assumed this finding to be correct in FII. In the latter case Lord Sumption acknowledged that there were reasons for doubting DMG, but declined to reopen the debate.58 Identifying the content and scope of the Woolwich rule remains difficult although the Supreme Court addressed their minds to this in FII. One question before the court was whether it forms part of the rule that the claimant’s payment must have been made pursuant to an official demand, as the Court of Appeal had previously held in NEC SemiConductors Ltd v IRC.59 In their leading judgments, Lord Walker and Lord Sumption held that it does not, but that it must somehow have been communicated to the taxpayer that a payment was required in circumstances where the tax authority had no power to require the payment.60 They both considered that for these purposes it suffices for a statute to exist which apparently requires the payment, although in fact the payment is not due.61 According to Lord Sumption, however, it does not suffice for a taxpayer to pay money in the mistaken belief that a statute requires him to do so because he has miscalculated his tax liability or has forgotten that he has already paid it.62 In such a case, though, the taxpayer only pays because he believes that he is required to do so by the legislation. So why should this situation fall outside the scope of the Woolwich rule as stated in FII? Presumably Lord Sumption conceives the point of the rule to be that public authorities should be deterred from acting beyond their powers, and thinks that this consideration does not arise in cases where a public authority defendant has taken no positive step to mislead the taxpayer. Against this, Williams argues that public authorities lack the power to receive payments that are not legally due, and that such payments should therefore be recoverable under the Woolwich principle whether or not the defendant authority has taken positive steps to induce the claimant’s belief that the payment is due. On this understanding of the principle, it looks identical to a rule that money paid as tax can be recovered simply on the ground that it is not due, and it is possible that Lord Walker had this in mind when he stated in FII that eliminating the ‘demand’ requirement would not be a decisive step towards a general ‘absence of basis’ principle in place of the ‘unjust factors’ approach that has prevailed in the past. It would merely be creating . . . a rather larger island of recovery in respect of undue tax.63 O’Reilly v Mackman [1983] 2 AC 237 (HL). Clark v University of Lincolnshire and Humberside [2000] 1 WLR 1988 (CA). 58 FII (SC) (n 11) [168]. 59 NEC Semi-Conductors Ltd v Inland Revenue Commissioners [2006] EWCA Civ 25, [2006] STC 606; not considered on appeal: [2007] UKHL 25, [2007] 1 WLR 1386. The case is referred to by this name in the FII judgments, but it is reported as Boake Allen Ltd v HM Revenue & Customs. 60 FII (SC) (n 11) [64]–[81] and [171]–[174]. 61 ibid [79] and [174]. 62 ibid [186]. 63 ibid [81]. 56 57

Introduction 13

Lord Walker’s use of the word ‘general’ suggests that he may have meant to say that an ‘absence of basis’ principle does now govern the recovery of undue tax payments. However, Lord Sumption does not seem to have understood Lord Walker in this way, or if he did, then he disagreed with him, since he said that the FII decision has not changed the law ‘by allowing an action for the recovery of payments on the simple ground that they were not due’.64

(4) Defences Various defences have been raised in response to claims for overpaid tax, limitation defences prominent among them, but also including the defence of change of position, which is discussed by Niamh Cleary in chapter seven.65 She reviews several first instance cases in which this defence has been allowed,66 and argues that in each case the court adopted an unrealistic and overly generous approach when accepting the proposition that the government commits to public expenditure because it anticipates particular levels of tax income. Cleary also argues that proponents of the view that public authority defendants should be permitted to rely on change of position, and variants such as ‘fiscal chaos’ and ‘exhaustion of benefits’, point to policy concerns that favour the state’s retention of overpaid tax but marginalise, or completely ignore, policy concerns that favour restitution. She contends that this is contrary to principle, and also contrary to Article 1 of the First Protocol to the European Convention on Human Rights, in accordance with which the state should be required to prove that denying restitution to the taxpayer by appealing to such defences is a justified and proportionate interference with the taxpayer’s right to property.

C. EUROPEAN LAW

Given that recent developments in England have – perhaps more than in any other legal system – been prompted by the case law of the CJEU, Section III of the book looks at the specific impact of European principles on the interpretation and application of the English law governing restitution of overpaid taxes. There are various levels at which European law can become relevant. First, individual Treaty provisions, Regulations and even untransposed Directives may be ‘directly effective’ so as to take automatic precedence over inconsistent national rules and thus help enforce citizens’ rights against the offending Member State. Secondly, the concept of ‘sympathetic construction’ requires national courts to interpret domestic law as far as possible in a way which makes it compatible with EU requirements. Thirdly, where it is not possible either to read national law in an EU-friendly ibid [162]. For additional discussion of defences in this area see E Bant, ‘Change of Position as a Defence to Restitution of Unlawfully Exacted Tax’ [2012] Lloyd’s Maritime and Commercial Law Quarterly 122; J Lee, ‘Defences to Claims for Restitution of Overpaid Tax’ [2012] Restitution Law Review 13. 66 FII (Ch) (n 47) (not considered on appeals to the CA and SC); Bloomsbury International Ltd v Sea Fish Industry Authority [2009] EWHC 1721 (QB), [2010] 1 CMLR 12; Littlewoods Retail Ltd v HM Revenue & Customs [2010] EWHC 1071 (Ch), [2010] STC 2072. 64 65

14 Steven Elliott, Birke Häcker and Charles Mitchell way or to disapply it altogether, the Member State concerned will often be held liable to make up for the resulting breach by providing a monetary remedy to the party affected by it. In the tax context, English courts are frequently faced with the task of determining whether one statutory provision or another is – or can be sympathetically construed to be – compliant with EU law. While the preliminary reference procedure enables them to obtain some guidance on the exact scope and meaning of the latter (a hope not always fulfilled in the past), the CJEU cannot rule on the substantive content and interpretation of domestic law as such. Against this background, the approach adopted by English courts towards sympathetic construction specifically in relation to statutory remedies for overpaid taxes is critically analysed by Catherine Barnard and Julian Ghosh in chapter nine. They argue that English courts have gone further in stretching the meaning of domestic legislation in order to make it EU-compliant than required by the CJEU ever since the Marleasing case67 or warranted by previous House of Lords authority. They illustrate the dangers of such an ‘over-expansive’ interpretative approach by reference, inter alia, to the Court of Appeal decision in the FII case,68 which in their view disregarded the limitation that statutes cannot be construed contra legem, offended the doctrine of ‘no reverse direct effect’ and also impinged on the principle of legal certainty. They therefore welcome the reassertion by the Supreme Court in FII 69 of a more orthodox and restrictive interpretative approach. Barnard and Ghosh also draw attention to the ‘potentially seismic’ implications of the CJEU’s recent jurisprudence on age discrimination,70 which – if translated into the tax context – could effectively impose on national courts a ‘general and absolute duty to confer a remedy [for the breach of EU law] in all circumstances’ by simply disapplying any domestic provision standing in the way of this goal. Yet, as EU law still stands at present, the duty to provide a remedy is neither absolute nor general. Where national rules, however sympathetically construed, prevent or impede citizens exercising their EU rights and freedoms, specific enforcement may yield to financial redress at a secondary level. The most wellknown form of such redress is constituted by a damages claim under the so-called Francovich principle.71 In practice, it is limited by the requirement that the breach of EU law committed by the defendant Member State must have been ‘sufficiently serious’. Less widely appreciated despite its venerable age (and indeed surprisingly often ignored by the general literature on EU law), but nevertheless of fundamental importance in the tax context, is a claim under the principle usually ascribed in England to the San Giorgio case72 demanding that charges levied in contravention of (what is now) EU law have to be refunded by the offending Member State. 67 Case C-106/89 Marleasing SA v La Comercial Internacional de Alimentación SA [1991] ECR I-4135 esp [8]: national courts are under a duty to interpret all national law (not merely implementing legislation) as far as possible ‘in the light of the wording and the purpose of [EU law] in order to achieve the result pursued by the latter’, such that even pre-existing domestic provisions must be construed by reference to later untransposed Directives (so-called ‘indirect effect’). 68 FII (CA) (n 33). 69 FII (SC) (n 11). 70 Especially Cases C-144/04 Mangold v Helm [2005] ECR I-9981 and C-555/07 Kücükdeveci v Swedex GmbH & Co KG [2010] ECR I-365. 71 Named after Joined Cases C-6/90 and C-9/90 Francovich and Bonifaci v Republic of Italy [1991] ECR I-5357 (hereafter ‘Francovich’). 72 San Giorgio (n 6), but note our comments in n 26.

Introduction 15

The incidents of San Giorgio-type claims are not, however, wholly determined by EU law. Pursuant to the principle of ‘national procedural autonomy’ the CJEU leaves it to the domestic legal system of each Member State to lay down the exact rules of recovery, and this includes the question of whether the claim is best characterised as restitutionary or compensatory.73 To ensure that the economic burden imposed on citizens by taxes or other charges levied in breach of EU law is sufficiently ‘neutralised’, the CJEU circumscribes national procedural autonomy by the twin principles of equivalence and effectiveness. The former maintains that the conditions under which claims may be brought for the recovery of offending charges must not be less favourable than those concerning similar domestic actions, while the latter stipulates that national law must not make the enforcement of an individual’s EU rights impossible or excessively difficult. As far as the principle of equivalence is concerned, the CJEU has so far been fairly generous with Member States. By contrast with its extremely strict jurisprudence on the recovery of state aid found to be incompatible with the Common Market (in a sense the reverse of the San Giorgio scenario),74 the Court does not invariably require national law to make its most favourable rules of recovery available in the context of taxes and charges violating EU law.75 A minimum standard is in any event secured by the principle of effectiveness. The impact of this principle on recent English cases is explored and assessed by Maximilian Schlote in chapter ten. He illustrates the interaction between the EU principle of effectiveness and the domestic English law of unjust enrichment with a particular view to defences that may be raised against San Giorgio-type claims, considers in detail the legal position of indirect taxpayers such as those in the ITC case,76 subjects the most recent statements of the CJEU77 on the availability of interest and the recovery of other sums ‘directly related to the overpaid tax’ to a critical analysis, and examines the question of concurrency of remedies raised by the Supreme Court’s FII decision. Schlote also puts into focus the sometimes tricky relationship between the principle of effectiveness and national limitation periods, especially where such limitation periods are specifically introduced to discourage potential claimants from pursuing their rights. This famously happened when, in the course of and following the DMG litigation,78 the UK government sought to counteract taxpayers relying on their mistake-based claim (as opposed to the Woolwich principle)79 in order to take advantage of the extended limitation period provided by the Limitation Act 1980, section 32(1)(c). Given that the principle of effectiveness itself competes with other principles of EU law, including that of legal certainty, the curtailment of limitation periods is not precluded wholesale. However, as the CJEU had already pointed out in its Marks & Spencer decision (another case from the UK),

73 The CJEU has so far not squarely faced up to the tensions arising from the fact that a Francovich claim requires there to have been a ‘sufficiently serious’ breach of EU law (see text to and following n 71), while a compensatory remedy under San Giorgio (n 6) would apparently not be subject to such a limitation. 74 See generally A Jones, Restitution and European Community Law (London, Mansfield Press/LLP, 2000) ch 5; T Köster, ‘Recovery of Unlawful State Aid’ in M Sánchez Rydelski (ed), The EC State Aid Regime: Distortive Effects of State Aid on Competition and Trade (London, Cameron May, 2006) 653; M Angeli, ‘The European Commission’s “New Policy” on State Aid Control: Some Reflections on Public and Private Enforcement of Recovery of Illegal Aid’ (2009) 30 European Competition Law Review 533. 75 cf Littlewoods (CJEU) (n 31) [31]. 76 ITC (n 30). 77 In Littlewoods (CJEU) (n 31). 78 DMG (n 12). 79 Woolwich (n 4).

16 Steven Elliott, Birke Häcker and Charles Mitchell the new limitation period must be reasonable and be flanked by adequate transitional arrangements so as not to deprive individuals of their right to repayment retroactively.80 It is remarkable that the UK legislature has repeatedly fallen foul of these criteria. Following the Marks & Spencer ruling, which concerned a time-cap on the recovery of output tax (without any transitional period), it took a House of Lords decision to establish that the input tax regime suffered from comparable defects.81 Most recently, the Supreme Court in FII unanimously held that the Finance Act 2007, section 107, contravened EU law, and by a majority of 5:2 found that the same was true of the Finance Act 2004, section 320, on the ground that claimants had to be given a free choice between relying on the Woolwich principle and bringing a mistake-based claim.82 In the light of the division of opinion, however, the Supreme Court considered that the latter was not acte clair and has thus referred it to the CJEU for a preliminary ruling. It has already been noted that the CJEU’s guidance to domestic courts on the appropriate remedy in tax cases has in the past not always been as helpful as it might have been, partly perhaps on account of the ambivalence between a loss- and gain-centred interpretation of San Giorgio-type claims. The hope expressed by Schlote that the CJEU will use the opportunity provided by FII to enunciate more clearly the principles governing the recovery of charges levied in breach of EU law is therefore to be endorsed.

D. COMPARATIVE LAW

The fourth and final section of the book adopts a comparative perspective and looks at how different jurisdictions deal with some of the issues raised by the English case law governing restitution of overpaid tax. Two chapters explore these questions from a Civilian angle, using German law as a point of contrast and comparison,83 while the remaining three consider the background and proper classification of Woolwich-type claims in the Republic of Ireland, Canada and Australia respectively. As far as German law is concerned, it is important to realise that many of the problems with which English courts have been grappling over the past decade or so could not have arisen in the same way in Germany. This is because German tax law (like German administrative law generally) embraces the doctrine of the so-called ‘administrative act’84 by virtue of which tax assessments – if unchallenged by the taxpayer – become final and binding very quickly. As a result, the tax payable in accordance with it will be treated as owing even if it subsequently emerges that the assessment was false or based on a legal provision which contravened EU law. Now and then, interestingly, arguments are advanced to challenge the orthodox legal position. It is claimed that national authorities are under an obligation to disregard or disapply tax assessments which offend EU law or – relying in part on the 80 See Marks & Spencer (n 29); to be distinguished from the other well-known Marks & Spencer case: Case C-446/03 Marks & Spencer plc v Halsey (HM Inspector of Taxes) [2006] STC 237, [36] ff, esp [38]. 81 Fleming (n 29). In consequence, the offending domestic provision (reg 29(1A) of the Value Added Tax Regulations 1995) was disapplied to the extent that this was necessary to enable taxpayers to seek recovery on the basis of accrued rights within an appropriate period. cf now Finance Act 2008, s 121. 82 FII (SC) (n 11), Lords Sumption and Brown dissenting. 83 For comparative analysis of the French approach see Williams (n 55) ch 6. 84 This doctrine is explained in chapter 12 of this volume (at pp 248–51).

Introduction 17

CJEU’s Kühne & Heitz ruling85 – at least to reopen the case and exercise a discretion to revoke the administrative act in question.86 So far, such arguments have largely fallen on deaf ears.87 This is because most contraventions of EU law are not serious enough to warrant treating a tax assessment as void ab initio 88 and because the CJEU itself recognises that the principle of effectiveness may be counter-balanced by the need for finality and legal certainty which is itself a value enshrined in EU law. Moreover, a provision has been introduced into the General Tax Code allowing for tax assessments to be made explicitly provisional where legal proceedings concerning the compatibility of a German tax statute with (inter alia) European law are pending at the time when the assessment is issued.89 In light of this, the present volume does not specifically consider the German law relating to restitution of overpaid tax.90 Instead, it has been thought more conducive to the comparative discussion of Woolwich-type cases to draw on the German experience more broadly defined with respect to two issues. First, there is the question raised by Woolwich, partially answered by DMG and still exercising their Lordships in the FII litigation,91 of whether English law would be well-advised to switch from the orthodox ‘unjust factor’ analysis of unjust enrichment to an ‘absence of basis’ approach similar to that operated by German law. And secondly, instructive lessons are to be learnt from exposing the parallels and divergences between the English post-Woolwich debate and corresponding developments pertaining to so-called ‘public law restitutionary claims’ in Germany. Accordingly, Section IV of the book starts with Anne Sanders considering in chapter eleven ‘absence of basis’ as a general concept of (German) enrichment law and critically assessing its suitability for transplantation into the English common law. She does not deny that cases such as Woolwich and FII could be decided under an absence of basis approach, but cautions that English law may thereby import a whole host of problems not evident at first sight. In order to corroborate her assessment, Sanders explains the historical background of the ‘absence of basis approach’ and how its understanding evolved under the German Civil Code (BGB); she closely scrutinises the notion of a ‘legal basis’ and its invalidity or voidability (pointing out that English law would struggle to rein in restitution even if it could come up with a satisfactory concept of ‘basis’); she outlines some German defences to restitutionary claims which look very much like harbouring unjust factors in disguise; and warns about the potential practical difficulties of adequately distributing the burden of proof and operating limitation periods. All in all, her conclusions suggest that

85 Case C-453/00 Kühne & Heitz NV v Produktschap voor Pluimvee en Eieren [2004] ECR I-837 (hereafter ‘Kühne & Heitz’). 86 But note that, by contrast to other areas of German administrative law (especially §§ 48 ff VwVfG), the General Tax Code (Abgabenordnung, abbreviated AO) in its current form does not give tax authorities much leeway in changing assessments once the relevant period for challenging them has expired: cf §§ 172 ff AO. 87 See only the recent spate of cases decided by the Federal Fiscal Court (BFH) on 16 September 2010, case refs V R 46/09, 48/09, 49/09, 51/09, 52/09, and 57/09, on which the Federal Constitutional Court (BVerfG) has now indicated that a preliminary reference to the CJEU was not warranted despite some uncertainty about the precise ramifications of the Kühne & Heitz ruling (n 85): case refs 1 BvR 1234/11, 1390/11, 1395/11, and 1403/11. 88 For the same reason, a compensatory claim under the Francovich principle (n 71) – to the extent that one regards such a claim as theoretically conceivable in the context of tax recovery – would typically fail for lack of a ‘sufficiently serious’ breach of EU law. 89 § 165 I 2 no 4 of the Abgabenordnung (AO). 90 The interested reader wishing to follow up on the substantive rules and their procedural backdrop may be referred to the monograph by U Lange, Der Anspruch auf Erstattung gemeinschaftsrechtswidrig erhobener Steuern (Berlin, Duncker & Humblot, 2008). 91 FII (SC) (n 11) [81] (Lord Walker) and [187]–[190] (Lord Sumption).

18 Steven Elliott, Birke Häcker and Charles Mitchell the rather frosty reception with which the absence of basis argument was met in FII is to be welcomed. Chapter twelve turns from the general German law of unjustified enrichment to the public law context. In it, Birke Häcker investigates similarities and differences between Woolwich-type claims in England and the German concept of specific ‘public law restitutionary claims’ as developed since the beginning of the twentieth century. She highlights striking parallels in the reasoning adopted by judges deciding the seminal cases and argues that although German law sees public law restitutionary claims as having been fully emancipated from their private law counterparts, the overall impression is nevertheless not far removed from the ‘hybrid approach’ advocated by Rebecca Williams in her monograph and in chapter two of the present volume. With regard to the incidents of public law restitutionary claims, Häcker demonstrates that German courts still accord special privileges to the fisc as far as disgorgement of the use value of money is concerned, which would be unthinkable under English law following the Sempra Metals decision.92 On the other hand, however, she finds the German categorical denial of the change of position defence in respect of state defendants preferable to the less clear-cut position under current English law. The picture conveyed by this book would not be complete without a side-glance at other Common law jurisdictions. Although their constitutional setup often differs greatly from that of the UK (not least because most countries with written constitutions have courts that can declare offending statutes null and void), the close connections these legal systems have with English law on the private law side make it particularly interesting to see how they cope with Woolwich-type cases and – to the extent that they allow recovery – how they classify the appropriate claims. Of the three jurisdictions considered in this collection, two (Ireland and Canada) have already openly grappled with the issue of granting restitution for unlawfully levied taxes, though only one (Canada) has tackled the classificatory problem head-on. The third jurisdiction (Australia) has yet formally to accommodate the idea that taxes exacted ultra vires have to be repaid as a matter of right. Since the Republic of Ireland – unlike Australia or Canada – is a Member State of the EU, the potential repercussions of the CJEU’s jurisprudence on the San Giorgio line of cases and its reception in England are likely to be felt more strongly than anywhere else in the Common law world. This is amplified by the fact that Irish courts have expressly endorsed the Woolwich principle as being a part of Irish law.93 At the same time, however, Ireland has its own leading authority on the recovery of unlawfully exacted taxes in the shape of a pre-Woolwich decision called Murphy v Attorney General.94 Against this background, Niamh Connolly in chapter thirteen examines how the Murphy/Woolwich right to restitution is best characterised as a matter of Irish domestic law. Despite a widespread tendency to subsume Murphy under an expansive definition of duress colore officii, she argues that the case can actually be understood as supporting a constitutional principle, based on the principle of legality, to the effect that prima facie: ‘State actions which are not mandated by law should . . . be undone and a remedy granted to persons adversely affected’. In her view, such a public law understanding also helps explain why, on the facts of Murphy, redress could (exceptionally) be denied on the ground that allowing too many claimants to recover would lead to fiscal chaos. In Murphy itself, the defence was described as being all about ‘change of position’, but – as Connolly points out – it would seem that it had less to Sempra (n 13). See O’Rourke v Revenue Commissioners [1996] 2 IR 1; Harris v Quigley [2006] 1 IR 165. 94 Murphy v Attorney General [1982] 1 IR 241 (hereafter ‘Murphy’). 92 93

Introduction 19

do with the state’s disenrichment than with broader public policy concerns. Whatever its domestic pedigree, it seems unlikely that the Murphy defence could ever succeed in a case where taxes levied in breach of EU law are at stake.95 In Canada, the ‘constitutional approach’ advocated by Connolly has already won the day. Very shortly after the House of Lords ruling in DMG, the Supreme Court of Canada handed down its judgment in the Kingstreet case,96 maintaining that the right to recover payments exacted ultra vires was a ‘public law remedy’ grounded in the rule of law and was thus distinct from the ordinary private law of unjust enrichment.97 Robert Chambers’ contribution in chapter fourteen discusses the pros and cons of the Kingstreet decision to segregate overpaid tax cases in this way and explains that although the Canadian rejection of unjust enrichment and its replacement with a form of public law restitution of ultra vires taxes appear to open a vast gulf between the English and Canadian law on the subject, the gap is much less significant than it seems.

He suggests that the difference between the law of unjust enrichment in England and Canada is not merely the latter system’s acceptance – in Garland 98 – of an ‘absence of juristic reason’ approach, but that the Canadian concept of unjust enrichment is strongly influenced by its case law concerning the proper division of the family home after the breakup of long-term relationships (cases which English law does not treat as part of the law of unjust enrichment at all). As a result, he argues, the Canadian law of unjust enrichment now finds it difficult to accommodate overpaid tax scenarios and, remarkably, even plain mistaken payment cases. Chambers sees the solution to this dilemma as lying in the adoption of the ‘hybrid approach’ put forward by Williams or – alternatively, but closer to Garland – in openly recognising that public law considerations may show that a tax was imposed without sufficient justification, thereby invalidating the juristic reason for the state’s enrichment. In chapter fifteen, Simone Degeling describes the Australian position with regard to unlawfully exacted taxes. Despite a number of sympathetic dicta, she notes that the Woolwich principle has not as yet been formally received into Australian law. The aim of her contribution is not primarily to discuss whether it should be so received, but to investigate what possible routes there are to allowing recovery. Besides the ‘convoluted web of indirect protections’ offered to affected taxpayers (such as specific statutory provisions or simple common law claims based on duress, the doctrine of colore officii, mistake, or ‘failure of basis’), Degeling makes out two main paths which would allow for the development of a Woolwich-style claim, and she explains them both by reference to and inferences from existing Australian authorities. The first would be to acknowledge the existence of common law claims that are ‘parasitic on some constitutional invalidity’. As she explains, the opportunity to establish such claims arose – but could ultimately not be seized – in the British American Tobacco case,99 a sequel to the Roxborough litigation.100 The second vehicle for recovery could lie in recognising a direct constitutional right to restitution (similar to See ch 13 of this volume text to fn 27. Kingstreet Investment Ltd v New Brunswick 2007 SCC 1, [2007] 1 SCR 3 (hereafter ‘Kingstreet’). Perhaps due to the short interval between the two decisions, the Supreme Court of Canada in Kingstreet did not refer to DMG (n 12). 97 ibid [40] (Bastarache J). 98 Garland v Consumers’ Gas Co 2004 SCC 25, [2004] 1 SCR 629. 99 British American Tobacco Australia Ltd v State of Western Australia (2003) 217 CLR 30. 100 Roxborough v Rothmans of Pall Mall Australia Ltd (2001) 208 CLR 516. 95 96

20 Steven Elliott, Birke Häcker and Charles Mitchell that established by the Supreme Court of Canada in Kingstreet), albeit that this is contrary to the orthodox understanding that private causes of action cannot be founded straight on the Australian Constitution and would therefore be potentially much more controversial than developing the common law via the first route. Yet, she concludes, if it is an important constitutional value that no tax be levied without proper parliamentary authorisation, then ‘[t]he corollary of this principle must surely be that such amounts are recoverable’.

2 Overpaid Taxes: A Hybrid Public and Private Approach REBECCA WILLIAMS*

A. INTRODUCTION

The recent explosion of litigation concerning unjust enrichment and public authorities, in particular Her Majesty’s Revenue and Customs (HMRC), has presented important challenges to both public and private law. The House of Lords has analysed these cases purely in terms of private law1 while the Supreme Court of Canada has used a public law framework2. Neither of these solutions is wholly successful,3 though it is no coincidence that such a schism of approach should exist. My position is that if we are to understand and decide these cases properly we must do so from a hybrid perspective which takes account of both these key aspects of the claims: public and private. Doing so will mean that we can more straightforwardly answer the questions raised in litigation, in a manner which leaves the law in a clearer, and thus more certain state, but perhaps more importantly it might also help to prevent further complex and costly litigation. The hybrid causes of action I propose would operate as follows: the existence of an enrichment, questions of whether it was at the claimant’s expense and many of the issues concerning defences should continue to be decided by the private law of unjust enrichment as usual, but the fourth requirement, the reason for restitution, should be understood as being a public law event.4

* Fellow and Tutor in Law, Pembroke College, Oxford and the Faculty of Law, University of Oxford. I am grateful to Eoin O’Dell, Steven Elliott, Richard Vallatt, Niamh Cleary and the other participants in the conference for the opportunity to take part in it and for their comments on my work. I am also grateful to John Armour. Of course, any remaining errors are mine alone. 1 Deutsche Morgan Grenfell Group plc v Inland Revenue Commissioners [2006] UKHL 49 (hereafter ‘DMG ’). 2 Kingstreet Investments Ltd v New Brunswick (Finance) 2007 SCC 1, [2007] 1 SCR 3. See also J Alder, ‘Restitution in Public Law: Bearing the Cost of Unlawful State Action’ (2002) 22 Legal Studies 165. 3 See further, R Williams, Unjust Enrichment and Public Law (Oxford, Hart Publishing, 2010). 4 ibid 36–39.

24 Rebecca Williams

B. A HYBRID CAUSE OF ACTION

(1) The Correct Interpretation of Woolwich Equitable Building Society v IRC Beginning in the obvious place with Woolwich,5 it is clear that their Lordships did not regard any of the existing private law reasons for restitution as sufficient to found recovery in that case,6 but there were several which were, as Lord Browne-Wilkinson put it, ‘in play’. Thus he and Lord Slynn referred to concepts of ‘implied compulsion’ and colore officii, though of course as Lords Keith, Goff and Jauncey pointed out, this was not duress in the usual private law sense. Lord Slynn referred to the fact that the payer and the payee in such cases were not on an equal footing, and Lords Goff and Browne-Wilkinson also referred to failure and absence of consideration as also being ‘in play’. To a public lawyer, this is highly reminiscent of the story of the blindfolded people asked to identify an object. Using only their sense of touch one says the object is like tree trunks and another that it is like a snake, whereas from a different perspective, with the blindfolds off, it is obviously an elephant. Similarly with Woolwich, from the public law point of view, it is possible to see how all these different potential grounds of review were ‘in play’.7 One of the key criteria for determining whether a body is ‘public’ or not, as established in the leading case of Datafin,8 is whether the agency is able to impose rules, rather than operating consensually. This explains why their Lordships in Woolwich could see issues of inequality or pressure as being relevant. As for illegality and incapacity – alternative candidates discussed in the subsequent academic literature9 – those two concepts are also at the heart of public law. The very essence of public law is that public bodies have simultaneously greater powers and more limited capacity than private entities.10 The whole of judicial review is devoted to charting the boundaries of those powers.

(2) Absence of Basis The hybrid approach can also offer an alternative perspective on the ‘absence of basis’ debate. As is well known, towards the end of his life Peter Birks suggested that a move to the civilian ‘absence of basis’ approach was necessary in order to accommodate public

5 Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] AC 70 (HL) (hereafter ‘Woolwich’). 6 See further Williams (n 3) 23–28. I refer here to the ‘reason for restitution’ in the more technical sense of a reason recognised by the courts as founding a claim in unjust enrichment (see also ibid 33). As for the second, more normative reason why a court might regard such a reason as justifying restitution (see also Charlie Webb’s contribution in chapter 5 of the present volume), it is equally difficult to regard any private law explanation as sufficient (see further Williams (n 3) 28–30), given that the public law invalidity remains the central issue. This is particularly so where that invalidity arises as a matter of EU law and restitution is simply required by the San Giorgio principle, as Lord Goff himself recognised in Woolwich (n 5) at 177, cf Case 199/82 Amministrazione delle Finanze dello Stato v SpA San Giorgio [1983] ECR 3595. 7 See further Williams (n 3) 36–39. 8 R v Panel on Take-overs and Mergers, ex parte Datafin Plc [1987] QB 815 (CA). 9 See Williams (n 3) at 28–29. 10 See, for example, Laws J in R v Somerset County Council ex parte Fewings [1995] 1 All ER 513 (HL) 524.

Overpaid Taxes: A Hybrid Approach 25

body cases.11 However, consideration of the hybrid approach reveals that the absence of basis alternative is neither desirable, nor even necessary. First, it is evident that the ‘no basis’ approach is not in fact adopted across the board in all civilian systems. Instead, as Anne Sanders demonstrates in her chapter, the German system contains a ‘mixed’ approach,12 and the same is true of France, which originally required evidence of a mistake in all instances of répétition de l’indu (repayment of an undue sum), and still does so in cases where the sum was owed, just not by that particular payer or not to that particular recipient (répétition de l’indu subjectif ).13 Secondly, in practice the difference between the civilian and common law approaches does not appear to be particularly great; certainly not to the extent that it would be worth English law undergoing any upheaval which might be inherent in switching to the ‘absence of basis’ system.14 Thirdly, there is no reason to suppose that the ‘absence of basis’ approach on its own provides a more satisfactory method of addressing the particular concerns of cases involving public authorities; rather it is the recognition of the hybrid public and private elements of such cases which allows the courts to deal with them in a satisfactory manner, without distorting private law concepts or mis-translating public law concepts by viewing them through ‘private law spectacles’.15 However, it is not surprising that courts have discussed the concept of ‘no consideration’ or ‘absence of basis’ in the public context. Boddington v British Transport Police16 establishes that unlawful administrative action is void ab initio, a position confirmed more recently by Secretary of State for the Home Department v JJ.17 In this sense, as Sanders puts it, the absence of basis approach can have some normative force. Without lawful authority to levy the sum in question there simply is no basis on which it can be levied; the decision to levy it is and always was, void. However, it is the fact that this basis is absent as a matter of public law which is crucial, a point supported by the fact that absence of consideration was only one of the factors considered in Woolwich alongside compulsion, inequality, etc, as discussed above. So my first argument is that instead of trying to mis-translate bits of the concept of public law in this way, or trying to see them through ‘private law spectacles’ as Burrows puts it,18 we should recognise that the reason for restitution in Woolwich is simply that a public body, with its simultaneously greater and more limited powers than private actors, has acted in excess of those powers and thus its action was void ab initio. In other words, that there has been a public law ultra vires event. It is important to understand the precise, and thus limited, role of this ‘absence of basis’ argument for two reasons. The first is that, as argued above, it avoids any confusion with the wider and more general private law absence of basis approach which is neither desirable nor necessary in this context. The second is that instead it allows us to focus our attention on what, precisely, was invalid in the particular situation. This is often vital, particularly in contexts involving P Birks, Unjust Enrichment, 2nd edn (Oxford, OUP, 2005) part 3, 108–28. See Anne Sanders’ contribution in chapter 11 of the present volume. 13 See further Williams (n 3) 192–98. 14 See Anne Sanders’ discussion in chapter 11 of the present volume. See also A Burrows, ‘Understanding the Law of Restitution: A Map Through the Thicket’ (1995) 18 University of Queensland Law Journal 149, 160; J Edelman and E Bant, Unjust Enrichment in Australia (Oxford, OUP, 2006) 159–65; Williams (n 3) 198–200. 15 The phrase is borrowed from A Burrows, ‘Public Authorities, Ultra Vires and Restitution’ in A Burrows (ed), Essays on the Law of Restitution (Oxford, Clarendon Press, 1991) 40. See further on the German system Birke Häcker’s contribution in chapter 12 of the present volume and, on the French system, Williams (n 3) 192–200. 16 Boddington v British Transport Police [1999] 2 AC 143 (HL). 17 Secretary of State for the Home Department v JJ [2007] UKHL 45, [2008] 1 AC 385 (HL). See further M Elliott, Beatson, Matthews and Elliott’s Administrative Law: Text and Materials, 4th edn (Oxford, OUP, 2011) ch 3. 18 cf n 15 and accompanying text. 11 12

26 Rebecca Williams the interrelationship of domestic and EU law. In Deutsche Morgan Grenfell (‘DMG’),19 for example, disagreement continues to exist over the scope of what exactly the ECJ held to be a breach of EU law in that case.20 My own view is that what was contrary to Community law (as it then was) was the fact that only domestically-resident companies could benefit from the ACT/MCT system then in force in the UK.21 Thus although the whole ACT/MCT scheme was not held to be contrary to Community law, and although it was the group income election provisions which were principally held to be discriminatory and thus contrary to Community law, the ECJ’s references to ‘benefit’ in para 76 of its judgment must refer not only to the ability to make the election, but also the ability to avoid paying the tax until later as a result of that election, since the ability to make the group income election is of no ‘benefit’ at all to the company without the resulting ability to defer payments of ACT. The tax itself was thus not invalid and therefore not repayable as a principal sum, but the premature charging of it was.22 Recognition of the specifically public law roots of this finding of unlawfulness explains why it may have a broader scope than, for example, the setting aside of a transaction in purely private law, to which Lord Scott compared it when the case reached the House of Lords. When it is recalled that, in some instances, public actions may be rendered ultra vires even though the illegality tainting the action has had no causative effect whatsoever,23 the suggestion that an illegality which has had such a causative effect may have rendered a further action (here the charging of ACT) invalid does not seem particularly strange. Indeed, in Fleming (trading as Bodycraft) and Condé Nast Publications Ltd v HM Revenue and Customs,24 the House of Lords, including Lord Scott himself, unanimously held that the invalidity of the relevant legislation in that case left taxpayers with an ‘accrued right’ even if they would not have taken advantage of this right had it been available to them. Lord Neuberger’s view was that the legislation in question was intended to be for the benefit of anyone who could take advantage of it, whether or not they in fact did so.25 It is submitted that this is perfectly in keeping with the discussion’s focus on the extent, under (what was then) EC law, of the Customs Commissioners’ public law powers to enact rules. To ask whether receipt of a tax can be set aside as if it were a transaction in private law is DMG (n 1). See eg Duncan Sheehan’s discussion in chapter 3 of the present volume and, more generally, R Stevens, ‘Justified Enrichment’ (2005) 5 Oxford University Commonwealth Law Journal 141, the judgments of Lord Scott in DMG (n 1) and Sempra Metals Ltd (formerly Metallgesellschaft Ltd) v HM Revenue & Customs [2007] UKHL 34, [2008] 1 AC 561, as well as R Chambers, ‘Deutsche Morgan Grenfell v IRC’ (2006) 6 Oxford University Commonwealth Law Journal 227, 234. 21 At para 76 of its judgment the ECJ held that ‘it is contrary to Article 52 [now Art 49 TFEU, ex-Art 43 EC] of the Treaty for the tax legislation of a Member State, such as that in issue in the main proceedings, to afford companies resident in that Member State the possibility of benefiting from a taxation regime allowing them to pay dividends to their parent company without having to pay advance corporation tax where their parent company is also resident in that Member State but to deny them that possibility where their parent company has its seat in another Member State’ (emphasis added). Joined Cases C-397/98 and C-410/98 Metallgesellschaft Ltd v Inland Revenue Commissioners and Hoechst AG v Inland Revenue Commissioners [2001] ECR I-1727 (hereafter ‘Metallgesellschaft’). 22 See further Williams (n 3) 77–81. 23 eg Dimes v The Proprietors of the Grand Junction Canal (1852) 3 HLC 759, 10 ER 301. In that case, Lord Campbell (ibid at 793, 315) held: ‘No-one can suppose that Lord Cottenham could be, in the remotest degree, influenced by the interests that he had in this concern’. Nevertheless, the fact that Lord Cottenham held shares in the respondent company meant that his decision on a case involving that company had to be set aside on grounds of financial interest and thus bias. 24 Fleming (trading as Bodycraft) and Condé Nast Publications Ltd v HM Revenue and Customs [2008] UKHL 2, [2008] 1 WLR 195. 25 ibid [97]. 19 20

Overpaid Taxes: A Hybrid Approach 27

therefore, with respect, neither here nor there.26 The real question is whether receipt of the ACT was beyond the powers of the taxing authorities and, for the reasons given here, it was. Stevens’ concern is that this is a ‘radical conclusion as it means that the effect of European Community Law is not merely to render legislation of the United Kingdom unlawful but also that it is, to the relevant extent, a “nullity’ ”.27 However, this concern falls away once one examines how precisely EU and national law interrelate. In Ministero Delle Finanze v IN.CO.GE.’90 Srl, the ECJ rejected the Commission’s argument that a national rule imposing unlawful charges should be treated as invalid rather than inapplicable.28 There is therefore no question of a decision of the ECJ making national law a nullity; all it can ever do is to identify potential incompatibility to be applied to the facts of the case by the national court, which if found would render the national law inapplicable during that country’s membership of the EU.29 And here, given that the denial of the benefit of the ACT/MCT scheme was incompatible with (what is now) EU law, the form of this denial (namely charging companies the ACT up front following a refusal of the group income election option) was, as a matter of UK law, ultra vires the Revenue on the basis that it contravened the will of Parliament as expressed in the European Communities Act 1972.30 The more difficult question is precisely how broad the scope of such a public law finding of illegality should be. Should it extend also to the levying of ACT from a company which, like DMG itself, could not have made the group income election because its parent was located abroad but which would in any event not have chosen to make the election even had it been available?31 This dilemma provides us with another example of an issue which may be novel to private law, but which is wholly familiar to the public law which ought also to be operating in this context, and which already contains a set of tools to deal with this problem.32 Taking a hybrid approach does not avoid the dilemma, but it does avoid the need for private lawyers to reinvent the wheel when an understanding of both the public and private aspects of the issue would allow them to draw on the experience of a body of law developed specifically to deal with such issues. Similarly, the question of the extent of invalidity is also raised by the ECJ decision in FII 33 and by the Thin Cap litigation.34 In all these cases we should be asking not only what the tests of causation and remoteness are in See further DMG (n 1) [80]. Stevens (n 20) 144. 28 Cases C-10-22/97 Ministero Delle Finanze v IN.CO.GE.’90 Srl [1998] ECR I-6307 [28]. 29 Case 106/77 Amminstrazione delle Finanze dello Stato v Simmenthal SpA [1978] ECR 629 [17]. For a recent domestic application of this reasoning, see the majority of the House of Lords in Fleming and Condé Nast (n 24). 30 Factortame Ltd v Secretary of State for Transport (No 2) [1991] 1 AC 603 (HL); Equal Opportunities Commission v Secretary of State for Employment [1995] 1 AC 1 (HL). 31 A case which appears to come close to these facts is that of Pirelli Cable Holding NV and others v Inland Revenue Commissioners [2006] UKHL 4, see also Pirelli Cable Holding NV v HM Revenue and Customs [2008] EWCA Civ 70. Here the House of Lords found that as a result of their inability (in contravention of EU law) to make a group income election, the claimants had received various tax credits which must be taken into account in establishing the degree of expense caused to them by the unlawful tax scheme. The cases appear to proceed on the assumption that the claimants would have made the group income election if available and thus that they would not have received the tax credits, but of course it is possible that the opposite may have been true, and even had they been able to make the election, the degree of credits which they in fact received by not making it might have deterred them from doing so. I am very grateful to David Ewart for bringing this case to my attention. 32 For a fuller statement of this position see Williams (n 3) 82–85. 33 Case C-446/04 Test Claimants in the Franked Investment Income Group Litigation v Inland Revenue Commissioners [2006] ECR I-11753 (hereafter ‘FII (ECJ)’). 34 Case C-524/04 Test Claimants in the Thin Cap Group Litigation v Inland Revenue Commissioners [2007] ECR I-2107 and Test Claimants in the Thin Cap Group Litigation v HM Revenue & Customs [2009] EWHC 2908 (Ch). 26 27

28 Rebecca Williams the private law of unjust enrichment, but also how far the underlying invalidity of the UK rules should extend as a matter of public law.35 In addition to these matters, adopting a hybrid public and private approach to such cases has various further important implications.

(3) A Certain, Coherent Interpretation of Woolwich and the Avoidance of Future Litigation First, the hybrid cause of action I propose allows us immediately to answer a lot of the questions left open after Woolwich,36 such as the subject-matter and types of bodies to which the reason for restitution in Woolwich extends. In short, it applies to all bodies subject to public law and to everything which administrative law would regard as an ultra vires public law event. For example, in FII the Supreme Court felt it had to decide whether a demand would be necessary for the so-called ‘Woolwich cause of action’ to be made out.37 Counsel for the Test Claimants had, somewhat counter-intuitively, suggested that a demand would be necessary in order to make out a ‘Woolwich claim’,38 arguing that to decide otherwise would be to shift English law closer to an absence of basis approach. The Supreme Court rejected this argument,39 holding that to dispense with the need for a demand would simply create a larger ‘island of recovery’, rather than moving the law towards absence of basis. But if a hybrid cause of action were to be adopted, none of these questions would have arisen in the first place.40 Recovery would be available whenever there has been an ultra vires event to which public law will respond; the presence or absence of a demand would simply be irrelevant; and, as explained above, there would be no question of any move towards absence of basis. Similarly, counsel for the Test Claimants had argued that abandoning the demand requirement would lead to uncertainty as to what amounts to a tax, an argument which again the Supreme Court rejected, but which equally would not have been open to the claimants to make in the first place on the basis of the argument here, since there is nothing to limit the public law unjust factor to tax; any ultra vires charge41 can be recovered using the hybrid cause of action with the public law reason for restitution.42 See further Williams (n 3) xvi–xvii and 110–13. See further ibid 40–55. 37 Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2012] UKSC 19, [2012] 2 AC 337 (hereafter ‘FII (SC)’). 38 The reason being, as Lord Walker puts it (ibid [40]), that by doing so they were seeking to ‘saw off ’ the availability of the ‘Woolwich cause of action’ in order to argue that the mistake-based claim was the only one available, thus bringing their case within the principle of effectiveness as discussed in Case C-62/00 Marks and Spencer plc v Customs and Excise Commissioners [2002] ECR I-6325 (hereafter ‘Marks & Spencer’). 39 FII (SC) (n 37) [81] (Lord Sumption). 40 See further Williams (n 3) xiii–iv and 40. 41 Or indeed payment: see text accompanying nn 52–53 and Williams (n 3) 56–69. 42 See further Williams (n 3) 52–53, 56–69. This does, of course, leave open the questions of when a body should be regarded as public, and whether there are any circumstances in which a public body should nevertheless be regarded as acting in a private context. However, (a) these are questions with which public law already has to contend, so that it already has a system for working out the answers to them, thereby reducing the potential for uncertainty, and (b) in any case this would be a relatively small area of uncertainty when contrasted with the degree of litigation which has taken place and no doubt will continue to take place concerning the precise interpretation and operation of the so-called ‘Woolwich’ unjust factor, as discussed here. That this is so is demonstrated by the fact that there are many recent cases dealing with the interpretation of Woolwich, including not least one 35 36

Overpaid Taxes: A Hybrid Approach 29

Two other questions arising in recent litigation can also be answered in a similar way. Thus in FII Lord Sumption suggested that a mistake claim would be necessary in a case such as Reemtsma,43 in which a German company purchased services from an Italian advertising agency and paid VAT to them which was not due. Lord Sumption thought that since the tax charged conformed entirely with the relevant Directives, the Italian tax authorities had charged no tax unlawfully and thus recovery would only be available on the basis of mistake were the case to arise in England.44 But the hybrid cause of action with the reason for restitution provided by public law is not confined to situations where the taxing legislation is unlawful. As long as receipt (or payment)45 of the money was beyond the powers of the public authority in the particular instance, recovery would be available on the basis of the hybrid cause of action.46 An overpayment by mistake would still be a payment of money that HMRC were not entitled to receive, and it would thus be recoverable on the basis of the public law reason for restitution. Similarly, in the Investment Trust case,47 Henderson J noted that there is still a bridge that needs to be crossed before it could be said that the Woolwich cause of action would be available to the claimants in the present case. It is true that their claims are to be characterised, for the purposes of EU law, as San Giorgio48 claims; but, so far at any rate, it has never been held that a Woolwich remedy is available to anybody other than the actual taxpayer who has paid the unlawful tax to HMRC.49

However, if the approach suggested here were to be adopted, this problem too would be resolved. San Giorgio claims ought to be regarded as a subset of ‘Woolwich claims’: relevant acts which are beyond the powers of the public body in question invoke the public law reason for restitution first recognised by the House of Lords in Woolwich, but only some of these claims will also be beyond the public body’s powers as a matter of European law, as in San Giorgio.50 If receipt of the money was, as a matter of public law, beyond the powers of the public authority in question, then provided that the other elements of an unjust enrichment claim in private law are made out (as indeed Henderson J had, perhaps unusually, to determine in Investment Trust) the hybrid cause of action is made out and the claimant can recover. Adoption of the hybrid approach suggested here can therefore provide a frame of reference to inform any similar future court decision about the boundaries and operation of the so-called ‘Woolwich cause of action’, thereby providing both clarity and certainty to this area of law. Conversely, as things stand, while there is no comprehensive understanding of decision by the House of Lords and one by the Supreme Court in the last six years, whereas it is difficult to think of a case which has or which might raise the question of whether a body ought to be regarded as public or whether, if public, it ought nevertheless to be regarded as having acted privately. 43 Case C-35/05 Reemtsma Cigarettenfabriken GmbH v Ministero delle Finanze [2007] ECR I-2425. 44 FII (SC) (n 37) [189]–[190]. 45 Where the public body is the claimant: see text accompanying nn 52–53. 46 See further Williams (n 3) 54–55. 47 Investment Trust Companies (in liq) v HM Revenue & Customs [2012] EWHC 458 (Ch) (hereafter ‘ITC ’). 48 Case 199/82 Amministrazione delle Finanze dello Stato v SpA San Giorgio [1983] ECR 3595. San Giorgio famously restated that ‘entitlement to the repayment of charges levied by a member state contrary to the rules of community law is a consequence of, and an adjunct to, the rights conferred on individuals by the community provisions prohibiting charges having an effect equivalent to customs duties or, as the case may be, the discriminatory application of internal taxes’ (Grounds, para 12). 49 ITC (n 47) [155]. The claimants had paid VAT to the Investment Managers, who then paid some of it on to HMRC. See also the discussion of the case by Charles Mitchell in chapter 6 of the present volume. 50 See further Williams (n 3) xiii and 105–9, esp 107.

30 Rebecca Williams precisely what ‘policy-motivated’ ‘Woolwich claims’ entail, any future question about the scope of ‘Woolwich’ will have to be tested by litigation, and the outcome of any such litigation may in turn create further uncertainty. For example, in FII Lord Sumption’s discussion of Reemtsma and how that case should be dealt with in national law51 may now require lines to be drawn in future cases between ‘pure mistake’ cases (which he regarded as falling outside ‘Woolwich’) and cases in which HMRC has somehow causally contributed to the payment such that it should be regarded as falling into ‘Woolwich’, a line which in the detailed functioning of taxation would surely prove difficult to draw.

(4) The Unification of Cases Where the Public Body is the Defendant and Those Where it is the Claimant The second further implication of the approach suggested here is that it allows us to consider, not just cases like Woolwich in which the public body is the defendant, but also those, such as Auckland Harbour Board v The King,52 in which it is the claimant, as well as the ‘swaps’ cases.53

(5) The Tailoring of Specific Defences The third implication is that it would also be possible to tailor defences in a manner that reflects the hybrid public and private nature of the claim.54 For various reasons, often to do with compatibility with EU law, it is actually very difficult to apply many of the usual restitutionary defences in cases involving public bodies.55 This makes it even more important to set a tailored time limit, subject to compatibility with EU law, as indeed has occurred to some extent via Schedule 39 of the Finance Act 2008.

C. A HIERARCHY OF REASONS FOR RESTITUTION

Other implications of a hybrid cause of action are, however, more controversial. My argument is not only that we can use the hybrid cause of action for restitution in all these cases, but actually we should do so. We should therefore have a hierarchy of reasons for restitution so that whenever there has been a public law ultra vires event which has led to a transfer, this should be the sole reason for restitution.56 This is to ensure that the law has access to both the public and the private aspects of the claim in those cases, which is vital if we are to ensure that, for example, the questions of the extent of invalidity or inapplicability of statutes discussed above are addressed accurately. See n 43 – 44 and accompanying text. Auckland Harbour Board v The King [1924] AC 318 (PC). 53 See further C Mitchell [2010] Public Law 747 and Williams (n 3) 56 – 69. 54 See further Williams (n 3) 123 – 61. 55 On which see esp text accompanying nn 150 –53. 56 See further Williams (n 3) 70 –122. 51 52

Overpaid Taxes: A Hybrid Approach 31

Obviously, there is a significant obstacle to the recognition of such a hierarchy, in the form of the House of Lords decision in DMG.57 As is well known, the Court of Appeal in that case sought to establish precisely this kind of hierarchy,58 but the House of Lords overruled them. This was, however, a lost opportunity in my view. The Court of Appeal and House of Lords, in particular Lord Hoffmann, addressed the question primarily on the basis of precedent, rather than principle, asking themselves whether Lord Goff’s speeches in previous cases were best read as having established such a hierarchy. Whether or not the Court of Appeal’s handling of precedent was correct, the House of Lords did not – and does not – need to treat itself as bound by them.59 Moreover, there exist compelling answers to each of the arguments from principle that were given by the House of Lords for rejecting the hierarchy. One such argument is that of Lord Scott, who of course held that if there were to be recovery at all it would have to be through a Francovich 60 action for damages, rather than an action for restitution. However, if the true basis of recovery in Woolwich is understood in the way I have suggested, this argument cannot hold. If an action is in contravention of EU law, it will automatically be ultra vires the public body and would thus be recoverable via a hybrid ‘Woolwich claim’, as described above. So if there has been a breach of EU law for the purposes of a Francovich damages action, there has by definition been a breach of EU law which would found an action for recovery in unjust enrichment on the basis of that public law, ultra vires event. It would not be possible to have the one without the other. Indeed, to make out a Francovich claim it is necessary to go on to prove the additional requirement of a ‘sufficiently serious’ breach. In the case of DMG itself this could be established fairly readily, but that will not always be the case, as was demonstrated by the FII decision.61 Lord Hope, on the other hand, was concerned that to create a hierarchy would be to build two exceptions into the ground of recovery for mistake of law established in Kleinwort Benson v Lincoln CC:62 ‘[T]he first exception would involve treating payments made under a mistake of fact differently from payments made under a mistake of law. The second would involve treating the Revenue differently from all other public authorities’.63 But once the true nature of the public law reason for restitution is properly understood, these difficulties fall away. The hierarchy I propose certainly would treat the public law unjust factor as taking precedence over both mistakes of fact and mistakes of law. However, the idea is not to carve out an arbitrary exception to one reason for restitution, but rather to place the hybrid cause of action above all private law reasons. Nor would this entail treating the Revenue differently from other public authorities. Indeed, the whole point is to recognise that it is a public authority and ground recovery on that basis, as it would be grounded for any other body subject to judicial review.

DMG (n 1). See also text accompanying n 19. Deutsche Morgan Grenfell Group plc v Inland Revenue Commissioners [2005] EWCA Civ 78 (CA). 59 Practice Statement (Judicial Precedent) House of Lords [1966] 1 WLR 1234. 60 Cases C-6/90 and C-9/90 Francovich and Bonifaci v Italian Republic [1991] ECR I-5357. 61 In FII (ECJ) (n 33), AG Geelhoed pointed out that not only could actions in unjust enrichment arise in circumstances where the ‘sufficiently serious’ breach condition would not be fulfilled, but also, for precisely this reason, the Member State’s obligation to provide an effective remedy for a breach of EU law might require it to provide ‘reimbursement’ instead (ibid [134]). See further Williams (n 3) 77–86. 62 Kleinwort Benson Ltd v Lincoln City Council [1999] 2 AC 349 (HL) (hereafter ‘Kleinwort Benson’). 63 DMG (n 1) [44]. 57 58

32 Rebecca Williams Nor is the hierarchy I propose, as Lord Hope fears, unprincipled. For even longer than the law has been moving towards ‘a principled statement of the law of restitution’,64 the law has been moving towards a principled statement of administrative law.65 The point is that both these areas of law are relevant to such claims. Lord Walker was also concerned that a hierarchy would create uncertainty in the law, but again, the first implication of my argument as outlined above is that to understand the ground of recovery correctly is actually to resolve a lot of the open questions and consequent uncertainty in the post-Woolwich law. Lord Hope’s second set of concerns, shared by Lord Walker, centred on the principle of concurrent remedies established in Henderson v Merrett.66 This, too, is a distortion caused by private law spectacles. From a public lawyer’s point of view there is nothing surprising about public law sometimes ‘trumping’ private law; that is precisely what the presumption of exclusivity in favour of public law in O’Reilly v Mackman 67 and Clark 68 establishes. In order to understand that the appropriate comparison is with Clark, rather than with Henderson v Merrett, it is worth taking a moment to reconsider the whole question of the procedure used to hear claims such as that at issue in DMG. The first thing to note is that it was essentially by chance that such claims ever came to be regarded as purely private law matters. Both the swaps cases and Woolwich actually entailed two sets of claims, one in public law69 and the other in the private law of unjust enrichment. This ‘bifurcation’ of claims was felt to be unsatisfactory and inefficient,70 and in British Steel plc v Customs and Excise Commissioners,71 it was established that both issues could be decided in the same case. However, given that in fact there were two relevant events in these cases (as their initial bifurcation demonstrated), one in public and one in private law, there was no principled ground to choose between the two procedures and no reason why one should be preferred to the other. There was, however, a pragmatic reason. Restitution did not become available as a remedy in public law until 2004, when section 31(4) of the Supreme Court Act 198172 was amended73 to give the High Court the power to award restitution or the recovery of a sum due, in addition to its existing power to award damages on an application for judicial review. At the time of British Steel, therefore, for purely fortuitous practical reasons the claim had to be brought using the private law procedure if restitution was to be recovered. This also fitted well with the fact that at the time there was an exception to the general procedural exclusivity in favour of public law, established in O’Reilly where ‘private rights’ were at stake. ibid. In R v Secretary of State for the Home Department, ex parte Fire Brigades Union [1995] 2 AC 513 (HL) 567, Lord Mustill (dissenting on the facts) pointed out that ‘[t]o avoid a vacuum in which the citizen would be left without protection against a misuse of executive powers the courts have had no option but to occupy the dead ground in a manner, and in areas of public life, which could not have been foreseen 30 years ago’. 66 Henderson v Merrett Syndicates Ltd [1995] 2 AC 145 (HL) (hereafter ‘Henderson v Merrett’). 67 O’Reilly v Mackman [1983] 2 AC 237 (HL). 68 Clark v University of Lincolnshire and Humberside [2000] 1 WLR 1988 (HL) (hereafter ‘Clark’). 69 R v Inland Revenue Commissioners, ex parte Woolwich Equitable Building Society [1987] STC 654 (QB) and Hazell v Hammersmith and Fulham London Borough Council [1992] 2 AC 1 (HL) respectively. 70 See further P Birks, ‘ “When Money is Paid in Pursuance of a Void Authority. . .” – a Duty to Repay?’ [1992] Public Law 580, 589, and J Beatson, ‘Public Law, Restitution and the Role of the House of Lords’ (1993) 109 Law Quarterly Review 1, 5. See also Williams (n 3) 49–52. 71 British Steel plc v Customs and Excise Commissioners [1997] 2 All ER 366 (CA). 72 Now known as the Senior Courts Act 1981 in order to avoid confusion, following the replacement of the Appellate Committee of the House of Lords by the Supreme Court in 2009. 73 By Civil Procedure (Modification of Supreme Court Act 1981) Order 2004 (SI 2004/1033). 64 65

Overpaid Taxes: A Hybrid Approach 33

However, since the reforms to the judicial review procedure which took place in 2000, the procedural exclusivity in favour of public law laid down in O’Reilly and exceptions to it should now be read in the light of cases such as Clark,74 which subsequently made clear that the initial choice of procedure will no longer be a watershed between the operation of public and private rules. Thus even within the private procedure, the court can now take account of public law imperatives such as the need for shorter time limits. In other words, the public/private divide is now more substantive rather than purely procedural,75 which of course strengthens the case for treating Clark as preferable to Henderson v Merrett. Indeed, two cases since these reforms, Jones v Powys Local Health Board 76 and Hemming v Westminster City Council,77 suggest that even British Steel itself might no longer be so straightforwardly applied, and that the judicial review procedure should perhaps be used for such claims instead. In Jones, a son, following his father’s death, sought to recover from the local health board the money paid for his father’s care on the basis that his father had had a legal entitlement to it, a fact which the son claimed had been missed because his father’s needs had been wrongly assessed by the health board’s review panel at the time of his father’s admission to the care in question. Plender J held that it was an abuse of process to try to bring this claim outside the judicial review procedure because the claimant could not succeed without proving that the review panel had erred, a question which was more suitable for judicial review, to which more stringent time limits and a permission requirement would have applied. Similarly in Hemming, the claimants sought to recover money paid to Westminster City Council as fees for sex shop licences, arguing that the fees had been charged ultra vires in the absence of a proper determination by the Council of their level. Keith J acknowledged that there were ‘comments’ in both British Steel and Woolwich suggesting that the private law procedure (with its longer time limits) could be used to bring such claims, but pointed out also that those comments had been, ‘made before 2004 when rule 54.3 of the Civil Procedure Rules was amended to permit a claim for judicial review to include a claim for restitution’,78 noting that in Jones v Powys this had caused Plender J to take a different view. Keith J therefore agreed with Plender J’s analysis in Jones that it was necessary to consider whether the claim is in substance ‘asserting an entitlement to a subsisting right in private law, which “may incidentally involve the examination of a public law issue” or whether the “primary focus” or “dominant issue” is to challenge a public law act or decision’.79 Applying that test, Keith J concluded that he had no doubt that the primary focus of the restitutionary claim is to challenge the equivalent of ‘a public law . . . decision’, i.e. the Council’s failure to determine a licence fee for the relevant years. Accordingly, if the restitutionary claim is to proceed, it must be treated as a claim for judicial review (because [private] proceedings would have amounted to an abuse of the court’s process). On that basis, the question of the claimants’ failure to bring the claim within the time limit for claims of judicial review has to be addressed.80 74 Above n 68. See also Carter Commercial Developments v Bedford Borough Council [2001] EWHC 669 (QB) (Admin) (2001) 34 EG 99; 75 See further Williams (n 3) at 14–15. 76 Jones v Powys Local Health Board [2008] EWHC 2562 (Admin). 77 Hemming (trading as Simply Pleasure Ltd) v Westminster City Council [2012] EWHC 1260 (Admin). I am very grateful to Charles Mitchell for bringing this case to my attention. 78 ibid [10]. 79 ibid. 80 ibid [12].

34 Rebecca Williams On the facts Keith J nevertheless decided that although the claims had not been brought within the time limits applicable to judicial review, he was prepared to extend those time limits because the claim had been brought within three months of the claimants receiving the information necessary for them to realise that they had a claim. On this basis the claim succeeded. There are several points to be made about Jones and Hemming. First, the test articulated by Plender J in Jones and subsequently applied by Keith J in Hemming illustrates the hybrid nature of these cases. It is in fact impossible to decide (as that test requires) whether the ‘dominant issue’ is one of public or private law, precisely because it is both. The private law claim in unjust enrichment derives its reason for restitution wholly from public law, but conversely, without that private law claim, public law alone would not give the claimant an entitlement to return of the money.81 Indeed it could even be argued that by choosing the public law procedure and then extending the time limit to allow the private law claim to succeed, Keith J was in fact instinctively trying to take a hybrid approach in Hemming. Secondly, as far as purely procedural questions are concerned, I have argued elsewhere that there are good reasons for thinking that it would be better for British Steel to continue to be followed.82 Nevertheless, the very existence of Jones and Hemming, and the happenstance that dictated the private procedure in British Steel should remind us that although the two events, public and private, can now be dealt with using one procedure, this does not mean that either of them has vanished. It is easy to see how use of the private law procedure has led to a substantive assumption that cases involving public law ultra vires actions are no different from alternative claims in private law based on mistake or failure of basis. Conversely, were Jones and Hemming to be followed more widely, there would be an equal and opposite danger that the private law aspects of the claim would be lost, with public lawyers arguing that the public law issues were ‘dominant’ and that such cases were no different from any other aspect of judicial review. Neither of these approaches would capture the true nature of these cases as would a hybrid public and private approach.83 And as far as the choice of causes of action is concerned, these cases must surely lead us to conclude that the appropriate comparison is with Clark and the need to accommodate both the public and private aspects of the claim, rather than the purely private approach of Henderson v Merrett. The private procedure may be used, but within this, in keeping with the new approach in Clark, we should prioritise the hybrid cause of action with its public law reason for restitution, which allows us to take account of the relevant public law event. Finally, returning to DMG, Lord Walker was concerned that a hierarchy of the kind proposed here would be in breach of the principles of effectiveness and equivalence in EU law,84 if recovery were limited so as to exclude an alternative concurrent remedy which would be available in a dispute between private citizens. However, this is not in fact the case. In Edis,85 the ECJ made it quite clear that it is perfectly compatible with EU law to have different rules for public bodies and for private parties, so long as the overall remedy See Williams (n 3) 31–35. ibid 52. 83 See further ibid 35 and 51. 84 See eg Case 33/76 Rewe-Zentralfinanz GmbH v Landwirtschaftskammer für das Saarland [1976] ECR 1989, referred to by the Supreme Court in FII (n 37) as ‘Rewe I ’. The principle of effectiveness requires just that, that the remedy be ‘effective’. The principle of equivalence requires that the means of enforcing EU rights at national level should not be less effective than a similar means of enforcing an analogous domestic right. For further detail of these two principles and their historical development see Williams (n 3) 236–41. 85 Case C-231/96 Edilizia Industriale Siderurgica Srl (Edis) v Ministero delle Finanze [1998] ECR I-4951. 81 82

Overpaid Taxes: A Hybrid Approach 35

is effective: the requirement of equivalence operates within each of the public and private contexts, not between them. Indeed, the hierarchy proposed here is as much about providing a remedy which would not be available in a dispute between private citizens as it is about excluding those which would be. This approach received some support from the decision of the Court of Appeal in the FII Group Litigation case,86 holding that sections 320 of the Finance Act 2004 and 107 of the Finance Act 2007 (which sought, retrospectively, to curtail the operation of section 32(1)(c) of the Limitation Act 1980, thereby removing the extended limitation period for mistake claims in relation to overpaid taxes) were not in breach of the European principles of equivalence, effectiveness, legal certainty and legitimate expectations, because the statutory provisions only limited mistake claims and mistake claims were not required by Community law. The ‘Woolwich cause of action’ provided ‘an effective remedy for all the Claimants’ San Giorgio87 claims’88 and this was unaffected by sections 320 and 107. In Littlewoods v HMRC,89 this approach was taken one stage further by Vos J.90 This case concerned sections 78 and 80 of the Value Added Tax Act 1994, which in combination were held to limit claimants to a claim for simple interest, excluding any other claims for interest outside the statutory scheme.91 The claimants argued that this was a breach of their EU right to an effective remedy,92 a question Vos J referred to the ECJ.93 86 Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2010] EWCA Civ 103, [2010] STC 1251. 87 cf n 48. 88 Above n 86 at [225]. This is not surprising, as explained above, text to n 50. 89 Littlewoods Retail Ltd v HM Revenue & Customs [2010] EWHC 1071 (Ch), [2010] STC 2072 (hereafter ‘Littlewoods (Ch)’. 90 Though he did regard his position as being supported by obiter dicta from the Court of Appeal in FII: ibid [85]. 91 ibid [45]–[62]. 92 On this point I have argued elsewhere that compound interest is necessary if transactions are to be fully unravelled as they ought to be, whether this requirement derives from domestic or European law: Williams (n 3) xi–xii and 44–49. It is therefore to be welcomed that the ECJ at least held that, ‘the national rules referring in particular to the calculation of interest which may be due should not lead to depriving the taxpayer of an adequate indemnity for the loss occasioned through the undue payment of VAT’ (Case C-591/10 Littlewoods Retail Ltd v HM Revenue and Customs (ECJ, 19 July 2012), [2012] ECR I-000, [2012] STC 1714, hereafter ‘Littlewoods (ECJ)’, at [29]), and it is to be hoped that at the national level it will be found that only compound interest would indeed be adequate. Yet arguably, for consistency with the decision in Metallgesellschaft (n 21), the ECJ should have gone even further and held more specifically that compound interest was required. Given the lack of clarity surrounding the precise extent and implications of the ECJ’s so-called ‘remedies jurisprudence’ (on which see Williams (n 3) xi–xii and 232–74), it is not surprising that it was unclear whether the right to recover the time value of money falls within that jurisprudence or not, but if compound interest is required when only the premature charging of the tax was in breach of EU law as in that case, then surely a fortiori compound interest must be required when the charging of the tax itself was invalid. That the ECJ was relatively reticent on this point is also out of keeping with some of its more interventionist recent judgments, on which see text accompanying nn 147–53. 93 In Littlewoods (Ch) (n 89), Vos J certified the following questions for reference to the ECJ: ‘Question 1: Where a taxable person has overpaid VAT which was collected by the Member State contrary to the requirements of EU VAT legislation, does the remedy provided by a Member State accord with EU law if that remedy provides only for (a) reimbursement of the principal sums overpaid, and (b) simple interest on those sums in accordance with national legislation, such as section 78 of the Value Added Tax Act 1994? Question 2: If not, does EU law require that the remedy provided by a Member State should provide for (a) reimbursement of the principal sums overpaid, and (b) payment of compound interest as the measure of the use value of the sums overpaid in the hands of the Member State and/or the loss of the use value of the money in the hands of the taxpayer? Question 3: If the answer to both questions 1 and 2 is in the negative, what must the remedy that EU law requires the Member State to provide include, in addition to reimbursement of the principal sums overpaid, in respect of the use value of the overpayment and/or interest? Question 4: If the answer to question 1 is in the negative, does the EU law principle of effectiveness require a Member State to disapply national law restrictions (such as sections 78 and 80 of the Value Added Tax Act 1994) on any domestic claims or remedies that would otherwise be available to the taxable person to vindicate the EU

36 Rebecca Williams However, the significant point for the present purpose is that Vos J held that even if sections 78 and 80 were in breach of EU law, this would only require the disapplication of those sections in relation to the ‘Woolwich claim’, not the mistake claim as well: [T]he cause of action most naturally and comprehensively giving effect to the San Giorgio right must be preferable to one that will only vindicate the right in certain circumstances . . . [T]he cause of action that most naturally and comprehensively gives effect to Littlewoods’ San Giorgio right to the use value of the repayments is the Woolwich claim, which is applicable, as we now know it to be, to all cases where the government has exacted tax, which was not lawfully due, whether or not a demand has been made The mistake-based restitutionary claims would only vindicate the San Giorgio right to the use value of money in some circumstances – namely where an appropriate mistake had been made. Such a claim would, therefore, be of less general application than the Woolwich claims.94

As a matter of principle this judgment is to be welcomed, since it provides further support for the hybrid approach for which this chapter argues, and does indeed confirm that the requirement of mistake is at best a ‘fifth wheel on the coach’95 and at worst a means of obscuring the hybrid nature of such claims and of separating those in which there is a mistake from those in which (like Woolwich) there is not.96 Even more importantly, however, the Grand Chamber of the ECJ did not think such an approach would be incompatible with EU law. On the contrary, it reiterated first that the principle of equivalence cannot be interpreted as requiring a Member State to extend its most favourable rules to all actions brought in a certain area of law. In order to ensure compliance with that principle, it is for the national court, which alone has direct knowledge of the procedural rules governing restitution actions against the State, to determine whether the procedural rules intended to ensure that the rights derived by individuals from EU law are safeguarded under domestic law comply with that principle.97

And then later, in response to the specific question of whether both ‘Woolwich’ and mistake of law must be available, the Grand Chamber held that when faced with a rule of law that is incompatible with directly applicable EU law, the national court is required to disapply that national rule, it being understood that that obligation does not restrict the power of the competent national courts to apply, amongst the various procedures of the internal legal order, those which are appropriate to safeguard the individual rights conferred by EU law.98

As far as the principle of equivalence is concerned, it will be argued below that this may not even be engaged in a case such as Littlewoods, where the restrictions on compound interest applied equally to both domestic and EU claims.99 Even if it were, it is clear from the ECJ’s law right established in the Court of Justice’s answer to the first 3 questions, or can the principle of effectiveness be satisfied if the national court disapplies such restrictions only in respect of one of these domestic claims or remedies? What other principles should guide the national court in giving effect to this EU law right so as to accord with the EU law principle of effectiveness?’ 94 Littlewoods (Ch) (n 89) [88] and [90]. 95 cf Häcker in chapter 12 of this volume at p 255. 96 See further Williams (n 3) 72–4. Assuming of course that the requirement of a mistake is a genuine requirement: ibid xvii–iii and 109–10. 97 Littlewoods (ECJ) (n 92) [31]. 98 ibid [33]. 99 See text accompanying nn 116 –22.

Overpaid Taxes: A Hybrid Approach 37

judgment that neither that principle nor the principle of effectiveness would prevent the national courts from holding that the ‘Woolwich cause of action’ is properly understood as a hybrid cause of action with a public law reason for restitution, and that it should therefore be used exclusively for such claims. The hybrid cause of action is different from the purely private mistake ground in terms of both its ‘purpose’ and ‘essential characteristics’, as the ECJ in Littlewoods requires, and it does ‘safeguard’ the ‘rights derived by individuals from EU law . . . under domestic law’,100 as the Supreme Court has recently acknowledged in FII. 101 And of course all this would be true a fortiori if, as a matter of English law, only the so-called ‘Woolwich cause of action’ were available across the board. The Supreme Court decision in FII, as noted above, concerned the compatibility of section 320 of the Finance Act 2004 and section 107 of the Finance Act 2007 with EU law, and in particular the EU principles of equivalence, effectiveness and protection of legitimate expectations. On the question whether or not the so-called ‘Woolwich cause of action’ would be sufficient to comply with the EU requirement to provide an effective remedy for breach of EU law, the Court held that this would in part depend on whether or not that cause of action depended on proof of a demand.102 And as also noted above, the Supreme Court unanimously held that it did not.103 The Court could therefore conclude, again unanimously, that the ‘Woolwich cause of action’ (which I have argued should be correctly understood as a hybrid cause of action) would in itself be sufficient to provide an effective remedy as required by EU law.104 Indeed the clearest support for my argument105 came from Lord Sumption who went so far as to state, that [i]f English law had never recognised a right to recover tax on the ground of mistake of law, but only on the basis of the principle in Woolwich Equitable, it is not disputed that that state of affairs would have satisfied the requirements of EU law.106

On this basis, Lord Sumption, with whom Lord Brown agreed, was able to hold that the cause of action in mistake was not necessary and thus, since sections 320 and 107 were only relevant to the mistake claim, just one challenge to those sections remained. This concerned their potential impact on the Test Claimants’ legitimate expectations.107 Here Lords Sumption and Brown concluded that since section 320 was enacted before the final decision of the House of Lords in DMG,108 at the time it was enacted the Test Claimants could not have had a legitimate expectation that their mistake claims would succeed, and thus section 320 did not infringe EU law.109 Section 107, by contrast, was enacted after the House of Lords definitively established the availability of mistake claims in DMG, and thus it did cf Littlewoods (ECJ) (n 92) [31]. FII (SC) (n 37). See also the text accompanying nn 104 and 107. 102 As Lord Walker put it, the Test Claimants also sought to ‘saw off ’ the availability of the so-called ‘Woolwich cause of action’ in order to leave mistake as the only route to recovery by arguing that Woolwich too was subject to s 32(1)(c) of the Limitation Act 1980: FII (SC) (n 37) [40], cf n 38. The Supreme Court unanimously rejected this proposition: ibid [10] (Lord Hope), [62]–[63] (Lord Walker), [121] (Lord Brown), [126] (Lord Clarke), [140] (Lord Dyson), [184]–[185] (Lord Sumption) and [246] (Lord Reed). 103 ibid [10] (Lord Hope), [79] (Lord Walker), [171]–[174] (Lord Sumption). Lords Brown, Clarke, Dyson and Reed agreed (ibid [121], [126], [140] and [246] respectively). 104 ibid [20] (Lord Hope), [39] (Lord Walker), [142] (Lord Sumption). Lords Brown, Clarke, Dyson and Reed agreed as above (n 103). 105 See further Williams (n 3) xii–xiii and 100–101. 106 FII (SC) (n 37) [199]. 107 ibid [198] (Lord Sumption) and [121]–[125] (Lord Brown). 108 cf DMG (n 1). 109 FII (SC) (n 37) [198]–[202] (Lord Sumption) and [121]–[125] (Lord Brown). 100 101

38 Rebecca Williams thwart the Test Claimants’ legitimate expectation that they would be able to use the mistake claim themselves.110 However, the majority of the Supreme Court in FII held that both sections 320 and 107 were in breach of EU law (a division in the court which thus necessitated a further reference to the ECJ),111 on the basis that both the so-called ‘Woolwich claim’ and the mistake claim were necessary for compliance with EU law. This was for various reasons. The majority (Lord Walker with whom, on this topic, Lords Hope Clarke, Dyson and Reed agreed),112 held that section 320 breached the principle of effectiveness as laid down in Marks & Spencer 113 because it did not contain any transitional provisions as required in that case. The majority were also unanimous that section 320 was in breach of the Test Claimants’ legitimate expectations.114 And, according to Lords Clarke and Reed,115 section 320 also infringed the principle of equivalence. On this last issue, the compatibility of section 320 with the principle of equivalence, it is submitted that it is the view held this time not only by Lords Sumption and Brown, but also Lord Walker,116 which is correct. When sections 320 and 107 were enacted, they were enacted in respect of all claims. As Lord Walker puts it, therefore, the principle of equivalence ‘is not applicable in this case because both of the statutory cut-off provisions applied to all claims for repayment of direct tax, whether or not the repayment was claimed because of an infringement of EU law’.117 The principle of equivalence118 requires only that ‘the procedural rule at issue applies without distinction to actions alleging infringements of Community law and to those alleging infringements of national law, with respect to the same kind of charges or dues’.119 If sections 320 and 107 had imposed the time limits for recovery of charges found to be in breach of EU law, but not for recovery of charges held to be in breach of purely domestic law, the principle of equivalence would have been infringed. As it is, however, neither section 320 nor section 107 does this. As noted above, a similar issue arises in both Littlewoods 120 and Investment Trust 121 in the sense that in both those cases the courts also have to decide whether a ‘Woolwich claim’ would be sufficient to fulfil the requirements of EU law on its own, or whether a mistake

ibid [203] (Lord Sumption) and [121]–[125] (Lord Brown). ibid [1] and [22] (Lord Hope), [120] (Lord Walker), [140] (Lord Dyson), [208] (Lord Sumption), [246] (Lord Reed). 112 ibid [112] (Lord Walker), [16] (Lord Hope), [129] (Lord Clarke), [140] (Lord Dyson), [209] (Lord Reed). 113 See n 38. 114 If and insofar as this amounted to a separate requirement: see n 131. For their Lordships’ conclusions that the Test Claimants’ legitimate expectations were, on the facts, breached, see further FII (SC) (n 37) [22] (Lord Hope), [100]–[115] (Lord Walker), [135] (Lord Clarke), [149] (Lord Dyson), [209] (Lord Reed). 115 FII (SC) (n37) [136] (Lord Clarke), [219]–[227] (Lord Reed). Lord Hope does suggest at [20] that the principle of equivalence was engaged, but in the event he stated himself to be in agreement with both Lords Reed and Walker (at [24]), who hold different views on the equivalence issue (see text accompanying n 117). Similarly Lord Dyson (at [140]) held himself to be in agreement with both Lords Walker and Sumption on the issues on which they agree (which includes the principle of equivalence), but also held himself to be in agreement with Lords Reed and Walker ‘on the DMG/Section 320 issue’. It is therefore not clear whether the majority view is that the principle of equivalence is engaged or not. 116 See n 115. 117 FII (SC) (n 37) [113]. See also Lord Sumption at [150]. 118 cf n 84. 119 Edis (n 85), citing also Joined Cases 66/79, 127/79 and 128/79 Amministrazione Delle Finanze dello Stato v Salumi [1980] ECR 1237, para 21. 120 cf Littlewoods (Ch) (n 89). 121 cf ITC (n 47). 110 111

Overpaid Taxes: A Hybrid Approach 39

claim would be required as well.122 In the Investment Trust litigation, Henderson J left the question open until the Supreme Court had ruled in FII and the ECJ had ruled in Littlewoods, as it now has. The situation is slightly different in Littlewoods and Investment Trust from FII, in that FII concerned the placing of a bar on the mistake cause of action while the Woolwich action remained open, whereas Littlewoods and Investment Trust both concern a bar on both the Woolwich and the mistake causes of action, and the question is therefore whether that bar must be lifted in relation to one or both claims. Nevertheless, the same point can be made in relation to all three cases; while purely domestic claimants are unable to choose between a mistake cause of action and one based on Woolwich, the EU principle of equivalence is not engaged if EU claimants are similarly denied that choice. And indeed, as noted above, the Grand Chamber of the ECJ has since held in Littlewoods that EU law does not necessarily require both ‘Woolwich’ and the mistake claim even if the principle of equivalence were engaged. As for the principle of effectiveness as discussed in Marks & Spencer,123 as noted above, the majority of the Supreme Court in FII upheld the Test Claimants’ challenge to sections 320 and 107 on this basis. The difficulty with this reasoning, however, as Lord Clarke explains, is that the problem . . . facing the Test Claimants is that English law provides two remedies for their claim that tax has been exacted from them contrary to EU law. If the only available remedy were the mistake claim, the position would be clear. It would fall within the principle in Marks and Spencer.124

As a matter of EU law therefore, the principle of effectiveness as discussed in Marks & Spencer is fulfilled by the availability of the ‘Woolwich cause of action’, and thus it is difficult to argue that any retrospective alteration of a wholly separate cause of action in mistake is a breach of the principle of effectiveness.125 This conclusion again perhaps receives a little further support from the ECJ’s decision in Littlewoods that it is up to the national courts to choose the cause of action which is ‘appropriate to safeguard the individual rights conferred by EU law’.126 In FII Lords Clarke and Reed sought to support the argument from effectiveness with one based on equivalence,127 in Lord Clarke’s case relying on Advocate General Sharpston’s opinion in an equivalence case as applying more generally in the context of effectiveness,128 while Lord Reed argued that since the principle of equivalence applied when the claimants made their claims it followed that the principle of effectiveness also applied and thus continued to apply until the claims were determined.129 The difficulty with these arguments is that, as noted above, the principle of equivalence is not engaged in FII. It is therefore hard to see how an inapplicable argument based on simple effectiveness can be strengthened by an equally inapplicable one based on equivalence. 122 See the text accompanying n 89. As also noted above, in Littlewoods (Ch), Vos J concluded that if ss 78 and 80 VATA 1994 were in breach of EU law, it would only be necessary to disapply them in relation to the so-called ‘Woolwich’ claim, not the mistake claim: text accompanying n 94. 123 See n 38. 124 FII (SC) (n 37) [130]. 125 This does not, of course, exclude the possibility that ss 320 and 107 were in breach of the doctrine of legitimate expectations, on which see text the accompanying n 131. 126 See Littlewoods (ECJ) (n 92) [31], set out in the text accompanying n 98. 127 FII (SC) (n 37) [136]–[138] (Lord Clarke) and [212]–[240] (Lord Reed). 128 ibid [137]–[138], the case being C-432/05 Unibet (London) Ltd v Justitiekanslern [2007] ECR I-2271 (ECJ), AG Sharpston. 129 FII (SC) (n 37) [227].

40 Rebecca Williams This, as Lords Brown and Sumption concluded, leaves open only the third basis for the challenge to sections 320 and 107, namely the doctrine of legitimate expectations. Their Lordships were unanimous that section 107 was in breach of the Test Claimants’ legitimate expectations, and a majority also found that the Claimants’ legitimate expectations were breached by section 320.130 There was perhaps some slight disagreement among their Lordships concerning the source of this principle,131 and a strict interpretation of EU law might suggest that once the principle of effectiveness has been satisfied (as suggested above), there is no further room for the doctrine of legitimate expectations to attach as a matter of EU law. However, it is unlikely that the ECJ would share this view, for three reasons. First, from a purely practical perspective, the European Commission has referred the UK to the ECJ as a result of the absence of proper transitional provisions in section 107,132 and thus the Commission clearly believes that at least that particular section is in breach of EU law. Secondly, it is worth noting that the principles of equivalence and effectiveness are themselves derived from a more general principle of cooperation between the Union and the Member States, now contained in Article 4 TFEU, but formerly protected by Article 10 (ex-Article 5) EC, and which is not necessarily exhausted by the principles of equivalence and effectiveness alone.133 Thirdly, as Lord Sumption points out, the principle of legal certainty lies at the heart of the EU legal order and entails (among other things) that those subject to EU law should be able clearly to ascertain their rights and obligations. One aspect of that principle is that within limits EU law will protect within its own domain legitimate expectations adversely affected by a change in the law.134

In as much as one can predict anything about the ECJ’s so-called ‘remedies jurisprudence’, therefore, the Court seems likely to take the view that once a particular remedy has become available for breach of EU law (whether that remedy was required in the first place by EU law or not), it would be a breach of the principles of legality and cooperation for that remedy to be removed retrospectively. As far as section 107 is concerned, it is hardly surprising that the Supreme Court’s conclusion on the legitimate expectation issue was unanimous. As Lord Sumption points out, the section was enacted after the decision of the House of Lords in DMG, by which time the British American Tobacco group and other companies in the same position had already been pursuing their claims through the English courts. And, even more significantly, they See nn 114 and 110 respectively. Lord Hope held that the question whether s 107 was a breach of legitimate expectations ‘does not raise any issue of EU law’ (FII (SC) (n 37) [22]), whereas Lord Walker (at [94]–[101] and [112]–[115]), Lord Brown (at [123]) Lord Clarke (at [135]), Lord Sumption (at [146] and [152]–[160], [198]) and Lord Reed (at [209], [214] and [241]–[245]) thought that it did. Lord Dyson agreed with both Lords Walker and Sumption on the issues on which they were agreed (at [140]). It is, however, worth noting that Lord Sumption (at [198]) doubted whether it made any difference whether the complaint was justified on the basis of the principle of effectiveness alone, or on the principle of legitimate expectations. 132 See further ibid [110] (Lord Walker). 133 As Lord Walker noted (ibid [98]), the question whether the UK legislation was specifically targeted at the Metallgesellschaft decision (cf n 21) was raised at first instance, but not discussed by Henderson J, not least because of ‘the difficult constitutional issues which would arise in inquiring into the legislative intention behind the amending legislation’. The point was not raised in the Court of Appeal or the Supreme Court, presumably for the same reasons. However, if there had been such a targeting, then this would arguably breach the more general principle of cooperation even if the specific remedy targeted was not itself necessary for fulfilment of the principles of effectiveness or equivalence. 134 FII (SC) (n 37) [146]. 130 131

Overpaid Taxes: A Hybrid Approach 41

had been expressly left to do so by their exclusion from the announcement on 8 September 2003 of what would become section 320, and its subsequent enactment in 2004.135 It would be difficult, therefore, not to conclude that section 107 of the Finance Act 2007 was in breach of the Test Claimants’ substantive legitimate expectations that they would be permitted to bring a mistake claim. It is, for the same reasons, equally unsurprising that the Commission chose to proceed against the UK in respect of that section.136 As for section 320 of the Finance Act 2004, however, it is submitted that the correct answer may in fact lie somewhere between the positions of the majority and minority in FII. It is clear that Lords Sumption and Brown rejected the claim of a legitimate expectation in the context of section 320 because they were looking for a potential substantive legitimate expectation on the part of the Test Claimants. In other words, the question Lords Sumption and Brown asked themselves was whether, as Lord Sumption puts it, the Test Claimants were ‘entitled to count on being able to recover on the ground of mistake of law with an extended limitation period’ (emphasis added). On that basis, he concluded that the highest that they can put their case is that they were entitled to do so in the seven week interval between [the decision of Park J in DMG on] 18 July and [the announcement of what would become section 320 on] 8 September 2003. Bearing in mind the brevity of the interval, the virtual certainty of an appeal and the uncertainty about its outcome, the argument that they had a legitimate expectation of the kind suggested seems to me to be unrealistic.137

And as Lord Brown points out, to find otherwise would be inconsistent with the finding of a mistake in DMG in the first place: [T]he situation even after Park J’s first-instance decision in [DMG] . . . was one of complete uncertainty as to whether tax could be re-claimed on the basis of a mistake of law – there being at least as much room for a mistake of law as to this as for the mistake of law which the majority of the House of Lords in DMG . . . held the taxpayers to remain under until the Court of Justice’s final authoritative decision in the [Metallgesellschaft and] Hoechst case.138

However, it is clear that at least Lords Hope and Clarke approached the question from a more procedural point of view. Thus Lord Hope held the following: There was no certainty at that time when s 320 FA 2004 was enacted that their claims based on mistake would succeed. But those claims were undoubtedly arguable . . . They were entitled to expect that the question whether their claims based on mistake were well founded would be decided by the courts, as there was a real issue to be tried. They were also entitled to expect, according to the principle of legal certainty, that this entitlement would not be removed from them by the state by the introduction without notice of a limitation period that was not fixed in advance.139

For the reasons given by Lords Sumption and Brown, it is difficult to argue that the Test Claimants had a legitimate expectation of substantive success, but it is far easier to argue, as ibid [203]. See n 132 and accompanying text. 137 FII (SC) (n 37) [200]. 138 ibid [123]. cf Metallgesellschaft (n 21). For a discussion of the existence or not of a mistake in Metallgesellschaft and other tax cases, see further Williams (n 3) 86–88, 94–96, 109–10 and 121. 139 FII (SC) (n 37) [19]. See also Lord Clarke (at [133]) holding that ‘the test claimants were entitled to have their mistake claim adjudicated upon by the English courts’. Though there is also evidence of the majority taking a more substantive approach: see eg Lord Walker (at [115]), Lord Hope (at [20]) and Lord Reed (at [236]). 135 136

42 Rebecca Williams Lords Hope and Clarke do, that they had a procedural legitimate expectation that their case would be heard. The difficulty is that such a procedural legitimate expectation is not particularly valuable to the claimants, since it need not lead to substantive success. Indeed, my argument is that the House of Lords in DMG should have upheld the Court of Appeal’s decision that only the so-called ‘Woolwich cause of action’ was available, not the mistake claim. Thus the opportunity taken away by section 320 was simply the opportunity for the Test Claimants to bring a claim which should, ultimately, have failed anyway. It is submitted, therefore, that the decision of the Supreme Court in FII provides yet further support for the view that the decision of the House of Lords in DMG is to be regretted. Without DMG there would have been no mistake claim at all, and thus none of the subsequent litigation over the relationship between ‘Woolwich claims’ and mistake claims in FII, Littlewoods or Investment Trust need have arisen in the first place. The public law reason for restitution would have provided a perfectly effective, exclusive remedy in both domestic and European law. On returning to the question of Lord Walker’s belief that the decision taken in DMG would help to prevent a breach of EU law, therefore, it is hard not to conclude that in fact all DMG has done is to open further questions of compatibility with EU law, particularly in the case of section 107 Finance Act 2007, whereas, given Edis, the establishment of a hierarchy of reasons for restitution in DMG would have been perfectly compatible with our EU law obligations. It is also worth reflecting that this tendency to perhaps over-comply with EU law is common to both the House of Lords’ DMG decision and the decision of the Supreme Court in FII. Thus just as Lord Walker in DMG believed that the principle of equivalence was engaged when in fact it was not, in FII the Supreme Court would, as outlined above, have been fully justified in holding that neither the principle of equivalence nor of effectiveness as laid down in Marks & Spencer140 was engaged, and that the Test Claimants’ legitimate expectations had only been breached by section 107 and to a very limited procedural extent by section 320. This would in turn have also avoided the need for a reference to be made to the ECJ. What will happen when the case reaches the ECJ is less clear. I have argued elsewhere that the very terminology involved in the ECJ’s so-called ‘remedies jurisprudence’ is unhelpful,141 and as both Lord Reed in FII 142 and Advocate General Trstenjak in Littlewoods143 point out, there is an ‘idiosyncrasy’ involved in describing causes of action as falling within the concept of national procedural autonomy. In FII Lord Walker states that claimants are not entitled to ‘a remedy arrived at by some precise formula furnished by EU law’,144 and as I have discussed elsewhere, this is indeed one model which could be adopted for arranging the division of labour between the national and EU courts.145 However, in recent judgments the ECJ has become increasingly interventionist in its approach in this context, moving towards the definition of an EU-level cause of action in unjust enrichment. Thus although in Metallgesellschaft the Court clearly stated that ‘it is not for the Court to assign a legal classification to the actions brought by the plaintiffs before the cf n 38. Williams (n 3) xii and 242. 142 FII (SC) (n 37) [217]. 143 Case C-591/10 Littlewoods Retail Ltd v HM Revenue & Customs (ECJ, 19 July 2012), [2012] STC 1714, Opinion of AG Trstenjak. 144 FII (SC) (n 37) [114]. 145 See further Williams (n 3) 242 ff. 140 141

Overpaid Taxes: A Hybrid Approach 43

national court’,146 in FII it held that it would nevertheless classify some of the claims as arising under the so-called San Giorgio principle (unjust enrichment), and others as arising under the principle of state liability in damages laid down in Francovich.147 In doing so Advocate General Geelhoed and the Court held that a claim would only be classified as a (strict liability) ‘San Giorgio’ claim rather than a (fault-based, sufficiently serious breach) Francovich claim if it was the ‘direct’ (AG) or ‘inevitable’ (ECJ) consequence of the unlawful charge.148 This rule obviously provides a significant disincentive to any attempts to mitigate the effects of an unlawful charge;149 there is no obvious reason for the adoption of such a rule, and yet very little explanation or reasoning was given for it. Similarly, in Lady & Kid 150 and Accor 151 the Court held, without detailed reasoning, and without hearing argument on any other defences,152 that the only defence available to a claim for the return of an unlawfully levied charge was the defence of passing on (Lady & Kid), and that this would only be available where the unlawful legislation itself required the passing on (Accor).153 One potential outcome of the FII reference, therefore, is that while the principles of effectiveness154 and equivalence do not currently prohibit legislation of the kind contained in sections 320 and 107, the ECJ could take the opportunity now presented by that preliminary reference to continue its interventionist approach by extending these principles, an outcome which is only a possibility as a result of the double over-compliance in DMG and FII. Even so, it would be difficult for the ECJ to go so far as to hold that there is a freestanding EU right to a mistake claim, and the ruling of the Grand Chamber in Littlewoods155 suggests that it is unlikely to try to do so. As outlined above, the lack of clarity surrounding the ECJ’s ‘remedies jurisprudence’ makes it difficult to state any principles in that area with absolute certainty, but at least Lord Sumption was of the view that EU law does not, indeed cannot, require that national law should recognise or create any particular cause of action or any particular remedy. It simply requires that whatever causes of action or remedies exist in national law must, taken as a whole, be effective and non-discriminatory.156

And, as also noted above, in Littlewoods the Grand Chamber confirmed that the obligation to comply with EU law ‘does not restrict the power of the competent national courts to apply, amongst the various procedures of the internal legal order, those which are appro Metallgesellschaft (n 21) [81]. On San Giorgio cf n 48 and on Francovich cf n 60. 148 FII (ECJ) (n 33) [133] (AG Geelhoed) and [207] (ECJ). 149 See further Williams (n 3) 263–64. 150 Case C-398/09 Lady & Kid A/S and others v Skatteministeriet (ECJ, 6 September 2011). 151 Case C-310/09 Ministre du Budget, des Comptes publics et de la Fonction publique v Accor SA (ECJ, 15 September 2011). 152 It is worth noting that the Court had only avoided hearing argument on the change of position defence because in Littlewoods (Ch) (n 89) Vos J refrained from sending a reference to the Court because it was not necessary to do so for the decision in the case in front of him. It is difficult therefore not to reach the conclusion that where there is a possibility that the ECJ might overlook a rule or principle of national law which might otherwise apply, courts should perhaps be alert to the possibility that a reference should be justified if at all possible, while, conversely, in other cases they should be aware of the possibility that a reference may give the opportunity for further extension of EU law to the exclusion of national law. 153 See further R Williams, ‘Unjust Enrichment and the Court of Justice of the European Communities, a loss of national competence and principle’ [2011] British Tax Review 631. 154 As discussed in Marks & Spencer (n 38), not including the principle of legitimate expectations. 155 Littlewoods (ECJ) (n 92). 156 FII (SC) (n 37) [148]. See also Lord Walker’s comment set out in the text accompanying n 144. 146 147

44 Rebecca Williams priate to safeguard’157 the relevant EU rights. All this suggests that the ECJ is unlikely specifically to require the mistake claim at all, but even if it did, it could only do so as a result of the continuing availability of the mistake claim under DMG. Thus, whatever the ECJ ultimately decides in FII, it will remain possible, consistently with EU law, to undo DMG, leaving the public law reason for restitution as the only basis for hybrid cases in both domestic and EU law in the manner I have suggested. It might be thought that even as a matter of domestic law such an overruling of DMG would be unlikely. It is true that their Lordships in FII were keen to point out that the contrary judgment reached by the Court of Appeal in FII was not just a ‘bump in the road’,158 and that the ultimate decision by the House of Lords is not ‘beyond academic controversy’.159 However, Lord Sumption also clearly stated that he did ‘not intend by making these points to reopen a debate which has been settled for more than five years’,160 and as a whole the Court’s decision in FII might well be taken as an implicit confirmation of the decision in DMG. And of course it need hardly be pointed out that even the Practice Statement which paved the way for (what was then) the House of Lords to overrule itself expressly refers to the need for legal certainty in precedent.161 Nevertheless, given that the decision in DMG is a continued source of costly and problematic litigation, especially in the context of EU law, it must at least be possible that its overruling, difficult as that might be, would now be the lesser of two evils. It is also clear that the situation in FII falls within the circumstances in which the House of Lords has previously chosen to overrule itself. If the so-called ‘Woolwich’ cause of action is reinterpreted as a hybrid claim with a public law reason for restitution, as I have argued that it should be, it would then be clear, indeed clearer than at present, how the law would operate, and there would thus be no problem with reversing the decision in DMG from the point of view of clarity.162 The area is one in which the law is in any case ‘the piecemeal creation of judges’, as Lord Sumption points out.163 Legislative intervention to change the result of the decision in DMG would not be impractical, but the attempts which were the subject of litigation in FII appear to have been unsuccessful, and given those attempts, were the Supreme Court now to overrule the decision in DMG this would clearly be in keeping with Parliament’s intent as expressed in those statutory provisions.164 Indeed the oddity of the situation is that any legislative action to undo DMG will always involve the removal of an existing right of some kind, and thus it will always be open to challenge on the basis of the EU principles of equivalence or effectiveness as in FII. Whereas, by contrast, a finding by the Supreme Court that the House of Lords in DMG took a wrong turning (as I argue it did) would amount to a finding that the mistake ground ought never to have been available in the first place, rather than a removal of that option. The situation is therefore one in which it is possible to conclude that, constitutionally, the Supreme Court Littlewoods (ECJ) (n 92) [33]. FII (SC) (n 37) [18] (Lord Hope) and [168] (Lord Sumption). 159 ibid [168] (Lord Sumption). 160 ibid. 161 cf n 59. 162 In Miliangos v George Frank Textiles Ltd [1976] AC 443 (hereafter: ‘Miliangos’), this was a factor which Lord Wilberforce (at 469, with whom Lords Cross, Edmund-Davies and Fraser agreed, Lord Simon dissenting) counted in favour of the House of Lords’ decision in that case to overrule the decision in In re United Railways of Havana and Regla Warehouses Ltd [1961] AC 1007 (HL), in order to allow judgments to be given by English courts in foreign currencies. 163 FII (SC) (n 37) [161]. The judge-made nature of the law in the context of the rule at issue in Miliangos (n 162) was a further reason for Lord Wilberforce’s decision to overrule the previous precedent in that case. 164 This again was a consideration which weighed in favour of Lord Wilberforce’s decision in Miliangos (n 162). 157 158

Overpaid Taxes: A Hybrid Approach 45

would not only be acting entirely consistently with the will of Parliament in overruling DMG, but that its freedom to do so is, unusually, greater than the ability of Parliament to execute that will, since the retrospective correction of DMG by the Supreme Court would be more obviously compatible with EU law than any similarly effective legislative intervention to remove DMG might be. Finally, it should be noted that the disadvantages and subsequent litigation resulting from the DMG decision were perhaps not fully foreseen by the House of Lords at the time of its decision in that case,165 and their Lordships were not asked to consider a principled argument in favour of the public law reason for restitution of the kind I have tried to present.166 Rather they were asked to rule on what Lord Goff had intended by his decision in Kleinwort Benson.167 In keeping with this it is plain, as noted above, that the Supreme Court has now changed its view on one of the key issues, holding that in principle the public law reason for restitution would be sufficient on its own to comply with EU law.168 It is therefore not impossible to suggest that if all these matters had been made clearer to their Lordships at the time of their decision in DMG, they could have been led to take a different view.169 Certainly, it is submitted, enough can be said in support of the overruling of DMG to make that possibility one which is still worthy of serious consideration, notwithstanding the decision of the Supreme Court in FII.

D. CONCLUSION

My conclusion, then, is that the ‘Woolwich cause of action’ ought properly to be regarded as a hybrid cause of action in which questions of enrichment and expense are dealt with by the private law of unjust enrichment, while the reason for restitution is recognised as being the existence of a public law ultra vires event. Adoption of this approach would have several important implications, many of which would have the potential to reduce substantially the extent and complexity of the costly on-going litigation in this area. First, it would provide a certain and coherent interpretation of what is otherwise regarded as the ‘policymotivated’ ‘Woolwich’ cause of action. This would explain why so many different factors were thought to be ‘in play’ in the Woolwich decision itself,170 and, even more importantly, it would answer many of the questions left open by Woolwich, such as the nature of the subject-matter or bodies to which that cause of action is applicable, the necessity of a demand or other action on the part of the charging authority, and so on.171 It would also provide a better alternative to the ‘absence of basis’ approach, focusing attention on the key

165 The degree of administrative and constitutional difficulties to which the present law gave rise was one of the reasons why the House of Lords also chose to exercise its ability to overrule itself in Vestey v Inland Revenue Commissioners (Nos 1 and 2) [1980] AC 1148 (HL) (hereafter ‘Vestey’), see esp Lord Wilberforce’s comments at 1178. 166 See Williams (n 3) 94–102. 167 cf Kleinwort Benson (n 62). 168 See the text accompanying nn 104–6. 169 The idea that the original court might have decided differently if the present arguments had been made before it was a factor which counted in favour of overruling in both Miliangos (n 162) 460 (Lord Wilberforce) and Vestey (n 165) 1198 (Lord Keith). 170 See the text accompanying nn 5–10. 171 See the text accompanying nn 36–55.

46 Rebecca Williams fact that the basis is absent specifically as a matter of public law,172 and it would explain why some legal systems regard such claims as wholly public while others regard them as wholly private.173 Indeed it is worth recalling in this context that the perception of such cases as purely private in the UK is essentially a matter of chance, arising from the choice of procedure used to hear such claims, a choice which has recently been questioned in two different cases.174 The hybrid approach would also allow us to consider cases where the public body is the claimant alongside those where it is the defendant,175 it would permit a proper examination of the extent of invalidity in a given case,176 and it would permit defences to be tailored to the specific public and private context of the claims.177 However, in addition to, or in some instances perhaps as a result of these implications, it is also my argument that the resulting hybrid cause of action should be used in preference to purely private law causes of action on the basis that only it can properly address both the public and private aspects of the claim.178 There are, of course, powerful obstacles to this approach in the form of the decision of the House of Lords in DMG,179 perhaps now implicitly reaffirmed by the decision of the Supreme Court in FII. Nevertheless, it should be noted that all the arguments presented in DMG against such a hierarchy are in fact answered once the true hybrid nature of the cause of action is properly understood.180 Indeed, once this understanding is achieved, it becomes clear that the exclusivity of the hybrid cause of action is not an unprincipled anomaly which is inconsistent with Henderson v Merrett, but rather a principled solution which is wholly in keeping with Clark.181 In addition, whilst it was feared in DMG that focusing exclusively on the ‘Woolwich cause of action’ would be a breach of EU law, in fact there was no such danger; conversely it was the legislative responses prompted by the DMG decision which have led to incompatibility with our EU law obligations and proceedings being brought by the Commission against the UK.182 Indeed, without the decision in DMG, there would have been no need for the further litigation on the relationship between the mistake and ‘Woolwich’ causes of action seen in FII, Littlewoods and Investment Trust, litigation which itself has the potential to raise yet further questions (such as the distinction between the cases which will fall within Woolwich in Lord Sumption’s view and those which will not).183 This in itself would be problematic enough, but the same tendency to do more than was necessary to comply with EU law in DMG has occurred again in FII,184 giving the ECJ opportunity, should it wish to take it, to continue the more interventionist approach it has demonstrated in cases such as FII itself, Lady & Kid and Accor (where the ECJ has chosen the cause of action which should be used, defined the test for causation in that cause of action, and specified exhaustively and in detail which defences are available).185 Nevertheless, even if the ECJ chose to specify See the text accompanying nn 11–35. See nn 1–3 and accompanying text. 174 See the text accompanying nn 66–83. 175 See the text accompanying nn 52–53. 176 See the text accompanying nn 19–35. 177 See the text accompanying nn 54 –55. 178 See the text accompanying n 56. 179 See the text accompanying nn 57–58. 180 See the text accompanying nn 60–61. 181 See the text accompanying nn 69–83. 182 See the text accompanying nn 84 –85. 183 See the text accompanying n 43. 184 See the text accompanying n 140. 185 See the text accompanying nn 141–56. 172 173

Overpaid Taxes: A Hybrid Approach 47

in more detail how the claim should operate at national level, there are indications in its ruling in Littlewoods that if the national courts were to hold that the hybrid cause of action proposed here were to constitute the only available route for recovery at national level, as I argue they should, this would not be incompatible with EU law in the way that some of the national legislative responses to DMG have been.186 The hybrid cause of action would then constitute not only a clear, principled and certain interpretation of the so-called ‘Woolwich cause of action’, and a means of allowing both the public and private aspects of such claims to be recognised, it would also, compatibly both with the intentions of the UK legislature and the requirements of EU law, provide a means of avoiding further costly litigation.

See the text accompanying nn 97–101.

186

3 Mistaken Overpayments of Tax DUNCAN SHEEHAN*

A. INTRODUCTION

This chapter explores the effects of a mistaken overpayment of tax. The overpayment may come about, for example, because of the invalidity of the taxing legislation or a misinterpretation of it, but the reason for the overpayment does not matter. The chapter aims to substantiate the view that there is, and should be, no important distinction drawn between mistaken overpayments of tax and mistaken overpayments of grocery bills. The chapter is divided into four main parts. In Part B it is briefly explained why this should be treated as a private law question. The view is taken that a claim to recover mistaken tax should be a private law action, contrary to the view taken by Rebecca Williams that it should be a hybrid action.1 Part C assesses the scope of common law actions for mistake, and their relation to actions founded on the Woolwich case,2 which established an additional unjust factor of ultra vires demand or receipt. The co-existence of these two actions in domestic English law was largely accepted in Test Claimants in the FII Group Litigation v HMRC.3 The chapter also argues that once the potentially retrospective nature of a mistake of law is grasped, there are situations where a claim for recovery of tax overpaid as a matter of EU law can only be grounded on mistake. The nature of a mistake of law also affects the subject addressed in Part D, which is whether a ‘settled understanding of the law’ defence should apply to overpaid tax cases even if it does not in other cases. The settled understanding defence may not be a true defence, but simply a function of how we conceptualise mistake. It has been suggested that the executive should be able to avail itself of defences that are not open to private parties. In principle, however, the executive should not be entitled to the protection of special defences. The details of this are not considered here; defences, including change of posi* Thanks to Steven Elliott, Charles Mitchell and Rebecca Williams for their comments on different drafts, to One Essex Court and Pump Court Tax Chambers for supporting both the conference and my attendance, and to Steven Elliott and Eoin O’Dell for organising it. Thanks are finally due to Charles Mitchell, Birke Häcker and Steven Elliott for editing the final volume. 1 R Williams, Unjust Enrichment and Public Law (Oxford, Hart Publishing, 2010) 36–39. See also chapter 2 in the present volume. 2 Named for Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] AC 70; see in Ireland O’Rourke v Revenue Commissioners [1996] 2 IR 1; Bank of Ireland Trust Services Ltd v Revenue Commissioners [2002] 4 IR 178. 3 Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2012] UKSC 19, [2012] 2 AC 337 (hereafter ‘FII (SC)’).

50 Duncan Sheehan tion, are discussed in Niamh Cleary’s chapter in the present volume and elsewhere.4 Some comments will be made about the change of position defence in Part E, though, as the underlying concern of that defence, at least as it has been constructed in Ireland, is the same as that of the settled understanding of the law defence.

B. A PRIVATE LAW REMEDY

Leaving aside the question of whether there should be any special defences, or modifications of defences, in these cases, this part seeks to show that we ought to treat the recovery of mistaken overpayments of tax as a private law question. There are two reasons for this. The first is that the reason why the law gives relief to a claimant who mistakenly overpays his grocery bills is the same as the reason why the law gives relief in the case of mistaken payments of tax. The second is that the cases see the recovery of money mistakenly paid as tax as a private law question. Taking the first point, the executive is in the normal course of events subject to the same remedies of private law as the rest of us.5 If the reason we want to reverse a mistaken payment applies both where the defendant payee is a private person and where it is the state, then it seems plausible as a starting point that private law rules should apply, unless there is a good reason not to apply those rules.6 Typically we are told in the unjust enrichment textbooks that the reason for giving the claimant relief in the case of a mistaken payment is that the payor’s intention to pay is vitiated.7 It is not obvious that this is quite right. There is no doubt that if I intend to hand over the £50 notes, the fact that I thought I owed the money when I did not does not mean that my intention to reach into my pocket and hand over the money was non-existent. I intended to pay; I paid intentionally. It is more likely that I paid on condition of discharge of the debt, and the condition was unfulfilled.8 Nevertheless, it seems clear that the effect on my intention of making the payment by mistake cannot depend on the identity of my payee. Williams, however, argues that there are significant public law factors in play in Woolwich that cannot be accommodated in a wholly private law framework. The case involved the Woolwich Building Society making a payment to the Revenue of a disputed tax debt. They paid believing that the tax debt was not owed, and protested as much. There was therefore no argument of mistake to be run. Essentially, perhaps in part to preserve their reputation, they chose to pay first and argue later. The House of Lords held that if an ultra vires tax was demanded then it must be repaid. Williams’ explanation of liability in that case is that an ultra vires event occurred to which an action for judicial review was a potential response,9 4 See chapter 7. See too E Bant, ‘Change of Position as a Defence to Restitution of Unlawfully Exacted Tax’ [2012] Lloyd’s Maritime & Commercial Law Quarterly 122; J Lee, ‘Defences to Claims for Overpaid Tax’ [2012] Restitution Law Review 13. 5 Crown Proceedings Act 1949, s 21. 6 P Birks, ‘Restitution from the Executive: A Tercentenary Footnote to the Bill of Rights’ in PD Finn (ed), Essays on Restitution (Sydney, Law Book Co, 1990) 164. 7 eg GJ Virgo, The Principles of the Law of Restitution, 2nd edn (Oxford, OUP, 2006) 171; in the Irish context see E O’Dell, ‘Bricks and Stones and the Structure of the Law of Restitution’ (1998) 20 Dublin University Law Journal 101, 137–38. 8 CE Webb, ‘Intention, Mistake and Resulting Trusts’ in C Mitchell (ed), Constructive and Resulting Trusts (Oxford, Hart Publishing, 2010) 315. 9 Williams (n 1) 37; on the need for a judicial review see also Jones v Powys Local Health Board [2008] EWHC 2562 (Admin); Hemming v Westminster City Council [2012] EWHC 1260 (Admin).

Mistaken Overpayments of Tax 51

and indeed the building society had brought a successful action to strike down the secondary legislation, the Income Tax (Building Societies) Regulations 1986, under which the tax had been demanded.10 In Kingstreet Investments Ltd v New Brunswick (Department of Finance)11 the Supreme Court of Canada was clear that an action to recover payments made pursuant to ultra vires tax demands differed from the normal private law action and should be treated as a purely public law phenomenon.12 Williams suggests by contrast that by treating public law as the reason for restitution, the law gains full access to the crucial event, thus allowing a truly hybrid approach,13 neither fully public or private. She means that it is impossible to explain Woolwich neatly using only illegality, incapacity, transactional inequality, or any of the other private law bases canvassed by Lord Browne-Wilkinson in the case.14 All are in play and all are relevant. However, it is a hybrid action because private law supplies the rules by which it is determined whether the defendant’s enrichment was gained at the claimant’s expense and whether there are any relevant defences.15 Yet the result of the public law rule is nullity; that the money was not owed. In other cases the law of unjust enrichment is uninterested in why the money was not owed – only that it was not owed. The private law of unjust enrichment simply does not bite until nullity has been decided, and the source of the nullity should not be relevant. There might then seem to be little difference between Williams’ position and mine other than a relatively trivial labelling matter. However, this is not the case. She argues for a hierarchy of reasons, which would rank mistake lower than the Woolwich principle.16 We return to the question of concurrency of Woolwich and mistake in the next Part, but it seems deeply embedded as a matter of authority. Nonetheless she gives the example of a tax based on the number of red-haired employees. A company pays the tax believing it has red-haired employees. In fact it does not. There are two problems. First, they are mistaken as to the applicability of the regulations in their own terms, but also the regulations are silly and void. Williams argues that the latter point might be missed were the company to focus on the mistake as to the number of redheads it employs. It is in the public interest, she suggests, that this type of illegality should be uncovered, yet it is hardly for the law to remove a perfectly good cause of action that fulfils the claimants’ needs, and to force them to engage in a more complex and more expensive enquiry. It may be that this would dissuade the company from suing if it thought the amount at issue not worth the effort. Further, she argues that the availability of defences should be different in cases where public law is the reason for restitution in preference to others,17 although questions of defences are largely left to one side here. To illustrate the point made at the beginning that the courts see this as a private law issue, reference can be made to Investment Trust Companies (in liq) v HMRC,18 where 10 In R v Inland Revenue Commissioners, ex parte Woolwich Building Society [1990] 1 WLR 1400; the immediate response was the Finance Act 1991, s 53, which retrospectively validated those regulations against all other claimants, and which itself triggered litigation in the European Court of Human Rights, in National & Provincial Building Society v United Kingdom (1997) 25 EHRR 127. 11 Kingstreet Investments Ltd v New Brunswick (Department of Finance) 2007 SCC 1, [2007] 1 SCR 3. 12 See Robert Chambers’ discussion in chapter 15 of the present volume. See also R Williams, ‘Recovery of Ultra Vires Taxes: A Wholly Public Approach?’ [2007] Restitution Law Review 130. 13 Williams (n 1) 73. 14 Woolwich (n 2) 196–200. 15 Williams (n 1) 37–38. 16 ibid 72–74. 17 ibid 125. 18 Investment Trust Companies (in liq) v HM Revenue & Customs [2012] EWHC 458 (Ch) [81]–[87] (hereafter ‘ITC ’).

52 Duncan Sheehan Henderson J discussed the prerequisites of an action in the private law of unjust enrichment. Although he ultimately deferred the question of whether a mistake claim could coexist with a Woolwich claim, he also said that the prerequisites for a claim in mistake had been satisfied. In the Republic of Ireland, too, these questions are dealt with through the private law of unjust enrichment. In Re Article 26 and the Health (Amendment) Bill 2004 19 the Government proposed to retrospectively validate approximately 30 years’ worth of invalid health charges. The Irish Supreme Court said that the statute would abrogate a property right20 – a chose in action, being the right to recover the money in unjust enrichment. As there was an insufficiently extreme financial crisis to be avoided the statute was struck down as unconstitutional. That unjust enrichment right, however, they characterised as lying in mistake (although to avoid the mistake of law bar which still exists in Ireland they also characterised it as an involuntary payment).21

C. THE SCOPE OF THE CAUSE OF ACTION FOR MISTAKE

This Part is divided into three sections. First, we examine the scope of the common law action and its relationship with the Woolwich claim. The second section looks at the important decision in Deutsche Morgan Grenfell Group plc v IRC,22 and the third at different statutory schemes. We see that although mistake and Woolwich are largely concurrent, there is still scope for a mistake claim where the Woolwich principle does not operate.

(1) The Common Law Action for Mistake and the Woolwich Claim We can break this section into a number of subsections. The first subsection sets out the way in which a mistake might be expected to work. The second subsection shows that a mistake claim can and should be concurrent with a Woolwich claim. The EU law principles of effectiveness and equivalence require that in cases where a breach of EU law has been committed, a mistake claim must be available, although they may not entirely require that the rules be the same as for mistaken payments between private parties. The third subsection argues that the combination of the retrospective nature of mistake of law and limitation allows a mistake claim to reach further than Woolwich. (a) Mistake Basics The usual claim is that the payment of tax was made by mistake as to liability to pay; this will usually be a mistake as to law. In England the mistake of law bar, the rule that such payments are irrecoverable, was removed by Kleinwort Benson Ltd v Lincoln City Council.23 19 Re Article 26 and the Health (Amendment) (No 2) Bill 2004 [2005] IESC 7, [2005] 1 IR 105. See too E O’Dell and G Whyte, ‘Is this a Country for Old Men and Women?’ (2005) 27 Dublin University Law Journal 368. 20 ibid 205–7. 21 ibid 178. See too Rogers v Louth County Council [1981] IR 265, 271; Dolan v Neligan [1967] IR 247. 22 Deutsche Morgan Grenfell Group plc v Inland Revenue Commissioners [2006] UKHL 49, [2007] 1 AC 558 (hereafter ‘DMG (HL)’). 23 Kleinwort Benson Ltd v Lincoln City Council [1999] 2 AC 349 (HL).

Mistaken Overpayments of Tax 53

After the House of Lords’ decision in that case, any causal mistake of law24 will now count in England. Kleinwort Benson did not involve tax payments, but payments under an ultra vires interest rate swaps agreement. The payments were made prior to the decision in Hazell v Hammersmith and Fulham LBC,25 which decided that such contracts were void. The claimants were out of time and therefore sought to overturn the mistake of law bar to enable them to make use of the Limitation Act 1980, section 32(1)(c), which would enable them to start the clock running at the time the mistake was discovered or could reasonably be discovered.26 I have previously argued that the bank was mistaken, and that the majority decision is largely correct.27 Essentially, relying on Dworkin’s theory of law and adjudication, I argued that it is always possible for Hercules to show that the given proposition being tested was wrong and another right, or that the case on which the payor relied was mistaken at the time of payment. Further it should be possible for a litigant to go through the same reasoning process as the overruling court and realise that the given proposition tested was wrong. This view that retrospective mistakes are possible was reaffirmed in Deutsche Morgan Grenfell Group plc v IRC.28 Accepting that there is a mistake in cases where a taxpayer relies on a past decision that is subsequently overruled, a further question, deferred for now, is whether a ‘settled understanding of the law’ defence should be available. Deutsche Morgan Grenfell involved a decision by the claimant bank to pay Advance Corporation Tax (ACT), believing that it could not make a group income election, which would have delayed the point at which tax was due. In fact the rule that the claimant could not make the election was contrary to EU law, and was later overturned in Metallgesellschaft Ltd v IRC.29 However, the decision that the provision barring an election was illegal under EU law did not change the fact that no election was ever made. The tax was in fact payable, and this meant that a Woolwich claim could not have been available. Nevertheless, the House of Lords held that the claimant had made a mistake and that a claim on this ground would lie against the Revenue,30 the quantification of which was then the subject of further litigation.31 Because of the limitation issue, the finding that there was an operative mistake was essential to the decision. Without a finding of mistake, time would have run out for the claim. Some support was canvassed for a change in the structure of the law of restitution to a framework where proof of nullity of obligation would suffice to justify restitution, but it seems that this is not English law as it currently stands.32 The causation test is given by Barclays Bank Ltd v WJ Simms, Son and Cooke (Southern) Ltd [1980] QB 677. Hazell and others v Hammersmith and Fulham London Borough Council [1992] 2 AC 1 (HL). 26 When applying the s 32(1)(c) rule, care must be taken to pinpoint the relevant mistake, which must be a necessary part of the cause of action: FJ Chalke Ltd v HM Revenue & Customs [2010] EWCA Civ 313, [2010] STC 1640 [44] (hereafter, ‘Chalke (CA)’), confirmed in FII (SC) (n 3) [184] (Lord Sumption); see also Phillips-Higgins v Harper [1954] 1 QB 411. 27 D Sheehan, ‘What is a Mistake?’ (2000) 20 Legal Studies 538. For an opposing view, see A Nair ‘ “Mistakes of Law” and Legal Reasoning: Interpreting Kleinwort Benson v Lincoln City Council ’ in R Chambers, C Mitchell, and J Penner (eds), Philosophical Foundations of the Law of Unjust Enrichment (Oxford, OUP, 2009) 373. 28 DMG (HL) (n 22) 570 (Lord Hoffmann); accepted in FII (SC) (n 3) [167] (Lord Sumption). 29 Joined Cases C-397/98 and C-410/98 Metallgesellschaft Ltd v Inland Revenue Commissioners and Hoechst AG v Inland Revenue Commissioners [2001] ECR I-1727, [2001] Ch 620. 30 DMG (HL) (n 22). See too Boake Allen Ltd v HM Revenue & Customs [2006] EWCA Civ 25, [2006] STC 626 [160] (Mummery LJ). 31 Sempra Metals Ltd v Inland Revenue Commissioners [2007] UKHL 34, [2008] 1 AC 561. 32 For comment on the absence of basis structure for restitution see D Sheehan, ‘Unjust Factors or Restitution of Transfers Sine Causa’ (2008) Oxford University Comparative Law Forum 1 at: ouclf.iuscomp.org; L Farrell, ‘The Future of the Law of Restitution in Ireland – the Unjust Question’ (2008) 15 Commercial Law Practitioner 239; there is a discussion of the absence of basis point in FII (SC) (n 3) [188]–[190] (Lord Sumption). 24 25

54 Duncan Sheehan (b) Concurrency of the Actions It has been suggested in the foregoing subsection that there are some cases where a mistake claim is essential, and that without the mistake claim the taxpayer has no remedy at all. However, the question also arises whether a mistake claim should be available at all in cases where a Woolwich claim can be brought. Some courts and commentators would give a negative answer, and would say that if the latter claim lies, then that should be the only cause of action open to the claimant. This requires acceptance of two points. First, Woolwich must cover the factual area. Secondly, there must be good reasons why concurrency of actions should be barred. Taking the first point, Henderson J suggested in FII that mistake may now be the primary right of action in this area because of its wider ambit.33 He considered that mistake had a wider ambit because Woolwich required there to be an actual demand by the HMRC. The decision in FII involved the question whether the claimants could reclaim additional tax paid on overseas income which would have been tax free if earned in the UK, and whether they could recover ACT, which they had been unable to offset against mainstream UK corporation tax, because all or most of their profits had come from sources taxed overseas rather than in the UK. The claimant companies argued that ACT had effectively become an unlawful secondary or double tax which penalised companies who invested abroad. As ACT is not subject to an assessment or demand by the Revenue, if Henderson J had been correct then only a mistake claim would have been available. The Court of Appeal, however, rejected Henderson J’s contention that an unlawful demand was required for a Woolwich claim to lie,34 and held that such a claim will lie whenever the tax payments were not due, regardless of whether HMRC made any formal demand. This finding has now been affirmed by the Supreme Court.35 Of course this still means the tax must have been not due. Moving to the second point, in the Court of Appeal in Deutsche Morgan Grenfell Jonathan Parker LJ held that the only relief available to the claimant was under the Woolwich principle. No mistake claim was available where a Woolwich claim could be brought.36 Because of the wider ambit of the latter claim, a mistake claim was unnecessary. That attracted support from Rebecca Williams,37 for whom this was consistent with her view that the law operates a hierarchy of claims, but it attracted less support from other academic sources.38 A couple of later cases equivocated as regards EU-related claims. In FJ Chalke Ltd v HMRC,39 the claimant motor dealers had paid VAT, which it turned out was not payable under EU law. They accepted that the Value Added Tax Act 1994, sections 78 and 80 were exhaustive as regards the recovery of the principal, but sought compound interest at common law. They pursued their claim in mistake even after the Court of Appeal decision in 33 Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2008] EWHC 2893, [2009] STC 254 [260] (hereafter ‘FII (Ch)’). 34 Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2010] EWCA Civ 103, [2010] STC 1251 [152]–[174] (hereafter ‘FII (CA)’). On mistake see [170]. 35 FII (SC) (n 3) [71]–[84] (Lord Walker), [171]–[174] (Lord Sumption), and [221] (Lord Reed). 36 Inland Revenue Commissioners v Deutsche Morgan Grenfell Group plc [2005] EWCA Civ 78, [2006] Ch 242 (hereafter ‘DMG (CA)’). 37 R Williams, ‘The Beginnings of a Public Law of Unjust Enrichment’ (2005) 16 King’s College Law Journal 194. 38 eg AS Burrows, ‘Restitution in Respect of Mistakenly Paid Tax’ (2005) 121 LQR 540. 39 Chalke (CA) (n 26).

Mistaken Overpayments of Tax 55

the FII case. They lost on the limitation point. The case involved two mistakes – one as to the invalidity of the time limit imposed in VAT cases and one as to the availability of compound interest. Neither availed the claimants, as their only operative mistake was whether the VAT was payable, and they were already recovering that principal plus simple interest. At first instance Henderson J said it was no part of the law on limitation to allow time to start anew every time a new claim appears.40 Etherton LJ therefore declined to comment on appeal as to whether the claimants ought to have framed their claim as a Woolwich claim rather than mistake.41 In Littlewoods Retail Ltd v HMRC,42 Vos J then held that the sections were exhaustive, and only allowed simple interest. The question was therefore whether the EU requirement that there be an effective remedy required that compound interest be available. He suggested that if sections 78 and 80 were inconsistent with EU law then they should be disapplied only to allow the Woolwich claims, but not mistake. In short Vos J wanted to confine the claimants to the Woolwich action, but then equivocated by referring the question of the compatibility of this position with EU law to the ECJ. Advocate General Trstenjak gave her opinion on this in January 2012.43 Her view was that this was a matter of national procedural autonomy. Having decided that effectiveness was not an issue, she did not fully commit to a view as to the principle of equivalence.44 However, she said that if domestic law allowed a choice between two actions, so that a domestic claimant could elect for one with more advantageous procedural (or other) rules, EU law required that choice to be extended to EU-related claims, although it did not require that the best national rules always apply to EU cases. On this approach, Vos J could not disapply the sections only as regards one or other claim. However, were a claim for compound interest available, even if only for Woolwich claimants, they would be substantively better off than domestic claimants; this privileges EU claimants45 over domestic ones and there has been some disquiet at a European level about this substantive privileging of European rights. The principle of equivalence is, however, concerned with procedural equivalence,46 not equivalence of substantive outcome per se. Nor is the Advocate General without authoritative support. There is ECJ authority, albeit in the equal pay context, that in deciding by reference to what actions the equivalence comparison is to be made, the court can look more widely than the immediate factual context.47 This seems particularly 40 FJ Chalke Ltd v HM Revenue & Customs [2009] EWHC 952, [2009] 3 CMLR 14 [163] (hereafter ‘Chalke (Ch)’). 41 Chalke (CA) (n 26) [75]–[76]; he did not discuss the limitation point in detail, but upheld Henderson J at [44]. 42 Littlewoods Retail Ltd v HM Revenue & Customs [2010] EWHC 1071 (Ch), [2010] STC 2072 (hereafter ‘Littlewoods (Ch)’). 43 Case C-591/10 Littlewoods Retail Ltd v HM Revenue & Customs (ECJ, 19 July 2012),[2012] ECR I-000, [2012] STC 1714. 44 ibid [33] and [58]–[61]. See too M Schlote, ‘Littlewoods Retail Ltd v HMRC: Compound Interest as a Matter of EU Law’ [2012] British Tax Review 144, 150–51. 45 On the privileging issue, see P Craig and G de Burca, EU Law: Text, Cases and Materials, 5th edn (Oxford, OUP, 2011) 240. 46 R Schütze, European Constitutional Law (Cambridge, CUP, 2012) 382–83. 47 ibid 386–87; Case C-326/96 Levez v Jennings (Harlow Pools) Ltd [1998] ECR I-7835, but see Case C-231/96 Edilizia Industriale Siderurgica Srl (Edis) v Minsterio delle Finanze [1998] ECR I-4951; the test is whether the actions are similar as regards their purpose, cause of action and essential characteristics Case C-78/98 Preston v Wolverhampton Healthcare NHS Trust [2000] ECR I-3201, [88]. This is not the authority the Advocate General in fact relied on. She relied on Case C-233/08 Kyrian v Celnı´ úrˇad Tabor [2010] ECR I-177 [67], for example, which seems directed to the effectiveness of EU law and might rather support the proposition that the choice of action is needed to give full effect to EU rights.

56 Duncan Sheehan important where, as in equal pay, or VAT, the relevant rights are largely derived from EU law. This seems to be what the Advocate General did when she cast her net wide enough to include the more general question of concurrency of action. In the Investment Trust Companies case48 Henderson J explained how the mistake cause of action worked in the overpaid tax context, but deferred the question whether mistake could co-exist with Woolwich until after the Supreme Court decision in FII.49 That decision is now with us. It confirms that concurrent claims are a common feature of English private law;50 irrespective of any EU requirements, such as those considered in the previous paragraph, one would expect concurrency not to be a problem. Indeed, as Lord Hope said, it was common ground between the parties in FII that both mistake and Woolwich were available in domestic cases.51 The majority of the Supreme Court correctly held that Woolwich was not an effective remedy for EU purposes and that mistake should also be available. However, because of the division in the court on this question, a reference was made to the ECJ. Lord Reed suggested, like Trstenjak AG, that the EU principle of equivalence would be breached if the mistake claim were dropped because the claimant would not then have the same freedom of choice as in other domestic overpaid tax cases.52 Lord Hope had a similar view that the taxpayer is entitled to use the most advantageous remedy he can, and agreed that the principle of equivalence required both actions to be available.53 But Lord Sumption held otherwise.54 (c) The Combination of Limitation and Retrospective Mistake Although the causes of action are largely concurrent, there is an extra hurdle in a mistake claim, and so San Giorgio rights would appear to be vindicated in only some cases by a mistake claim.55 It might appear then that the Woolwich action is the most advantageous action. A mistake of law claim, however, attracts advantages under the Limitation Act 1980, section 32(1)(c). These advantages were removed retrospectively in the tax context by the Finance Act 2004, section 320, and the Finance Act 2007, section 107. In the lower courts in FII, the main challenge to the latter provisions was on the basis of effectiveness. Henderson J barred the Revenue from invoking the sections where the claim is one of mistake of law relating to EU law,56 in line with similar rulings related to the retrospective imposition of limitation periods in VAT cases.57 However, while accepting that sections 320 and 107 would be disapplied under EU law were a mistake claim required to give effect to San Giorgio, one of the consequences of the Court of Appeal’s decision that Woolwich covered the whole ground was that the sections did not need to be disapplied, because an adequate – and therefore effective in EU terms – remedy was available.58 If that were cor ITC (n 18) [74]–[87]. ibid [172]. 50 See for example the concurrency of claims in contract and tort in Henderson v Merrett Syndicates Ltd [1995] 2 AC 145 (HL). 51 FII (SC) (n 3) [5]. 52 ibid [220]. 53 ibid [16]–[21]; see also [114]–[115] (Lord Walker), [133] (Lord Clarke), and [140] (Lord Dyson). 54 ibid [142]. See too Lord Brown [123]. The terms of the reference are at [23] (Lord Hope). 55 Littlewoods (Ch) (n 42) [88]–[92] (Vos J). 56 M Chowdry, ‘The Revenue’s Response: A Time Bar on Claims’ (2005) 121 LQR 546; FII (Ch) (n 33) [427] (Henderson J). 57 Fleming v HM Revenue & Customs [2008] UKHL 2, [2008] 1 WLR 195; the Government responded by enacting the Finance Act 2008, s 121. 58 FII (CA) (n 34) [217]–[229]. 48 49

Mistaken Overpayments of Tax 57

rect, and the Supreme Court held it was not, although there has been a reference to the ECJ, then it would substantially reduce the advantages accruing to a mistake claim. In the Supreme Court, the curtailment was also attacked on the grounds of legitimate expectations. As Lord Walker explained, what this meant was that although the claimants were not entitled to the indefinite continuation of the entire panoply of remedies available at any given time, any curtailment would and could not retrospectively and without warning deprive a claimant of all its claims,59 when it had a legitimate expectation of having adequate notice of the law changing. It is important to stress that it was not the changed time limit itself that was objectionable under European law: very strict time limits are not a problem per se and have in the past been held acceptable under EU law.60 The problem rather lay in the retrospective nature of the changes and lack of adequate transitional provision.61 This was also relevant to the principle of effectiveness, as Lord Walker pointed out,62 and it was beside the point that the claimants might have had no complaint if section 32(1)(c) had never existed. Ultimately the Supreme Court agreed that section 107 should be struck down, but referred to the ECJ the question of the compatibility of section 320 with EU law, although the preferable view is that of Henderson J that both sections are incompatible. Two points should finally be made. First, but for sections 320 and 107, the Limitation Act 1980, section 32(1)(c) would allow otherwise concurrent mistake-based claims where the limitation period relating to Woolwich has expired and mistake is therefore the only San Giorgio claim applicable. This was a point made by Lord Reed in FII. He commented that Woolwich and the mistake claim are not identical. The mistake is an additional element, but allows the claimant to take advantage of the extended limitation period, unavailable to Woolwich claimants because mistake is not part of that cause of action.63 This works because of the peculiar nature of a mistake of law. Suppose that a taxpayer makes an overpayment in 2001. A decision that the ‘obligation’ under which the payment was made was void is handed down in 2006. In 2012 the taxpayer has a subsisting cause of action in mistake because time starts to run in 2006. For the purposes of a Woolwich claim, however, it starts to run in 2001 and so this claim is out of time. Lord Sumption put it as follows: If tax was overpaid under a mistake of law, then provided that a claim to recover it was brought before six years had elapsed from the judgment establishing the correct legal position, there was no limit upon how far back the claim could go. In the present cases, it goes back to the accession of the United Kingdom to the Common Market in 1973. If it had arisen from a mistake of purely domestic law, it might have gone back to the inception of corporation tax in 1965.64

A second point is this. At present sections 320 and 107 are disapplied in cases with an EU component. In those cases, therefore, the Limitation Act 1980, s 32(1)(c) still applies, but not in purely domestic cases, thus privileging the EU-related cases in terms of their substantive outcome. 59 FII (SC) (n 3) [111]–[115]; see also [15] (Lord Hope), [133]–[135] (Lord Clarke), [198] (Lord Sumption), and [241]–[245] (Lord Reed). 60 A Jones, Restitution and European Community Law (Oxford, Mansfield Press, 2000) 78. 61 Case C-62/00 Marks & Spencer plc v Customs & Excise Commissioners [2002] ECR I-6325, [2003] QB 866; M Chowdry, ‘Coulda Woulda Shoulda: Shortening Time Limits for the Recovery of Overpaid VAT’ [2006] EC Tax Review 226. 62 FII (SC) (n 3) [111]–[112]. 63 FII (SC) (n 3) [224]–[225]; on the requirement that mistake be a necessary part of the cause of action for section 32(1)(c) to operate see Phillips-Higgins v Harper [1954] 1 QB 411, affirmed in FII (SC) (n 3) [184] (Lord Sumption). 64 FII (SC) (n 3) [167].

58 Duncan Sheehan (2) Deutsche Morgan Grenfell v IRC The position so far is that English law allows the concurrency of Woolwich and mistake claims. The principle of equivalence requires that a taxpayer be no worse off procedurally in a case with an EU dimension than in one without such a dimension; mistake of law claims therefore run side-by-side with Woolwich and the principles of effectiveness and legitimate expectations require that the Limitation Act 1980, section 32(1)(c) should continue to be applicable, at least until adequate transitional arrangements for changes in the limitation periods are in place. Here we return to Deutsche Morgan Grenfell and deal with some of criticisms that have been levelled against the bank’s recovery in that case. It could not have fallen under the Taxes Management Act 1970, section 33, which provided that a person who has paid too much income or capital gains tax under a mistaken assessment could recover within six years, but that the Revenue may give such restitution as is just and equitable. The reason was a technical one relating to the mechanism for collection of ACT, which was not in the usual way subject to assessment.65 Lord Hope said the mistake in Deutsche Morgan Grenfell was a mistake ‘that group relief could not be claimed which led inevitably to the liability to pay ACT’.66 Robert Stevens has correctly pointed out that the money was actually due,67 something about which Lord Hope seemed unconcerned. Robert Chambers also points out that it is a case of restitution despite a valid obligation to pay.68 This means that the requirement to pay itself was not unlawful, and this means in turn that there could have been no Woolwich claim. This is important. Even though a demand is not a necessary condition for a Woolwich claim, such a claim must be precluded if a lawful demand has been made. Williams has suggested that all receipts to which the Revenue is not entitled are necessarily ultra vires the executive.69 However, this is not how the phrase is usually interpreted, and was not Jonathan Parker LJ’s view in the Court of Appeal: there he assumed that mistakes could be made justifying relief from the state even though the demand is lawful,70 Williams has, however, rejected the argument that there was a valid obligation to pay; she argues that a second mistake as to the lawfulness of the ACT was generated.71 The fact of the unlawfulness of the bar on election, according to Williams, meant that the particular demand for the ACT was unlawful, a position which Lord Hoffmann seems to have taken as well. The Court of Appeal even seems to have thought, albeit probably obiter, the whole regime to be illegal.72

65 Amendments to the 1970 Act made by the Finance Act 2009, s 100 and sch 52 have now removed the prerequisite for an assessment of tax before section 33 can apply. 66 DMG (HL) (n 22). 67 R Stevens, ‘Justified Enrichment’ (2005) 5 Oxford University Commonwealth Law Journal 141, 144–45. See too M Chowdry and C Mitchell, ‘Tax Legislation as a Justifying Factor’ [2005] Restitution Law Review 1, 16–18. But contrast Williams (n 1) 80–82. 68 R Chambers, ‘Deutsche Morgan Grenfell v IRC’ (2006) 6 Oxford University Commonwealth Law Journal 227, 234. 69 Williams (n 37) 198–99. 70 DMG (CA) (n 36) 316–17. 71 Williams (n 37) 199; but for the view that unlawfulness requires the demand to be ultra vires, see British Steel plc v Customs & Excise Commissioners [1997] 2 All ER 366, 376. 72 DMG (CA) (n 36) 320 (Jonathan Parker LJ); but see Test Claimants in the ACT Litigation v HM Revenue & Customs [2010] EWHC 359 (Ch), [2010] STC 1678 [2] (Henderson J).

Mistaken Overpayments of Tax 59

Stevens’ views are preferable. Williams’ views sit uneasily with the fact that only compound interest,73 and not the principal, was repayable. The tax itself did not have to be repaid. Also, the overturning of the rule that no election could be made makes it possible to say there was a mistake. It does not make it possible to re-characterise history by retrospectively creating an election that was never made. Williams does this by arguing that the form taken by the denial of the claimant’s EU rights was the charging of tax. This is because she reads the ECJ as saying that the breach of EU law was the denial of the benefit of the election.74 The language supports such a reading, but it is problematic if it implies the repayment of the principal, which might be required under some (but not these) circumstances. The form the denial took was in fact refusing to allow an election; the indirect consequence was the charge of tax. Allowing recovery meant allowing an election, which might not have been made. Lord Hoffmann also made this error when he said: ‘The mistake was about whether DMG was liable for ACT. The election provisions were purely machinery, which DMG would undoubtedly have used, by which it could enforce its right to exemption from liability’.75 His Lordship compounded the problem by saying that the mistake had a special nature, which impacted on the application of the Limitation Act 1980, section 32(1)(c).76 Time starts to run when the claimant could with reasonable diligence have discovered the mistake. This, he said, was impossible until the true state of affairs was there to be discovered. True, but this means as well that there was no mistake prior to the point when the true state of affairs appeared. Lord Brown, by contrast, took the view that DMG ceased to be acting under a mistake of law when they became aware of the fact that Metallgesellschaft and Hoechst were bringing proceedings to challenge the legislation.77 This is unlikely to be correct, but Lord Brown is right that the mistake became reasonably discoverable at that point. However, once this is understood the need for a special mistake of law limitation rule becomes unnecessary. The fact that, despite the assumption to the contrary by most academics and judges, a mistake might be reasonably discoverable before the overruling decision backdates the point at which section 32(1)(c) bites quite considerably. Returning to the question of what the mistake was, it was precisely these difficulties in finding an operative mistake that led Lord Scott to argue that the claimant’s money should be recoverable only if the EU law decision rendered the ACT unlawful.78 In his view, however, the tax was not unlawful, and so the only possible action open to the claimant was one for compensation for breach of EU law. That action was time barred, and in any case damages for breach of Community rights are only available in cases of a ‘sufficiently serious breach’.79 This will not always be so – even in cases where restitutionary remedies are 73 On the requirement under EU law that compound interest be available, see Chalke (Ch) (n 40) [108]–[116] (Henderson J); the matter was left hanging unsatisfactorily on appeal: Chalke (CA) (n 26) [41] (Etherton LJ). See too Sempra (n 31); M Dougan, ‘Cutting Your Losses in the Enforcement Deficit: A Community Right to the Recovery of Unlawfully Levied Charges’ [1998] Cambridge Yearbook of European Legal Studies 233, 260–62. 74 Williams (n 37) 198–99; Metallgesellschaft (n 29) 645; ACT could in some cases not be fully offset against mainstream corporation tax, even in domestic cases, and so it would presumably be possible for cases to emerge where the principal would have to be repaid because there would never have been an obligation to pay. In Metallgesellschaft the tax would have had to be paid later as Mainstream Corporation Tax (MCT); the obligation to pay always existed, and only the timing was at issue. These possibilities need clearly to be separated. 75 DMG (HL) (n 22) 572–73. Henderson J also took this approach in FII (Ch) (n 33) [261]. 76 ibid 572. 77 ibid 614–15; Williams (n 1) 94–97; on the nature of the mistake in DMG see ITC (n 18) [77]–[79] (Henderson J). 78 ibid 589–93. 79 Damages here are under the Factortame principle: FII (Ch) (n 33) [211] and [237]–[245]; FII (CA) (n 34) [196]–[206]; Test Claimants in the Thin Cap Litigation v HM Revenue & Customs [2009] EWHC 2908, [2010] STC

60 Duncan Sheehan required. Only by discovering a mistake, and declaring it operative as a necessary part of the cause of action, was the majority able to justify relief. In previous writing80 I have therefore implied that restitutionary recovery was unacceptable, although it may be possible to argue that, as the mistake removed the possibility of an election and forced a compulsory interest-free loan, an exception should be allowed. The Court of Appeal in FII took the view that there is an outright bar on restitution of tax lawfully due;81 the principal (and interest) was irrecoverable under Woolwich and so it is difficult to accept that a mistake claim is never required to satisfy a San Giorgio claim. There is one further difficulty. That is the causation question raised by Robert Chambers and Charles Mitchell. The argument is that if DMG had tried to pay and the IRC had known the true position, the system would have been changed. More importantly, if DMG had refused to pay and the IRC were mistaken then the IRC would have taken enforcement action. This forms the counterfactual proving that only where both parties are mistaken can the cause of action bite. Indeed one might argue that the IRC’s mistake is more important. If DMG refused to pay because they knew the position, they would find themselves having to pay anyway. Their mistake would not be a ‘but-for’ cause of the payment; it would have happened anyway.82 The argument can be countered in two ways. First, we might argue that the but-for test is inappropriate and that a test of ‘a cause’ might be more so. Secondly, the more important argument is that if the taxpayer disputes the payment, the IRC will seek to enforce and it is always possible that the taxpayer will be successful. It cannot be assumed that if DMG had resisted, knowing the truth, that they would have been unsuccessful. This is to confuse the question of whether the bank was mistaken with the question whether they would be successful in persuading a court to overrule the settled view, and further to assume they would not be successful in that. It confuses ultimately questions of what the law is with what it is thought to be.

(3) Relationship with Statutory Recovery Schemes Nobody argues that the regime for recovery of tax is not different from recovery of other overpayments. That is apparent to all;83 although the common law regime prima facie applies, there has been significant statutory intervention. For example, prior to its amendment by the Finance Act 2009, the Taxes Management Act 1970, section 33 provided that a person who has paid too much income or capital gains tax under a mistaken assessment can recover within six years, but that the Revenue may give such restitution as is just and equitable. That covered some ground, but has been criticised judicially as being inconsistent with the modern concentration on rights,84 and would not cover the case of a taxpayer

301; R Shiers and R Williams, ‘FII GLO (Chancery) and FJ Chalke: Tax and Restitution Developing Hand-inHand’ [2009] British Tax Review 365, 367–68. 80 Sheehan (n 32), text to n 53. 81 FII (CA) (n 34) [181]. 82 Chambers (n 68) 230–32; C Mitchell, ‘Restitution of Ultra Vires Tax Payments’ [2007] Restitution Law Review 123. 83 DMG (n 22) (HL) 567 (Lord Hoffmann). 84 Eagerpath Ltd v Edwards [2001] STC 26.

Mistaken Overpayments of Tax 61

– in a modification of Chase Manhattan Bank NA v Israel-British Bank (London) Ltd 85 – paying twice because he has forgotten he has paid already. This is because there is no mistake in the assessment in those cases.86 Consequently section 33 played only a small part in the FII litigation. Nonetheless it was relevant to the claim for repayment of tax charged under the Income and Corporation Taxes Act 1988, section 18. The Court of Appeal put a somewhat tortured construction on the section to make it compatible with EU law;87 essentially they said that it applied subject to the proviso that it would be limited by the UK’s treaty obligations, a proviso that would not apply if such a construction went against the grain of the legislation. The Supreme Court took an easier route, saying it was clear that the section could not be an exclusive right to recover tax overcharged contrary to EU law.88 However, where the Act does apply in purely domestic cases, it is exclusive. This is because the Act has a set of specific policies which are essential to its operation and not present at common law.89 The most important of these is the prevailing practice defence in section 33(2A), which has been there since the Act’s inception at a time before the mistake of law bar was removed. These statutory schemes are all unsatisfactorily inconsistent both with each other and the common law scheme.90 Some schemes have ‘passing on’ defences, such as the VAT Act 1994, section 80(3), which provides for a defence that the claimant taxpayer would be unjustly enriched by the repayment of VAT; others, like the 1970 Act, provide only a discretionary recovery regime. The statutory grounds for relief, whether for or against the state, need not displace the common law of mistake in its entirety.91 This depends on the terms of the statutory scheme. The Social Security Administration Act 1992, section 71, for example, excludes the common law rights of the Secretary of State for Work and Pensions to recover mistaken overpayment of benefits and so it covers the whole possible ground for recovery.92 There might, however, be inconsistency in some cases between the Taxes Management Act 1970, section 33 and the common law action for mistake, particularly in respect to limitation. Although these regimes will not operate on the same set of facts it is hard to identify a significant difference between cases where the 1970 Act applies and others where it does not. In the Court of Appeal in DMG Jonathan Parker LJ therefore described the possibility of different limitation periods as an unsatisfactory state of affairs.93 The common law action will allow the limitation period to be extended under the Limitation Act 1980, but this is not possible Chase Manhattan Bank NA v Israel-British Bank (London) Ltd [1981] Ch 102 (CA). FII (Ch) (n 33) [436] (Henderson J). 87 FII (CA) (n 34) [64]; C Mitchell, P Mitchell, and S Watterson (eds), Goff and Jones: The Law of Unjust Enrichment, 8th edn (Sweet and Maxwell, London, 2011) para 22.06; this is the Marleasing interpretation. See Case C-106/89 Marleasing SA v La Comercial Internacional de Alimentacion SA [1991] ECR I-4135. 88 FII (SC) (n 3) [10] (Lord Hope), [117]–[119] (Lord Walker), [127] (Lord Clarke), and [204]–[205] (Lord Sumption). 89 Monro v HM Revenue & Customs [2008] EWCA Civ 306, [2009] Ch 69; ITC (n 18) [100] (Henderson J); see M Jones, ‘Monro v HMRC: Defining the Scope of Section 33 TMA 1970’ [2008] British Tax Review 318; the VAT Act 1994, ss 78 and 80 are also exclusive: Littlewoods (Ch) (n 42) [62] (Vos J). 90 J Beatson, ‘Restitution of Taxes, Levies and Other Imposts: Defining the Scope of the Woolwich Principle’ (1993) 109 LQR 401, 430; Goff and Jones (n 87) para 22.02. 91 British Steel v Customs & Excise Commissioners [1997] 2 All ER 366; E O’Dell, ‘Restitution, Debt and Taxes: Woolwich in Ireland’ (2002) 24 Dublin University Law Journal 295. 92 R (Child Poverty Action Group) v Secretary of State for Work and Pensions [2010] UKSC 54, [2011] 2 AC 15; on recovery by public bodies, see generally C Mitchell, ‘Recovery of Ultra Vires Payments by Public Bodies’ [2010] Public Law 747; Goff and Jones (n 87) ch 23. 93 DMG (CA) (n 36) 316. 85 86

62 Duncan Sheehan under the 1970 Act, which also has a more restrictive three-year limitation period. This difference may still arise in domestic cases, where the Finance Act provisions, barring the application of section 32(1)(c), cannot themselves be disapplied. Additionally, and slightly tangentially, the Finance Act 2004, section 320 does not in its terms apply to mistakes of fact; the Limitation Act 1980, section 32(1)(c), which allows time to start running when the mistake was reasonably discoverable, still applies to mistake of fact cases in the tax context. This causes its own problems, not least that it perpetuates the need to distinguish between mistakes of law and fact that so bedevilled English law before Kleinwort Benson.94 That said, the need to draw the distinction remains in some cases of illegal contracts95 and in criminal law, so if there is a pressing need for making the distinction the difficulty in doing so should not be decisive. What is certain, however, is that there is scope and need for some rationalisation of the differing statutory schemes to iron out undesirable and arbitrary inconsistencies both between them and with the common law.

D. THE ‘SETTLED UNDERSTANDING OF THE LAW’ DEFENCE

We move in this, and in the final substantive part, to deal with possible defences that the state may have. Like anyone else the state will also have the benefit of limitation rules and such rules may provide the only way forward to protect the Revenue. The first question in this section is what the settled understanding of the law defence is and how it should be juridically understood. We then examine the views of the English and Scottish Law Commissions and whether the defence is a sound one.

(1) The Scope of the ‘Defence’ In Kleinwort Benson Lord Browne-Wilkinson96 and Lord Lloyd97 in the minority believed that the settled understanding of the law defence was no real defence at all, but simply a recognition that the claimants had no cause of action, because there is no mistake where a party relies on a view of the law that is only shown to be incorrect by a subsequent case. Over 50 years ago in a discussion of the New Zealand legislation, Cameron similarly commented that there was no mistake if one paid in reliance on a decision later overruled.98 Irish commentators also consider there is no such thing as retrospective mistake and that the minority in Kleinwort Benson were correct,99 despite the apparent judicial acceptance of D Sheehan, Mistakes of Law (Oxford University DPhil Thesis, 2002) 100–105. D Sheehan, ‘Reconsidering the Defence of Illegality in Unjust Enrichment’ [2009] Lloyd’s Maritime & Commercial Law Quarterly 319, 330–35. 96 Kleinwort (n 23) 359. 97 ibid 393–94. 98 BJ Cameron, ‘Payments Made under Mistake of Law’ [1959] New Zealand Law Journal 4, 5; see New Zealand Judicature Act 1908, s 94A. 99 E Gannon, ‘The Rise and Fall of the Mistake of Law Rule’ (2000) 3 Trinity College Law Review 73, 91; K Spencer, ‘Restitution of Monies Paid Pursuant to a Mistake of Law’ (2004) 12 Irish Student Law Review 153, 166; L Farrell, ‘The Overpaying Taxpayer and the Restitutionary Remedy’ (2007) 14 Commercial Law Practitioner 214, 218. 94 95

Mistaken Overpayments of Tax 63

retrospective mistakes in Ireland by Casey v Irish Sailors and Soldiers’ Land Trust.100 There the Irish Supreme Court held that rent paid by the plaintiffs from 1924 to 1933, when the trust’s demands for rent were declared illegal, were paid under mistake of law, despite the payments occurring prior to the decision the demand was unlawful. Giving the main majority speech in Kleinwort Benson, Lord Goff discussed the settled understanding defence, having also rejected prospective overruling.101 He said: I recognise that the law of restitution must embody specific defences which are concerned to protect the stability of closed transactions . . . But the proposed ‘settled understanding of the law defence’ is not, overtly, such a defence. It is based on the theory that a payment made on that basis is not made under a mistake at all. Once that reasoning is seen not to be correct, the basis for the proposed defence . . . is undermined.102

Lord Goff therefore focused on the question whether there is a mistake at all. Any defence properly so-called would have to rest on independent policy concerns. Those concerns are that there would be too great a scope for unravelling presumed settled transactions. He took the view that other defences available to the payee would solve any security of receipt issues.103 The presence of retrospective mistake has now been reaffirmed in Deutsche Morgan Grenfell. Indeed the idea of retrospective mistakes has never been seriously challenged by counsel in that or other subsequent cases. The House said in DMG that there had been a mistake, despite the payments taking place prior to the ECJ decision that the rules were partially unlawful. Lord Walker declared that any settled understanding defence, being based on independent policy issues, should be introduced by legislation.104 He referred approvingly to Lord Hoffmann’s discussion in Kleinwort Benson where his Lordship concluded that the adoption of the defence would be a legislative act, which the removal of the mistake of law bar was not.105 This seems right; there are no cases unequivocally in favour of such a settled understanding bar, which cannot be explained as applications of the general mistake of law bar.106 Lord Goff was, however, prepared to accept that there might be a distinction between private and public transactions. In tax cases ‘large numbers of taxpayers’ might be affected, meaning that a settled understanding defence might be appropriate for policy reasons107 – essentially because the volume of claims and the size of recoverable sums might otherwise be too great. However, that defence may only be available in mistake cases if Lord Hoffmann in Deutsche Morgan Grenfell v IRC 108 is right to argue that Lord Goff did not turn to the Woolwich action at that point in his judgment, but was looking purely at mistake. This is doubtful. Lord Goff had just distinguished between the Woolwich scheme and the ‘private transactions’ scheme. He then went back to talk about overpaid tax. It is more likely that he intended his settled understanding or prevailing practice defence to benefit claimants under Woolwich. The statement he made did not admit of qualification in the way of cause Casey v Irish Sailors and Soldiers’ Land Trust [1937] 1 IR 208. Kleinwort (n 23) 381–84. 102 ibid 382. 103 ibid 382–83; see also 414–15 (Lord Hope); Goff and Jones (n 87) para 9.83. 104 DMG (HL) (n 22) 609–10. See too DMG (CA) (n 36) 321 (Jonathan Parker LJ); and R Allen, ‘Restitution of Payments Made under a Mistake of Law and the Security of Completed Transactions’ (2004) 63 CLJ 27. 105 Kleinwort (n 23) 401. 106 See eg Derrick v Williams [1939] 2 All ER 559 (CA); Henderson v Folkestone Waterworks Co (1885) 1 TLR 329 (CA). 107 Kleinwort (n 23) 382, a passage discussed in FII (SC) (n 3) [166] (Lord Sumption). 108 DMG (HL) (n 22) 568. 100 101

64 Duncan Sheehan of action. Kleinwort Benson as a private transaction Lord Goff did exclude. Lord Goff did, however, suggest that the Law Commission re-examine the question of the impact of mistake of law claims on the executive with particular reference to the limitation point.109

(2) Experience of Other Jurisdictions In the UK, as we have seen, the Taxes Management Act 1970, section 33(2A) establishes a defence of prevailing practice into tax cases. In Monro v HM Revenue & Customs,110 Arden LJ said that the subsection fulfils a pressing need, because it ensures that past settled transactions are not susceptible to being reopened. That section may, however, after the Supreme Court decision in the FII case itself be susceptible to being side-stepped by recourse to the common law, albeit only in cases where EU law is engaged. In New Zealand there is a general defence. The New Zealand Judicature Act 1908, section 94A(2)111 provides that: Relief shall not be given in respect of any payment where the law at the time held or was commonly thought to hold that the payment was required or allowed, by reason only that the law has been changed or has ceased to be what it was commonly understood to be.

In Western Australia, the Law Reform (Property, Perpetuities and Succession) Act 1962, section 23(2) was enacted in identical terms to the New Zealand provision.112 This section was considered in Bell Bros Pty Ltd v Shire of Serpentine-Jarrahdale,113 where the claimant company carried on business as a road construction contractor. It sought quarrying licences from the defendant local authority, which was empowered by by-laws to charge fees for the licences, which the company paid. It transpired that the by-laws were ultra vires and the company tried to recover the fees. Negus J said that most ordinary individuals who read the by-laws would assume that they were valid, and therefore for the purposes of section 23 they had to be considered valid.114 The Full Court in Bell Bros Pty Ltd v Shire of Serpentine-Jarrahdale upheld this,115 and made only minor changes to Negus J’s formula. The case was appealed to the High Court of Australia116 where recovery was permitted as the money was demanded colore officii. Nothing was said in the High Court about section 23(2), which raises questions about its ultimate effectiveness if it can be sidestepped by resort to a different cause of action. The defence has not therefore caused any problems; in the 50 years since its enactment in New Zealand there has been no litigation. However, this may be in spite of what the Law Commission describes as its poor drafting and conception.117 After all, Negus J gave an unhelpful explanation of the Western Australian provi Kleinwort (n 23) 389. Monro (n 89) 73; see also DMG (CA) (n 36) 311 (Jonathan Parker LJ); Finance Act 2009, s 100 and sch 52 controversially left the defence intact. 111 Inserted by the New Zealand Judicature Amendment Act 1958, s 2. 112 Since re-enacted as Property Law Act 1969, s 124(2). 113 Bell Bros Pty Ltd v Shire of Serpentine-Jarrahdale [1969] WAR 104. 114 ibid 111. 115 Bell Bros Pty Ltd v Shire of Serpentine-Jarrahdale [1969] WAR 155, 158–60 (SCWA). 116 Bell Bros Pty Ltd v Shire of Serpentine-Jarrahdale (1969) 121 CLR 137 (HCA). 117 Law Commission, Restitution: Mistakes of Law and Ultra Vires Public Authority Receipts and Payments (Law Com No 227, 1994) para 5.11. 109 110

Mistaken Overpayments of Tax 65

sion, merely replacing ‘commonly understood’ with the equally unclear ‘understanding of the ordinary individual’.118

(3) The Debate in the UK (a) The Law Commissions The Law Commission proposed a statutory defence of a settled understanding of the law.119 Originally the Scottish Law Commission supported a similar statutory defence.120 It has since recanted121 and now believes that no such defence is necessary. That is, even though the position is slightly different in Scotland, cogent support for believing such a defence need not be introduced. In England, time does not run until the claimant discovered, or could reasonably have discovered the mistake. In EU-related mistake of law cases involving the Revenue, that rule should properly apply. We saw that Henderson J’s view that the extended limitation period ought to remain available is preferable to that of the Court of Appeal. In Scotland the limitation period runs from the date of payment for all cases whether private or public bodies are involved and whether mistake of law or fact. There is no delay of the point from which time begins to run in cases of mistake, only in those of misrepresentation.122 It was mainly the limitation issue that caused disquiet about Kleinwort Benson, and much less disquiet in Scotland,123 because of the lesser scope for reopening transactions. The Law Commission proposed the defence, because they did not believe that a payor who relied on a settled view subsequently declared incorrect was mistaken.124 They did, however, propose an exception to the general rule. The settled view of the law defence would not apply where the claim is for the repayment of sums paid under an ultra vires statutory instrument.125 On this basis, Bell Bros would be decided differently under the Law Commission proposals. This still, however, leaves considerable scope for the defence to operate.126 The Commission, in coming to their conclusion on the availability of a settled understanding of the law defence, relied on Derrick v Williams.127 There the claimant was killed in a road accident as a result of the fault of the defendant. At the time it was thought no damages for loss of expectation of life could be recovered. The claimant sued under the Law Reform (Miscellaneous Provisions) Act 1934 and later settled. After a House of Lords Bell Bros (n 113) 111. Law Com No 227 (n 117) paras 5.2–5.13. 120 Scottish Law Commission, Recovery of Benefits Conferred under Error of Law (Scot Law Com DP 95, 1993) vol 1 paras 2.122–2.123. 121 Scottish Law Commission, Judicial Abolition of the Error of Law Rule and its Aftermath (Scot Law Com DP 99, 1996) para 3.50; Scottish Law Commission, Report on Unjustified Enrichment, Error of Law and Public Authority Receipts and Disbursements (Scot Law Com No 169, 1999) paras 2.47–2.49. 122 Scot Law Com No 169 (n 121) paras 2.29–2.32; Prescription and Limitation (Scotland) Act 1973, s 6, sch 1 para 1. 123 Morgan Guaranty Trust Corporation of New York v Lothian Regional Council 1995 SLT 299 (Inner House) 320; Scot Law Com DP 99 (n 121) para 2.31. 124 Law Com No 227 (n 117) para 5.3, citing Scot Law Com DP 99 (n 121) para 2.116; see also Scot Law Com No 169 (n 121) para 3.20. 125 Law Com No 227 (n 117) para 10.24. 126 Virgo (n 7) 421. 127 Derrick v Williams [1939] 2 All ER 559 (CA). 118 119

66 Duncan Sheehan decision128 that damages could be recovered for loss of expectation of life he sued again. Greene MR accepted that there was a mistake and rested his judgment on questions of finality. He said: It would be an intolerable hardship on successful litigants if, in circumstances such as these, their opponents were entitled to harass them with further litigation because their view of the law had turned out to be wrong.129

This is clearly consistent with a settled understanding of the law defence, but in fact need reflect no such defence and Lord Browne-Wilkinson commented in Kleinwort Benson that the basis of the decision was unclear.130 It must first be remembered that at the time mistaken payments could only be recovered if the mistake was of fact. More importantly Derrick is equally consistent with a separate defence of settlement and there is no reason to believe that defence is not open to the executive.131 To illustrate the defence of settlement, reference can be made to Bank of Credit and Commerce International SA (in liq) v Ali (No 1),132 where the respondent bank had made a number of employees redundant and invited them to sign a form of release, by which they released all their claims against the bank, in consideration for extra payments on termination of their employment. The employees signed, accepting the extra payments in full and final settlement of all claims against the bank. It then became clear after Malik v Bank of Credit and Commerce International SA (in liq)133 that a claim could exist at common law for what have become known as ‘stigma damages’. The claimants in BCCI v Ali attempted to sue for such damages. The bank resisted, arguing that the claimants had released the claim. The claimants could have argued that common mistake of law as to the claim’s availability justified the nullity of the releases.134 However, the House of Lords held that the releases were in fact not drafted to include claims the parties had not conceived to be possible.135 This seems implausible. The phrase ‘all claims’ meant precisely that – although maybe impliedly limited to claims for financial loss arising out of the employment relationship. It is hard to see how much clearer it could have been, and Lord Hoffmann in dissent took this view 136 – perhaps mindful of the dangers of combining the majority decision with Kleinwort Benson. Thus the Law Commission relied on an inconclusive authority on a different point in adopting a settled understanding defence. (b) Conceptual (Dis)Advantages Peter Birks claimed that a major problem resulting from the decision in Kleinwort Benson would be the disruption of the judicial function as potential restitutionary claims inhibited

Rose v Ford [1937] AC 826 (HL). Derrick (n 127) 565. 130 Kleinwort (n 23) 359–60. 131 But see J Beatson, ‘Unlawful Statutes and Mistake of Law: Is there a Smile on the Face of Schrödinger’s Cat?’ in A Burrows and Lord Rodger (eds), Mapping the Law (Oxford, OUP, 2006) 163, 172; Woolwich (n 2) 174 (Lord Goff) 132 Bank of Credit and Commerce International SA (in liq) v Ali (No 1) [2001] UKHL 8, [2002] 1 AC 251. 133 Malik v Bank of Credit and Commerce International SA (in liq) [1997] AC 20 (HL). 134 D Sheehan, ‘Unconscionability and Mistake in the Court of Appeal’ [2001] Journal of Business Law 107, 110–11. 135 K Wheat, ‘Bank of Credit and Commerce International v Ali’ (2002) 65 MLR 425. 136 Ali (n 132) 271. 128 129

Mistaken Overpayments of Tax 67

judicial creativity.137 Essentially his point was that when a case is overruled the courts would worry about a risk of floods of actions to recover money, or rescind contracts. Judges would be reluctant to overrule decisions. However, because of the cost few litigators are likely to be employed to pore over judgments to see if their clients can recover.138 The forecast flood of cases has therefore never arrived. The argument goes as well that the possibility of a flood of cases might also affect public finances. It was in fact as a direct response to the first instance decision in Deutsche Morgan Grenfell139 that the Finance Act 2004, section 320 was passed. One obvious problem with the formulation of the defence is reflected in the following question. How many people must believe something to be the law, before it becomes a settled or common understanding? If the ‘settled’ view of the law has changed through litigation it cannot have been a universally held view. Somebody had to believe the law to be different in order to challenge the prior understanding of it. How many people must want to challenge it before it ceases to be commonly understood? What effect does an unsettled view of the law have? The answers to these questions, as the Scottish Law Commission has observed, are obscure.140 The Law Commission was aware of the problem, but unfortunately was content to leave the matter to the judiciary.141 The judiciary, however, have similar concerns. Lord Goff referred to this obscurity about the ‘settledness’ of settled understandings as one of the reasons for his rejection of the defence.142 It would simply cause too much uncertainty. The New South Wales Law Revision Commission, when considering the area, suggested a change in the law defence because of the obscurity of the answer to the question how settled the understanding must be.143 However, such a defence is redundant; if there is a change in the law there is no mistake and no cause of action.144 In cases where no previous decision needed to be overruled, there is another objection to the defence, which is that there would in effect be an additional source of law, namely the common understanding of the law. Lord Browne-Wilkinson acknowledged the point; he accepted the settled understanding defence in Kleinwort Benson, but admitted that it would be wrong to say that a uniform practice of the profession made the law.145 The minority confused the question of what the law was with the question of what the law was thought to be. The law actually was different; there was a pre-existing right answer and that answer was recognised where it had not been before.146 These difficulties of course apply just as much to the more limited defence in the tax context as to the more general defence.

P Birks, ‘Mistakes of Law’ (2000) 53 Current Legal Problems 205, 215–16. C Lamont, ‘The Abolition of the Mistake of Law Bar and its Significance for Landlord and Tenant Practitioners’ [1999] Landlord & Tenant Review 12, 13. 139 DMG [2003] EWHC 1779, [2003] STC 1017. 140 Scot Law Com No 169 (n 121) paras 3.34–3.37. 141 J Beatson, ‘Mistakes of Law and Ultra Vires Public Authority Receipts: The Law Commission Report’ [1995] Restitution Law Review 280, 284–86. Beatson, as a Law Commissioner, helped produce the report. See also G Virgo, ‘Striking the Balance in the Law of Restitution’ [1995] Lloyd’s Maritime & Commercial Law Quarterly 362, 365–66. 142 Kleinwort (n 23) 383–84. 143 New South Wales Law Revision Commission, Restitution of Benefits Conferred under Mistake of Law (LRC 53, 1987) paras 5.25–5.28. 144 Scot Law Com No 169 (n 121) para 3.33; Scot Law Com DP 95 (n 121) para 2.6. 145 Kleinwort (n 23) 363. 146 Sheehan (n 94) 180. 137 138

68 Duncan Sheehan There may be other problems as well. The principle of effectiveness puts limits on the Taxes Management Act 1970, section 33(2A),147 and requires its disapplication where directly effective Community rights – to the repayment of overpaid taxes, for example – are at issue. It will also prevent the application of any general common law or statutory defence in EU-related cases, but will not have any impact on purely domestic cases.148 The European Court of Justice itself decides the temporal effect of its own judgments. It has in the past limited the retrospective nature of its judgments,149 but the ECJ does not permit national courts or legislatures to limit the effect of its judgments. Concerns have, however, as we have seen, been expressed at European level that developments in EU remedial law should not substantively privilege EU rights over purely national ones, a development that we may well be seeing in this area where national statutes, such as the Taxes Management Act 1970, section 33(2A), the Finance Act 2004, s 320, and the Finance Act 2007, section 107 are at issue.150 While the principle of equivalence is strictly about procedural equivalence not equivalence of outcome, where the national and EU rights are essentially the same, it raises legitimate questions of consistency. A possible defence the applicability of which at common law is at best controversial should not therefore be introduced if it is definitely inapplicable in EU cases.

E. CHANGE OF POSITION

Some brief comments are made here about change of position because of the similarity between the values underlying one proposed version of this defence and the settled understanding defence. It is worth bearing in mind that it is open to question whether change of position should be available to public authorities at all. One argument is that only authorities of relatively limited means will be able to show the required reliance for the defence. The state will always, however, spend the money raised in taxes; taxes are raised for no other reason than to fund expenditure. Birks and Virgo have argued, writing separately, that where the government spends on its normal programmes, we can conclude that if it did not spend this money it would have spent other money instead and therefore does not change its position in reliance on the receipt.151 That is not a given, however. It may be that a government facing a deficit would impose cuts in spending. It may be possible for the state to show that it would have made cuts if it did not have the tax receipts, and to order 147 M Bhandari and C Mitchell, ‘Lessons of the Metallgesellschaft Litigation’ [2008] Restitution Law Review 1, 17; Case C-188/95 Fantask A/S and others v Industriministeriet (Erhvervsministeriet) [1997] ECR 1-6783; Chalke (Ch) (n 41) [176]; M Bhandari, ‘Recovering Mistakenly Paid Tax: Taxes Management Act 1970, section 33 and the Prevailing Practice Defence’ [2008] British Tax Review 335; FII (CA) (n 34) [256]. 148 Monro (n 89) 88 (Arden LJ); this will also be true in Ireland despite the support for such a general defence in the academic commentary. 149 Jones (n 60) 86–87; Case 149/77 Defrenne v Sabena [1978] ECR 1365; see Murphy v Attorney General [1982] IR 241 for the use of this idea in Irish law. 150 R Craufurd Smith, ‘Remedies for Breaches of EU Law in National Courts: Legal Variation and Selection’ in P Craig and G de Burca (eds), The Evolution of EU Law (Oxford, OUP, 1999) 287, 297–98; W van Gerven, ‘Bridging the Gap between Community and National Laws: Towards a Principle of Homogeneity in the Field of Legal Remedies’ (1995) 32 Common Market Law Review 679, 700–702. 151 G Virgo, ‘The Law of Taxation Is No Island – Overpaid Taxes and the Law of Restitution’ [1993] British Tax Review 443, 459–60; Birks (n 6) 201; Williams (n 1) 142–45. See too Niamh Cleary’s discussion in chapter 7 of this volume.

Mistaken Overpayments of Tax 69

repayment would force it to borrow money – maybe at a higher rate of interest. Factually, however, this will be exceedingly hard to demonstrate. In terms of the substance, in Murphy v AG 152 parts of the Irish income tax code relating to married women were struck down as unconstitutional. The overpaid tax was recoverable as demanded colore officii.153 However, the Supreme Court allowed an expansive interpretation of change of position to provide the state with a defence. Henchy J said that the claimant’s rights to recover the tax only arose in the year in which they objected – and no taxpayer who did not object could recover. Up to that point the state was entitled to assume that the tax was valid and could be collected and spent. In effect the state was entitled to different defences in each year. Henchy J buttressed his argument by saying that in each year the taxpayers are different and at the end of the year the state is entitled to be able to close its books. If restitution was allowed it could only be effected by imposing fresh taxation on a new set of taxpayers.154 This is a similar principle – a presumption of validity155 – to that used by Negus J in Bell Bros, although references to that case were confined to the High Court decision that the payments were colore officii. In O’Rourke v Revenue Commissioners 156 Keane J argued that because the numbers of claimants affected by the decision were very small – only 80–90 people were affected by a misinterpretation by the Revenue Commissioners of a valid statute – the distortion of state finances was minimal. He therefore distinguished Murphy,157 and rejected the defence of change of position. Henchy J then had used the policies of fiscal chaos and finality of transaction to construct a change of position defence. He did not therefore introduce a pure change of position defence,158 but one infused with the values of the settled understanding defence. In England, Henderson J said in FII 159 that where the claim is one that relates to charges imposed in violation of EU law, change of position is not available to the Revenue. This is so whether the claim is a Woolwich claim or a mistake-based claim if that is needed to ground recovery required by EU law.160 FII in fact involved a concession by HMRC that the defence of change of position was unavailable. Henderson J subsequently rejected the Revenue’s later attempts to argue that the defence might be available on a case-by-case basis in FJ Chalke Ltd v HMRC.161 Henderson J accepted the defence would not apply to a Woolwich claim, whether the unlawfulness derived from EU law or not, because the state would then be a wrongdoer. This seems an undesirable extension of our understanding of wrongdoer. The Woolwich claim is an unjust enrichment claim which is proven in the absence of the state’s having committed a wrong, and indeed, in the absence of any need for a demand by the executive, necessarily having done anything at all. However, in purely domestic mistake claims against the Revenue, Henderson J’s interpretation of the defence in FII is very wide; he argued that if everyone assumed the validity of the tax and the state therefore spent the money it would be inequitable to require the executive to repay the Murphy v Attorney General [1982] IR 241. ibid 317. 154 ibid 318–19. 155 Law Com No 227 (n 117) para 10.23. 156 O’Rourke v Revenue Commissioners [1996] 2 IR 1. 157 ibid 17–19. 158 Law Comm No 227 (n 117) para 11.14; Williams (n 1) 142–43. 159 FII (Ch) (n 33). 160 ibid [304] and [339]–[340]; a position accepted in FII (CA) (n 34) [191], and followed in Littlewoods (Ch) (n 42) [105] (Vos J); nothing was said about change of position in FII (SC) (n 3). 161 Chalke (Ch) (n 40) [173]–[176]; this was not addressed on appeal: Chalke (CA) (n 26) [74] (Etherton LJ). 152 153

70 Duncan Sheehan money.162 Common sense, he said, dictated that departmental budgets would be set on a yearly basis predicated on the availability of tax receipts which would be used up in a relatively short period. Two final points are worth making. First, although Henderson J did not, as Henchy J did, assume a year-on-year closure of budgets, the idea is much the same. If so, and Lord Goff’s settled understanding defence were introduced applying to both causes of action, there would be undesirable inconsistency in the defences’ application. In the form suggested by Henchy J and Henderson J, the two defences must therefore stand or fall together. As we have concluded that the settled understanding of the law defence should fall, so should change of position. Secondly, even aside from this, where change of position applies, Henderson J’s very wide formulation neuters the cause of action. That explains how the principle of effectiveness can disapply the defence in EU-related cases. If the defence is inapplicable in EU cases, it ought also to be so in domestic cases, at least in the very wide terms in which Henderson J framed it. If it is to be available, its width should be no greater than between private parties.

F. CONCLUSION

Despite the Woolwich claim in both England and Ireland there is still scope for mistake claims in the context of overpaid tax either concurrent with the Woolwich claim or as the sole available claim; indeed the nature of mistake of law makes this inevitable. Alongside this, the settled understanding of the law defence is unavailable as a matter of EU law, and to introduce it into purely domestic cases creates undesirable inconsistencies. Given that change of position in this context may rest on similar concerns, it too should not be introduced or widened to cases where it would not have been available to a private party. It may be thought that this will give rise to significant uncertainty for the Revenue, but the ECJ has indicated that there are no problems with very restrictive limitation periods, and a flat limitation period with no possibility of extension may be the best way forward, and it seems to be strengthened by EU law’s acceptance of the need to balance the interests of the state and taxpayers.163 This will leave the executive in much the same position as private payees subject to a mistake claim and as indicated at the start of this chapter, mistaken overpayments of tax should be treated no differently to those of grocery bills.

162 FII (Ch) (n 33) [344]; the applicability of change of position and questions of how departmental budgets would have to be affected for the Government to rely on change of position is discussed at Littlewoods (Ch) (n 42) [111]–[125] (Vos J). 163 Bhandari (n 147) 341.

4 Restitution from Public Authorities: Any Room for Duress? NELSON ENONCHONG

A. INTRODUCTION

Under European Union law, a person who has paid taxes or duties levied by a Member State in breach of European Union law is in principle entitled to recover the payment from the Member State (the San Giorgio principle).1 However, under English domestic law restitution for money unlawfully collected by a public authority or the state can be a matter of some complexity. Prior to the landmark decision of the House of Lords in Woolwich Equitable Building Society v IRC 2 the restitutionary remedies available at common law3 were rather limited.4 The payer could recover in an action for money had and received on the grounds of duress5 (including duress colore officii)6 or mistake of fact.7 In the Woolwich 1 See Case 199/82 Amministrazione delle Finanze dello Stato v SpA San Giorgio [1983] ECR 3595; Case C-264/08 Belgische Staat v Direct Parcel Distribution Belgium [2010] ECR I-731 [45]; and Case C-398/09 Lady & Kid A/S v Skatteministeriet (ECJ, 6 September 2011) [17]. See further chapter 10 of this volume. 2 Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] AC 70 (HL) (hereafter ‘Woolwich’). 3 This chapter is not concerned with recovery under statute. 4 A number of possible grounds of restitution remained insecure. For example, recovery on the ground of an implied contract to refund the money (as developed by Nolan J in Woolwich Equitable Building Society v Inland Revenue Commissioners [1989] 1 WLR 137 (Ch) 147, following Vaisey J’s decision in Sebel Products Ltd v Commissioners of Customs & Excise [1949] Ch 409) was not regarded as a sound proposition of law (Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] AC 70 (HL) 150–51 (Lord Keith)) and is now considered obsolete: Deutsche Morgan Grenfell Group plc v Inland Revenue Commissioners [2006] UKHL 49, [2007] 1 AC 558 (hereafter ‘DMG’) [133] (Lord Walker). 5 eg Maskell v Horner [1915] 3 KB 106. 6 The colore officii principle was considered by most of the judges in the Woolwich case as a species of duress: (n 4) 143 and 146 (Nolan J); (n 2) (CA) [1993] AC 70, 97 (Glidewell LJ), 131 (Ralph Gibson LJ), 139 (Butler Sloss LJ); (HL) [1993] AC 70, 161 (Lord Keith), 164 (Lord Goff) and 194 (Lord Jauncey). See also Sargood Brothers v The Commonwealth (1910) 11 CLR 258, 276. However, in Woolwich (at 201 and 204) Lord Slynn appeared to treat the colore officii principle as a ground of restitution separate from duress. See also Law Commission, Restitution: Mistakes of Law and Ultra Vires Public Authority Receipts and Payments (Law Com No 227, 1994) [6.3], [6.13]–[6.15]. 7 eg Meadows v Grand Junction Waterworks Co (1905) 21 TLR 538. There was no recovery on the ground of mistake of law. However, the general rule that money paid under mistake of law was irrecoverable was subject to an exception in the case of money paid to an officer of the court such as a trustee in bankruptcy: Ex parte James (1874) LR 9 Ch App 609; Ex parte Simmonds (1885) 16 QBD 308. In Blackpool and Fleetwood Tramroad Co v Bispham with Norbreck Urban District Council [1910] 1 KB 592 this exception was extended to allow a ratepayer who had overpaid to set off the overpayment against further liability for rates owed to the same rating authority. But in Reg v Tower Hamlet London Borough Council, ex parte Chetnik Ltd [1988] 1 AC 858 (HL) 877, Lord Bridge, whose judgment commanded the support of all the other members of the Appellate Committee, disapproved of

72 Nelson Enonchong case the House of Lords recognised a right of recovery on the ground that the payment was unlawfully exacted by a public authority (the Woolwich principle) and this was followed, a few years later, by recognition of mistake of law as a further ground of recovery.8 The emergence of the Woolwich principle (which allows recovery even where there is no duress or mistake) has resulted in uncertainty as to the status of duress as a ground of restitution from public authorities, since the Woolwich principle now performs the function which duress would have discharged in this context. It is not entirely clear whether duress, or at least duress colore officii, has been discarded as a ground of restitution. As the Law Commission observed in its report on Restitution: Mistakes of Law and Ultra Vires Public Authority Receipts and Payments, ‘[i]t is not clear whether the colore officii principle is now subsumed within the Woolwich principle, or whether it still has an independent existence’.9 Even if, in principle, duress survives as an independent ground of restitution some doubt whether it has any significance in practice.10 This chapter seeks to argue that, in principle, duress (including duress colore officii) survives as an independent ground of restitution from public authorities. It endeavours to show that although the Woolwich principle now covers much of the ground once occupied by duress, so that, to that extent, duress is no longer an important ground of restitution, duress continues to have some practical relevance because the Woolwich principle does not embrace all the cases to which duress is applicable. As there are cases falling beyond the scope of the Woolwich principle but within the sphere of duress, the distinction between the two grounds of restitution should continue to be maintained. The discussion is divided into five main parts. After this introduction, part B considers the nature of duress as a ground of restitution against public authorities as it had developed by the time of the decision in Woolwich. Part C examines the Woolwich principle and the main justifications advanced for it. In part D the limits of the Woolwich principle and the extent to which duress remains a relevant ground of restitution from public authorities are examined. The conclusions are stated in part E.

B. THE NATURE OF DURESS

(1) Duress in General In general, two requirements have long been regarded as necessary and sufficient for a claim based on duress; these are that (a) there must be pressure amounting to compulsion of the will of the complainant and (b) the pressure must be illegitimate.11 In the case of payments unlawfully collected by public authorities, the requirement that the pressure must this extension on the ground that it produced an anomaly because in effect it allowed a ratepayer to recover a mistakenly overpaid rate by way of set off against a subsequent rate liability, when the general rule precluded any direct right of recovery. 8 DMG (n 4); Kleinwort Benson Ltd v Lincoln City Council [1999] 2 AC 349 (HL). 9 Law Commission (n 6). See also C Mitchell, P Mitchell and S Watterson (eds), Goff & Jones: The Law of Unjust Enrichment, 8th edn (London, Sweet & Maxwell, 2011) 310. 10 eg A Burrows, The Law of Restitution, 3rd edn (London, Butterworths, 2011) 499, where duress as a ground of restitution from public authorities is dismissed as ‘presumably no longer of practical importance’. 11 Universe Tankships Inc of Monrovia v International Transport Workers’ Federation [1983] AC 366 (HL) 400.

Public Authorities: Room for Duress? 73

be illegitimate is normally satisfied because of the unlawfulness of the demand. For this reason, in such cases, once it is established that there was no lawful authority for the demand, the dispute is focused on the question whether the payment was made under pressure amounting to compulsion. The question is whether the pressure caused the victim to make the payment.12 In determining whether the pressure caused the complainant to make the payment, the court considers whether there was a practical alternative open to the complainant. In some cases the fact that the claimant protested at the time of payment was regarded as evidence of compulsion.13 However, it is now accepted that protest is neither necessary nor sufficient as evidence of compulsion.14 Traditionally restitution on the ground of duress has been confined to limited categories of compulsion which, as Lord Goff said, should not be regarded as closed.15 These include restitution for (a) payments made as a result of actual or threatened interference with a person, his goods or his economic interests (this is normally the situation in the case of horizontal claims between citizens), (b) payments made for the performance of a legal duty which the payee (who is not a public official) is bound to perform for nothing or for less than the amount demanded, and (c) payments made to a public official to obtain the performance of a legal duty to which the claimant is entitled for nothing or for less than the amount charged (the colore officii cases). Payments made to a public authority otherwise than to obtain the performance of a legal duty (such as payments made by way of tax) were recoverable under (a). In other words, in this category, the law made no distinction between duress by a citizen (in horizontal claims) and duress by a public authority or the state (in vertical claims). In discussing duress as a ground of restitution from public authorities, it is helpful to distinguish between cases where the payment is made to obtain the performance of a duty (the colore officii cases) and cases where the payment is made by way of tax or other levy because in the former the courts had recognised the need to protect the citizen by means of a broad concept of compulsion whereas in the latter, at the time of the Woolwich decision, the English courts were still attached to a narrow concept of compulsion. (2) Duress Colore Officii Prior to the decision in Woolwich it was settled law that a payment made to a public official to obtain performance of a legal duty to which the payor was entitled for nothing or for less than the amount charged was recoverable as money paid under duress. Examples include excessive charges paid to make use of pier facilities,16 excessive tolls levied by a turnpike keeper,17 excessive fees charged by a sheriff on executing a writ of fieri facias,18 payments 12 Dimskal Shipping Co SA v International Transport Workers’ Federation [1992] 2 AC 152 (HL) 165. Although the pressure must have caused the victim to make the payment, it need not have been the only reason for making the payment. See eg Barton v Armstrong [1976] AC 104 (PC) 121; Pao On v Lau Yiu Long [1980] AC 614 (PC) 635. 13 Hills v Street (1825) 5 Bing 36, 130 ER 973; cf Maskell (n 5). 14 See eg Sargood (n 6) 276; Kingstreet Investments Ltd v New Brunswick (Department of Finance) 2007 SCC 1, [2007] 1 SCR 3 [58]. See also Universe Tankships (n 11) 400. 15 Woolwich (n 2) 165. 16 Queens of the River Steamship Co Ltd v Conservators of the River Thames (1899) 15 TLR 474. 17 Parsons v Blandy (1810) Wight 22, 145 ER 1160. 18 Longdill v Jones (1816) 1 Stark 345, 171 ER 492. See also Jons v Perchard (1796) 2 Esp 507, 170 ER 432 (taking more than the legal fees to which he was entitled on giving bail); Lovell v Simpson (1800) 3 Esp 153, 170 ER 570 (taking more than was allowed on an arrest); Dew v Parsons (1819) 2 B & Ald 562, 106 ER 471 (excessive charge for

74 Nelson Enonchong made to a mayor in order to obtain a licence,19 payments made to a parish clerk in order to take extracts from a burials and baptisms register,20 excessive payments made to a returning officer by a candidate for parliamentary election,21 payment made to the Shipping Controller in order to obtain a licence to sell a British ship to a foreign purchaser22 and payment made to the Commissioners of Inland Revenue in order to obtain an excise licence.23 This category of duress has been referred to as ‘duress colore officii’,24 that is to say duress ‘by colour of office’.25 The words colore officii or, in some cases, colore officii sui were used in the old authorities in a derogatory sense to describe a wrongful act done under the cover of an office but which was actually not within the authority conferred on the office.26 The term was used as the opposite of the words virtute officii sui which were used in bonam partem to refer to cases where the act was done pursuant to the office and within the lawful authority conferred on the official.27 There is duress in the colore officii cases because there is illegitimate pressure in the form of a threat (whether express or implied) by the public official to refuse the service requested unless the claimant pays the money demanded. The threat amounts to compulsion of the will of the payer because the parties are not in a position of equality and the payer, who cannot otherwise obtain the service he requires, has no alternative but to pay the sums demanded. In Morgan v Palmer,28 for example, the plaintiff paid a sum of money to the Mayor of Great Yarmouth as a condition for the Mayor granting him his annual licence as a publican. It turned out that the payment was demanded without lawful authority. The court was unanimous in the view that the payment was not voluntary, that is to say, it was induced by duress. Abbott CJ said that the payment might have been voluntary if the parties were on equal terms. But where, as in this case, one party has the power of saying to the other, ‘that which you require shall not be done except upon the conditions which I choose to impose’, no person can contend that they stand upon anything like an equal footing.29 the issue of warrants). The report of the judgments in this case is very brief and none of the three judges who delivered judgments expressly put their decision on the basis of duress colore officii. But the idea is apparent in the judgments, especially that of Best J, who made it clear that the case was one where the payment was not voluntary. In Woolwich (n 2), Lord Keith (at 155A) and Lord Jauncey (at 180D–F) considered that Dew v Parsons was a case of a demand colore officii, although Lord Goff (at 166G) and Lord Browne-Wilkinson (at 197F) regarded the case as one where recovery was available on the ground that the payment was made without consideration. Lord Slynn expressed no view on the case. 19 Morgan v Palmer (1824) 2 B & C 729, 107 ER 554. 20 Steele v Williams (1853) 8 Ex 625, 155 ER 1502. 21 Martin v Tomkinson [1893] 2 QB 121. 22 T and J Brocklebank Ltd v The King [1924] 1 KB 647. 23 Malkin v The King [1906] 2 KB 886. 24 Woolwich (n 2) 159A (Lord Keith). 25 Mason v New South Wales (1959) 102 CLR 108, 140. See W Cornish, ‘ “Colour of Office”: Restitutionary Redress Against Public Authority’ (1987) 14 Journal of Malaysian and Comparative Law 41. 26 However, in some of the authorities the term colore officii was used in a positive sense in the construction of statutory provisions giving a public official protection from legal action. In this context if the public official acted within the powers conferred on his office he was said to have acted colore officii and so had the protection conferred by statute but if he acted outside the authority conferred on the office he was said not to have acted colore officii, and therefore did not enjoy the protection conferred by the relevant statute, and consequently an action for the refund of money collected by him could be maintained: eg Irving v Wilson (1791) 4 Term Rep 485, 100 ER 1132; Greenway v Hurd (1792) 4 Term Rep 553, 100 ER 1171; Waterhouse v Keen (1825) 4 B & C 200, 107 ER 1033. 27 Lewis v Maningham (1550) 1 Plowden 60, 68; 75 ER 96, 109, where Lord Montagu CJ likened conduct colore officii to extortion and even robbery. 28 Morgan (n 19). 29 ibid 735.

Public Authorities: Room for Duress? 75

Bayley J was entirely in agreement with the Lord Chief Justice that the payment was by no means voluntary. Littledale J also said that the payment was not voluntary because the plaintiff was merely passive, and submitted to pay the sum claimed, as he could not otherwise procure his licence.30 An express threat was not required. The demand for an unlawful payment was enough. The demand carried with it an implied threat that the service requested would not be performed if the payment was not made. The concept of implied threat applied even where the complainant made the payment after receiving the service, provided that the demand for payment had been made before the service was rendered. In Steele v Williams31 the plaintiff applied to the defendant, a parish clerk, for liberty to search the register of burials and baptisms. He told the clerk that he did not want certificates, but only to make extracts. The defendant said the charge would be the same whether he made extracts or had certificates. The plaintiff searched through the register and made extracts. The defendant then charged him for the extracts and he paid the charge. In the plaintiff’s action to recover the payment it was argued that the payment was voluntary since it was paid only after the extracts were made (after the service had been rendered), rather than to obtain a service, as in Morgan v Palmer. But that contention was rejected. Parke B said that on a true construction of the evidence, the payment in that case was not voluntary, because, in effect, the defendant told the plaintiff ‘that if he did not pay for certificates when he wanted to make extracts, he should not be permitted to search’.32 Platt B said that inasmuch as before the search began the defendant told the plaintiff that the charge would be the same whether he made extracts or had certificates ‘and under that pressure the extracts were obtained, and it would have been most dishonourable for the party, after having got the extracts, to refuse to pay, the money so obtained may be recovered back’.33 In the Woolwich case, Lord Goff said34 that both Parke and Platt BB treated this case as one where there was an implied threat by the defendant to deprive the plaintiff of his right to take extracts from the register for no charge. Since the threat was made before the plaintiff was allowed to search through the register it made no difference that the payment was made after the plaintiff had completed the search.35 However, in Twyford v Manchester Corporation36 the court took a different approach. The defendant corporation owned a cemetery and charged fees for permission to cut, recut, re-paint or re-gild inscriptions on stones or monuments in the cemetery. The plaintiff, a monumental mason, repeatedly went to the cemetery to do work of that nature and was charged the fees appropriate to each occasion, which he paid under protest. Eventually he brought an action for a declaration that the corporation was not entitled to make the charges and for recovery of the amounts which he had paid. It was held that the corporation was not entitled to charge fees for permission to re-cut, re-paint and re-gild inscriptions. Yet Romer J held that the plaintiff could not recover because the payment was voluntary. He said that the plaintiff did not suggest in evidence that the registrar of the ibid 739. See also Traherne v Gardner (1856) 5 El & Bl 914, 943; 119 ER 721, 732. Steele (n 20). 32 ibid 630. 33 ibid 631. 34 Woolwich (n 2) 167A–B. 35 cf Bell Bros Pty Ltd v Serpentine-Jarrahdale Shire (1969) 121 CLR 137, recovery allowed for unlawful payment demanded as a condition for a licence to excavate gravel although the money was paid after the licence had been given. 36 Twyford v Manchester Corporation [1946] Ch 236. 30 31

76 Nelson Enonchong cemetery threatened that, unless the plaintiff paid, he would be excluded from the burial ground in future, or that any other unpleasant result would follow. In other words, the judge assumed that there must be an express threat for a payment to be regarded as not voluntary.37 This decision is open to criticism because, as Lord Goff has pointed out, the court overlooked the point that the threat which constitutes compulsion may be an implied one.38 The accepted position is that in the colore officii cases a broad concept of duress is applied based on implied threats. As the headnote to the report of Steele v Williams39 states, ‘when money is paid under an illegal demand, colore officii, the payment can never be voluntary’. In effect, in the colore officii cases, the unlawful demand for payment gives rise to a presumption of duress.40

(3) Payments Made by Way of Tax or Similar Levies In relation to claims for the recovery of money paid by way of tax or similar levies the English courts were slow to apply the concept of implied threat or presumption of duress. However, some judges in other common law jurisdictions were prepared to move in that direction. (a) Narrow Concept of Duress in the English Authorities In the case of payments made by way of tax or similar levies recovery on the ground of duress was only available where the payment was made in response to a threat made by the public authority (other than a threat to institute legal proceedings). Where the payee had no power to enforce a demand for payment by extra-judicial measures, such as seizure of goods, and could only enforce payment by means of legal action which the claimant could resist, a threat of legal proceedings was not regarded as sufficient to give rise to a wellfounded fear that could amount to compulsion.41 The idea was that since the public authority could only enforce payment by legal proceedings, the payer could take the opportunity of challenging the legality of the demand by legal process. If he decided instead to give in to the unlawful demand, the payment was regarded as voluntary, so that recovery on the ground of duress was not available. In William Whiteley Ltd v The King 42 the restitution claim failed on this ground. The Inland Revenue had demanded duties which were not due. The payer knew that if he did not pay the duties the Inland Revenue could take no action except by legal proceedings and it was open to the payer in such proceedings to raise 37 A similar approach was adopted in Slater v Mayor of Burnley (1888) 59 LT 636. There the defendant public authority had power to cut off the complainant’s water supply for failure to pay the rates, but they had not done so, had not threatened to do so and had not taken legal proceedings to recover the rates. It was held that payment of excess rates by the complainant was voluntary in the circumstances. Wills J said (at 639) that ‘there was nothing in the nature of a threat used; it is simply the ordinary case of a person raising a contention when a demand is made upon him. This is not sufficient to constitute duress, so as to prevent a payment being a voluntary one’. 38 Woolwich (n 2) 165C. The same criticism applies to the decision in Slater (n 37). 39 Steele (n 20). 40 A similarly broad concept of duress was applied by the High Court of Australia in Bell Bros (n 35) 145 (Kitto J). See also Eadie v Township of Bradford [1967] SCR 573. 41 William Whiteley Ltd v The King (1909) 101 LT 741, 745. See also Werrin v The Commonwealth (1938) 59 CLR 150, 157–59 (Latham CJ); Mason (n 25) 135 (Menzies J), 144 (Windeyer J). 42 Whiteley (n 41).

Public Authorities: Room for Duress? 77

the question whether the duties were payable. Instead of refusing to pay and having the matter settled by legal proceedings, the payer decided to pay. Walton J dismissed the claim on the ground that the moneys were paid voluntarily and therefore could not be recovered back on the ground of duress. He concluded that ‘there was nothing in this case which amounted to compulsion’.43 Where the public authority had powers to enforce the unlawful demand for payment by summary measures and had actually exercised such powers or had threatened to do so before the payment was made, it was accepted that the payment was made under compulsion.44 However, where, although the public official had powers to enforce the unlawful demand by self-executing measures, the powers had not been used and no express threat to use them had been made, the English courts did not accept that a payment made in response to the demand could be regarded as made under compulsion based on the idea of implied threat. The implied threat in such a case is that the claimant paid because he anticipated the action which the public official would take (exercising his self-executing powers) if payment was not made. In other words, in such cases, although there is no express threat by the public official, there is an implied threat flowing from the provisions of the statute giving the official powers to take summary measures against the claimant. The case of Hooper v Corporation of Exeter 45 may be seen to have recognised the concept of implied threat based on the payee’s statutory powers of extra-judicial enforcement. The defendant corporation was empowered by statute to charge dues upon imported limestone. An exception was made in favour of limestone imported for the purpose of being burnt into lime. The relevant Act gave the Corporation a power of distress for non-payment. The plaintiff, a builder, was unaware of the exception and paid dues on limestone imported to be burnt into lime. The payment was made without any threat by the Corporation to use its power of immediate distress. The plaintiff simply did not know of the exception and therefore did not declare the purpose of the limestone imported. When he became aware of the exception he demanded a refund of the dues paid on limestone falling within the exception. It was argued for the plaintiff that the payment could not be considered a voluntary payment since the Corporation had received the payment whilst acting under a statute which gave it a power of absolute and immediate distress. It was held that the plaintiff was entitled to recover the money on the ground that, as Lord Coleridge CJ put it, ‘such payment could not be considered as voluntary’.46 It may be argued that since the statutory power of the Corporation was relied upon by counsel for the claimant in arguing that the payment was not voluntary, by accepting that the payment was not voluntary the court recognised the notion of implied compulsion by the statute. However, since in the brief report of the case there is no express reference by Lord Coleridge or Smith J to the defendant’s power of immediate distress, the real basis of the decision in that case remained unclear. It does not appear that subsequent English authorities considered the case as extending the concept of implied threat, which applied in the colore officii cases, to cases where the payment was made by way of tax or duty. 43 ibid 745. A similar position was adopted in New Zealand (Julian v Mayor of Auckland [1927] NZLR 453) and the High Court of Australia (in Werrin (n 41) especially 159). 44 eg Umphelby v M’Lean (1817) 1 B & Ald 42, 106 ER 16; Hills (n 13). In the case of a payment made for the release of goods seized it was a case of duress of goods; cf Maskell (n 5). 45 Hooper v Corporation of Exeter (1887) 56 LJQB 457. 46 ibid 458.

78 Nelson Enonchong (b) Broader Concept of Duress in Other Jurisdictions Whereas English law appeared to take a narrow view of duress, insisting on an express threat, judges in some common law jurisdictions were developing a wider concept of duress, based on implied compulsion, along the lines of that developed by the English courts in the colore officii cases. In the United States case of Atchison, Topeka & Santa Fe Railway Co v O’Connor,47 for example, it was held that the payment of unconstitutional tax by a company could not be deemed voluntary because failure to pay the tax would have exposed the company to the risk of having its contracts disputed, and its business injured by finding the tax greatly increased in case it finally had to pay. Holmes J said, in a passage which is often quoted, that when, as is common, the state has a more summary remedy, such as distress and the party indicates by protest that he is yielding to what he cannot prevent, courts sometimes perhaps have been a little too slow to recognise the implied duress under which payment is made. But even if the state is driven to an action, if at the same time the citizen is put at a serious disadvantage in the assertion of his legal, in this case of his constitutional, rights, by defence in the suit, justice may require that he should be at liberty to avoid those disadvantages by paying promptly and bringing suit on his side. He is entitled to assert his supposed right on reasonably equal terms.48

Holmes J not only recognised compulsion by the self-executing provisions of the statute, he went further and recognised compulsion due to enforcement by legal proceedings where the claimant would suffer serious disadvantages if he sought to assert his legal rights by defending such proceedings. Similarly, in Gaar, Scott & Co v Shannon49 the United States Supreme Court said that there could be implied compulsion where the statute contains self-executing provisions the application of which will cause the complainant serious disadvantage. Examples given by the court included a statute which declares that where a franchise tax is not paid by a certain date a penalty of 25 per cent shall be incurred, the licence of the company shall be cancelled and the right to sue shall be lost. It was stated that payment made to avoid such consequences was not voluntary but was a payment made ‘under compulsion of a statute, whose self-executing provisions amount to duress’.50 The broad concept of duress based on implied compulsion was also adopted by the High Court of Australia in Sargood Brothers v The Commonwealth.51 There the claim was to recover money unlawfully exacted from the claimants as duties of customs in respect of certain goods. The payment was demanded and paid without any threat from the Customs officials that the claimants’ goods would be detained if the payment was not made. However, the payment was made with knowledge on both sides that the Controller had the power to keep the claimant’s goods until the payment was made. The High Court of Australia allowed the claim on the ground that the money was paid under duress.52 O’Connor J said that the money was paid with the knowledge on both sides that Customs control over goods imported may be exercised in support of illegal as well as of legal demands of duty. He continued: Atchison, Topeka & Santa Fe Railway Co v O’Connor 223 US 280 (1911). ibid 285–86. 49 Gaar, Scott & Co v Shannon 223 US 468 (1911). 50 ibid 471. 51 Sargood (n 6). 52 ibid, Isaacs J dissenting. 47 48

Public Authorities: Room for Duress? 79 The principle of law applicable in such cases is well recognized. Where an officer of Government in the exercise of his office obtains payment of moneys as and for a charge which the law enables him to demand and enforce, such moneys may be recovered back from him if it should afterwards turn out that they were not legally payable even though no protest was made or question raised at the time of payment. Payments thus demanded colore officii are regarded by the law as being made under duress.53

In other words, since the claimants knew that the Customs Department had the power to detain the claimant’s goods if the money demanded was not paid, a payment made in response to a demand from Customs was presumed to be a payment made under compulsion. It was not necessary to show that the Customs officials had threatened to exercise their power. This presumption of compulsion is based on the idea that a reasonable person in the position of the claimant with knowledge of the powers of the public authority to take swift action will have a well-founded fear that failure to pay will result in the use of the powers which could cause the claimant significant damage.54 In Canada, a divergence of judicial opinion on the scope of the concept of duress emerged in Air Canada v British Columbia.55 In that case the majority of the Supreme Court of Canada, in holding that there was no right to recover tax paid under a statute later found to be unconstitutional, did not endorse the concept of compulsion by the statute itself. In addressing an argument that the tax had been paid under compulsion, La Forest J said that for a payment to be regarded as made under compulsion in this context, there must be some natural or threatened exercise of power possessed by the party receiving it over the person or property of the taxpayer for which the taxpayer had no immediate relief than to make the payment.56 However, in a much quoted dissenting judgment, Wilson J adopted a very broad view of compulsion. She said that payments made under unconstitutional legislation are not ‘voluntary’ in a sense which should prejudice the taxpayer. The taxpayer, assuming the validity of the statute as I believe it is entitled to do, considers itself obligated to pay. Citizens are expected to be law-abiding. They are expected to pay their taxes. Pay first and object later is the general rule. The payments are made pursuant to a perceived obligation to pay which results from the combined presumption of constitutional validity of duly enacted legislation and the holding out of such validity by the legislature. In such circumstances I consider it quite unrealistic to expect the taxpayer to make its payments ‘under protests’. Any taxpayer paying taxes exigible under a statute which it has no reason to believe or suspect is other than valid should be viewed as having paid pursuant to the statutory obligation to do so.57

This view was subsequently endorsed by the Supreme Court of Canada in Kingstreet Investments Ltd v New Brunswick (Finance).58 Wilson J’s concept of compulsion is not based on the summary powers of enforcement conferred by statute on the public authority collecting the payment. The compulsion arises simply from the fact that the taxes were exigible under the statute (later found to be invalid). It is not necessary that the collector ibid 276. As Higgins J said (at 309), the claimants ‘had to pay or they might lose their market’. 55 Air Canada v British Columbia (Attorney General) [1989] 1 SCR 1161; 59 DLR (4th) 161. 56 ibid 199. Following this approach, in Re Eurig Estate [1998] 2 SCR 565 [45], the Supreme Court of Canada allowed recovery of money collected by a public authority pursuant to ultra vires regulations because the claimant had challenged the validity of the regulation imposing the tax from the outset and she paid the fee in order to fulfil her legal obligations (as executor of the estate) only after a lower court held that the regulation was legally valid. 57 Air Canada (n 55) 169. 58 Kingstreet (n 14) [55] (Bastarache J), who delivered the judgment of the Court. 53 54

80 Nelson Enonchong should have had self-executing powers of enforcement. It is enough that the payment was made pursuant to a perceived obligation to pay. Therefore, by the time of the Woolwich decision an expanded concept of compulsion had received judicial expression in the United States of America, Australia and Canada. The House of Lords could have decided the Woolwich case by expanding the concept of duress. But the House opted instead to base its decision on a different and new principle.

C. THE WOOLWICH PRINCIPLE

In order fully to appreciate the scope of the Woolwich principle and therefore the extent to which it has affected duress as a ground of restitution against public authorities, it may be helpful first to identify the principle and the policy justifications advanced for it.

(1) The Principle In Woolwich Equitable Building Society v IRC 59 the claimant building society decided to challenge the validity of Regulations60 under which it had to pay certain tax to the Inland Revenue. However, to avoid the danger of being charged penalties and to avoid adverse publicity which might arise if it was the only building society failing to pay in accordance with the Regulations, it made payments to the Revenue as required by the Regulations. After it was held that the Regulations were ultra vires and void, the building society brought an action in restitution to recover the money paid with interest. One of the grounds of restitution advanced was that the money was paid under compulsion. In this case, in addition to enforcement through legal proceedings, the Inland Revenue had self-executing powers to enforce demands for payment. These included the powers to charge interest which was penal in its effect and to distrain. However, the Revenue had not threatened to use these powers. In the absence of any express threat from the Revenue the court could not find duress. Lord Slynn said that ‘There was no duress in a sense of an actual or threatened interference with the person or property of Woolwich . . . (though I am of the view that the notion of duress or coercion should not be narrowly confined)’.61 By a majority of three to two, the House of Lords, encouraged by academic commentary,62 preferred to allow the claim on the basis of a wider principle that money paid by a citizen to a public authority pursuant to an ultra vires demand by the public authority is prima facie recoverable as of right.63 This Woolwich principle is not concerned with the reason why the payment was made (whether duress or mistake). The simple fact that tax has been exacted unlawfully is enough to ground restitution.

Woolwich (n 2). The Income Tax (Building Societies) Regulations 1986 (SI 1986/482). 61 Woolwich (n 2) 201. 62 Cornish (n 25); P Birks, ‘Restitution from the Executive: A Tercentenary Footnote to the Bill of Rights’ in P Finn (ed), Essays on Restitution (Sydney, Law Book Co, 1990). 63 Woolwich (n 2) 177. 59 60

Public Authorities: Room for Duress? 81

(2) The Policy Grounds A number of reasons were advanced by the majority as the policy justifications for adopting the Woolwich principle. Amongst these is the inadequacy of duress as a ground of restitution in cases, such as the present, where there is a need to protect the citizen against the power of the state. Lord Goff said that a citizen to whom an unlawful demand for tax has been made is facing the Revenue armed with the coercive powers of the state, including a power to charge interest which is penal in its effect.64 He explained that when the Revenue makes a demand for tax that demand is implicitly backed by the coercive powers of the state and may well entail (as in the present case) unpleasant economic and social consequences if the taxpayer does not pay.65 Similarly, Lord Browne-Wilkinson expressed the view that the colore officii cases are merely examples of a wider principle that where parties are on an unequal footing so that money is paid by way of tax or other impost in pursuance of a demand by some public officer, these moneys are recoverable since the citizen is, in practice, unable to resist making the payment save at the risk of breaking the law or exposing himself to penalties or other disadvantages.66 He said that compulsion, arising from the inequalities and powers of the two parties (the citizen and the state), was one of the ‘sound reasons’ for recognising the wider principle.67 Lord Slynn also based his decision on the policy to protect against compulsion or pressure. He said that although the facts of the case did not fit easily into the existing category of duress or of claims colore officii, they shaded into them. There was ‘a common element of pressure which by analogy can be said to justify a claim for repayment’.68 Other policy reasons were advanced for the principle. These include the constitutional principle enshrined in article 4 of the Bill of Rights 1688, that taxes should not be levied without the authority of Parliament,69 the consideration that the law-abiding citizen, who pays his taxes pursuant to a perceived obligation to pay, should not be penalised,70 the public law principle of legality,71 the requirement of common justice,72 comparison with the position of the Crown that has the absolute right to recover ultra vires payments made out of the Consolidated Fund,73 and the desirability for English law to be aligned with EU law, under which restitution was already available for taxes collected under national legislation in breach of EU law.74 (3) Scope of the Principle The extent to which duress remains a relevant ground of restitution against public authorities depends on the scope of the Woolwich principle. The greater the extent to which the ibid 171. ibid 172. 66 ibid 198H. 67 ibid 198C. 68 ibid 204E. 69 ibid 172. 70 ibid 172. See also 175–76, with reference to the dissenting judgment of Wilson J in Air Canada (n 55). 71 Woolwich (n 2) 172, the need for a public authority not to retain the fruit of its unlawful action. 72 ibid 172. 73 ibid 177. 74 ibid. 64 65

82 Nelson Enonchong Woolwich principle applies to cases of duress the smaller the room left for duress. It is accepted that the Woolwich principle has a very wide scope. It extends beyond cases where the legislation imposing the tax is ultra vires to cases where the legislation is valid but the demand for tax is unlawful because it is based on a mistaken view of the facts.75 It allows restitution for unlawfully exacted tax whether or not the tax was paid as a result of compulsion or mistake. Moreover, after a period of doubt and uncertainty, the Supreme Court has now decided, in Test Claimants in the FII Group Litigation v HMRC,76 that a demand for payment is not a requirement for a Woolwich claim. The principle therefore applies to tax, such as advanced corporation tax, that is paid without a demand by the Revenue. In FII,77 when holding that a demand is not a requirement for the Woolwich principle, the Court of Appeal said that the underlying principle is that the Revenue should repay tax that has been ‘exacted’ without legal justification.78 In other words, for a Woolwich claim to succeed, it is sufficient that the state has exacted tax, which was not lawfully due, by voluntary compliance by the taxpayer with the legislative imposition of the tax.79 The problem with this formulation is that the requirement of tax ‘exacted’ does not take us very far from the requirement of tax ‘demanded’, since to exact payment is to demand and enforce payment or to insist on and obtain payment. Indeed in the Supreme Court Lord Walker said80 that the two words have much the same meaning. He even went on to say in parenthesis that ‘indeed, arguably “exaction” sounds rather more coercive’. Lord Walker, with whose judgment on this point the majority of the members of the Supreme Court agreed,81 restated the Woolwich principle so as ‘to cover all sums paid to a public authority in response to (and sufficiently causally connected with) an apparent statutory requirement to pay tax which (in fact and in law) is not lawfully due’.82 The difference between the formulation of the Court of Appeal in terms of payments ‘exacted’ and that of the Supreme Court is that the Woolwich principle restated by the Supreme Court is broader. For example, a Woolwich principle which applies only in the case of ‘exacted’ taxes, will not cover certain payments made as a result of the misapplication of a valid statute. Thus if the Revenue takes a mistaken view of the facts and makes a demand for tax in an amount which exceeds the amount for which the taxpayer is liable British Steel plc v Commissioners of Customs & Excise [1997] 2 All ER 366 (CA). Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2012] UKSC 19, [2012] 2 AC 337 (FII), affirming the decision of the Court of Appeal on this point: see [2010] EWCA Civ 103, [2010] STC 1251, distinguishing NEC Semi-conductors Ltd v HM Revenue & Customs [2006] EWCA Civ 25, [2006] STC 606. In the FII case the Court of Appeal said that the language used by the majority in Woolwich supports the view that a demand is not a requirement. However, it is doubtful whether the language of the majority unequivocally points in that direction. The Court of Appeal identified a number of passages in the judgment of Lord Goff in which he expressed the Woolwich principle in terms of ‘exaction of tax’, instead of demand of tax. But there are a number of passages in Lord Goff ’s judgment in which the principle is expressed in terms of payment made pursuant to a ‘demand’ (eg at 171F–G, 172B–C and 177F). Lords Browne-Wilkinson and Slynn also referred to payment made pursuant to ‘a demand’ (at 198C, 199D, 201D–E and 205H). In the Supreme Court Lord Walker noted (at [ 69]) the ‘frequent and consistent’ use of the word ‘demand’ in the speeches of the members in the House of Lords majority in Woolwich and said that the occasional variant uses of the word ‘exaction’ carried little weight since the two words have much the same meaning. 77 FII (CA) (n 76). 78 ibid [157]. 79 ibid [158]. 80 FII (SC) (n 76) [69]. 81 Only Lord Walker and Lord Sumption addressed the question whether a demand was a requirement for a Woolwich claim. Lord Sumption came to the same conclusion as Lord Walker that a demand was not a requirement for a Woolwich claim. On this point Lord Reed agreed with Lord Walker and Lord Hope, Lord Brown, Lord Clarke and Lord Dyson agreed with both Lord Walker and Lord Sumption. 82 ibid [79]. 75

76

Public Authorities: Room for Duress? 83

under the legislation, the Woolwich principle applies since the overpaid amount has been exacted by the Revenue through its demand. But if the taxpayer himself made a mistake and miscalculated the amount of the tax to be paid and paid it without any demand or assessment by the Revenue, it will be difficult to say that the overpayment was ‘exacted’ by the Revenue. A claim to recover it will therefore fall outside the Woolwich principle of the Court of Appeal.83 However, the claim will be covered by the Woolwich principle as reformulated by the Supreme Court since the overpayment was made in response to an apparent statutory requirement to pay tax which (in fact and in law) was not lawfully due.

C. POSSIBLE SPACE FOR DURESS

Since, as indicated above, the Woolwich principle has a very wide scope, which covers a large part of the cases that fall within the scope of duress, and since compulsion is not a requirement of the Woolwich principle, many claims will now be based on the Woolwich principle instead of duress. To that extent, the Woolwich principle now discharges the duty which was previously performed by duress. The question then is whether duress, especially duress colore officii, survives as a ground of restitution from public authorities. And, even if in principle duress survives as an independent ground of restitution, a separate question is whether it has any practical importance. In other words, is there any room left for duress? This part seeks to advance the view that duress survives as an independent ground of restitution so that a claimant may have concurrent causes of action. It also ventures to argue that, although in many cases where a claim in duress is available against a public authority the Woolwich principle will also apply, there are some cases to which duress is applicable but which fall outside the scope of the Woolwich principle. The distinction between the two grounds of restitution therefore remains significant in this class of cases.

(1) Duress Has Not Been Discarded by Woolwich It is now accepted that the decision in Woolwich was not intended to displace duress (or the other traditional grounds of restitution).84 In Woolwich, Lord Goff remarked85 that, ‘[w]e may expect that in any event the common law principles of compulsion, and indeed of mistake, will continue to develop in the future’. In the subsequent decision in Deutsche Morgan Grenfell Group plc v IRC 86 the House of Lords rejected a submission that the Woolwich principle was the only ground of restitution of overpaid taxes. The House of Lords went on to recognise mistake of law as a further ground of restitution. Lord Walker said87 that if the Woolwich principle was the only ground of restitution of overpaid taxes, it would exclude not only an alternative cause of action based on mistake of law, as contended for by the claimant in that case, but also one based on mistake of fact. He rejected that idea because it involved a departure from the well-established principle, which existed Although restitution may nevertheless be available on the ground of mistake. See also Law Commission (n 6) [6.3]. 85 Woolwich (n 2) 173. 86 DMG (n 4). 87 ibid [138]. 83 84

84 Nelson Enonchong before the Woolwich decision, that mistake of fact was a ground of restitution against public authorities. The same reasoning applies to duress. Similarly, Lord Hoffmann pointed out that in Woolwich Lord Goff provided88 a list of the then recognised grounds of restitution (which included duress). Lord Hoffmann then went on to state that the Woolwich decision itself added another.89 The Deutsche Morgan Grenfell case added yet a further ground of restitution (mistake of law). If the colore officii ground is but a species of duress then it survives the emergence of the Woolwich principle since duress continues to be available as an alternative ground of restitution. In a case where the facts allow the claimant to base his claim for restitution either on duress or on the Woolwich principle, it will be for the claimant to choose between the concurrent grounds of liability as best suits his interests.90 In such cases of concurrent causes of action it will normally be to the claimant’s advantage to base his claim on the Woolwich principle rather than duress (which requires compulsion). For this reason, it may be thought that duress no longer has any practical relevance as a ground of restitution against public authorities and does not merit separate treatment.91 That would be the position if the Woolwich principle is so wide that it extends to all cases covered by duress. However, it is open to doubt whether the Woolwich principle has such a wide scope.

(2) Payments Other Than Taxes and Similar Charges (a) Not Covered by Woolwich It is not entirely clear whether the Woolwich principle applies to any or all types of payment made to public authorities. The language of the majority in Woolwich suggests that it does not apply to all payments. Lord Slynn only referred to money paid by way of tax.92 Lord Browne-Wilkinson referred to payments made ‘by way of tax or other impost’.93 Lord Goff stated the principle as applying to money paid to public authorities ‘in the form of taxes or other levies’.94 The Law Commission took the view that although the language of the majority suggests a narrow scope for the principle, there is no positive indication that they intended the principle to have a narrow scope, as nothing turned on the boundary of the principle in the case itself. The Law Commission concluded that the Woolwich principle is not limited to payments of tax to Governmental or quasi-governmental bodies. They said, ‘[w]e believe that the crucial element is that the payment is collected by any person or body which is operating outside its statutory authority, that is, it is acting ultra vires’.95 In other words, the suggestion is that the Woolwich principle applies to all payments to public authorities.96 However, the courts do not appear to have accepted that the Woolwich principle has or was intended to have such a wide scope. In Waikato Regional Airport v Attorney Woolwich (n 2) 164–65. DMG (n 4) [21]. 90 Henderson v Merrett Syndicates Ltd [1995] 2 AC 145 (HL) 193–94; DMG (n 4) 607. 91 A Burrows, The Law of Restitution, 3rd edn (London, Butterworths, 2011) 266–67. 92 Woolwich (n 2) 201, 204. 93 ibid 198. 94 ibid 177. 95 Law Commission (n 6) [6.42]. 96 Assuming of course that the payment was collected without lawful authority. 88 89

Public Authorities: Room for Duress? 85

General,97 the Privy Council considered the issue of the precise limit of the Woolwich principle and observed that it was not clear whether Lord Goff intended his reference to ‘other levies’ to be limited to levies similar to taxation. In Kleinwort Benson Ltd v Lincoln City Council,98 Lord Goff again referred to the Woolwich principle as applying to ‘taxes and other similar charges’. In the Deutsche Morgan Grenfell case, Lord Walker said that that phrase was imprecise.99 He highlighted the difficulty of drawing a clear dividing line within a spectrum which stretches from central government taxes and duties through rates, community charge, drainage rates and charges, special levies and licence fees imposed by statute on different industrial and commercial activities, and charges made by statutory undertakers.

More recently in FII,100 when dealing with the question whether a demand for payment is a requirement for a Woolwich claim, Lord Walker again stated the Woolwich principle in terms that limit it to the payment of tax. He said that, [w]e should restate the Woolwich principle so as to cover all sums paid to a public authority in response to (and sufficiently causally connected with) an apparent statutory requirement to pay tax which (in fact and in law) is not lawfully due.101

He responded to counsel’s submission that to follow that course would lead to uncertainty as to what amounts to tax by saying the expression should be generously construed and continued, but there are bound to be borderline cases (the Foreign and Commonwealth Office is said to be engaged in a constant dialogue with foreign embassies in London as to whether the congestion charge is a tax). Borderline cases of that sort will arise whether or not a demand is needed.102

Although in both Deutsche Morgan Grenfell and FII Lord Walker did not provide a test by which to identify the dividing line, he recognised that a dividing line was to be drawn somewhere separating payments by way of tax from other payments to public authorities. In other words, Lord Walker did not regard the Woolwich principle as applying to all payments to public authorities. Some payments fall outside the scope of the principle. The difficulty is in identifying the precise point where the line is to be drawn. Once the dividing line is drawn, thus marking the limit of the Woolwich principle, recovery of payments falling outside that limit will have to be based on duress.103 In Waikato Regional Airport v Attorney General,104 for example, Wild J allowed a claim for the refund of money paid to the New Zealand Ministry of Agriculture and Forestry (the Ministry) on the ground that the money was exacted colore officii. The Ministry, in carrying out its responsibilities for border control at international airports under the Biosecurity Act 1933, decided to recover the costs of providing those services at regional airports approved under the 1933 Act as places of first arrival of international flights. At the same time border control services were being provided free of charge at the metropolitan [2003] UKPC 50, [2004] 3 NZLR 1. Kleinwort Benson (n 8) 381–83. 99 DMG (n 4) [140]. See also Lord Hope at [46]. 100 FII (SC) (n 76). 101 ibid [79]. 102 ibid [80]. 103 Or mistake. 104 Waikato (n 97) [79]. 97 98

86 Nelson Enonchong airports through a parliamentary approbation for border biosecurity services. From the outset the regional airports objected to the differential charging scheme. However, they paid the charges because the Ministry insisted that they were due and, as the Privy Council pointed out, the regional airports saw no alternative to paying. In proceedings brought by the regional airports to recover the charges paid, it was held that the charges were excessive and, to that extent, unlawful. Wild J considered both the Woolwich principle and the doctrine of money exacted colore officii. However, he allowed the claim for restitution of the excess charges on the ground of money exacted colore officii. The New Zealand Court of Appeal reversed the decision on the ground that the charges were not unlawful. Consequently, it expressed no view on the question of restitution. However, the Privy Council allowed an appeal from the decision of the Court of Appeal and restored the order of Wild J. In arriving at that result, the Privy Council considered the Woolwich principle and observed, as already indicated, that it was not clear whether Lord Goff intended his reference to ‘other levies’ to be limited to levies similar to tax. But the Privy Council noted that it was not suggested in argument that the charges constituted a tax. Since the payment in this case was not a tax or similar levy, it was beyond the reach of the Woolwich principle but within the scope of duress,105 and in particular duress colore officii. In England, the decision of the Court of Appeal in Norwich City Council v Stringer,106 supports the view that not every payment unlawfully collected by a public authority falls within the Woolwich principle. A local authority had made an overpayment of housing benefit to a landlord (on behalf of a tenant) and the landlord repaid the overpayment in response to a demand by the local authority for the refund of the overpayment. However, the demand did not comply with the relevant statutory requirements and so it was unlawful. The Court of Appeal held that the landlord was not entitled to recover the money repaid on the ground of the Woolwich principle. One reason for the decision that the claim fell outside the Woolwich principle was that the local authority’s demand for a refund was not a demand in the nature of a demand for tax which was backed by the coercive power of the state and which put the citizen at a disadvantage with the state. It was explained that the demand for payment in that case was different from a tax demand which a citizen is bound to pay on pain of penalty. However, although the payment fell outside the scope of the Woolwich principle, it was within the scope of duress. If the payment had been made because the local authority had made threats of unlawful action to back its unlawful demand for payment then it may have been possible to establish duress. The distinction between a claim based on duress and one based on the Woolwich principle may also be significant in terms of defences. This may be the case, for example, if the law were to recognise special defences to the Woolwich principle107 as a response to concerns about serious disruption to public finances resulting from Woolwich claims.108 Since such defences are likely to be confined to the recovery of tax or similar charges within the 105 In arriving at its decision the Privy Council also referred (at [80]) to the well-known case of South of Scotland Electricity Board v British Oxygen Co Ltd (No 2) [1959] 1 WLR 587 (HL). 106 Norwich City Council v Stringer (2001) 33 HLR 15 (CA). 107 eg a shorter limitation period, as suggested by Lord Goff in the Woolwich case: (n 2) 174, although he later (at 176) expressed doubt about ‘the advisability of imposing special limits upon recovery in the case of “unconstitutional or ultra vires levies”’. 108 Concerns expressed by the majority of the Supreme Court of Canada in Air Canada (n 55) 193 and by Isaacs J in Sargood (n 6) 303 and accepted by Lord Goff in Woolwich (n 2) 174. See also United States v Butler, 297 US 1 (1936). Contra Kingstreet (n 14) [25], [28]–[29]. In Waikato (n 97) [82] the Privy Council did not decide on the special defence since the amount at stake was not such as to present a risk of disruption of public finances.

Public Authorities: Room for Duress? 87

Woolwich principle, they may not apply to claims based on duress to recover money paid otherwise than by way of tax or similar charges. (b) The Experience in Canada The view that the Woolwich principle does not apply to all payments to public authorities is in line with the approach taken by the Supreme Court of Canada to the effect a Woolwichtype public law remedy is confined to restitution of taxes, so that the ordinary principles of unjust enrichment apply in other cases. In Kingstreet Investments Ltd v New Bruswick 109 the Supreme Court of Canada held that restitution for taxes collected without legal authority is based, as a public law remedy, on constitutional principles rather than the ordinary principles of unjust enrichment.110 The court said that the ordinary principles of unjust enrichment should not be applied to claims for the recovery of monies paid pursuant to a statute held to be unconstitutional.111 Therefore in Canada the ordinary grounds of unjust enrichment (such as duress or mistake) no longer apply to claims for the recovery of taxes collected pursuant to ultra vires statute.112 In the Kingstreet case, the claim was by taxpayers seeking restitution of user charges paid pursuant to ultra vires legislation. Bastarache J, who delivered the judgment of the Supreme Court of Canada, said that the doctrine of duress or compulsion was simply not applicable to cases of that kind. He said that the doctrine of protest and compulsion should be discarded in the context of claims to recover payments made to public authorities, whether pursuant to unconstitutional legislation or as the result of a misapplication of otherwise valid legislation.113 However, the Kingstreet doctrine is limited to claims for restitution of ‘taxes’ levied under an ultra vires statute. Therefore, as the Supreme Court of Canada has recently explained,114 Kingstreet leaves open the possibility of suing for unjust enrichment in other circumstances. Thus where a claim against a public authority is not for restitution of taxes paid under an ultra vires statute it can be advanced on the basis of unjust enrichment. In Alberta v Elder Advocates of Alberta Society,115 for example, the claimants alleged that the government had inflated accommodation charges required of elderly patients in long-term care facilities in order to subsidise medical expenses which were properly the responsibility of government. They claimed, inter alia, for unjust enrichment. The Supreme Court of Canada rejected a submission that Kingstreet precluded a claim for unjust enrichment against the government and allowed the claim to proceed to trial. Since duress is one of the grounds of recovery in an unjust enrichment claim, it remains available in cases outside the Kingstreet principle, where unjust enrichment claims against public authorities are possible as in the Elder Advocates case.

Kingstreet (n 14). ibid [40]. 111 ibid [39]. 112 See Alder, ‘Restitution in Public Law: Bearing the Cost of Unlawful State Action’ (2002) 2 Legal Studies 165, where it had been argued that restitution of money exacted ultra vires should be seen as an exclusively public law right. Contra, R Williams, Unjust Enrichment and Public Law (Oxford, Hart Publishing, 2010), who argues for a hybrid approach. 113 Kingstreet (n 14) [57]. 114 Alberta v Elder Advocates of Alberta Society 2011 SCC 24, [2011] 2 SCR 261 [91]. 115 ibid. 109 110

88 Nelson Enonchong (c) A Broad Concept of Duress in Future? If, as suggested above, the Woolwich principle does not extend to all payments made to public authorities, then in a case where a payment which falls outside the Woolwich principle was made for a service116 or some other performance of a legal duty, recovery will be based on the colore officii line of cases and therefore the broad concept of duress, which includes implied threats, will apply. However, in cases where the payment is not made for the performance of a legal duty, although recovery may be based on duress, it will not fall within the colore officii category of duress. In such a case, it is not clear whether the broad concept of duress, which is applicable in the colore officii cases, will also be applied. The majority in Woolwich117 were in favour of a broad concept of duress in such cases, although exactly how the concept of duress should be broadened in such cases was not specified. Three possible ways of developing duress into a wider concept may be identified. The first is duress based on an implied threat due to the self-executing powers of enforcement of the public authority (compulsion in the statute). On one view, there should be implied compulsion where the public authority had powers to take swift action. However, that view will be too broad and may not be justified in certain cases, such as where the claimant knew that it was highly unlikely that the public authority would exercise its powers. In Woolwich, for example, although the Revenue had such powers the court took the view that there was no basis for implied compulsion because the claimant could not be said to have made the payment as a result of a fear that the Revenue would exercise its power, which was ‘drastic and highly unusual’.118 Lord Goff explained that ‘since the possibility of distraint by the Revenue was very remote, the concept of compulsion will have to be stretched to the utmost to embrace the circumstances of such a case as this’.119 Lord Goff therefore appears to favour a broad concept of duress based on implied compulsion, but one where compulsion is implied only where there was a real possibility that the public authority would exercise its powers to take summary measures.120 Another approach to a broad concept of duress is the idea of compulsion due to serious disadvantages arising from defending legal proceedings commenced by the public authority to enforce payment or instituting legal proceedings to challenge the right of the public authority to the payment.121 In this case, the compulsion is not due to apprehension of the exercise of the self-executing power of enforcement conferred on the public authority. It is based on apprehension of the practical consequences which the claimant would suffer as a result of refusing to make the payment and instead engaging in legal proceedings either to defend an action by the authority or prosecuting its own proceedings to challenge the authority’s right to demand payment. If the adverse consequences of taking either step cf the charges in Waikato (n 97) [80]. Woolwich (n 2) 173, 198. See also 203 where Lord Slynn stated that he found Holmes J’s statement about implied duress to be ‘particularly persuasive when considering the principle of law to be applied’. The minority were unwilling to embrace Holmes J’s expanded concept of duress. Lord Keith (at 162) did not find in Holmes J’s statement any reasoning persuasive of a view contrary to that which he had formed in the light of the English and Scottish authorities. Lord Jauncey (at 192) said that Holmes J’s expanded concept of duress was inconsistent with the reasoning in the existing English authorities. 118 Woolwich (n 4) 146 (Nolan J). 119 Woolwich (n 2) 173. See also 119 (Ralph Gibson LJ). The decision in Henderson v Folkstone Waterworks Co (1885) 1 TLR 329 is also explicable on this ground. 120 Sargood (n 6) is perhaps an example of a case where compulsion was found because it was likely that the public authority would exercise its powers if payment was not made. 121 Atchison (n 47). 116 117

Public Authorities: Room for Duress? 89

would be serious, then that is considered to be a menace to the claimant’s interests which could amount to compulsion. Some may take the view that in Woolwich it was possible to find compulsion on this ground since the claimant could not refuse to pay without suffering serious damage. The payment was made to avoid serious damage to its reputation which it feared would result if it was the only building society that failed to pay the sums claimed by the Revenue. Moreover, if it failed to make the payment and challenged the Revenue’s right to demand it but ultimately failed in the legal action, it could be exposed to considerable liability to pay interests. Nolan J accepted that the fear of these adverse consequences was significant. He said that the scales in this respect were tilted heavily in favour of the Revenue. And he accepted that, ‘as a practical matter, Woolwich had little choice but to make the three payments’.122 In the House of Lords, Lords Goff and Browne-Wilkinson also accepted that in practice if Woolwich had refused to pay, and engaged in legal proceedings instead, it would have exposed itself to unpleasant economic consequences or other disadvantages.123 Yet the courts did not accept that the practical compulsion which was found was enough to constitute duress in that case. It may be that in future the courts will recognise serious disadvantages of the kind with which Woolwich was confronted, arising from a demand for payment by a public authority, as indicating that the claimant made the payment under implied duress. A third way of expanding the concept of compulsion in this context is to adopt the idea of compulsion arising from the obligation to obey the law.124 On this view there is compulsion where the claimant makes a payment because it considered itself obligated to pay under legislation which turns out to be invalid. However, such an expanded concept of compulsion may be considered to be too wide. In any event, where the claimant pays because he believed that the statute under which the payment was required was valid, a claim to recover the payment may also be based on mistake of law.

(3) Payment Made to a Person Who is Not a Public Official in Excess of What is Permitted by Statute As the Woolwich case was concerned with recovery from a public authority, the House of Lords did not address the question whether the principle applies equally to the recovery of payments made to private bodies and, if so, to which private bodies. There is much uncertainty on the point.125 The Law Commission has expressed the view that the principle should apply to payments collected by ‘any person or body’ which is operating outside its statutory authority.126 However, as the Scottish Law Commission observed, although some ibid. Woolwich (n 2) 171, 172 and 198. 124 See Air Canada (n 55) 169, Wilson J, quoted above in the text to n 57. 125 eg J Beatson, ‘Restitution of Taxes, Levies and Other Imposts: Defining the Extent of the Woolwich Principle’ (1993) 109 LQR 401, arguing that the Woolwich principle applies to taxes, duties and imposts levied ultra vires by (i) governmental bodies, (ii) other public bodies whose authority to charge is subject to public law principles and (iii) other bodies whose authority to charge is solely the product of statute. Contra, A Burrows, The Law of Restitution, 3rd edn (London, Butterworths, 2011) 508, where it is argued that the test should be whether the payment falls within the public law doctrine of ultra vires and hence susceptible to judicial review. If yes, the Woolwich principle applies. 126 Law Commission (n 6) [6.42]. 122 123

90 Nelson Enonchong of the justifications for the Woolwich principle apply well enough to powerful governmental bodies such as the Revenue, they may not be appropriate when applied to less powerful semi-public bodies and may appear absurd when applied to a small private trader.127 It is therefore not obvious that the Woolwich principle was intended to apply to payments made to private bodies or that it is appropriate to extend the principle to such payments.128 If the Woolwich principle is confined to restitution from public authorities, recovery of payments made to persons other than public authorities, including privatised companies providing utilities, in excess of the amount permitted by statute, will continue to be based on the ground of duress rather than the Woolwich principle. Lord Goff recognised this ground of recovery in Woolwich.129 He treated it as a category of duress that is distinct from the colore officii cases. This is because in these cases the money is demanded by and paid to a private entity rather than a public authority.130 However, since in many cases where recovery is sought from a private body that has overcharged the claimant the circumstances are similar to those of the colore officii cases – in that the payment is made to obtain a service to which the complainant is entitled for nothing or for less than the amount paid – the courts apply the concept of implied duress as in the colore officii cases.131 This is especially so where the person making the unlawful demand is in a monopolistic position132 since, in such a case, the complainant will have no choice but to pay in order to obtain the services required. In Parker v Great Western Railway,133 for example, a railway company refused to carry the complainant’s goods except at rates in excess of those they were legally entitled to demand. The complainant paid under protest and later recovered the excess. That case was approved in Great Western Railway Co v Sutton134 where Willes J explained this category of duress by saying that when a man pays more than he is bound to do by law for the performance of a duty which the law says is owed to him for nothing, or for less than he has paid, there is a compulsion or concussion in respect of which he is entitled to recover the excess.135

In Canada, where the Woolwich equivalent ground of restitution is confined to restitution of taxes levied under an ultra vires statute, it has been said that in cases not involving payments made to public authorities pursuant to unconstitutional legislation or the misapplication of an otherwise valid law the courts should continue to insist on proof of compulsion in fact.136 In other words, duress is still relevant in such cases.

127 Scottish Law Commission, Unjust Enrichment, Error of Law and Public Authority Receipts and Disbursements (Scot Law Com No 169, 1999) [2.62]. 128 It is recognised that the distinction between public and private bodies is difficult to draw in certain cases. However, the hard cases on the borderline will be few. 129 Woolwich (n 2) 165. 130 It should be noted that, as the editors of Goff & Jones (n 9) point out (at [10.018]), it is debatable whether some of the payees from whom restitution has been available under the colore officii ground can properly regarded as public authorities (eg arbitrators and stewards of the manor). 131 eg Hills (n 13) (demand by a broker); Fernley v Branson (1851) 20 LJQB 178 (demand by an arbitrator). 132 South of Scotland (n 105). 133 Parker v Great Western Railway Co (1844) 7 Man & G 253, 135 ER 107. 134 Great Western Railway Co v Sutton (1869) LR 4 HL 226, 249. 135 See also South of Scotland (n 105). 136 Kingstreet (n 14) [58].

Public Authorities: Room for Duress? 91

E. CONCLUSION

Prior to the Woolwich decision, duress was the principal ground on which payments exacted by public authorities without lawful justification could be recovered. Since the decision in Woolwich, duress has lost its dominance in this field. It is recognised that although, in principle, duress continues to exist as an independent ground of restitution against public authorities, in many cases the Woolwich principle will be available on facts giving rise to a claim based on duress and, in such cases of concurrent causes of action, in practice claimants will prefer to advance their claims on the ground of the Woolwich principle rather than duress. However, this chapter has endeavoured to show that the Woolwich principle does not apply to all payments to public authorities and in particular that it does not apply to payments other than taxes or other similar levies. In the case of such payments, falling outside the scope of the Woolwich principle, duress (including duress colore officii) continues to be an important ground of restitution against public authorities. Therefore, at least to that extent, it might not be safe to assume that duress no longer serves any useful purpose and that it will simply wither away. In Canada, where the ordinary principles of unjust enrichment have been discarded in favour of an exclusively public law remedy for the recovery of taxes collected under ultra vires statutes, unjust enrichment principles (including duress) are still applicable in other cases of restitution against public authorities. In Woolwich Lord Goff expressed the expectation that the common law principles of duress and mistake will continue to develop in the future. Mistake has since developed so that restitution of unlawfully collected taxes is now available on the ground of mistake of law. Perhaps duress will also be developed in future so that the concept of implied compulsion, recognised in the colore officii cases, will be extended to other cases of restitution against public authorities.

5 Reasons for Restitution CHARLIE WEBB

A. WOOLWICH v IRC

Woolwich Equitable Building Society v IRC concerned a claim to recover interest upon sums totalling approximately £57 million paid by Woolwich to the Inland Revenue to meet a tax demand under the Income Tax (Building Societies) Regulations 1986.1 The building society had doubted the validity of the Regulations and brought judicial review proceedings seeking to establish that the provisions supporting the demand were void. These proceedings were successful. The House of Lords declared that the relevant Regulations were invalid, making the tax demand ultra vires.2 Notwithstanding their doubts as to the validity of the Regulations, Woolwich had paid the sums demanded by the Inland Revenue, on the basis that this was the only commercially viable course of action. Pending the outcome of the judicial review, Woolwich issued a writ seeking repayment of the money it had paid plus interest. However, when the building society succeeded in the judicial review proceedings at first instance, the Inland Revenue agreed to make an ex gratia repayment of the capital sums paid by Woolwich plus interest from the date of that judgment. This left Woolwich without interest on those sums for the period between their initial payment to the Inland Revenue and that judgment. This ‘missing’ interest was agreed by the parties to come to £6.73 million. The question the court had to answer was whether Woolwich was entitled to recover this sum from the Revenue. This in turn depended on whether Woolwich had a right to repayment of the capital sums, upon which interest could be awarded from the date of the initial payment under the discretion provided by section 35A of the Supreme Court Act 1981. Woolwich’s claim failed at first instance in front of Nolan J,3 but succeeded, by a bare majority, in both the Court of Appeal and House of Lords. That the money had been paid by Woolwich to the Revenue was clear. The only question was whether Woolwich could identify some ground for its recovery. In other words, it was plain that the Revenue had been enriched; the issue was whether there was a sufficient reason to require restitution of that enrichment. Here Woolwich’s claim faced a problem. Though English law recognised a number of situations in which money paid by a claimant to a defendant was recoverable – or to put it in now familiar terminology, though the courts recognised a variety of ‘unjust factors’ – none fitted the facts of Woolwich. Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] AC 70 (HL) (hereafter ‘Woolwich’). R v Inland Revenue Commissioners, ex parte Woolwich Equitable Building Society [1990] 1 WLR 1400 (HL). 3 Woolwich Equitable Building Society v Inland Revenue Commissioners [1989] 1 WLR 137 (QB). 1 2

94 Charlie Webb So, it was well established that payments made as a result of a mistake of fact by the claimant could ground recovery. But Woolwich was not mistaken – it believed that the tax demand was ultra vires – and even if it had been wrong about the lawfulness of the demand this would have been a mistake of law, not of fact and, at that time, a mistake of law provided no basis for a claim. Payments made under duress were another established category of claim. Here, however, there was no illegitimate threat or pressure coming from the Revenue – just the prospect of Woolwich being sued if it failed to pay. A further series of cases concerned demands for payment from officials to secure their performance of public functions which they were obligated to provide for nothing or for less. But the Inland Revenue was not charging Woolwich for a service or the performance of some statutory function and so, despite the factual similarities, the claim fell outside this line of authority too. All this meant that Woolwich could succeed only if the courts were to recognise a new ground or reason for restitution; a new class of restitutionary claim. And this is just what the House of Lords did. So what exactly was this new reason for restitution and what does this decision tell us about the shape and nature of the law of restitution more generally?

B. REASONS FOR RESTITUTION

Central to the justification of any restitutionary claim must be the identification of a reason why we should require that the relevant enrichment be given up. Reasons for restitution aren’t quite the same thing as unjust factors. An unjust factor – at least as that phrase is typically used – is some fact or set of circumstances upon which a claimant can base a claim in restitution. So, for instance, we know that mistake is an unjust factor because a claimant can obtain restitution of money he has paid by showing that the payment was made as a result of a mistaken belief he held. The mistake marks the situation out as one where a restitutionary claim will arise. The same is true of other unjust factors. But the bare facts of the claimant’s mistake along with its causal connection to the payment he made and so to the enrichment of the defendant which he now seeks to reverse do not in themselves constitute a reason for requiring the defendant to make restitution. These facts, without more, do not tell us why giving the claimant a restitutionary claim is the right thing to do. What we need is a principle. Principles identify some good or value to which the law and/or people generally should be committed and which should guide our actions.4 Principles tell us not what is but what ought to be. Within the law of restitution we are looking for principles which justify the imposition of duties to give up enrichments received at another’s expense. So in relation to mistaken payments, we require a principle which tells us why the facts on which a claim is based – a payment by A to B, a mistake made by A but for which he would not have made the payment etc – justify a particular legal conclusion – B being ordered to repay that or an equivalent sum of money to A. Where do these principles come from? Though the authorities may embody certain principles and though particular principles may be thought to best explain the law as it stands, the weight and force of such principles – whether they really do identify good 4 Here I use the term ‘principle’ to embrace all relevant standards and so including not just principles of morality and justice but also what others have labelled ‘policies’: cf R Dworkin, Taking Rights Seriously (London, Duckworth, 1977) 22–28, 82–84.

Reasons for Restitution 95

reasons for action, whether they really should be followed and applied – is not a matter of positive law. Rather these are matters of justice, of morality or policy. The law, both as an institution and as a body of decisions and rules, can be justified only by reference to standards which are, in this sense, external to it. So, though the cases will tell us when restitutionary claims are/have been allowed, and hence what, as a matter of positive law, are treated as (good) reasons for restitution, they cannot settle what truly are (good) reasons. This is to say no more than that though precedent tells us what the law does do, it cannot alone tell us what the law should do. Of course, because of this, claims about what principles the law should follow will always be contestable. There is no means of authoritatively resolving whether the law should, say, be guided by a principle of wealth maximisation or whether it should recognise individuals as having basic rights which should be protected even where this is economically inefficient. But this doesn’t mean these questions can be ducked. They must be addressed every time a court is faced with a claim where the existing authorities send mixed messages or no message at all; they must be addressed every time a court or legislature is faced with the choice of reforming the law; and they must be addressed by all of us concerned with seeing that the law – and the coercive mechanisms which back it up – treats people justly. Restitution lawyers haven’t done a very good job of identifying reasons for restitution in this sense. This may simply be a side-effect of the chequered history of the subject. Those who took on the job of marking out a distinct territory and structure for a body of claims long left wallowing in the legal backwaters feared that progress would be threatened if their case depended on the adoption of a particular normative stance. So rather than seeking to identify the moral or political foundations of restitutionary claims, restitution lawyers immersed themselves in the cases, showing that there simply are claims arising out of mistaken payments and that these cases simply cannot be said to derive from any contract between the parties or the commission of a wrong. From this purely descriptive and largely uncontroversial premise, we could then draw out the basic (and again descriptive) features of these claims: an enrichment received by the defendant, coming in some way from the claimant, arising in circumstances which (as a matter of positive law) entitled the claimant to recover that enrichment. This was the basis of Birks’ claim that the notion of ‘injustice’ employed in the concept ‘unjust enrichment’ was purely downward looking, focused on the reality of the case law and not on moral principle or public policy.5 And often looking down to the cases tells us all we need to know. Lawyers arguing cases, judges deciding them, lecturers giving classes and tutorials, students sitting exams will get a long way simply by knowing what the law is, by which I mean knowing on what sets of facts claims will arise and how the courts have presented these conclusions. So, to this extent, it doesn’t matter why mistaken payments (etc) give rise to restitutionary claims, what the reason for restitution is here, so long as we know – as we do – that proof of a mistaken belief which caused the payment will ground such a claim. Of course, this is not to say that there is not some moral or political principle which justifies restitutionary claims following mistaken payments. If such claims are justified – as few people, if any, doubt – there must be. It is simply that, so long as we agree that mistaken payments do and should lead to restitutionary liability, and so long as we know how the authorities deal with such claims, we do not need to go back to first principles to decide cases, advise clients and the like.

P Birks, An Introduction to the Law of Restitution, rev edn (Oxford, Clarendon Press, 1989) 23–24.

5

96 Charlie Webb But, once again, as soon as we face a question not clearly determined by prior authority – as courts often do – we need to turn to the question of principle. Similarly, any time a court or legislature considers whether the law should change direction, there is no alternative but to consider the principles which (we think) should shape the law, and which the law should reflect and embody. And this question will always take us outside the authorities. This was the position in which the House of Lords found itself in Woolwich. Though there were authorities which shed light on whether an unlawful tax demand supported a claim to restitution – authorities which (at the very least) suggested that no such claim would lie – this was not the end of the matter. The question facing the court was whether the law should be developed – changed – so as to provide a claim on such facts.6 To answer this the court had to consider whether there was, notwithstanding the authorities, a (good) reason for requiring restitution of the payment. The fact that a majority allowed the claim shows that they thought there was. More difficult is to tell what they considered this reason to be. Later cases make frequent reference to the ‘Woolwich principle’, but the way this ‘principle’ is typically formulated does no more than restate the conclusion reached by their Lordships in that case, leaving their reasons for coming to this conclusion – the real principle underlying the decision – wholly out of view.7 The textbooks, however, are in little doubt as to what the reason for restitution was. Peter Birks writes, ‘the unjust factor was that it was essential that the servants stay within their powers and, within that, especially important that the principle of no taxation without Parliament be upheld’.8 James Edelman and Elise Bant state, ‘[t]he right to restitution was based on a policy against unlawful taxation’.9 Graham Virgo tells us that the payment of an ultra vires tax demand is a ground of restitution ‘because of fundamental constitutional principles, arising from the Bill of Rights, that no public authority can retain money which it has no authority to receive’.10 As such, Woolwich is treated as a core instance of ‘policy-motivated restitution’, in contrast to the bulk of unjust enrichment claims which are grounded in the claimant’s defective or impaired consent to the transfer.11 6 ‘I now turn to the submission of Woolwich that your Lordships’ House should, despite the authorities to which I have referred, reformulate the law so as to establish that the subject who makes a payment in response to an unlawful demand of tax acquires forthwith a prima facie right in restitution to the repayment of the money. This is the real point which lies at the heart of the present appeal’. Woolwich (n 1) 105 (Lord Goff). 7 Take, as an example, the statement of the ‘Woolwich principle’ offered by Lord Walker in Deutsche Morgan Grenfell Group plc v Inland Revenue Commissioners [2006] UKHL 49, [2007] 1 AC 558 (hereafter ‘DMG’) [121]: ‘The majority [in Woolwich] upheld the simple principle that tax paid in response to an unlawful demand could be recovered, regardless of compulsion’. The same goes for the reformulation of the ‘principle’ in Test Claimants in the Franked Investment Income Group Litigation v HMRC [2012] UKSC 19, [2012] 2 AC 337 (hereafter ‘FII’), [79] (Lord Walker): ‘We should restate the Woolwich principle so as to cover all sums paid to a public authority in response to (and sufficiently causally connected with) an apparent statutory requirement to pay tax which (in fact and in law) is not lawfully due’. See too A Burrows, The Law of Restitution, 3rd edn (Oxford, OUP, 2010) 499, taking as a statement of the ‘Woolwich principle’ the following quote from Lord Goff ’s speech in Woolwich (n 1) 177: ‘money paid by a citizen to a public authority in the form of taxes or other levies paid pursuant to an ultra vires demand by the authority is prima facie recoverable by the citizen as of right’. 8 P Birks, Unjust Enrichment, 2nd edn (Oxford, Clarendon Press, 2005) 134. See too P Birks, ‘“When Money is Paid in Pursuance of a Void Authority” – A Duty to Repay?’ [1992] Public Law 580. 9 J Edelman and E Bant, Unjust Enrichment in Australia (Oxford, OUP, 2006) 297. 10 G Virgo, Principles of the Law of Restitution, 2nd edn (Oxford, OUP, 2006) 411. cf the (slightly) more cautious account given in A Burrows, Understanding the Law of Obligations (Oxford, Hart Publishing, 1998) 56: ‘This was in essence to recognise a special public law ground for restitution based on the policy or constitutional principle that there should be no taxation without Parliament’ (emphasis added). 11 P Birks and R Chambers, The Restitution Research Resource (Oxford, Mansfield Press, 1997) § 251; Birks (n 8) 133; Edelman and Bant (n 9) 294, 297; Virgo (n 10) 124, 411.

Reasons for Restitution 97

I’m far from sure that the majority speeches back these claims up, meaning that, though the scope of the rule set down in Woolwich is tolerably clear, its rationale is not. This matters since a clear understanding of the reason for restitution in this case might reveal a lot about the nature of claims in unjust enrichment and the future direction of the subject.

C. THE RATIONALE OF WOOLWICH

Despite this being the only issue at stake in the case, the majority speeches are in fact rather opaque when it comes to identifying the precise reason for allowing Woolwich’s claim. The question the court asked itself was whether a claimant who had paid pursuant to an ultra vires tax or unlawful demand was prima facie recoverable as of right. The answer given by the majority is clear: yes. Equally clear is the fact that all their Lordships considered that such a right could not simply be read from the existing authorities. Since the cases, at the very least, left the issue in doubt, the court approached the question as a matter of principle – should the law recognise such a right? So why did the majority consider that it should? I’ll start with Lord Goff, who gave what is regarded as the leading majority speech and with whose reasoning Lord BrowneWilkinson expressly agreed. Having surveyed the authorities, Lord Goff’s basic view on the question of principle is contained in the following passage: The justice underlying Woolwich’s submission is, I consider, plain to see. Take the present case. The revenue has made an unlawful demand for tax. The taxpayer is convinced that the demand is unlawful, and has to decide what to do. It is faced with the revenue, armed with the coercive power of the state, including what is in practice a power to charge interest which is penal in its effect. In addition, being a reputable society which alone among building societies is challenging the lawfulness of the demand, it understandably fears damage to its reputation if it does not pay. So it decides to pay first, asserting that it will challenge the lawfulness of the demand in litigation. Now, Woolwich having won that litigation, the revenue asserts that it was never under any obligation to repay the money, and that it in fact repaid it only as a matter of grace. There being no applicable statute to regulate the position, the revenue has to maintain this position at common law. Stated in this stark form, the revenue’s position appears to me, as a matter of common justice, to be unsustainable; and the injustice is rendered worse by the fact that it involves . . . the revenue having the benefit of a massive interest-free loan. I then turn from the particular to the general. Take any tax or duty paid by the citizen pursuant to an unlawful demand. Common justice seems to require that tax to be repaid, unless special circumstances or some principle of policy require otherwise; prima facie, the taxpayer should be entitled to repayment as of right.12

Two points are worth noting at this stage. First, there is very little, if anything, in the way of argument here. The injustice of the Revenue retaining the money is treated as so plain that all that is needed by way of demonstration is a simple restatement of the facts. So, while certain aspects of the factual background are emphasised – Woolwich’s lack of alternative options, the power wielded by the Revenue, the windfall the Revenue would receive if there was to be no claim – what actually makes this situation unjust is seen to go without saying. (I should stress that my suggestion is not that Lord Goff was wrong to think there Woolwich (n 1) 171–72.

12

98 Charlie Webb was a manifest injustice here, simply that there is little here to explain what the nature of the injustice is.) Secondly, there is no (explicit) reference here to the policy or constitutional principle that taxes should not be levied without Parliamentary authority that the textbook writers identify as the basis of the claim. The special position of the Revenue – with its power to demand taxes and with the state’s enforcement machinery at hand to back this up – is given emphasis, but the purpose of this appears to be to bring out the impossible position Woolwich found itself in rather than to invoke some distinct (and unspoken) public law principle. It is only later in his speech, when dealing with possible objections to ‘the simple call of justice’, that this constitutional principle is given any attention: [The] law might have developed so as to recognise a condictio indebiti – an action for the recovery of money on the ground that it was not due. But it did not do so. Instead, as we have seen, there developed common law actions for the recovery of money paid under a mistake of fact, and under certain forms of compulsion. What is now being sought is, in a sense, a reversal of that development, in a particular type of case; and it is said that it is too late to take that step. To that objection, however, there are two answers. The first is that the retention by the state of taxes unlawfully exacted is particularly obnoxious, because it is one of the most fundamental principles of our law – enshrined in a famous constitutional document, the Bill of Rights 1688 – that taxes should not be levied without the authority of Parliament; and full effect can only be given to that principle if the return of the taxes exacted under an unlawful demand can be enforced as a matter of right. The second is that, when the revenue makes a demand for tax, that demand is implicitly backed by the coercive powers of the state and may well entail (as in the present case) unpleasant economic and social consequences if the taxpayer does not pay.13

I shall return to Lord Goff’s reference to the question of recovering money simply on the grounds that it is not due shortly. Here I want to draw attention to the very much secondary role that the constitutional principle plays in Lord Goff’s reasoning. For a start, as noted already, it comes in only after Lord Goff has set out what he considers to be the basic justice of Woolwich’s claim. Moreover, it is invoked – in response to the suggestion that the simple fact that the money was not owing is not, as English law had developed, a sufficient ground for recovery – to show why it might be thought particularly bad for this defendant to retain money it was not due. But the clear implication is that the retention of money not owing is unjust in any case, an injustice that exists independently of the Bill of Rights point.14 Lord Goff’s second response to the ‘no condictio indebiti’ objection draws an analogy with cases of compulsion. However, though he was prepared to countenance the development of the forms of compulsion upon which a restitutionary claim may be grounded, Lord Goff considered that, in the absence of any real possibility of Woolwich having its property seized in the event of non-payment, ‘the concept of compulsion would have to be stretched to the utmost’ to embrace Woolwich’s claim.15 As such, ‘[i]n the end, logic appears to demand that the right of recovery should require neither mistake nor compulsion, and that the simple fact that the tax was exacted unlawfully should prima facie be enough to require its repayment’.16

ibid 172. See too ibid 166. 15 ibid 173. 16 ibid 173. 13 14

Reasons for Restitution 99

The ‘no tax without Parliament’ principle did not even have a supporting role in Lord Browne-Wilkinson’s speech. He stated his basic position as follows: In the present case, the concept of unjust enrichment suggests that the plaintiffs should have a remedy. The revenue demanded and received payment of a sum by way of tax alleged to be due under regulations subsequently held by your Lordships’ House to be ultra vires. The payment was made under protest. Yet the revenue maintains that it was under no obligation to repay the wrongly extracted tax and in consequence is not liable to pay interest on the sum held by it between the date it received the money and the date of the order of Nolan J. If the revenue is right, it will be enriched by the interest on the money to which it had no right during the period. In my judgment this is the paradigm of a case of unjust enrichment.17

Lord Browne-Wilkinson considered that there were two alternative routes, both with support in the cases, by which this conclusion could be reached. First, the payment was without consideration: ‘money paid on the footing that there is a legal demand is paid for a reason that does not exist if that demand is a nullity’.18 Secondly, and unlike Lord Goff, Lord Browne-Wilkinson took the view that the concept of compulsion could, and should, be extended to cases such as this: In my judgment, the colore officii case are merely examples of a wider principle, viz. that where the parties are on an unequal footing so that money is paid by way of tax or other impost in pursuance of a demand by some public officer, these moneys are recoverable since the citizen is, in practice, unable to resist the payment save at the risk of breaking the law or exposing himself to penalties or other disadvantages.19

The constitutional principle is also missing from Lord Slynn’s speech. His reasons for allowing the claim are set out here: Although as I see it the facts do not fit easily into the existing category of duress or of claims colore officii, they shade into them. There is a common element of pressure which by analogy can be said to justify a claim for repayment . . . I find it quite unacceptable in principle that the common law should have no remedy for a taxpayer who has paid large sums or any sum of money to the revenue when those sums have been demanded pursuant to an invalid regulation and retained free of interest pending a decision of the courts.20

Where does this leave us? Lord Slynn and Lord Browne-Wilkinson both considered that the claim may be justified by analogy with, or extension of, the compulsion cases, but this approach was explicitly rejected by Lord Goff. Lord Browne-Wilkinson also thought that the same result could be supported on the basis that the payment was made without consideration, an approach which also gains some support from Lord Goff’s speech, but to which Lord Slynn makes no reference. Finally, the constitutional principle that taxes should not be levied without Parliamentary authority is referred to only by Lord Goff (though Lord Browne-Wilkinson does express agreement with Lord Goff’s reasons) and then, it seems, only as a secondary argument. Now, none of this means that the constitutional principle identified by the textbook writers may not provide the soundest basis for such a claim – that this might not be the best rationalisation of this line of authority. This remains to be established. It does, however, ibid 197. ibid 197. 19 ibid 198. 20 ibid 204. 17 18

100 Charlie Webb suggest that the claim this was the basis the court gave for the decision is at the very least exaggerated, if not simply wrong. So why have restitution lawyers been so keen to push the constitutional principle explanation for the decision and play down these other arguments which are given greater emphasis in the majority speeches? Why should we consider this the best possible interpretation of the case? We can begin by considering the alternative bases given for allowing Woolwich’s claim.

D. COMPULSION

It might seem surprising that a compulsion-focused rationale for the decision in Woolwich has not found greater favour. Not only did it receive the express support of two of the majority but, of all the possible rationales, it ties in most closely with existing authority, at least in so far as compulsion is a well-established unjust factor. As against this, Lord Goff rejected this as a possible basis for the decision in his leading speech (though the time he spent setting out the case for a compulsion-based rationale suggests he considered that it had at least some plausibility). A further reason – beyond those offered by Lord Goff – sometimes given for rejecting a compulsion-focused rationale for the claim is that, though Woolwich was clearly put in a very difficult position by the Revenue’s demand, with no other reasonable option but to pay and take action later, the law only allows recovery of money paid under compulsion where this results from illegitimate pressure. An express or implied threat to take legal proceedings against a claimant should he fail to accede to a demand is not, however, illegitimate.21 Since the compulsion rationale has not been picked up either in later cases22 or by commentators, and since it raises fewer interesting questions than the other bases for the decision proposed in the majority speeches, I shall make only a brief comment on the possibility of compulsion as a ground for Woolwich’s claim and, in particular, on the requirement described in the previous paragraph that the pressure be illegitimate. The need for the pressure to be illegitimate has clear support in the cases.23 Nonetheless, it has been challenged and indeed challenged by restitution lawyers. As a matter of principle, whether recovery should be limited to cases of illegitimate pressure should turn on why it is that pressure of any sort is thought to justify a restitutionary claim, or in other words what the reason for restitution is in duress cases.24 One possible rationale focuses on the defendant’s wrongful conduct in subjecting the claimant to threats. On this basis, we might say that we allow claimants to recover payments made, and indeed to escape from contracts entered into, as a result of a defendant’s duress precisely to ensure that the defendant does not gain from his wrongful conduct. If so, or to this extent, it is right to have a requirement that the pressure applied be illegitimate, since only then will this principle be called into action. (Moreover, on this approach, it would seem to follow that we should require, 21 See eg Burrows (n 7) 500–501; P Birks, ‘Restitution from the Executive: A Tercentenary Footnote to the Bill of Rights’ in P Finn (ed), Essays on Restitution (Sydney, Law Book Co, 1990) 174–75. 22 The last trace of this analysis – a possible requirement that the tax be paid in response to a ‘demand’ – has now been removed by the Supreme Court in FII (n 7). 23 See eg Universe Tankships Inc of Monrovia v International Transport Workers Federation (The Universe Sentinel) [1983] 1 AC 366 (HL) 384 (Lord Diplock), 400 (Lord Scarman); R v Attorney-General of England and Wales [2003] UKPC 22, [2003] EMLR 24 [15] (Lord Hoffmann). 24 See generally S Smith, ‘Contracting Under Pressure: A Theory of Duress’ (1997) 56 CLJ 343.

Reasons for Restitution 101

at most, a basic ‘but for’ causal connection between the pressure and the claimant’s decision to pay or enter the contract; and that, so long as such a causal link exists, we do not need to inquire into what other courses of action the claimant had open to him.25) Alternatively, we might see the justification for duress pleas as lying in the effect the pressure has on the claimant’s autonomy or freedom of choice. On this view, the reason that we allow the claimant to recover the payment or avoid the contract is because he did not truly or freely consent to it (and that true, free consent is necessary for a fully effective transfer of one’s property and when entering into a contract). Now, this consent-focused understanding of duress went out of fashion for a while, following Atiyah’s criticism of the view that duress leads to a claimant’s will being ‘overborne’.26 No doubt, a claimant with a gun put to his head does make a choice and, whatever duress does do, it does not deprive an agent of the capacity to make choices. Nonetheless, there is a clear sense in which the choice of a claimant under extreme compulsion is not the equal of the choice of a claimant with an array of reasonable options in front of him. It is for this reason that the criminal law is right to recognise a defence of duress (a position which cannot be reached through the application of the principle of preventing wrongdoers profiting from their wrongful conduct). On this view, though severe compulsion – and so a significant reduction of the claimant’s freedom of choice – will often be the result of another’s wrongful conduct, equivalent pressure may, on occasion, also arise in the absence of wrongdoing. As such, illegitimate pressure is not a pre-condition for the application of the principle. Moreover, there is no reason to think that we need to choose between these two principles – of preventing wrongdoers gaining from their wrongdoing and of respecting the choice or autonomy of the claimant in respect of dealings with his property and the contracts he enters into. Rather, though they may happen both to apply on a particular set of facts, they provide independent reasons for entitling a claimant to recover money paid under compulsion. As such, while the need for the defendant to be implicated in the compulsion to which the claimant is subject, and hence the requirement that the pressure be illegitimate, might be defensible in contract cases – where the question is not so much ‘did the claimant truly consent?’ as ‘can the defendant reasonably rely on the appearance of the claimant’s consent?’ – it should be dispensable in cases of non-contractual restitution, where the defendant’s interests are not so much at stake (and, in so far as they are, are protected by the defence of change of position).27 On this basis, Woolwich could be seen – or could have been seen – as providing important recognition of this second, consent-focused principle.

E. ABSENCE OF CONSIDERATION

Though initial response to the suggestion in Lord Goff and Lord Browne-Wilkinson’s speeches that the money was recoverable on the basis that there had been no consideration cf Barton v Armstrong [1976] AC 104 (PC). P Atiyah, ‘Economic Duress and the “Overborne Will” ’ (1982) 98 LQR 197. 27 This difference in the way consent is relevant as between the law of contract and the law of unjust enrichment is seen most clearly in their respective responses to mistakes. While a simple causative mistake will entitle a claimant to restitution of money paid (see Barclays Bank Ltd v WJ Simms, Son and Cooke (Southern) Ltd [1980] QB 677 (QB)), this is not sufficient to enable a claimant to escape a contract (see Smith v Hughes (1871) LR 6 QB 597 (CA)). If, however, the defendant is responsible for the claimant’s mistake, then the contract can be avoided (Redgrave v Hurd (1881) LR 20 Ch D 1 (CA); cf Barclays Bank plc v O’Brien [1994] 1 AC 180 (HL)). 25 26

102 Charlie Webb for the payment was mixed,28 this aspect of the case might now be considered ahead of its time. Whether English law should adopt – or, on one view, whether it already has adopted – an absence of consideration or absence of basis approach to unjust enrichment claims is now seen to be one of the pressing questions for the law of restitution. Though Birks – whose volte-face on the issue precipitated this on-going debate – did not identify Woolwich as one of the cases which, to him, demonstrated that English law had now committed itself to an ‘absence of basis’ approach,29 Woolwich lends greater support, unquestionably more explicit support, for this view than the swaps cases. Certainly Woolwich provides a useful backdrop against which the ramifications of this switch can be considered. There has been much discussion of what exactly is meant by an absence of ‘basis’ or ‘consideration’. Lords Goff and Browne-Wilkinson only refer to the payment of money that is not due, yet clearly this cannot be a complete statement of the concept, since it cannot be that all payments which the payer is under no obligation to make are recoverable. Instead of addressing this question directly here, I want to say something about how an absence of basis approach – however it is to be understood – connects to reasons for restitution. As noted already, the typical ‘unjust factors’ listed in the restitution textbooks are not themselves reasons for restitution. Mistakes, duress and the like are not reasons or principles which justify recovery. They are simply facts. If we want to know what we have reason to do – what ought to be done – then we need to identify some good or value which should rightly guide our actions.30 So we have reason to recognise restitutionary claims in cases of mistake because there is a principle which, when applied to such facts, tells us that requiring the defendant to give up his gain to the claimant is what should be done. It is this principle which marks out these facts as normatively significant and directs us to the appropriate normative response. (I shall say a little more about what the principle in such cases may be shortly.) It doesn’t follow from this that we should not speak of or seek to identify unjust factors. On the contrary, we, as lawyers and citizens, need to know what rights we have, in what circumstances we are accorded legal claims and when we come under legal duties. And this is done by communicating on what facts those rights, claims and duties arise. So, though the law can be justified – and novel questions of law answered – only by examination of the principles which should guide it, the law as it is (meaning something like: the 28 See eg Birks, ‘ “When Money is Paid in Pursuance of a Void Authority” – A Duty to Repay?’ (n 8) 587–88; J Beatson, ‘Restitution of Taxes, Levies and Other Imposts: Defining the Extent of the Woolwich Principle’ (1993) 109 LQR 401, 411; A Burrows, The Law of Restitution (London, Butterworths, 1993) 351–52; cf G Virgo, ‘Restitution of Overpaid Tax – Justice at the Expense of Certainty’ (1993) 52 CLJ 22, 23. 29 Birks, Unjust Enrichment (n 8) 133–34. The idea that the ‘Woolwich principle’ might or should be seen as an example of recovery for a payment made without basis was recently floated before the Supreme Court in FII (n 7). The Court’s response to this was rather tentative, noting both the echoes of absence of basis thinking in some of the Woolwich speeches and that the view that a demand was not necessary for a claim to succeed was consistent with an absence of basis approach. Nonetheless the fact that the Court didn’t seek to ground its decision in a commitment to an absence of basis understanding of either Woolwich or unjust enrichment claims generally is a clear indication that it didn’t consider that there was anything in Woolwich or in FII’s restatement of the ‘Woolwich principle’ which could be made sense of only in absence of basis terms. 30 John Finnis illustrates the point in this way (J Finnis, ‘On Hart’s Ways: Law as Reason and as Fact’ (2007) 52 American Journal of Jurisprudence 25, 44–45): ‘The fact that it is raining is in itself no reason to carry an umbrella, no reason at all, even in conjunction with the fact that without an umbrella I’ll get wet. But facts like these can play their part in the reason, the warranted conclusion (that I should [had better] carry an umbrella) which gets its directive or normative element from some practical, evaluative premise such as: it’s bad for one’s health to get wet or: it’s bad for one’s ability to think and function to get uncomfortably wet and cold. By virtue only of that or some similar truth (as one supposes) about good and bad, the plain fact that an umbrella can prevent these evils by keeping me dry can contribute to the normative conclusion that I have reason to, ought to, carry an umbrella’.

Reasons for Restitution 103

rules posited as authoritative by competent officials within the legal system) is reducible to a series of propositions detailing the legal consequences that follow from particular sets of facts. In light of this, we ought to question not just the wisdom but the very possibility of switching away from an inquiry into unjust factors to an absence of basis approach. For so long as we have a law of restitution there will be factual situations to which the law responds by requiring a defendant to give up his gains. And so, whether or not we use the term ‘unjust factor’ to describe the factual circumstances in which a payment (etc) will give rise to a restitutionary right, not only will it remain possible to list the factual circumstances in which such claims arise, but such an account will remain essential for anyone wanting to know what the law is on this question. Similarly, anyone going to court seeking restitution will still have to plead and prove the facts upon which their claim is grounded – the facts from which, under the rules of law set down in that legal system, a restitutionary claim follows (or the facts from which, in the absence of such rules, principle requires such a claim to be granted). An absence of basis approach changes none of this. I think it is clear that ‘absence of basis’ is not a fact or factual status like a mistake or compulsion. But nor does it identify a reason for restitution, a principle which provides a justification for requiring a particular gain to be given up.31 Rather it describes or states the conceptual, technical (as opposed to practical) consequences of the application of certain rules of law. As such, ‘absence of basis’ inhabits a middle ground between the real world facts from which restitutionary claims arise, the rules by which those facts are identified (legally, practically) and the principles which justify those rules and hence the recognition of such claims on such facts. So viewed, it is clear that restitution lawyers have been right to say that ‘absence of basis’ or ‘absence of consideration’ is not an unjust factor which aligns with mistake, duress and the like. It is not an unjust factor, in this sense, at all. But then an inquiry into whether there is consideration or a basis for the enrichment cannot do away with an inquiry into the facts upon which restitutionary claims will arise nor an inquiry into the principles which justify recognising such claims. In other words, even within an absence of basis model, we will still need to ask when and why restitutionary rights and duties arise. The answer to the ‘when?’ question will (ultimately) be a set of various fact combinations (identifying, amongst other things, a list of ‘unjust factors’). The answer to the ‘why?’ question will be a set of principles. Absence of basis has no obvious role in either inquiry, and indeed seems only likely to distract from them.32 Not only this, 31 This was clear to Birks, who suggested that the move to an absence of basis model would mean that all unjust enrichment claims would have a common rationale: ‘that in the events which have happened the claimant did not intend the recipient to be enriched’ (Birks, Unjust Enrichment (n 8) 108; see too ibid 104, 116). Now, as I have argued elsewhere, I think Birks was wrong to say that an absence of basis model makes all unjust enrichment claims explicable by reference to the claimant’s defective or conditional intention in making the transfer (see C Webb, ‘What is Unjust Enrichment?’ (2009) 29 Oxford Journal of Legal Studies 215, 240–42). But the point for now is that this argument shows that Birks did not consider that absence of basis was itself a reason or justification for a restitutionary claim. See too below n 32. 32 As Birks put it before his conversion (P Birks, ‘No Consideration: Restitution after Void Contracts’ (1993) 23 University of Western Australia Law Review 195, 232): ‘As for the grounds for restitution, the common law asks for, and expects the plaintiff to show, a specific unjust factor, rooted in fact and not in legal abstraction, such as mistake, duress, failure of basis, and so on. The three great families into which, at a higher level of generality, these specific grounds fall [are that the plaintiff did not intend the defendant to have the enrichment, unconscientious receipt, and that there is some other policy which requires that restitution be granted]. At each level of generality the common law’s answer is given in down-to-earth terms. No useful purpose would be served, indeed this useful approach would be subverted, if the attempt to pin down the abstract term “unjust” began to be conducted

104 Charlie Webb since every restitutionary claim will arise out of some or other combination of facts, every (real and hypothetical) instance of restitutionary liability is explicable through the notion of unjust factors. An unjust factor is simply a description of (some salient aspect of) the factual context in which the payment was made and which (by virtue of a relevant principle or reason for restitution) supports a claim to its recovery, and every successful restitutionary claim will derive from such a context. That there may be cases of successful restitutionary claims which do not fit within our existing list of unjust factors simply means that this list is incomplete and requires addition. It is not a sign that an inquiry into unjust factors – into the circumstances in which the payment was made and upon which the claimant relies in seeking restitution – is redundant or dispensable. So what is gained by adopting an absence of basis model? One answer is that it would give greater elegance, simplicity or unity to the law of unjust enrichment.33 But this would be only the most superficial elegance and simplicity for, behind the apparent unity of a single inquiry into the basis of the enrichment, we would still have the same mess of factual circumstances (and so of ‘unjust factors’) from which such claims arise and the same old diversity of reasons for recognising rights to restitution.34 Another suggestion is that an absence of basis approach is more likely to lead us to the right answers or at least steer us away from the wrong ones.35 But, aside from rightly highlighting that there should be no restitution of a payment which was due, I doubt absence of basis approaches hold the advantage here. Though a list of unjust factors will not tell us why we should recognise restitutionary claims on such facts or tell us where and how that list should be added to, it does at least identify the factual situations in which the law does grant such claims. This is in clear contrast to an absence of basis model. To repeat: an inquiry into the basis of an enrichment addresses neither the when nor the why of restitutionary liability directly. As such, it is highly unlikely to improve our chance of answering those questions correctly. Returning to Woolwich then, in so far as the majority speeches relied on a notion of absence of consideration or absence of basis, this does not in itself identify either an unjust factor (in the sense in which mistake and compulsion are unjust factors) or a genuine reason for restitution (in the sense of a principle which justifies recognising a restitutionary claim on a given set of facts). And yet Lords Goff and Browne-Wilkinson must have meant something by this. One possibility is that they were (tacitly) invoking the constitutional principle pushed by the textbook writers, to which I now turn. I note now, however, that a second possibility exists, one to which I shall return at the end of this chapter.

through another abstraction such as “absence of consideration” or, synonymously, “insufficient juridical cause” ’. See too ibid 23: ‘ “No consideration” will be a conclusion, and very probably a conclusion which, masquerading as a reason, will conceal the need for reasons. That is the real danger. The question whether an enrichment at the expense of the plaintiff is an unjust enrichment cannot be usefully mediated through a second tier of abstraction, but that second level of abstraction can become a powerful agent of mystification’. 33 See eg R Stevens, ‘Is there a Law of Unjust Enrichment?’ in S Degeling and J Edelman (eds), Unjust Enrichment in Commercial Law (Sydney, Law Book Co, 2008); T Baloch, Unjust Enrichment and Contract (Oxford, Hart Publishing, 2009) 76; DMG (n 7) [158] (Lord Walker). 34 And if we don’t think there is any such diversity, then there is no reason to think that an absence of basis approach has any advantage in terms of simplicity and elegance. On the multiplicity of reasons for restitution, see Webb (n 31) especially at 238–42. As I argue there, as well as for the reasons given above (text to n 31), moving to an absence of basis model makes no change to the (true, moral, political) reasons for restitution. 35 See eg Baloch (n 33) 78–84.

Reasons for Restitution 105

F. NO TAX WITHOUT PARLIAMENT

That there is a legal rule that taxes should not be levied without the authority of Parliament is clear and has been since 168936 when the Bill of Rights was enacted. Now, this in itself does not tell us what we are to do in the event that Parliament does levy unlawful taxes but, as Lord Goff noted,37 the same principle as supports the primary rule – don’t levy unlawful taxes – also supports the secondary rule that any money so raised be returned. So there is no doubt that this constitutional principle is capable of justifying the decision in Woolwich and the broader ground of restitution to which the decision has given rise. As noted already, my doubt is as to whether this really was the reason for restitution the court had in mind. However, what really matters is not whether the court identified an adequate justification for the claim but that such claims are justified and, going forward, here we have a principle which does the job. Indeed, it seems clear that this is how the courts now understand and justify the rule coming out of Woolwich.38 The question I want to examine briefly here is what this view of the basis of the decision might mean for our broader understanding of the law of restitution. As is well known, it was once academic orthodoxy (at least among restitution lawyers) that all instances of restitutionary liability were to be understood as a response to the defendant’s unjust enrichment. There was, in Birks’ terminology, a perfect quadration between the two: all restitutionary claims arise out of and to reverse an unjust enrichment, all unjust enrichments give rise to a right to restitution.39 Now, so long as unjust enrichment was understood to be simply ‘the generic conception’ of all events which trigger restitution40 – and so as meaning no more than ‘gains which the law requires to be given up’ – all was well. It was clear that unjust enrichment played no role in explaining or justifying when and why restitutionary claims would arise – an unjust enrichment was no more than an enrichment which must be given up, and so added nothing to the idea of restitution as the giving up of a gain – but at least we knew what the terms meant and what the relationship between them was. Two related problems, however, began to emerge which threatened this. First, despite Birks’ attempts to ensure that ‘unjust enrichment’ – in particular the ‘unjust’ component – was understood as merely descriptive of the existing case law and not as involving an appeal to principles of justice,41 there proved to be an irresistible impulse, especially on the part of the courts, to use the term normatively, as a (purported) justification or reason for ordering restitution. This can be seen in Woolwich, in Lord BrowneWilkinson’s statement that ‘the concept of unjust enrichment suggests that the plaintiffs should have a remedy’.42 On Birks’ original understanding of unjust enrichment this makes See Birks (n 21) 165 on the question of the correct date of the Act. Woolwich (n 1) 172; quoted above, text to n 13. 38 See most recently FII (n 7). This understanding of the basis of such claims is not only stated in the judgments but is the only way of squaring the court’s conclusion that no demand is needed for such a claim to lie with its view that this does not entail any ‘decisive step’ towards an absence of basis/consideration approach to the law of unjust enrichment: see ibid [81] (Lord Walker). 39 Birks (n 5) 17. 40 ibid 16. 41 ibid 23–24. 42 Woolwich (n 1) 197; quoted above, text to n 17. Similarly, see the statement (ibid) that the facts disclose a ‘paradigm of a case of unjust enrichment’. By this, it seems clear that Lord Browne-Wilkinson meant that this was 36 37

106 Charlie Webb no sense. Unjust enrichments are those enrichments which the law requires to be given up, those which lead to a restitutionary remedy. Unjust enrichment, in this sense, can play no role in determining whether a particular gain is one the law should require to be given up, whether the claimant should have a remedy. Indeed, we cannot say whether an enrichment is an unjust enrichment until we have answered this question. Secondly, once we start looking for reasons for restitution, it becomes clear that there is no single, common principle – of unjust enrichment or anything else – which justifies all instances of restitutionary liability. This is clearest in the distinction between, on the one hand, those restitutionary claims which are grounded in the defendant’s wrongful conduct in making the gain and, on the other, those arising from mistaken payments and the like where the focus is not the defendant’s conduct but the claimant’s consent to the transfer. There is no doubt that these two classes of cases are materially different, notwithstanding that they both support a restitutionary response. This then gives us our reason for distinguishing cases of restitution for wrongs from cases of restitution for (autonomous) unjust enrichment. But as Woolwich shows – at least if explained on the basis of the constitutional principle – we can’t stop here.43 Even among those cases where the restitutionary claim arises out a payment or other transfer from claimant to defendant and where the ground of restitution is something other than the defendant’s breach of some duty owed to the claimant, we still find a diversity of reasons for restitution, giving us materially different classes of case. For, whatever the principle underlying standard cases of mistake (for instance), it clearly isn’t the constitutional principle of no tax without Parliament. What this shows in turn is that, even when we excise the wrongs cases from the category ‘unjust enrichment’, we still have within this class a range of diverse claims resting on distinct principles.44 Woolwich also provides a good example of the redundancy of unjust enrichment when it comes to justifying claims for restitution. Though Lord Browne-Wilkinson makes reference to it at the beginning of his speech, it disappears from view as soon as he starts to consider more concrete bases for the claim. The only references to unjust enrichment (and its variants) in Lord Goff’s speech come when setting out the impact of other decisions,45 and it appears in Lord Slynn’s opinion only in a quote from Lord Wright’s speech in Fibrosa.46 This is hardly surprising. Even once ‘unjust enrichment’ is decoupled from the authorities – so that it becomes ‘upward’ and not merely downward looking – it fails to identify any particular moral or political principle, any particular reason for a restitutionary claim. As such, when a court is faced with the question of whether to identify a new ground of restitution, it is to be expected that they would look beyond unjust enrichment a clear situation in which the law should grant restitution, not that this was a case which fell squarely within the existing precedents which set out when restitution is granted (notwithstanding that he thought that such precedents could be extended to this case). 43 See generally Webb (n 31) 238–42 and C Webb, ‘Property, Unjust Enrichment, and Defective Transfers’ in R Chambers, C Mitchell and J Penner (eds), Philosophical Foundations of the Law of Unjust Enrichment (Oxford, OUP, 2009) 331–34. 44 It is for this reason that some have suggested that Woolwich and other cases where the unjust factor is something other than a defect in or the failure of a condition attached to the claimant’s intention to enrich the defendant be removed from the category ‘unjust enrichment’. See eg Edelman and Bant (n 9) 294, 317. 45 Woolwich (n 1) 175, 177. The cases discussed are Air Canada v British Columbia [1989] 1 SCR 1161 and Case 199/82 Amministrazione delle Finanze dello Stato v SpA San Giorgio [1983] ECR 3595. 46 Woolwich (n 1) 202. The quote is from Fibrosa Spolka Akcyjna v Fairbairn Lawson Combe Barbour Ltd [1943] AC 32 (HL) 61.

Reasons for Restitution 107

and instead in the direction of principles which are in fact capable of justifying such a result. This can be seen too in the textbook treatment of Woolwich. Though, as I have argued, the prioritisation given to the ‘no taxes without Parliamentary authority’ principle may not in fact reflect the content of the majority speeches, it does at the least reveal an understanding that something more needs to be said to account for the decision beyond simply stating that the defendant was liable because unjustly enriched.

G. INTENTIONS AND CONDITIONS

I noted earlier that the references in Lord Goff and Lord Browne-Wilkinson’s speeches to the money being recoverable because there was no consideration for the payments must mean something. They must have been intended to identify, or at least to invoke, some reason for restitution. The difficulty is in saying what this reason is. One possibility, favoured by most commentators, is the constitutional principle discussed already, though this interpretation of their speeches is to my mind rather strained. There is, however, a second possibility; one which has been almost wholly overlooked in discussion of the case. It is clear that Woolwich could not have recovered on the basis that they paid by mistake, for the simple reason that they were not mistaken. Woolwich knew, or at least strongly suspected, that the demand was unlawful. However, though the issue is complicated slightly by the mistake of law bar which, at the time, remained part of English law, there is no doubt that if Woolwich had been able to identify a relevant mistake then recovery on this basis would have been uncontroversial. Now, as I have argued already, mistakes are not in themselves reasons for restitution. Rather the reason for restitution is some principle which dictates that mistaken payments and other transfers should be reversed. What could this principle be? In an earlier article I argued that the principle is one entailed by the notion of private property: that owners of assets – including money – have an exclusive interest in determining their disposition, which means that, as owner, it is me and me alone who gets to decide how the asset is to be used and hence who is to receive and take the benefit of it.47 Where an asset to which I was entitled at the point of transfer comes into your hands without my full, effective consent, then my interest in that asset provides a reason for enabling me to recover it, or at least its value, from you. Mistakes are therefore significant because they demonstrate that the claimant’s consent to the transfer – his intention to transfer (title to) that asset to the defendant – was defective or incomplete. I, of course, think this account is a good one and, as I sought to show in that article, the numerous objections raised to this and similar arguments are misplaced. But even if we don’t accept it, I think it is plain that the principle which justifies restitutionary claims in cases of mistake must in some way connect to the importance of the claimant’s consent to the transfer; more specifically, to the quality and scope of that consent. In other words, mistakes are normatively significant and ground restitutionary claims because, for some as yet unsettled background reason, transfers of (certain) assets and/or benefits are fully effective – and hence irreversible – only if the claimant properly consented to them. Or to put it

Webb (n 31).

47

108 Charlie Webb another way, mistaken payments are recoverable because the mistake reveals some defect in or limit to the claimant’s consent to the payment.48 As soon as we see this, it becomes clear that mistake cases involve just one application of the relevant principle. A claimant’s consent to a transfer can be impaired or limited in other ways too. As noted above, there is a good argument that some forms of compulsion have this effect and so support a restitutionary claim on an analogous basis. This will also be the case where the claimant makes the transfer subject to a condition which subsequently fails. This same line of reasoning tells us that there should, in principle, also be unjust factors of ignorance and powerlessness.49 So, the important question, in so far as we are concerned with reasons for restitution, is not whether the claimant was mistaken but whether he (truly, fully) intended/consented to the transfer. And while mistake is one way – perhaps the most common way – of a claimant establishing that his consent was not full and true, it is not the only way he can do this. Now, returning to Woolwich we have to ask a different question: not were Woolwich mistaken (or indeed suffering from compulsion), but rather did Woolwich truly, fully intend that the money be received and retained by the Revenue? At one level, of course they intended the payment. The payments were not the result of a slip of the hand; they were chosen. Woolwich gave careful thought to whether to make the payments and did so with its eyes open. Nonetheless it should be plain that its intention – its consent to the transfer – was not unconditional. It clearly didn’t intend that the Revenue be free to keep the money even if it turned out that the demand was unlawful and the money was not due. It did not set out to make a gift to the Revenue, intending that it should retain the money irrespective of the outcome of the judicial review proceedings.50 Rather, the lawfulness of the demand – and hence that the money was in fact due – was a condition to which the claimant’s consent to the money remaining with the Revenue was subject.51 When the demand was held to be unlawful this condition failed. As such, though it was not mistaken, Woolwich no more intended the Revenue to retain the money in the event that the Regulations were held to be ultra vires than a claimant who had paid on the mistaken and unquestioned assumption that the demand was lawful and the money owing. The fact that Woolwich’s state of mind when making the payments couldn’t be fitted squarely within any of the intent-focused unjust factors simply shows that these factors don’t cover the whole field of non-intended benefits.52 There was a clear sense in which Woolwich did not intend that the Revenue should be free to keep the 48 I analyse what mistakes tell us about a claimant’s consent/intentions and quell one possible objection to this analysis in C Webb, ‘Intention, Mistakes and Resulting Trusts’ in C Mitchell (ed), Constructive and Resulting Trusts (Oxford, Hart Publishing, 2010). 49 See eg Birks (n 5) 140–42, 174. For the argument that these are all applications of a broader principle or reason for restitution, see ibid 100–103. 50 Vaisey J expressed just this conclusion in a claim arising from very similar facts (Sebel Products Ltd v Customs and Excise Commissioners [1949] Ch 409 (Ch) 413): ‘What is, to my mind, incredible is that [a reasonable person] could ever have supposed that the money was being either paid by the plaintiffs or received by the defendants with the intention or on the footing that the defendants were to keep it in any event’. (Vaisey J went on to hold that this supported a finding of a contract between the parties for the return of the money in that event. This conclusion has been rightly criticised, but this gives us no reason to doubt the accuracy of his observations about the claimant’s intentions.) 51 See too Birks, Unjust Enrichment (n 8) 134–35. 52 A similar problem of fitting a clear case of non-intended transfer into the existing list of unjust factors can be found in Nurdin & Peacock plc v DB Ramsden & Co Ltd [1999] 1 WLR 1249 (Ch), and is discussed in Webb (n 48) 325–27. As before, the fact that courts and commentators sometimes misidentify or fail to spot unjust factors is not a reason for doubting the usefulness – indeed the necessity – of an inquiry into unjust factors.

Reasons for Restitution 109

money. And, accordingly, the same principle as justifies recovery in cases of mistake also justifies the claim here. The reason for restitution in each case is one and the same.

H. CONCLUSION

This intention-focused rationale is, I suspect, far nearer to what motivated Lords Goff and Browne-Wilkinson when they spoke of recovery being grounded in the absence of consideration for the payment or in the fact that the money simply was not due. What is the relevance of the money not being due? It can only be that Woolwich paid the money only on the basis (meaning footing, rather than assumption – this after all wasn’t their assumption) that it was due, which is just another way of saying that its consent to the payment (or at least to the Revenue’s retention of the money) was conditional on the money being due. This also seems to me to be a plausible interpretation – certainly more plausible than an invocation of the constitutional principle – of the ‘common justice’ which Lord Goff considered supported Woolwich’s claim. As such I see no basis for the conclusion that ‘private law cannot . . . single-handedly provide the reason for restitution in cases such as . . . Woolwich’.53 On the contrary, we do not need to look to the Revenue’s status as a public body to ground its liability. The better view then is that we can identify alternative reasons for restitution in Woolwich, each of which operates independently and each of which is sufficient to justify the Revenue’s liability.54 This is not to say that the Revenue’s status as a public body can be ignored when considering such claims. To know whether a particular claim is justified, we need to consider not just what reasons point in favour of allowing such a claim but also what reasons we may have to deny or restrict recovery. And no doubt there are reasons which are unique or at least particularly salient to public bodies, given their distinctive functions and responsibilities, which may, on occasion, justify placing restrictions on the claims which may be brought against them. These reasons will apply and must be factored in whatever positive reason for restitution we identify. Returning to Woolwich itself, though the immediate practical impact of the decision is both important and clear – the recognition that money paid as tax which has not been lawfully levied may be recovered as of right – what it tells us about the structure and normative foundations of the law of restitution is a matter of some doubt, for the simple reason that it is unclear what Woolwich, in more general terms, stands for. Nonetheless, though it may not provide the answers, the questions raised by Woolwich – What are (good) reasons for restitution? What role does unjust enrichment play in explaining and justifying instances of restitutionary liability? What is at stake in the proposal that English law should adopt an absence of basis model? – go to the heart of our understanding of the law of restitution and remain as vital today as they were then.

R Williams, Unjust Enrichment and Public Law (Oxford, Hart Publishing, 2010) 33. cf Birks, Unjust Enrichment (n 8) 133–35.

53 54

6 Restitutionary Claims by Indirect Taxpayers CHARLES MITCHELL*

A. INTRODUCTION

Investment Trust Companies (in liq) v HMRC 1 concerned a group of claims against HMRC to recover money paid in the mistaken belief that it was due as Value Added Tax. Previous English cases on the recovery of money mistakenly paid as VAT have all concerned the statutory and common law rights of businesses which have paid HMRC in their capacity as ‘taxable persons’ under the Value Added Tax Act 1994.2 The claims in the ITC case were therefore unusual, as they were made not by taxable persons but by consumers.3 The claimants received services from fund management companies to which they paid money in the mistaken belief that it was due as VAT. The managers accounted for this money to HMRC as output tax, deducting sums which they had paid as input tax on associated supplies, and paid HMRC the net amount. It then became clear that VAT should not have been charged on the services supplied by the managers to the claimants. The managers therefore recovered some of their payments to HMRC under VATA, section 80, and they paid these amounts over to the claimants. However the claimants were still out of pocket after this had been done, and so they brought their own common law claims in unjust enrichment against HMRC to recover the balance. These claims required Henderson J to answer some interesting but difficult questions: could the ingredients of a common law claim in unjust enrichment against HMRC be established on the facts, was such a claim excluded by statute, did EU law give the claimants a directly effective right to reimbursement from HMRC, and if so, what impact would this have on limits placed on the claimant’s rights under English law?

* I am grateful to Max Schlote for his helpful comments on a draft version of this chapter. 1 Investment Trust Companies (in liq) v HM Revenue & Customs [2012] EWHC 458 (Ch), [2012] STC 1150 (hereafter ‘ITC ’). 2 Hereafter ‘VATA’. The case law is discussed in R Davies and D Rudling, Tolley’s Value Added Tax 2011–12, 2nd edn (London, LexisNexis, 2011) paras 51.7–51.18; De Voil Indirect Tax Service, looseleaf edn (London, LexisNexis, 2011) paras V5.159B–V5.159C; C Mitchell, P Mitchell and S Watterson (eds), Goff & Jones: The Law of Unjust Enrichment, 8th edn (London, Sweet & Maxwell, 2011) paras 22.11–22.12 and 32.02–32.10. 3 For claims of the latter kind under EU law, see Case C-35/05 Reemstma Cigarettenfabriken GmbH v Ministero delle Finanze [2007] ECR I-2425 and Case C-94/10 Danfoss A/S & Sauer-Danfoss ApS v Skatteministeriet CJEU 20 October 2011.

112 Charles Mitchell Henderson J held that a common law claim could be established on the facts, but that this was excluded by VATA, section 80(7). He also held that EU law required English law to provide the claimants with a direct remedy against HMRC, subject to limitation defences analogous to those that would have applied to claims against HMRC by the managers. However, he deferred final resolution of the EU law issues in the case, pending the CJEU’s judgment on the reference in Littlewoods Retail Ltd v HMRC 4 and the Supreme Court’s judgment in Test Claimants in the FII Group Litigation v HMRC.5 The English law issues in the ITC case are discussed in this chapter, and the EU law issues are discussed by Maximilian Schlote in chapter ten.

B. VALUE ADDED TAX

VAT is charged by virtue of directly effective provisions of EU law, implemented in the United Kingdom by VATA. Hence EU law plays a leading role in the interpretation of VATA and the practical application of VAT. The key EU legislation is the Sixth VAT Directive,6 which has been recast as the 2006 Directive.7 The San Giorgio principle8 also provides that taxpayers have a directly enforceable right under EU law to the repayment of money paid as VAT that was charged in breach of EU law. VAT is a consumption tax assessed on the value added to goods and services. It is levied indirectly, by requiring registered businesses to collect the tax from their customers and pay it to HMRC. VAT is not intended to be a charge on businesses, and it is intended that the economic cost of the tax should be borne by final consumers, ie those who buy goods or services for non-business purposes. Businesses sometimes absorb the cost of VAT into their overheads rather than passing this on to their customers, eg because they want to make their prices competitive. However they are not required to do this, and in the words of the CJEU, the sole requirement imposed on them, when they take part in the production and distribution process prior to the stage of final taxation . . . is that, at each stage of the process, they collect the tax on behalf of the tax authorities and account for it to them.9

VAT is collected fractionally via a system of partial payments. Registered businesses must charge VAT on their sales (‘output tax’) and pay VAT on purchases from their own suppliers (‘input tax’). The amount that a business must pay to HMRC is calculated by subtracting the amount of input tax that it has paid on its purchases from the amount of output tax that it has charged on its sales, where the input tax was a cost component of making the onward supplies on which the output tax was charged. If the output tax charged by a business 4 The reference was made by Vos J in Littlewoods Retail Ltd v HM Revenue & Customs [2010] EWHC 2771 (Ch), [2011] STC 171. The court has now given its judgment: Case C-591/10 Littlewoods Retail Ltd v HM Revenue & Customs, CJEU, 19 July 2012, [2012] ECR I-000, [2012] STC 1714. In this chapter, references to the CJEU (ie the Court of Justice of the European Union) include the ECJ (ie the European Court of Justice). 5 Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2012] UKSC 19, [2012] 2 AC 337. 6 EC Council Directive 77/388 of 17 May 1977 on the harmonisation of the laws of the Member States relating to turnover taxes. 7 EC Council Directive 2006/112 of 28 November 2006 on the common system of Value Added Tax. 8 Named for Case C-199/82 Amministrazione delle Finanze dello Stato v SpA San Giorgio [1983] ECR 3595. 9 Case C-317/94 Elida Gibbs Ltd v Customs & Excise Commissioners [1996] ECR I-5339, [1997] QB 499 [22].

Claims by Indirect Taxpayers 113

exceeds the input tax that it has paid then it must pay the difference to HMRC; if its input tax exceeds its output tax, then HMRC must refund the difference. By this device VAT is collected along the chain of supply by reference to each supplier’s contributions of labour and mark-up (its ‘value added’), and the cost of the tax is ultimately borne by the final consumer of the goods or services within the price that it pays at the end of the chain. These principles may be illustrated by a simple example of the manufacture and sale of wooden figurines. Raw materials are sold to a manufacturer, which makes the figurines and sells them to a retailer, which packages them and sells them to a final consumer. Assuming that all of the supplies in this chain are subject to VAT at the same 20 per cent rate, and that each supplier is entitled to full credit for its input tax, VAT would be charged along this chain of supply as follows. Price £ inc VAT at 20%

Output Tax £

Input Tax £

VAT £ Paid to HMRC

Supplier to manufacturer

60

10

—

10

Manufacturer to retailer

180

30

10

20

Retailer to consumer

360

60

30

30 60

It is no coincidence that the VAT element of the price paid by the final consumer in this example exactly equals the total of the VAT collected by HMRC from each business along the chain of supply. This is the effect of the system of deductions, the purpose of which is to ensure the neutrality of the collection machinery so far as taxable persons are concerned. As stated by the CJEU, a basic feature of the VAT system is that VAT is chargeable on each transaction only after deduction of the amount of VAT borne directly by the cost of the various price components of the goods and services. The procedure for deduction is so arranged that only taxable persons are authorised to deduct from the VAT for which they are liable the VAT which the goods and services have already borne.10

Reviewing the VAT system in the ITC case, Henderson J made the following additional comments with regard to the relationship between HMRC, taxable persons, and the final consumers of goods and services on which VAT has been charged: [At] a fairly high level of generality, [it can be said] that for the purposes of VAT the final consumer is properly to be regarded as the taxpayer, and that the role of the intermediate taxable persons in the chain of supply is to collect the tax and account for it to the tax authorities . . . [However] the final consumer is not in any ordinary sense a taxpayer vis-à-vis HMRC. It is the supply of goods or services to the final consumer which triggers the liability to pay and account for VAT on the supply, and the supplier is liable accordingly, whether or not he passes on the tax 10 ibid [23], citing Case 15/81 Gaston Schul Douane Expediteur BV v Inspecteur der Invoerrechten en Accijnzen, Roosendaal [1982] ECR 1409 [10].

114 Charles Mitchell charge to his customer. Conversely, the customer is under no liability to HMRC for the VAT, even if the supplier fails to pay it, and the customer’s obligation to pay the VAT to the supplier rests only in contract. If, for whatever reason, the supplier fails to charge VAT to the customer, he will have no remedy against the customer unless the customer is contractually bound to pay it . . . [Hence] the liability of the supplier to pay and account for output tax is a primary one, and in legal terms the taxable person is the supplier, not the customer.11

C. THE FACTS OF THE ITC CASE

The claimants were closed-end investment funds, constituted as limited liability companies. They bought fund management services from fund managers. The parties’ contracts typically provided that the managers would be remunerated by the payment of fees plus VAT ‘if applicable’. It was generally understood at the time that VAT was chargeable on the services supplied to the claimants, although fund management services supplied to authorised unit trust schemes were exempted from VAT from 1 January 1990 onwards, and this exemption was extended to fund management services supplied to open-ended investment companies from 1997 onwards. The managers therefore charged VAT at the standard rate on the services supplied to the claimants from 1990 onwards, and the claimants paid the sums charged as VAT believing that the tax was due. In an unconnected case, JP Morgan Fleming Claverhouse Investment Trust plc v HMRC,12 the failure of the UK legislation to exempt fund management services supplied to closedend investment trust companies was then successfully challenged in the CJEU, on the ground that it was incompatible with Article 13B(d)(6) of the Sixth Directive.13 Following the CJEU’s disposal of this case, HMRC published Business Brief 65/07, in which they accepted that fund management services supplied to closed-end investment trust companies are exempt from VAT, and VATA was amended to exempt such services from VAT with prospective effect from 1 October 2008. Following HMRC’s response to the Claverhouse decision, it was also clear that supplies of fund management services by the managers in the ITC case should have been exempt from VAT under Article 13B(d)(6) from 1 January 1990 onwards; that Article 13B(d)(6) had had direct effect at all material times; that it had not been correctly transposed into UK legislation between 1 January 1990 and 1 October 2008; and that the managers could therefore make claims under VATA, section 80 for the crediting and repayment of sums for which they had accounted as output tax on the services which they had supplied. The managers successfully made such claims against HMRC, and paid their recoveries over to the claimants.14 However, this did not result in the claimants being fully reim ITC (n 1) [50]. Case C-363/05 JP Morgan Fleming Claverhouse Investment Trust plc v HM Revenue & Customs [2007] ECR I-5517. 13 In 2011, a reference was made to the CJEU by the First-tier Tribunal (Tax Chamber) to determine whether the provision of fund management services to the trustees of defined benefit pension schemes should also have been exempt: Wheels Common Investment Fund Trustees Ltd v HM Revenue & Customs [2011] UKFTT 534 (TC), [2011] Pens LR 387; Case C-424/11 [2011] OJ C311/23. If it is found that such services should also have been exempt then this will trigger a further set of claims against HMRC raising identical issues to those which have been litigated in the ITC case. 14 If the managers had not made legally binding arrangements with the claimants to pay over the fruits of their recoveries from HMRC, then their claims against HMRC would have failed by reason of the statutory passing on defence contained in VATA, s 80(3). 11 12

Claims by Indirect Taxpayers 115

bursed, for two reasons. One was that some of the managers’ claims against HMRC were time-barred. The other was that they could not recover the whole amount of the output tax for which they had accounted, but could only recover the net amount which they had paid, ie the amount of output tax for which they had accounted, less the amount of their input tax credits. The first of these problems was created by VATA, section 80(4). At the relevant times, this sub-section provided that HMRC should not be liable on a claim in respect of an overpayment where the claim was made more than three years after the end of a relevant accounting period.15 In early 2004, when the Claverhouse litigation began and was publicised, the managers claimed refunds in respect of VAT accounting periods from 2001 to 2004, but made no claims in respect of any earlier accounting period because of the limitation bar established by section 80(4). A complication then ensued. The three-year limitation period had been introduced by legislation on 4 December 1996, replacing the limitation period that had previously applied to section 80 claims. This had given claimants six years in which to claim, but in cases of mistaken payments, the period had not started to run until the date when the mistake had been known to the claimant or had been reasonably discoverable. When the new limitation rule was enacted in 1996, no transitional arrangements were put in place to protect those claimants whose accrued rights would be affected by the change, and in 2008 the House of Lords held in Fleming v HMRC 16 that the introduction of the three-year period was therefore incompatible with EU law and had to be disapplied in respect of rights that had accrued by 4 December 1996. Following this decision, the managers therefore made further claims for refunds in respect of accounting periods ending before 4 December 1996. However they still could not claim refunds in respect of accounting periods ending after 4 December 1996 but before the dates in 2001 that were three years before the date of their first set of claims. One of the claimants’ objectives in bringing their own actions against HMRC in the ITC case was to recover in respect of money paid as VAT during the ‘dead periods’ between 1996 and 2001. Of course, the claimants could do this only if their own actions were brought in time. However HMRC conceded that if the claimants had a common law right to restitution, then their actions would be governed by the Limitation Act 1980, section 32(1)(c), under which they would have six years in which to claim, running from the date when they could first reasonably have discovered their mistake in thinking that the money they had paid as VAT had been legally due.17 HMRC also conceded that this would be the date when the CJEU handed down its decision in Claverhouse.18 This concession was in line with the approach previously taken by the House of Lords to a similar question in Deutsche Morgan Grenfell plc v IRC.19 The second problem was created by VATA, section 80(2A). To illustrate the effect of this sub-section, the claimants posited a hypothetical situation in which a claimant investment trust company pays £100 as VAT to a manager. The manager accounts to HMRC for £100 of output tax and £25 of associated input tax, and makes a net cash payment to HMRC of £75. The manager’s own supplier incurs zero input tax, and so makes no deduction from 15 Section 80(4) was later amended with prospective effect by the Finance Act 2008, Sched 39, para 36, to give claimants four years from the relevant date in which to claim. 16 Fleming v HM Revenue & Customs [2008] UKHL 2, [2008] 1 WLR 195. 17 ITC (n 1) [80]. 18 ibid [80]. 19 Deutsche Morgan Grenfell plc v Inland Revenue Commissioners [2006] UKHL 49, [2007] 1 AC 558 [31], [71] and [144].

116 Charles Mitchell the £25 of output tax for which it must account, and so pays £25 cash to HMRC. The upshot is that HMRC receives £75 + £25 = £100. It then transpires that the £100 output tax for which the manager accounted to HMRC was wrongly charged because the relevant services were exempt. However, VATA, section 80(2A) does not permit the manager to recover more than £75 from HMRC, because it provides that the amount of input tax deducted by the manager in relation to the supplies on which VAT was wrongly charged must be deducted from the amount of output tax that the manager can recover from HMRC under section 80. Hence the second reason why the claimants brought their own actions against HMRC was to recover the equivalent of the £25 shortfall, ie the amounts which the managers could not recover under section 80 because they corresponded to the amounts of the managers’ input tax credits. VATA provides final consumers with no statutory right against HMRC to recover money paid as VAT that is not due.20 Hence the claimants’ only hope of recovery was to argue that the common law of unjust enrichment provided them with a right to restitution. To determine whether they had such a claim Henderson J adopted the four-part analysis of actions in unjust enrichment described by Lord Steyn in Banque Financière de la Cité v Parc (Battersea) Ltd.21 This requires the court to consider four questions: (1) was the defendant enriched, (2) was the enrichment at the claimant’s expense, (3) was the enrichment unjust, and (4) are there any defences? Henderson J gave a positive answer to the first three questions but denied the claim because he considered that it was excluded by VATA, section 80(7). He tested each question against the facts of the claimants’ hypothetical example, and his analysis of each question will be considered here in turn. Like Henderson J’s judgment, this discussion will proceed by reference to the facts of the example.

D. WERE HMRC ENRICHED?

In respect of the dead periods, the claimants alleged that HMRC received £100 worth of value, consisting of the £75 cash and the £25 input tax credit which the manager used to discharge the balance of its supposed output tax liability. In respect of the other periods the claimants had already been repaid the £75 cash, which had been recovered by the managers from HMRC and passed back to the claimants. Hence in respect of the other periods, the claim related to the £25 input tax credit only. So far as the £75 cash payment was concerned, HMRC conceded that they were enriched when they received this money,22 but denied that this enrichment was gained at the claimants’ expense. Hence the parties’ arguments with respect to the £75 were focused not on the question whether HMRC were enriched, but on the questions whether the law of unjust enrichment considers that benefits can ever be received ‘at the expense of’ remote rather than immediate enrichers, and, if so, whether the £75 was received at the claimants’ expense on the facts of the ITC case. These arguments are considered in Part E. 20 As confirmed by the Edinburgh VAT & Duties Tribunal in Aberdeen Estates Ltd VTD 13622, 28 September 1995. 21 Banque Financière de la Cité v Parc (Battersea) Ltd [1999] AC 221 (HL) 227. See too Cressman v Coys of Kensington (Sales) Ltd [2004] EWCA Civ 47, [2004] 1 WLR 2775 [22]; Chief Constable of Greater Manchester v Wigan Athletic AFC Ltd [2008] EWCA Civ 1449, [2009] 1 WLR 1580 [38] and [54]. 22 ITC (n 1) [41]: HMRC ‘did not dispute that [they] were enriched by the net payments of £75 which they actually received’.

Claims by Indirect Taxpayers 117

So far as the £25 input tax credit was concerned, HMRC assumed that the enrichment in respect of which the claim was made was the right acquired by HMRC against the supplier when the manager paid £25 to the supplier and thereby rendered the supplier liable to account to HMRC for £25 output tax. HMRC argued that this was not a recoverable benefit for the purposes of a claim in unjust enrichment because it was a benefit to which HMRC were legally entitled, the Claverhouse mistake not having affected the fact that the £25 paid by the manager to the supplier was legally chargeable as VAT.23 However, Henderson J held that this argument was beside the point because HMRC’s right to receive £25 as output tax from the supplier was not the enrichment in respect of which the claim was made.24 The claimants argued, and Henderson J accepted, that HMRC were enriched for the following, different, reasons. A taxable person is entitled to an input tax credit when it pays VAT to its supplier and this payment is a cost component of making onward supplies to a customer on which further VAT is charged and accounted for as output tax. In these circumstances HMRC is placed under an obligation. This may be an obligation to pay the taxable person a sum in cash representing the difference between the input tax which it can reclaim and the lesser amount of output tax which it owes (for example, because its onward supplies are zero-rated).25 Or HMRC may owe an obligation to let the taxable person set off the amount of its input tax payment against its output tax liability in respect of associated supplies, where its output tax liability is greater than its input tax credit.26 In the ITC case, the manager accounted to HMRC for the £25 received from the claimant as output tax, and HMRC were enriched because they accepted the manager’s input tax credit in lieu of cash payment for its £25 output tax liability, with the result that they were discharged from the obligation that they would otherwise have owed, to let the manager use this credit to pay for a legally valid output tax liability. Thus, Henderson J held that HMRC were ‘indirectly’ enriched because The VAT which the investment trusts paid to the Managers . . . reduced the amount of [the Managers’ output tax payments] by being set against and absorbing the related input tax which HMRC would otherwise have been obliged to pay or credit to the Managers.27

The judge also held that this analysis was consistent with VATA, section 25(3) and Articles 17(1) and (2), 18(2) and 22(5) of the Sixth Directive, which established that HMRC should . . . be regarded as enriched by the £25, because although the £25 was not paid to HMRC by the Managers, it was nevertheless used to give them a credit, at HMRC’s expense, for the input tax attributable to their investment management services which was wrongly thought to be deductible on the footing that the services were not exempt from VAT. HMRC therefore ended up out of pocket to the extent of the input tax.28

This passage is over-compressed and consequently rather hard to understand, but Henderson J’s essential point was that HMRC were enriched by the discharge of the obligation ITC (n 1) [41]: ‘the £25 was properly payable to HMRC in any event’. ibid [45]: ‘simply irrelevant’. 25 ibid [42] and [44]. 26 cf Times Right Marketing Limited (in liq) v HM Revenue & Customs London VAT and Duties Tribunal, 13 March 2008 [30]: a taxable person can be said to have ‘accounted for and paid [VAT] . . . by way of input tax credit’. 27 ITC (n 1) [40]. 28 ibid [45]. 23 24

118 Charles Mitchell which they owed the manager. This analysis is consistent with cases in the law of unjust enrichment which treat the discharge of an obligation as a valuable benefit. Most of these concern the situation where a claimant pays off a defendant’s liability to a third party,29 but closer to the facts of the present case is Gibb v Maidstone and Tunbridge Wells NHS Trust,30 where the Court of Appeal held that in principle a claim in unjust enrichment would lie to recover the value of a legal obligation previously owed by the defendant to the claimant, where the claimant had surrendered her corresponding legal right against the defendant under a void compromise agreement.31 In this case the defendant would be enriched because it would no longer have to perform its obligation and would have given the claimant nothing in exchange for this benefit. By analogy, one can argue that HMRC are better off if they accept an input tax credit in (part) payment of a non-existent output tax liability, because they will not thereafter be obliged to accept the tax credit in reduction of some other valid output tax liability owed by the taxable person, which will accordingly have to be paid in full.32 As Maximilian Schlote points out, however, there is a significant problem with Henderson J’s analysis.33 This is that the manager’s right to an input tax credit under VATA, sections 25 and 26 depended on the manager having made associated taxable supplies to the claimant. Thus section 26(1) and (2) provide that a credit can be claimed for ‘input tax . . . attributable to . . . taxable supplies’.34 It was assumed by the parties that the manager was entitled to do this, but they were mistaken. They would have been right if the CJEU had held in Claverhouse that the services were zero-rated taxable supplies. But in fact the court held that they were exempt supplies,35 and this meant not only that output tax could not have been charged on these services, but also that no input tax credit could have been offset against any output tax liability that the manager supposedly owed. As provided by the VAT Regulations 1995, regulation 101(2)(c), [no] part of the input tax on such goods or services as are used or to be used by [a taxable person] exclusively in making exempt supplies shall be attributable to taxable supplies.

And as the CJEU said in BLP Group plc v Customs & Excise Commissioners, [where] a taxable person supplies services to another taxable person who uses them for an exempt transaction, the latter person is not entitled to deduct the input VAT paid.36 29 eg Filby v Mortgage Express (No 2) Ltd [2004] EWCA Civ 759 [62] (May LJ): ‘the defendant is enriched if his financial position is materially improved . . . as here where the defendant is relieved of a financial burden’. See too Moule v Garrett (1872) LR 7 Ex 101, 104; Brook’s Wharf and Bull Wharf Ltd v Goodman Bros [1937] 1 KB 534 (CA) 544; Regional Municipality of Peel v Canada [1992] 3 SCR 762, 790; Littlewoods Retail Ltd v HM Revenue & Cusoms [2010] EWHC 1071 (Ch), [2010] STC 2072 [121]. 30 Gibb v Maidstone and Tunbridge Wells NHS Trust [2010] EWCA Civ 678, [2010] IRLR 786 [24]–[37]. 31 See esp Laws LJ’s statement that there is no difference between ‘a benefit consisting in money paid and a benefit consisting in a claim foregone’: ibid [30]. 32 A similar argument was made in relation to the use of tax reliefs to reduce unlawful tax liabilities in Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs but was rejected by the CJEU and the CA: [2007] 1 CMLR 35; [2010] EWCA Civ 103, [2010] STC 1251. The courts’ reasons are criticised by Maximilian Schlote in chapter 10 of the present volume, at pp 202–06, and by Goff & Jones (n 2) paras 2.12 and 6.07–6.09. 33 See chapter 10 of the present volume, at p 195. 34 See too VAT Regulations 1995 (SI 1995/2518), reg 101(2)(b). 35 cf VATA, s 4(2): ‘A taxable supply is a supply of goods or services made in the United Kingdom other than an exempt supply’. 36 Case C-4/94 BLP Group plc v Customs & Excise Commissioners [1996] 1 WLR 174 [28], construing Article 17 of the Sixth Directive. See too Birmingham Hippodrome Theatre Trust Ltd v HM Revenue & Customs [2011] UKFTT 117 (TC) [4]: ‘a trader who makes taxable output supplies is entitled to deduct his input tax; a trader who makes exempt output supplies is not entitled to deduct his input tax’.

Claims by Indirect Taxpayers 119

Another way of putting this point is to say that when the manager paid the £25 cash to the supplier it acquired no more than a potential right to set off an equivalent amount as input tax against any amount for which it was liable to account as output tax in respect of associated taxable supplies. This was a ‘potential right’ in the sense that if the manager provided associated taxable supplies then its right of set-off would crystallise, and the amount of output tax payable to HMRC would be reduced by the amount of the input tax credit. However, if the manager never provided associated taxable supplies to anyone then it would never become liable for output tax against which an input tax credit could legitimately be set off, and so it would effectively be a final consumer of the goods or services received from the supplier.37 Schlote argues that the best way to resolve the ITC case would have been to allow the manager to recover the whole of the £100 output tax for which it accounted to HMRC, and to pass this back to the claimant. He also argues that it would have been appropriate to short-circuit this process and to let the claimant recover directly from HMRC in order to comply with the EU law requirement that recovery by the claimant of the sums paid in breach of EU law should not be excessively difficult. Finally, he says, HMRC should then have been allowed to recover £25 from the manager on the ground that they made a mistake when they allowed the manager an input tax credit in the belief that the management services were chargeable supplies. While accepting that these steps would have put all the parties into the final position that they ought to have occupied, both as a matter of EU law and as a matter of English law, my own view is that the claimant should not have been entitled to recover £25 from HMRC. The manager never had a valid legal right to use a £25 input tax credit to reduce its supposed £100 output tax liability, because the manager’s right to an input tax credit was conditional on it owing a valid output tax liability, and this condition was not satisfied. By the same token, HMRC never owed the manager a liability corresponding to a £25 input tax credit, and so HMRC cannot have been enriched by the discharge of any such liability. Probably this also explains why VATA, section 80(2A) did not permit the manager to recover its mistaken ‘payment’ by use of its input tax credit: HMRC were not enriched when the manager purported to use an input tax credit to pay its supposed output tax liability. Certainly, one would expect the law to give consistent answers to the enrichment questions raised by the manager’s statutory claims under VATA and the claimant’s common law claims, and it is a significant objection to Schlote’s (and Henderson J’s) approach that it would hold HMRC to have been enriched for the purposes of one set of claims, but not for the purposes of the other. If the foregoing analysis is correct, then the question obviously arises whether there was anyone apart from HMRC from whom the claimant could have recovered its missing £25? The answer is that the claimant should have been able to recover this sum from the manager. Henderson J considered this point but concluded that no such claim could lie because the manager would have a cast iron defence of change of position, having accounted to HMRC for the entirety of the tax as output tax, and having retained no benefit from it.38

37 cf British Eventing Ltd v HM Revenue & Customs [2010] UKFTT 382 (TC) [175]: ‘some supplies are exempt, putting a taxable person in the same position as a consumer: unable to recover input tax on purchases’. 38 ITC (n 1) [69]. See, too, his comments at [135]–[139].

120 Charles Mitchell This was correct so far as the £75 was concerned. The manager paid HMRC £75 cash and HMRC were better off and the manager worse off to that extent. However Henderson J’s finding in relation to the £25 was incorrect. The alleged discharge of HMRC’s obligation never took place because the management services were not associated chargeable supplies and so the manager was not entitled to an input tax credit. Hence, the value of the £25 never passed from the manager to HMRC, and although no claim could have been made against HMRC, a claim might therefore have been possible against the manager, whose position had not changed because it had never validly used its potential right to claim an input tax credit. Furthermore, the manager could not have made the different argument that it had changed its position when it failed to recoup the cost of its £25 payment to the supplier by charging the claimant an inflated price for its services, because the evidence showed that even if the manager had known that these services were exempt supplies it would have absorbed this cost, and would not sought to pass it on to the claimant.39

E. WAS HMRC’S ENRICHMENT GAINED AT THE CLAIMANTS’ EXPENSE?

It follows from the foregoing discussion that this question should only have arisen with regard to the claims to recover the £75 cash received by HMRC, since the claims for the £25 input tax credits should have failed on the basis that HMRC had not been enriched. However, Henderson J held that enrichment had been established for the purposes of all of the claims, and went on to give a single set of reasons for his further conclusion that the benefits gained by HMRC had all relevantly been gained at the claimants’ expense. The question whether this was so divided into two questions. The first was a question of pure law: what test is used to determine whether an enrichment has been gained at a claimant’s expense in multi-party cases where the defendant is the remote recipient of a benefit (ie where he receives a benefit which ultimately comes from the claimant, but which passes through the hands of an intermediate party from whom the defendant directly receives it)? The second was a question of fact: applying the appropriate legal test, were the £75 cash and the (supposed) discharge of HMRC’s £25 obligation relevantly gained by HMRC at the claimants’ expense? The claimants argued that English law takes a liberal approach when asking whether a benefit has been received ‘at the claimant’s expense’ in multi-party cases, using a causal test subject to restrictions, and that applying this test to the facts would produce the conclusion that both benefits were received at the claimants’ expense. HMRC argued that English law does not generally permit claims against remote recipients, and that although it recognises exceptions to this general prohibition, neither benefit fell within the scope of an exception. Henderson J took the more restrictive view of the legal test to be applied, holding that as a general rule benefits acquired by a defendant are not legally gained ‘at the expense of’ anyone other than the person from whom they have been directly received. However, he also held that on the facts of the example both benefits were relevantly acquired at the claimants’ expense because the circumstances of their receipt brought them within the ibid [84] and [137].

39

Claims by Indirect Taxpayers 121

scope of an exception that allows such a finding to be made where this reflects ‘the underlying commercial reality of a transaction’.40 The English law governing claims in unjust enrichment against remote recipients is obscure. There are few cases in which the courts have expressly considered when such claims should be permitted, and academic opinion is divided: some scholars argue that the law should operate a ‘directness’ or ‘privity’ rule subject to exceptions,41 while others (including the present author) consider that a causal test subject to restrictions should be used to determine the ‘at the expense of’ question in multi-party cases.42 All the participants in this debate agree that there are good reasons why many ‘indirect’ transfers should not be reversed. Various bad effects can flow from this, including conflicts with contractual arrangements entered by the parties,43 contradiction of the pari passu principle of insolvency law,44 increased risks of double recovery and double liability,45 and the practical difficulties of managing litigation between multiple interconnected parties.46 However, there is a consensus that forbidding every claim against a remote recipient would also be undesirable, because this would deny restitution to meritorious claimants. As Stephen Watterson has written, the dispute between proponents of the two approaches therefore boils down to a question of strategy:47 would the law be clearer and fairer if it adopted a narrow test subject to exceptions, or a wide test subject to limitations? In the ITC case Henderson J favoured the first approach,48 on the ground that this would be consistent with obiter dicta of the Court of Appeal in Kleinwort Benson Ltd v Birmingham CC,49 and that taking the second approach would require a ‘reformulation’ of the law, notwithstanding obiter remarks to the opposite effect by May LJ in Filby v Mortgage Express (No 2) Ltd.50 He also pointed to ‘the difficulty in framing the general rule in acceptable terms if it is not confined to direct recipients’.51 Henderson J overstated matters when he suggested that the ‘direct enrichment’ rule represents the current ‘formulation’ of the law: the truth is that judicial dicta on the point are sparse and generally unreasoned because the question has infrequently been presented to the courts in a way that enables them to identify and explore the policy issues in play. So far as the difficulty of framing a wider test is concerned, Henderson J was right to identify this as a problem for proponents of this approach, but wrong to imply that this is not also a problem for proponents of the ‘direct enrichment’ rule. As Watterson has observed, there ibid [72]. eg G Virgo, The Principles of the Law of Restitution, 2nd edn (Oxford, OUP, 2006) 105–12; B McFarlane, ‘Unjust Enrichment, Property Rights and Indirect Recipients’ [2009] RLR 37, esp 56–9; A Burrows, The Law of Restitution, 3rd edn (Oxford, OUP, 2011) 69–85. 42 eg P Birks, ‘“At the Expense of the Claimant”: Direct and Indirect Enrichment in English Law’ in D Johnston and R Zimmermann (eds), Unjustified Enrichment: Key Issues in Comparative Perspective (Cambridge, CUP, 2002) 493; P Birks, Unjust Enrichment, 2nd edn (Oxford, OUP, 2005) 86–98; C Mitchell, ‘Liability Chains’ in S Degeling and J Edelman (eds), Unjust Enrichment in Commercial Law (Sydney, Thomson Reuters, 2008) 131; Goff & Jones (n 2) paras 6.12–6.62; S Watterson, ‘“Direct Transfers” in the Law of Unjust Enrichment’ (2011) 64 Current Legal Problems 435. 43 cf Costello v MacDonald [2011] EWCA Civ 930 [23]. 44 cf Yew Sang Hong Ltd v Hong Kong Housing Authority [2008] HKCA 109 [13] and [30]–[32]. 45 cf Trustor AB v Smallbone (CA) 9 May 2000 [63] (double recovery); Greatworth v Sun Fook Kong Construction Ltd [2006] HKCFI 356 [52] (double liability). 46 cf SmithKline Beecham plc v Apotex Europe Ltd [2006] EWCA Civ 658, [2007] Ch 71 [41]–[42]. 47 Watterson (n 42) 448. 48 ITC (n 1) [67]. 49 Kleinwort Benson Ltd v Birmingham City Council [1997] QB 380, 393, 395 and 400. 50 Filby v Mortgage Express (No 2) Ltd [2004] EWCA Civ 759 [46]. 51 ITC (n 1) [67]. 40 41

122 Charles Mitchell is no consensus among those who use this language as to which types of transfer count as ‘direct’ transfers and which do not, nor is there a consensus as to which types of transfer should count as admissible exceptions to the general rule and which should not.52 Henderson J’s judgment makes matters no clearer in this respect. He did not try to justify the existence of the ‘direct enrichment’ rule from first principles, nor did he explain the circumstances in which the rule applies, simply assuming that the rule existed and applied to the facts of the example.53 He correctly said that ‘no exhaustive list of criteria for the recognition of exceptions’ has been offered by proponents of the ‘direct enrichment’ rule, and proceeded on the basis that exceptions can be recognised in a ‘pragmatic and step by step fashion’.54 He identified some bad effects that can flow from admitting an exception to the rule (risk of double recovery, conflict with contractual arrangements, etc), found that none would arise on the facts of the example, and held that it was therefore open to him to treat these facts as exceptional, and appropriate to do so since this would ‘give due weight to the economic reality which explains and underpins [the structure of the VAT legislation]’.55 Henderson J’s analysis should be applauded to the extent that he spelt out the policybased calculations that led to his conclusion. Such calculations are not always made explicit in the cases and there is no hope of understanding the law unless judges explain their reasoning. However, his analysis and conclusion were so closely tied to the facts of the case that there is little to be taken from them by a future court confronted with different facts.56 His reasoning went as follows.57 The claimants were not directly liable to pay money as tax to HMRC, and were only contractually liable to pay money ‘as VAT’ to the managers. Nevertheless ‘in economic terms’ it was the claimants and not the managers who were intended by the legislation to bear the cost of paying the tax received by HMRC.58 It was also commonly assumed by the parties that this is what would happen. Hence there was a sufficiently close link between the claimants’ payments to the managers, and the benefits acquired by HMRC, to make it appropriate to allow the claimants a direct claim against HMRC. Thus, Henderson J concluded that the scheme of VAT . . . is to impose the burden of the tax on the final consumer, and to make the suppliers of the goods or services the collectors of the tax on behalf of the tax authorities. In other words, VAT is a tax on the consumer, collected by the supplier, and paid or accounted for to HMRC. Viewed in this way, the nexus between the consumer and HMRC could hardly be closer Watterson (n 42) 442. cf Armstrong DLW GmbH v Winnington Networks Ltd [2012] EWHC 10 (Ch), [2012] Bus LR 1199 [97], where Stephen Morris QC, sitting as a deputy High Court judge, asserted that ‘a claim in unjust enrichment is only generally available where the benefit has been provided directly by the claimant to the defendant, and not where it has been provided indirectly via a third party’. In the latter case, he thought, ‘the defendant will have been enriched at the third party’s expense’. This assumed, incorrectly, that where an enrichment has passed from a claimant to a third party to a defendant, it cannot be said that the defendant has been enriched at the expense of the claimant and of the third party. 54 ITC (n 1) [68]. 55 ibid [72]. 56 Compare the very different circumstances of two cases in which the same question has arisen since the ITC case was decided: Menelaou v Bank of Cyprus plc [2012] EWHC 1991 (Ch) esp [22]; Relfo Ltd (in liq) v Varsani [2012] EWHC 2168 (Ch) [86]–[87]. 57 ITC (n 1) [72]. See also his comments at [50], cited in the text to n 10. 58 This is borne out by the rule in VATA, s 80(3), which provides that taxable persons can only recover money paid as VAT that is not due if they have entered legally binding arrangements to return the money back up the chain of payments to their customers, where the customers have borne the cost of paying the money. 52 53

Claims by Indirect Taxpayers 123 or stronger, and in economic terms the person at whose expense unlawful VAT is paid to HMRC is indubitably the consumer.59

F. WAS HMRC’S ENRICHMENT UNJUST?

HMRC conceded that ‘in so far as any VAT may have been paid by an investment trust to a manager in circumstances where the VAT was not properly charged and due as a matter of law, the VAT was paid by mistake’.60 The parties also agreed that payments are generally recoverable if they are made pursuant to a causative mistake, and that the question whether such a mistake was made is tested by asking whether the payment would have been made but for the payor’s mistake.61 The wording of HMRC’s concession was designed to leave room for an argument that although the claimants made a mistake, they were obliged to pay the money as a matter of contract in any event and so the payments were not caused by the mistake. However, this argument fell away in the course of proceedings when HMRC accepted that on their true construction the contracts did not oblige the claimants to make the VAT payments unless the tax was legally due. Henderson J therefore concluded that HMRC’s enrichment had been unjust, stating that: the same mistake about the lawfulness of the VAT charges caused both the payments by the investment trusts to the Managers and the corresponding enrichment of HMRC . . . [and] if the true legal position had been appreciated, no such payments would have been made by the investment trusts, and the enrichment of HMRC would not have occurred.62

Two points can be made about these observations. The first is that it is easy to elide two questions when discussing claims in unjust enrichment founded on causative mistake: one is whether the claimant’s mistake caused him to decide to confer a benefit on the defendant, and the other is whether his actions caused the defendant to become enriched. The first question is relevant because it determines whether the claimant made a mistake of a kind that is sufficiently serious to ground a claim; the second is relevant because it may determine whether any benefit received by the defendant was acquired at the claimant’s expense. Henderson J appears to have elided these two questions in the passage of his judgment that is quoted above. Given his definite view that HMRC were enriched at the claimants’ expense, this probably did not matter much in the context of the ITC case. Nevertheless – and contrary to Henderson J’s view that there may be a ‘considerable degree of overlap’ between the different elements of a claim in unjust enrichment63 – it is important that these elements should be kept separate, or confusion and uncertainty are bound to follow. An example is provided by the Court of Appeal’s decision in Test Claimants in the FII Group Litigation v HMRC.64 Here Arden LJ seems to have held that a ‘but for’ causation test should ITC (n 1) ibid [81]. 61 Consistently with Barclays Bank v WJ Simms (Southern) Ltd [1980] QB 677; Banque Financière de la Cité v Parc (Battersea) Ltd [1999] 1 AC 221 (HL) 227 and 231; Lloyds Bank plc v Independent Insurance Co [2000] QB 110 (CA) 115–16; Dextra Bank & Trust Co Ltd v Bank of Jamaica [2001] UKPC 50, [2002] 1 All ER (Comm) 193 [28] ff; Deutsche Morgan Grenfell Group plc v IRC [2006] UKHL 49, [2007] 1 AC 558 [59]–[60] and [143]–[144]. 62 ITC (n 1) [87]. 63 ibid [39]. 64 Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2010] EWCA Civ 103, [2010] STC 1251 [178]–[182], discussed in Goff & Jones (n 2) paras 6.07–6.08. 59 60

124 Charles Mitchell always be used to determine whether an enrichment has been gained at the claimant’s expense in two-party cases. However it is hard to be sure if that was her intention because she did not clearly distinguish this issue from the separate issue of whether a ‘but for’ causative mistake is a sufficiently serious type of mistake to ground liability. The second point is that it is currently unclear what ‘unjust factor’65 a claimant can and should invoke in a case involving a liability chain, ie a case where C confers a benefit on X, as a result of which X confers a benefit on D, and the circumstances of both transfers make the recipient’s enrichment unjust. If C can show that D is enriched at his expense, then the question arises whether C can and should invoke the same unjust factor against D as he would have invoked in an action against X? Or can and should C be treated as though he has taken over X’s action by subrogation and invoke the unjust factor that X would have invoked against D if he were bringing the action for his own benefit? In the ITC case, the answer to this did not matter because the unjust factor was the same for both parts of the chain: the claimant (C) and the manager (X) made the same mistake with respect to the legality of the VAT charge imposed on the manager’s services. But in cases where the unjust factor is different, this issue may have to be resolved. Suppose, for example, that C pays X money for a consideration that fails and that D then forces X to pay him the money under duress. Can C bring an action against D that is grounded on failure of consideration, or duress, or some combination of the two? And what are the consequences of this for the defences that D can raise in reply to the claim, given that eg change of position is not available to defendants who have acted in bad faith?

G. WERE THERE ANY DEFENCES?

The only issue considered under this final head was whether the claimants’ common law claims against HMRC were excluded by VATA, section 80(7). This states that Except as provided by [section 80], the Commissioners shall not be liable to credit or repay any amount accounted for or paid to them by way of VAT that was not VAT due to them.

The parties argued for different interpretations of these words. According to the claimants, the sub-section applies only to taxable persons who account for or pay money to HMRC as VAT that is not due, and who are given limited recovery rights by section 80. This is the most natural reading of the words ‘repay any amount . . . paid to [HMRC] by way of VAT’: under the scheme of VATA, taxable persons are the only people who pay money as VAT to HMRC, and so the sub-section can only apply to them. Hence it does not exclude common law claims against HMRC by final consumers. HMRC replied that this argument ignored the underlying economic reality of VAT, which (they observed) formed the starting point for the claimants’ successful arguments with regard to the question of whether HMRC had been enriched at the claimants’ expense. This economic reality is that final consumers pay money as VAT that finds its way into HMRC’s hands through the actions of taxable persons. To reflect this, the sub-section 65 An ‘unjust factor’ is a reason why a defendant’s enrichment at the claimant’s expense is unjust. For this terminology, see eg Banque Financière de la Cité v Parc (Battersea) Ltd [1999] 1 AC 221, 227; Kleinwort Benson Ltd v Lincoln City Council [1999] 2 AC 349, 363, 386, 395, and 409; Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2012] UKSC 19, [2012] 2 AC 337.

Claims by Indirect Taxpayers 125

should therefore be read as applying to final consumers as well as taxable persons, since there is a real sense in which final consumers are the payors of money as VAT that is received as such by HMRC. Henderson J found for HMRC on this issue. He was clearly attracted by the argument that he should interpret the sub-section in a way that was consistent with the economic reality of VAT, given the manner in which he had arrived at the conclusion that a common law claim in unjust enrichment would lie against HMRC. However, his main reason was that Parliament had carefully limited the statutory recovery rights of taxable persons under section 80, and it was inconceivable that Parliament could have intended a more generous regime to be available to the end customers by whom the economic burden of the tax was actually borne. It would make no sense to limit recovery by the tax collector, but to expose the Exchequer at the same time to far more extensive claims by the ‘real’ taxpayer.66

Henderson J considered that there was a bright-line distinction between a case such as the ITC case, where the ‘only issue’ was the extent of an express statutory exclusion, and a case like Monro v HMRC,67 where ‘any exclusion of a concurrent common law remedy has to be found by a process of necessary implication’.68 Clearly there is an important difference between the two, since argument in the former class of case, unlike argument in the latter, necessarily turns on the interpretation of words that are present in the statute. However, once the interpretative process in the former class of case turns to purposive considerations, the arguments in both classes of case start to look very similar. Thus, Henderson J’s reasoning in the passage quoted above bears a strong resemblance to the reasoning of Lord Brown in R (Child Poverty Action Group) v Secretary of State for Work and Pensions.69 There the question arose whether the Secretary of State for Work and Pensions could bring a common law claim in unjust enrichment to recover money mistakenly paid as social security benefits in a case that fell outside the scope of the Social Security Administration Act 1992, s 71 (which gives him a statutory right to recover such payments in limited circumstances). Lord Brown held that he could not because it was inconceivable that Parliament would have contemplated leaving the suggested common law restitutionary route to the recovery of overpayments available to the Secretary of State to be pursued by way of ordinary court proceedings alongside the carefully prescribed scheme of recovery set out in [section 71].70

The courts are often persuaded by arguments of this kind to deny common law claims in unjust enrichment, but such arguments do not always succeed. The main reason is that the statutory lacunae and ambiguities which give rise to such points frequently result from the inattention or ignorance of the legislature, rather than considered policy, and the courts can baulk at depriving claimants of their common law rights where there is little or no evidence that Parliament positively intended to take them away. Thus, Brooke LJ has warned that ‘[f]irst principles in statutory interpretation . . . rule out the dismantling of judge-made law

ITC (n 1) [104]. Monro v HM Revenue & Customs [2008] EWCA Civ 306, [2009] Ch 69. 68 ITC (n 1) [102]. 69 R (Child Poverty Action Group) v Secretary of State for Work and Pensions [2010] UKSC 54, [2011] 2 AC 15. 70 ibid [14]. 66 67

126 Charles Mitchell by stealth’,71 and Sir John Dyson SCJ stressed in the Child Poverty Action Group case that the court should not be too ready to find that a common law remedy has been displaced by a statutory one, not least because it is always open to Parliament to make the position clear by stating explicitly whether the statute is intended to be exhaustive.72

Unfortunately, it is hard to predict how much weight a court will give to this consideration when faced with an argument that Parliament must have meant to prevent claimants from relying on their common law rights when it introduced a statutory recovery scheme.73 One might have thought that a court would be more likely to favour such an argument where its effect is to leave a claimant with some limited statutory rights than where its effect is to leave the claimant with no rights at all, because the statutory scheme does not give him any. However, that assumption is not borne out by Henderson J’s findings in the ITC case.

H. FINAL THOUGHTS: THE EUROPEAN DIMENSION

Henderson J’s disposal of the English law issues in the ITC case did not resolve the parties’ dispute, as a set of EU law issues also had to be addressed. The judge’s analysis of these issues led him to conclude that there were good reasons for thinking that EU law, and in particular the principle of effectiveness, might require section 80(7) to be disapplied, to the extent that it prevented the claimants from bringing their common law claims. However, he also considered that these claims should probably be subject to the same time limits as would have affected statutory claims by the managers, although he deferred final resolution of this issue. In chapter ten of the present book, Maximilian Schlote argues that Henderson J’s findings, both in relation to the English law issues discussed in this chapter, and in relation to the disapplication of section 80(7), are consistent with a proper understanding of the EU law principles of effectiveness and legal certainty. In my view, Schlote is right to support the judge’s findings in relation to the disapplication of section 80(7) and the imposition of an analogous limitation period, but wrong to support his finding that the claimants could make out the elements of all their common law claims in unjust enrichment against HMRC. The San Giorgio principle requires only that Member States should make restitution of benefits which they have received, and although the value of the £75 paid by the claimants came into HMRC’s hands, the value of the £25 did not. The reasons for this have been explained. If this analysis is correct, then it follows that neither English law nor European law required HMRC to make restitution of the latter sum.

71 Green v Associated Newspapers Ltd [2005] QB 972 (CA) 991–92, citing Black-Clawson International Ltd v Papierwerke Waldhof-Aschaffenburg AG [1975] AC 591 (HL) 614; Re F (a minor) (publication of information) [1977] Fam 58 (CA) 99; R (Rottman) v Metropolitan Police Commissioner [2002] UKHL 20, [2002] 2 AC 692 [75]. 72 Child Poverty Action Group (n 69) [34]. 73 Consider the courts’ inconsistent responses to such arguments, made in connection with the same statutory recovery scheme, in Legal Services Commission v Henthorn [2011] EWHC 258 (QB) (not considered on appeal: [2011] EWCA Civ 1415, [2012] 1 WLR 1173) and Legal Services Commission v Loomba [2012] EWHC 29 (QB), [2012] 1 WLR 2461.

7 Property, Proportionality, and the Change of Position Defence NIAMH CLEARY*

A. INTRODUCTION

In claims for restitution of unlawfully exacted taxes, the courts must balance a number of conflicting principles. First, there is the taxpayer’s personal interest in recovering money paid as tax that was unlawfully exacted. Secondly, there is the public interest in ensuring that the state should not retain taxes to which it is not lawfully entitled.1 Thirdly, these imperatives in favour of restitution must be balanced against the public interest in preserving the integrity of the public finances.2 The ‘change of position’ defence is one area of the law governing restitution of unlawfully exacted taxes where this clash of imperatives is particularly evident. The critical importance of the defence is due in part to the structure of the law of restitution. The law’s generous approach to restitutionary causes of action is offset by broadly drawn defences to restitutionary claims.3 However, where restitution is sought of unlawfully exacted taxes, it is not clear that the liability of the state should be offset by generous defences. There are both practical and principled reasons why the state cannot, and should not, be treated like a private individual where it seeks to defend a claim for recovery of unlawfully exacted tax.4 Recent cases in which Her Majesty’s Revenue & Customs (HMRC) have relied on change of position as a defence to claims for restitution of overpaid tax highlight the difficulties that the courts face when applying change of position to claims against the state. In the light of these difficulties, it is unfortunate that neither the Supreme Court nor the Court of Appeal has had the opportunity to address this issue in any detail. Part B of this chapter considers three High Court decisions in which the change of position defence has been considered in the context of claims to recover money paid to public bodies as tax: Test Claimants in the FII Group Litigation v HMRC,5 Bloomsbury International Ltd v Sea Fish * I would like to thank James Lee, Charles Mitchell, and Eoin O’Dell for their comments in respect of earlier drafts of this chapter. This research was funded by the Irish Research Council for the Humanities and Social Sciences. 1 Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] AC 70 (HL) 172. 2 G Virgo, Principles of the Law of Restitution, 2nd edn (Oxford, OUP 2006) 401–2. 3 Lipkin Gorman (a firm) v Karpnale Ltd [1991] 2 AC 548 (HL) 581. 4 cf J Lee, ‘Defences to Claims for Restitution of Overpaid Tax’ [2012] Restitution Law Review 13. 5 Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2008] EWHC 2893 (Ch), [2009] STC 254 (hereafter ‘FII’).

128 Niamh Cleary Industry Authority,6 and Littlewoods Retail Ltd v HMRC.7 These cases demonstrate that the traditional change of position defence, which is founded on concepts such as expenditure, reliance, and causation, cannot be applied successfully to claims against the state. Part B concludes that any defence that protects the state’s security of receipt must therefore be derived from public policy, rather than from traditional restitutionary principles. Part C considers three policy-motivated defences that could be raised in response to claims for restitution of overpaid tax: the traditional ‘fiscal chaos’ defence, a modified form of the fiscal chaos defence, and the ‘exhaustion of benefits’ defence canvassed in Littlewoods. It argues that none of these defences can adequately balance the competing interests involved in a claim for restitution of tax payments. One striking deficiency in the current debate on restitution of overpaid tax is the failure to consider how the taxpayer’s right to property affects his right to restitution of unlawfully exacted tax. Part D considers how the right to restitution of unlawfully exacted tax can be reconceptualised in terms of property rights. It argues that the change of position defence, as currently applied to restitution claims against the state, does not accord due respect to the taxpayer’s right to property, and suggests that any policy-oriented defence to claims for restitution of unlawfully exacted tax must protect taxpayers against a disproportionate interference with their right to property.

B. CHANGE OF POSITION AND RESTITUTION OF UNLAWFULLY EXACTED TAX

(1) Change of Position and Causation A claim in restitution will be defeated where the defendant’s position ‘is so changed that he will suffer an injustice if called upon to repay or to repay in full’.8 The change of position defence therefore seeks to: strike a fair balance between the claimant’s interest in restitution and the defendant’s interest in making spending decisions freely, without fear that a claim in unjust enrichment might later invalidate the assumptions that he makes about the means at his disposal.9

Two rationales have been suggested for the change of position defence.10 First, change of position can be explained as a defence that denies an essential element of the claimant’s cause of action. A defendant who has dissipated an enrichment can no longer be said to be enriched, so that liability in unjust enrichment cannot be made out. This approach has been adopted by the High Court of Australia.11 6 Bloomsbury International Ltd v Sea Fish Industry Authority [2009] EWHC 1721 (QB), [2010] 1 CMLR 12 (hereafter ‘Bloomsbury’). 7 Littlewoods Retail Ltd v HM Revenue & Customs [2010] EWHC 1071 (Ch), [2010] STC 2072 (hereafter ‘Littlewoods’). 8 Lipkin Gorman (n 3) 579. See, generally, E Bant, The Change of Position Defence (Oxford, Hart Publishing, 2009). 9 C Mitchell, P Mitchell and S Watterson (eds), Goff and Jones: The Law of Unjust Enrichment, 8th edn (London, Sweet and Maxwell, 2011) para [27-03]. 10 cf Bant (n 8) 9 and 132–34. 11 David Securities Pty Ltd v Commonwealth Bank of Australia (1992) 175 CLR 353, 385–86. See, also, Lipkin Gorman (n 3) 563, where Lord Templeman held that the defence should succeed where the defendant’s change of position meant that he was no longer enriched.

Property, Proportionality, etc 129

In contrast, the English courts have adopted a justice-focused conception of the defence,12 which considers that change of position seeks to prevent injustice rather than to deny an element of the unjust enrichment principle: [W]here an innocent defendant’s position is so changed that he will suffer an injustice if called upon to repay or to repay in full, the injustice of requiring him so to repay outweighs the injustice of denying the claimant restitution.13

This justification is closely related to the competing interests that the change of position defence seeks to balance. The defendant’s receipt of money affects his spending and budgeting decisions. Change of position recognises that it would be an infringement of the defendant’s personal autonomy to require him to make restitution where he has incurred expenditure, which, but for the receipt of the enrichment, he would not otherwise have incurred. However, it will not be unjust to require the defendant to make restitution where his expenditure was not caused by his enrichment at the claimant’s expense. Therefore, a defendant must establish a causal link between his enrichment and his change of position. Causation in the context of change of position is expressed in two different ways. First, it is often said that the expenditure relied on by the defendant must not have been expenditure that he would have incurred in the ordinary course of things; to establish the change of position defence, the expenditure incurred by the defendant must be ‘extraordinary’.14 Secondly, the courts require a causal link between the receipt of the enrichment and the change of position.15 Although they are sometimes considered as separate elements of the defence,16 in reality both requirements serve the same purpose of establishing that receipt of the enrichment caused the defendant to change his position.

(2) Change of Position, Causation, and Claims against the State Two approaches to causation might be taken in respect of change of position, one broad and one narrow.17 The narrow approach requires that the defendant changed his position in reliance upon receipt of the enrichment.18 The wide approach simply requires that the change in position was causally linked to receipt of the enrichment. The wide approach has been adopted by the English courts.19 They have also adopted a generous standard of proof in establishing causation,20 holding that it is unnecessary to establish a precise link between

12 Lipkin Gorman (n 3) 579 (Lord Goff); Dextra Bank & Trust Co Ltd v Bank of Jamaica [2002] 1 All ER (Comm) 193 (PC) [38]; Charles Terence Estates Ltd v Cornwall Council [2011] EWHC 2542 (QB), [2012] 1 P & CR 2 [99]. 13 Lipkin Gorman (n 3) 579 (Lord Goff); Haugesund Kommune v Depfa ACS Bank [2010] EWCA Civ 579, [2012] Bus LR 1 [110] and [152]. See, also, Garland v Consumers’ Gas Co 2004 SCC 25, [2004] 1 SCR 629 [65]; J Lee, ‘Changing Position on Change of Position’ [2007] Restitution Law Review 135, 137–38. 14 Philip Collins Ltd v Davis [2000] 3 All ER 808 (Ch) 827. See too Dextra Bank (n 12) [38]; Barons Finance Ltd v Kensington Mortgage Co Ltd [2011] EWCA Civ 1592 [28]. 15 ibid 827. 16 ibid 827. 17 Scottish Equitable plc v Derby [2001] EWCA 369, [2001] All ER 818, 827. See, also, Cressman v Coys of Kensington (Sales) Ltd [2004] EWCA Civ 47, [2004] 1 WLR 2775 [41]; Bant (n 8) 143–51. 18 Storthoaks (Rural Municipality) v Mobil Oil Canada Ltd [1976] 2 SCR 147; David Securities (n 11). 19 cf Goff and Jones (n 9) paras [27-27]–[27-28]. 20 Scottish Equitable (n 17) 827; Commerzbank AG v Price-Jones [2003] EWCA Civ 1663 [43] and [58].

130 Niamh Cleary the defendant’s expenditure and his receipts.21 This approach has recently been applied in cases where public bodies have invoked the change of position defence, without considering whether such a standard of proof is appropriate: these cases simply assume that the defence is made out whenever the public body has a programme of expenditure in place. The judgment of Henderson J in FII demonstrates a particularly expansive approach to causation where the state relies on the change of position defence.22 In this case, HMRC sought to rely on change of position to defeat a claim for restitution of surplus Advance Corporation Tax (ACT) paid by the claimants, as well as a claim for restitution of the use value of ACT that had been paid prematurely. Henderson J noted that the courts will not impose an exacting standard of proof on defendants who rely on the defence.23 He held that government spending plans are predicated upon expected tax revenues and that the expenditure relied on by the defendant would therefore not have been incurred at the level at which it was incurred if receipt of the unlawfully exacted ACT had not been relied upon.24 Henderson J’s expansive approach in FII was followed by Hamblen J in Bloomsbury International Limited v Sea Fish Industry Authority. He held that: A causal connection must be shown, on at least a ‘but for’ basis between the receipt and any expenditure or change of position upon which the defendant wishes to rely. However, in the context of government expenditure, there is no need to demonstrate a precise link between particular receipts and particular items of expenditure and that it is reasonable to infer that planned expenditure would not have taken place at the level which it did but for the availability of the tax receipts which were taken into account in fixing departmental budgets.25

Hamblen J concluded that it was sufficient that the defendant could establish that it had ‘arranged its budgeting and planned its expenditure on the basis that it was entitled to receive and spend those monies’.26 However, HMRC have not always succeeded in making out a change of position defence on this basis. In Littlewoods Retail Ltd v HMRC,27 Vos J accepted expert evidence tendered on behalf of the claimant which established that the state’s budgeting and expenditure decisions were unlikely to have been affected by the expected receipt of tax revenues from the claimant.28 This suggests that the presumption of expenditure adopted in FII and Bloomsbury can be rebutted in some circumstances. The approach to causation adopted in FII, Bloomsbury, and Littlewoods is questionable in principle and undesirable as a matter of practice. There are three practical objections to this approach. First, the approach adopted in FII and Bloomsbury is derived from the generous approach to causation that has been adopted in claims between private parties.29 Recent English cases have held that defendants who wish to rely on change of position do eg Charles Terence Estates (n 12) [98]. FII (n 5). This aspect of Henderson J’s decision was not considered on appeal: [2010] EWCA Civ 103, [2010] STC 1251; [2012] UKSC 19, [2012] 2 AC 337. Note that Henderson J explicitly denied that fiscal chaos considerations had anything to do with his formulation of the defence at [347]. 23 ibid [343]. 24 ibid [344]. 25 Bloomsbury (n 6) [137]. 26 ibid [141]. 27 Littlewoods (n 7). 28 ibid [114]. 29 FII (n 5) [346]. See also, R Shiers and R Williams, ‘FII GLO (Chancery) and FJ Chalke: Tax and Restitution Developing Hand in Hand’ [2009] British Tax Review 365, 376. 21 22

Property, Proportionality, etc 131

not need to ‘demonstrate precisely’ where a specific disbursement went.30 The approach adopted in claims against HMRC has been compared31 to the generous approach to causation in cases such as Philip Collins Ltd v Davis,32 where the courts recognised that allowing one’s expenditure to expand in line with one’s income constitutes a valid change of position for the purposes of the defence. However, the same rationale cannot be applied to the state’s budgetary and spending decisions. The relationship between government receipts and government spending should not be overstated. It is wrong to assume that the state only spends money when it has tax revenues available to it. Thain and Wright have noted that spending ‘has a momentum independent of the declared aims of policy’.33 They continue: While the Treasury has authority to allocate among spending departments a total agreed to collectively by Cabinet (not imposed upon them), that allocation process is constrained in practice by the independent authority (and obligation) of each Secretary of State derived from statute to provide services, benefits, and goods.34

Many of the state’s financial obligations must be met, come what may. Thus, unlike cases where private defendants allowed their expenditure to ‘drift upwards’ to match their income,35 the state is often unable to avoid much of its anticipated expenditure and so cannot simply tailor its outgoings to match its income. Yet continuing to do ‘business as usual’ when making his spending decisions does not entitle a defendant to rely on the change of position defence.36 Thus, although the courts have taken a more generous approach to causation in the context of change of position, necessary and unavoidable expenditure incurred by the state cannot fall within the more generous approach to the defence. Secondly, there is disjuncture between governmental spending and revenue raising measures.37 Government spending plans under Labour were made on a three-yearly basis.38 The current coalition Government has adopted a five-year spending plan to coincide with its proposed term of government.39 In contrast, taxation measures are generally imposed on a short term or annual basis. Indeed, many taxes, such as income tax and corporation tax, can only be imposed on an annual basis.40 Thus, the Government enters into spending commitments for the future without necessarily considering the precise basis on which it Charles Terence Estates (n 12) [98]. Y Hu, ‘Change of Position and the Revenue’ [2011] Restitution Law Review 113, 130–31. 32 Philip Collins Ltd (n 14). See also Scottish Equitable (n 17). 33 C Thain and M Wright, The Treasury and Whitehall: The Planning and Control of Public Expenditure 1976– 1993 (Oxford, Clarendon Press, 1995) 2. 34 ibid 505. 35 Goff and Jones (n 9) para [27-06]. 36 Barons Finance (n 14) [28]. See, also, Barros Mattos Junior v MacDaniels Ltd [2004] EWHC 1188 (Ch), [2005] 1 WLR 247 [15]. 37 See, generally, J McEldowney, ‘Public Expenditure and the Control of Public Finance’ in J Jowell and D Oliver (eds), The Changing Constitution, 7th edn (Oxford, OUP, 2011) 351–53; Thain and Wright (n 33). 38 See, for example, HM Treasury, Modern Public Services for Britain: Investing in Reform. Comprehensive Spending Review: New Public Spending Plans 1999–2002 (Cm 401, 1998); HM Treasury, Meeting the Aspirations of the British People: 2007 Pre-Budget Report and Comprehensive Spending Review (Cm 7227, 2007). In addition to these Comprehensive Spending Reviews, spending reviews were carried out in 2000, 2002 and 2004. See, generally, T Prosser, ‘ “An Opportunity to Take a More Fundamental Look at the Role of Government in Society”: The Spending Review as Regulation’ [2011] Public Law 596, 598–600; C Turpin and A Tompkins, British Government and the Constitution: Text and Materials, 7th edn (Cambridge, CUP, 2011) 646–47. 39 HM Treasury, Spending Review 2010 (Cm 7942, 2010). 40 Such taxes include income tax and corporation tax. See O Hood Philips, P Jackson and P Leopold, O Hood Phillips and Jackson: Constitutional and Administrative Law, 8th edn (London, Sweet and Maxwell, 2001) 256–57; J Jaconelli, ‘The “Bowles Act” – Cornerstone of the Fiscal Constitution’ (2010) 69 CLJ 582, 605–7. 30 31

132 Niamh Cleary will generate the income to meet those commitments. Even if a generous approach is taken to anticipatory spending decisions,41 the relationship between long-term spending commitments and short-term revenue raising measures is not sufficiently close to establish a causal link between taxation and Government expenditure.42 Thirdly, even if Henderson J was correct to conclude in FII that Government spending is currently contingent upon Government receipts, this is not true for the entire period during which ACT was imposed. ACT was first introduced in 1973.43 It cannot be said that, from 1973 to the abolition of ACT in 1999, revenue projections influenced governmental expenditure plans in every year in which the tax was imposed. For example, Thain and Wright note that: In some years, 1985 and 1986 for example, there were no revenue projections at all. It was only in 1992, on the initiative of Chancellor Lamont, that the Treasury accepted the need for a unified budget, bringing together revenue and expenditure decisions.44

Thus, even if governmental budgetary decisions can constitute a relevant change of position, this should only be the case for the years following the introduction of the unified budget in 1992, and the change of position defence ought not have been available in respect of preceding years. In addition to these practical difficulties, the approach adopted in recent cases is also objectionable as a matter of principle. In FII, it was accepted between the parties that the change of position defence was not available in response to claims under the Woolwich principle.45 Henderson J reached this conclusion by characterising the Woolwich claim as one involving the commission of a legal wrong. As a result, the defendant fell within the ‘established wrongdoer’ exception to change of position46 and could not rely on the defence.47 Similarly, in Littlewoods, Vos J held that the defendant’s wrongdoing stemmed from the unlawful nature of the tax.48 This conclusion is questionable for several reasons. First, it not clear that the state should be regarded as a wrongdoer for the purposes of the change of position defence. In Elise Bant’s words, ‘[o]ften, the fact that a tax is not supported by valid legislation will simply mean that the tax is invalid, not that it inherently constitutes a breach of some independent duty’.49 In contrast to FII, in Murphy v Attorney General,50 the Irish Supreme Court expressly accepted that the state had received overpaid taxes in good faith and held that the constitutional invalidity of the tax did not preclude the state from relying on change of position. Henderson J’s focus on the ‘wrongful’ nature of Dextra Bank (n 12) [38]; Charles Terence Estates (n 12) [99]. See Law Commission, Restitution: Mistakes of Law and Ultra Vires Public Authority Receipts and Payments (Law Com No 227, 1994) para [11.12]; E Bant, ‘Change of Position as a Defence to Restitution of Unlawfully Exacted Tax’ [2012] Lloyd’s Maritime and Commercial Law Quarterly 122, 132. 43 See Finance Act 1972, pt V. 44 Thain and Wright (n 33) 235. 45 FII (n 5) [304]. See also Bloomsbury (n 6) [134]. On this concession, see R Williams, Unjust Enrichment and Public Law (Oxford, Hart Publishing, 2010) 137–38. 46 In Lipkin Gorman (n 3) 580, Lord Goff held that it is ‘the defence is not open to one who has changed his position in bad faith . . . and it is commonly accepted that the defence should not be open to a wrongdoer’. 47 FII (n 5) [339]. 48 Littlewoods (n 7) [105]–[107]. 49 E Bant, ‘Restitution from the Revenue and Change of Position’ [2009] Lloyd’s Maritime and Commercial Law Quarterly 166, 172. See also Bant (n 42) 137; A Goymour and G Virgo, ‘Unlawful Tax: Tantalizing Glimpses of Restitution’ (2010) 69 CLJ 232, 235; Shiers and Williams (n 29) 375; Williams (n 45) 148–49; Hu (n 31) 123. Compare Lee (n 4). 50 Murphy v Attorney General [1982] IR 241, 319. 41 42

Property, Proportionality, etc 133

the tax is unfortunate, as it obscures the public law principles engaged by claims for restitution of overpaid tax. As a result, restrictions on the change of position defence in such claims remain focused on the nature of the cause of action relied on by the claimant, rather than on the nature of the defendant. However, Henderson J also acknowledged the importance of public law principles in tax cases. Thus, in addition to characterising the Woolwich principle as based on wrongdoing, he also acknowledged that the public law principles underpinning the Woolwich claim should bar the change of position defence where restitution was sought under the Woolwich principle, on the basis that ‘the right of a private citizen to recover tax which has been unlawfully levied by the executive arm of government should, as a matter of principle, be unfettered’.51 This justification is entirely consistent with the principles that underlie restitution from public authorities. There is a strong public interest in restitution of unlawfully exacted taxes. If this principle is sufficiently important to require restitution as of right under the Woolwich principle, the change of position defence should not subvert it.52 In Auckland Harbour Board v R, the Privy Council adopted a restrictive approach to defences where restitution was sought by the state for money paid by the state ultra vires.53 This approach should be mirrored in cases where a private citizen seeks restitution from the state. It was by a similar analogy with Auckland Harbour Board that Lord Goff held that ultra vires payments could be recovered by private citizens as of right.54 This reasoning was adopted in the context of the passing on defence in Commissioner of State Revenue (Victoria) v Royal Insurance Australia Ltd.55 Mason CJ held that the passing on defence should not apply where it would subvert the constitutional principles that support the Woolwich principle: It would be subversive of an important constitutional value if this Court were to endorse a principle of law which, in the absence of such circumstances, authorized the retention by the executive of payments which it lacked authority to receive and which were paid as a result of causative mistake.56

Like the passing on defence, change of position allows the state to retain money exacted without Parliamentary authority. The principles underpinning the Woolwich principle should likewise require a restrictive approach to change of position. The second principled difficulty with the approach in FII is that Henderson J held that change of position would not be available where the Woolwich principle was relied on by litigants, but that it would be available in respect of mistake claims. He offered two justifications for this conclusion, both of which are questionable. First, Henderson J held that the defendant will only be precluded from relying on the change of position defence where the cause of action relied on by the claimant involves an allegation of wrongdoing: When Lord Goff said it is commonly accepted that the defence ‘should not be open to a wrongdoer’, it seems to me fairly clear from the context that what he had in mind was cases where FII (n 5) [340]. See also Bant (n 8) 169, 187, and 203. 53 Auckland Harbour Board v R [1924] AC 318 (PC) 326; Attorney General v Gray [1977] NSWLR 406, 410–11; Commonwealth of Australia v Burns [1971] VR 825, 829. This point appears to have been overlooked in Charles Terence Estates (n 12) [97]–[98]. 54 Woolwich (n 1) 177. 55 Commissioner of State Revenue (Victoria) v Royal Insurance Australia Ltd (1994) 182 CLR 51. 56 ibid 69. 51 52

134 Niamh Cleary the cause of action relied upon by the claimant was itself founded upon wrongdoing by the defendant.57

Commentators have argued that this interpretation of Lord Goff’s speech is inconsistent with previous authority, where the defence has not been so narrowly construed.58 For example, in Barros Mattos Junior v MacDaniels Ltd, Laddie J held that the defendants could not rely on the change of position defence,59 notwithstanding express acknowledgement by counsel for the claimants that the action for money had and received relied upon was not based on wrongful conduct.60 However, in Barros Mattos, the defendant’s change of position was itself unlawful, as it involved a breach of Nigerian exchange controls.61 Thus, FII and Barros Mattos could be reconciled by holding that the change of position defence is not available where the wrongdoing is related to the cause of action relied upon or where the change of position itself was unlawful. However, the English courts have accepted a justicebased conception of change of position. This would suggest that, once the defendant’s wrongdoing is related to his transaction with the claimant, it should preclude reliance on the defence, even where the wrong is not directly related to the claimant’s cause of action or to the defendant’s change of position. If this more expansive interpretation of the defence is correct, the defendant in FII should have been precluded from relying upon change of position in respect of the claimants’ mistake claim, as well as their claims under the Woolwich principle. Secondly, Henderson J held that the public law principles that justified disallowing the defence to Woolwich claims were less potent where the claimant based his claim on mistake: The position is very different, however, where a taxpayer pleads his claim in mistake, and particularly where his motive in doing so is to take advantage of a more generous limitation period. As a matter of principle, it seems to me entirely appropriate that the defence of change of position should then be available, and that the Revenue should not be obliged to make restitution of mistakenly paid tax if it can establish the necessary factual ingredients of the defence.62

However, the constitutional principles that require restitution of unlawfully exacted tax are engaged in all cases where recovery is sought, irrespective of the cause of action by which this is sought to be achieved. If the public law principles behind the Woolwich principle justify a total restriction on the change of position defence where this principle is relied on, they also justify a restrictive approach to change of position where restitution is sought from the state on the basis of other unjust factors.63

(3) Conclusions on Change of Position and the State Recent decisions such as FII, Bloomsbury, and Littlewoods demonstrate that the traditional change of position defence cannot apply to claims against the state. The approach adopted FII (n 5) [320]. Hu (n 31) 123; Goff and Jones (n 9) paras [27-43]–[27-44]. 59 Barros Mattos (n 36) [44]. 60 ibid [10]. 61 However, see Goff and Jones (n 9) paras [27-40]–[27-42]. 62 FII (n 5) [341]. 63 See, also, Shiers and Williams (n 29) 374; Goff and Jones (n 9) para [27-51]. 57 58

Property, Proportionality, etc 135

in these cases distorts the law in two ways. First, the courts presume that the state has incurred expenditure. Even in cases where the change of position defence was rejected, such as Littlewoods, this appears to have been because the claimant positively established that the receipt of the enrichment had no effect on the Government’s budgetary decisions.64 This is inconsistent with the traditional conception of the defence, which requires defendants seeking to rely on change of position to establish the elements of the defence.65 Thus, the approach adopted in recent decisions turns the burden of proving causation on its head in claims against the state, and creates a presumption in favour of change of position by assuming that all government expenditure is caused by the receipt of the enrichment.66 Secondly, change of position traditionally operates pro tanto, that is, only to the extent that the defendant’s position has changed as a result of receiving the enrichment.67 However, if the state is deemed to have changed its position by relying on the receipt when planning its programme of expenditure, the defence cannot operate pro tanto, as the state will have relied on those receipts in their entirety. Thus, recent cases in which public bodies have relied on the change of position defence indicate that the courts will assume that the money was spent in full, so that change of position will always be a complete defence where it is relied upon by the state. These advantages to the state are not the result of a conscious decision to favour state entities over private defendants. Instead, the standards of proof and causation that apply to individual defendants have been transposed without thought into cases in which the state is the defendant.68 Yet FII, Littlewoods, and Bloomsbury demonstrate that traditional concepts of causation and expenditure cannot be tailored to situations to where the state seeks to rely on change of position. Where a generous standard of proof is adopted, as in FII and Bloomsbury, claims for restitution will automatically be defeated by change of position if claimants cannot bring their claims within the Woolwich principle.69 On the other hand, a more stringent standard of proof, such as that adopted in Littlewoods, is also problematic, as the state will rarely be able to establish that its programme of expenditure was premised on the receipt of particular revenues.70 Therefore, applying the change of position defence to state bodies is likely to be unsatisfactory, no matter which approach is taken. This analysis suggests that the change of position defence does not reflect the unique position of the state and should be abandoned in claims for restitution of unlawfully exacted tax. Instead of relying on concepts such as expenditure and causation, the courts should adopt a policy-motivated defence that looks to the public nature of the defendant, rather than the nature of the cause of action relied on by the claimant. Part C considers policy-motivated defences that have been proposed in lieu of the change of position defence.

Littlewoods (n 7) [120]. Philip Collins Ltd (n 14) 827. 66 Bant (n 49) 170–71; Bant (n 42) 132–34. 67 Dextra Bank (n 12) [38]; Lipkin Gorman (n 3) 580–81. 68 See also, Williams (n 45) 142–46. 69 See G Virgo, ‘The Law of Taxation Is Not an Island – Overpaid Taxes and the Law of Restitution’ [1993] British Tax Review 442, 458–60. 70 However, in AJ Seversen Inc v Qualicum Beach [1980] 3 WWR 375 [32]–[35], a decision of the British Columbia Supreme Court, such a defence was successful, as the money exacted by the defendant council was not intended for general expenditure but instead was paid into a special fund for the purpose of installing a suitable water supply to the plaintiff ’s lots and others in the area. 64 65

136 Niamh Cleary

C. SUGGESTED ALTERNATIVES TO THE CHANGE OF POSITION DEFENCE

The courts have traditionally been reluctant to consider policy-motivated defences in claims for restitution of unlawfully exacted taxes. However, the difficulties experienced by the courts in applying traditional defences to claims against the state may render them more amenable to adopting such defences. Moreover, the development of policymotivated defences may be considered less objectionable when viewed in light of the development of policy-motivated causes of action, such as the Woolwich principle.71 This Part considers three policy-oriented defences: the traditional ‘fiscal chaos’ defence, a modified form of the fiscal chaos defence proposed by Ying Hu,72 and the ‘exhaustion of benefits’ defence canvassed in Littlewoods.

(1) Fiscal Chaos The ‘fiscal chaos’ defence was first advocated by La Forest J in Air Canada v British Columbia.73 La Forest J considered that several considerations favoured a general rule against the recovery of unconstitutional taxes. These considerations included the protection of the treasury, and a recognition of the reality that if the tax were refunded, modern government would be driven to the inefficient course of reimposing it either on the same, or on a new generation of taxpayers, to finance the operations of government.74

The rule proposed by La Forest J appeared to operate as a blanket rule against the recovery of unconstitutional taxes. Although it was unclear what circumstances would engage the defence, La Forest J cautioned that the defence was ‘an exceptional rule, and should not be construed more widely than is necessary to fulfil the values which support it’.75 This defence has not met with approval in any other jurisdiction. In Australia, in Royal Insurance, Mason CJ was reluctant to recognise ‘a vague and amorphous defence based on the notion of avoiding disruption of public finances’.76 Although, in Woolwich, Lord Goff left open the possibility that specific limits could be imposed on restitution of unlawfully exacted taxes,77 the English courts have been unwilling to adopt a fiscal chaos defence.78 The defence has also been overturned in Canada.79 There appears to be judicial consensus in most jurisdictions that the formulation of such a defence is a task best left to the legislature.80 Lee (n 4). Hu (n 31). 73 Air Canada v British Columbia (Attorney General) [1989] 1 SCR 1161. 74 ibid [94]. 75 ibid. 76 Royal Insurance (n 55) 68. See, also, A Burrows, ‘Public Authorities, Ultra Vires and Restitution’ in A Burrows (ed), Essays on the Law of Restitution (Oxford, Clarendon Press, 1991) 58. See also Virgo (n 69) 464. 77 Woolwich (n 1) 176. See also Deutsche Morgan Grenfell Group plc v Inland Revenue Commissioners [2005] EWCA Civ 78, [2006] Ch 243 [271] (Buxton LJ). 78 Mallusk Cold Storage Ltd v Dept of Finance and Personnel [2003] NIQB 58, [2003] RA 370 [25]. 79 Kingstreet Investments Ltd v New Brunswick (Department of Finance) 2007 SCC 1, [2007] 1 SCR 3 [21]. See, also, Sorbara v Canada (Attorney General) (2008) 304 DLR (4th) 470 (Ont SCJ) [32]. 80 Woolwich (n 1) 200; Mallusk Cold Storage (n 78) [25]; Kingstreet (n 79) [25]; O’Rourke v Revenue Commissioners [1996] 2 IR 1 (HC) 18. 71 72

Property, Proportionality, etc 137

Judicial reluctance to embrace ‘fiscal chaos’ as a defence is understandable. The defence advocated by La Forest J prioritises pragmatic considerations of efficacy over constitutional principle, placing as it does the state’s interest in retaining the money over the constitutional right of the claimant to its return.81 As Birks has noted, ‘inconvenience and disruption are usually ignored in establishing liabilities’.82 A broadly defined ‘fiscal chaos’ defence, such as that adopted in Air Canada, undermines the important constitutional imperative in favour of restitution of unlawfully exacted taxes.83 In practical terms, it confers a qualified immunity on the state where restitution of taxes is sought, subject to a requirement of fiscal disruption.84 As Williams has noted, if the fiscal chaos defence is recognised, then it will operate as a complete bar to restitution claims: ‘Public bodies specifically raise money in order to spend it in the public interest and could therefore always argue that return of the money would upset their finances’.85 The problems underpinning the fiscal chaos defence suggest that if such a defence is to be adopted, it should, at the very least, be confined to cases of severe fiscal disruption. Indeed, supporters of the fiscal chaos defence have recommended that it is limited to exceptional circumstances, where there is a real risk of serious disruption to the public finances.86 However, it is almost impossible to determine what will constitute ‘fiscal chaos’.87 In light of the principled objections to such a defence, as well as the uncertainties that attend its scope, application and operation, it should not be adopted in claims for restitution of unlawfully exacted tax.

(2) A Modified Fiscal Chaos Defence A second policy motivated defence has been proposed by Ying Hu.88 Although Hu presents the defence as a form of change of position, it is more accurately viewed as a sophisticated variety of the fiscal chaos defence.89 Hu proposes two presumptions, which will operate where the state establishes ‘that a policy response is necessary if full repayment of the ultra vires tax is required’.90 First, there will be a strong presumption in favour of allowing the state to refuse repayment of taxes levied before the commencement of the first claim to recover the money. Secondly, there will be a strong presumption in favour of restitution of taxes levied after the commencement of the first claim to recover the money, See Virgo (n 2) 422. P Birks, ‘Restitution from the Executive: A Tercentenary Footnote to the Bill of Rights’ in PD Finn (ed), Essays on Restitution (Sydney, Law Book Co, 1990) 202. 83 FII (n 5) [340]; Royal Insurance (n 55) 69 (Mason CJ); Air Canada (n 73) [15] (Wilson J). See also G Virgo, ‘Striking a Balance in the Law of Restitution’ [1995] Lloyd’s Maritime and Commercial Law Quarterly 362, 367; JD Merralls, ‘Recovery of Taxes After the Royal Insurance Case’ in M McInnes (ed), Restitution: Developments in Unjust Enrichment (Sydney, LBC Information Services, 1996) 122. 84 Marcotte v Longueuil (Ville) 2009 SCC 43, [2009] 3 SCR 65 [102] (Deschamps J). 85 Williams (n 45) 159. 86 J McCamus, ‘Restitutionary Recovery of Moneys Paid to a Public Authority Under a Mistake of Law: Ignorantia Juris in the Supreme Court of Canada’ (1983) 17 University of British Columbia Law Review 233, 257; P Mitchell, ‘Restitution, “Passing On,” and the Recovery of Unlawfully Demanded Taxes: Why Air Canada Doesn’t Fly’ (1995) 53 University of Toronto Faculty Law Review 130, 141 and 143. Although Virgo is not a devotee of the fiscal chaos defence, he argues that the defence, if adopted, should be tightly limited: Virgo (n 69) 464. 87 Royal Insurance (n 55) 68. 88 Hu (n 31) 133. 89 Lee (n 4). 90 Hu (n 31) 133. 81 82

138 Niamh Cleary which would only be rebutted in extreme circumstances,91 for example, ‘where repayment would result in fiscal chaos’.92 This defence is considerably more sophisticated in its operation than the traditional ‘fiscal chaos’ defence. However, it is too generous to the state and fails to strike an appropriate balance between the state’s interest in the security of its receipt, the taxpayer’s interest in restitution, and the more general public interest in ensuring that the state is not enriched by unlawful gains. Hu stipulates that the defence will be available where the Crown establishes that a policy response is necessary to ensure repayment of overpaid taxes. These policy responses include the ‘imposition of new taxes, decrease in planned expenditure, and increase in public borrowing’.93 This is a considerably lower bar than the, admittedly uncertain, requirement of ‘fiscal chaos’. The state will frequently, if not always, need to impose such ‘policy responses’ when meeting its financial liabilities, irrespective of their nature and origin. It is unclear why the need for policy responses should entitle the state to defeat what would otherwise be a valid liability. Moreover, the frequency with which the state will need to invoke such responses means that payers would rarely be entitled to recover taxes exacted from them unlawfully. A further difficulty with the defence is that the presumption in favour of restitution will only operate until the first claim for recovery of the overpaid tax has been commenced.94 This approach is reminiscent of the Irish Supreme Court’s decision in Murphy, where Henchy J held that the state was entitled to retain unconstitutionally exacted taxes received before it had notice of the Murphys’ claim.95 However, the state will rarely receive taxes where it already has notice that these taxes are invalid,96 and it may receive taxes over a long period without being aware of their invalidity. For example, the incompatibility of the UK ACT regime with EU law was not conclusively established until 2001,97 two years after the tax had been abolished. Even if a presumption that the state is entitled to retain taxes where it has no notice of their invalidity is defensible in spite of these concerns, Hu’s proposed presumption will apply until the first claim for restitution of the tax has been commenced. This is considerably more generous than a ‘notice requirement’. Furthermore, it is not clear what will be necessary to rebut this presumption. This policy-motivated defence is therefore more generous to the state than any other defence proposed to date. Hu, in support of this defence, relies upon the policy concerns that are traditionally advanced in favour of a fiscal chaos defence. However, many of the arguments put forward in support of policy-motivated defences focus exclusively upon a concern to protect the public purse and fail to have regard to countervailing arguments in favour of restitution. Proponents of ‘fiscal chaos’ defences claim that it is unfair to require one generation of taxpayers to fund restitution of unlawfully exacted taxes, where the benefit of the tax was

ibid 133. ibid 133. 93 ibid 119. 94 ibid 133. 95 Murphy (n 50). 96 cf Re Article 26 and the Health (Amendment) (No 2) Bill 2004 [2005] IESC 7, [2005] 1 IR 105, where the Irish Government continued to levy charges for health care on elderly citizens, even though it was aware that those charges were unlawful. 97 Joined Cases C-397/98 and C-410/98 Metallgesellschaft Ltd v Inland Revenue Commissioners, Hoechst AG v Inland Revenue Commissioners [2001] ECR I-1727. See, also, Deutsche Morgan Grenfell Group plc v Inland Revenue Commissioners [2006] UKHL 49, [2007] 1 AC 558 [31] (Lord Hoffmann). 91 92

Property, Proportionality, etc 139

received by another generation.98 However, it is unclear why this should be considered unfair. The obligation of the state to make restitution of unlawfully exacted taxes is a liability like any other. The fact that taxpayers who did not benefit from the tax will ultimately meet the cost incurred by the state in meeting this liability is irrelevant. After all, it would not be suggested that the state should be exempt from meeting its tortious liabilities simply because the taxpayers who would fund that liability did not inflict the tort upon the victim. Similarly, suggestions that requiring the state to make restitution is a ‘waste of its administrative and financial resources’99 do not justify requiring taxpayers to bear a substantial loss, by forgoing restitution of taxes that they were never legally obliged to pay.

(3) Exhaustion of Benefits The ‘exhaustion of benefits’ defence was canvassed by HMRC in Littlewoods. The defence: asserts that the government’s annual budget procedures mean that tax receipts are used up in each fiscal year, and no benefit is derived by the government from those receipts thereafter. For that reason, the claimants’ measure of recovery should be limited to the benefit of the use of the principal sums for a single fiscal year.100

Exhaustion of benefits is closely related to the change of position defence. In Littlewoods, Vos J concluded that: ‘insofar as an exhaustion of benefits defence is a species of change of position, it is, as a matter of English law, available to a mistake-based restitutionary claim, but not to a Woolwich based claim’.101 However, he doubted whether the defence is really a defence as such at all. It seems to me to be more an argument as to how the benefit that has accrued to the Commissioners should be calculated for the purposes of assessing the enrichment that is to be the subject of an order for restitution.102

However, exhaustion of benefits cannot be wholly divorced from change of position for two reasons.103 First, change of position is not simply concerned with ‘disenrichment’; it is also concerned with identifying the enrichment surviving in the defendant’s hands, and recognises that this need not be the same as the value received by the defendant. Therefore, there is a strong link between the ‘value surviving’ concept of enrichment that is recognised in English law104 and change of position. Secondly, characterising exhaustion of benefits as being exclusively a matter for enrichment will have significant repercussions for the value that we ascribe to the use value of money. In particular, it is questionable whether dissipation of money negates the possibility of deriving further use from it. Presumably, if the state had not received the claimant’s money in Littlewoods, it would have been obliged to borrow the money and to pay interest on that sum for the duration of the loan. Therefore, 98 Air Canada (n 73) [88]; Hu (n 31) 120; B Fitzgerald, ‘Ultra Vires as an Unjust Factor in the Law of Unjust Enrichment’ (1993) 2 Griffith Law Review 1. 99 Hu (n 31). 100 Littlewoods (n 7) [97]. 101 ibid [111]. 102 ibid [110]. 103 See also Bant (n 42) 142–43; Lee (n 4). 104 Lipkin Gorman (n 3) 560 (Lord Templeman); P Birks, Unjust Enrichment (Oxford, OUP, 2005) 63; R Chambers, ‘Two Kinds of Enrichment’ in R Chambers, C Mitchell and J Penner (eds), Philosophical Foundations of the Law of Unjust Enrichment (Oxford, OUP, 2009) 247–49.

140 Niamh Cleary the use value derived from tax receipts, even if not carried forward by the state in assessing its budgetary objectives for the following year, does not necessarily cease to constitute an enrichment as understood by the law of restitution. In light of the problems of proof that arise where change of position is relied upon by the state, exhaustion of benefits may supplant change of position and come to be regarded as a unique species of change of position that applies to the state. However, exhaustion of benefits only provides a defence to claims for restitution of the use value of money. Therefore, it is a more limited defence than change of position. This is not, however, to deny the utility of such a defence. Many, if not most, high-profile claims for restitution of overpaid tax have concerned recovery of interest due on the tax, rather than recovery of the principal sum itself.105 The sums claimed have been substantial.106 However, this realisation does not alter the fact that the exhaustion of benefits defence is considerably more limited in scope than a more general change of position defence.

(4) Policy-Motivated Defences: Conclusions Two major difficulties attend the policy-motivated defences developed to date. First, it is difficult to develop a defence that is not wholly discretionary and unprincipled in its operation. Secondly, many of the policy-oriented defences that have been proposed are predicated solely on policy concerns that favour the state’s retention of overpaid tax. Policy concerns that favour restitution – such as the taxpayer’s interest in restitution, or the public interest in restitution – are marginalised, if not completely ignored. However, there is another layer of analysis applicable to restitution from public authorities which has not yet been fully explored. It is possible to reconceptualise restitution of unlawfully exacted taxes in terms of property rights. Part D considers the relationship between restitution and the right to property under Article 1 of the First Protocol to the European Convention on Human Rights (ECHR). Policy-motivated defences should require the state to prove that denying restitution to the taxpayer is a justified and proportionate interference with the taxpayer’s right to property.

D. RESTITUTION AND PROPERTY RIGHTS: AN ALTERNATIVE APPROACH TO POLICY-MOTIVATED DEFENCES

(1) Restitution and the Right to Property Where a public authority levies taxes or charges in the absence of lawful authority, this constitutes an unlawful interference with the property rights of the individuals affected by the charge.107 Therefore, the unlawful levying of taxes and other charges engages public law 105 See, for example, Woolwich (n 1); Sempra Metals Ltd v Inland Revenue Commissioners [2007] UKHL 34, [2008] 1 AC 561; DMG (HL) (n 97). 106 For example, the sums sought to be recovered in FII and Littlewoods are, respectively, £5 billion and £1 billion. 107 D Hoffman, ‘The Law of Restitution and the Human Rights Act 1998’ [2004] Restitution Law Review 64, 68–69.

Property, Proportionality, etc 141

protections of property rights, as well as private law principles of the law of restitution. Although some cases involving restitution from public authorities have adverted to the relevance of property rights,108 little consideration has been given to how these principles can, or should, shape the substantive law of restitution. In England, recent property rights jurisprudence has focused on Article 1 of the First Protocol to the ECHR, which guarantees the individual’s peaceful enjoyment of his possessions, subject to the state’s interest in controlling the right to property in the public interest.109 The Human Rights Act 1998 (HRA) requires courts to interpret primary and secondary legislation in a manner that is compatible with Convention rights, where possible to do so.110 In determining issues that affect Convention rights, courts must also have regard to the jurisprudence of the European Court of Human Rights (ECtHR).111 The HRA does not explicitly refer to common law doctrines. However, the courts have construed their obligation to interpret legislation in a manner that is compatible with Convention rights112 as extending to interpretation of the common law, and common law doctrines are assessed for compatibility with the ECHR.113 In assessing interference with property rights, the ECHR requires courts to consider whether the measure was lawful, whether it was justified by the public interest, and whether the means adopted were proportionate to the aim pursued by the Contracting State.114 In giving effect to Convention jurisprudence, the English courts recognise that property rights jurisprudence involves interaction between three Convention principles: the need to strike a fair balance between public interests and private rights, the principle of proportionality, and the doctrine of the margin of appreciation.115 The proportionality principle is the means by which courts determine whether a measure strikes a fair balance between individual rights and the general interests of the community. In applying the proportionality test, the courts will consider whether: (i) the legislative objective [of the impugned measure] is sufficiently important to justify limiting a fundamental right; (ii) the measures designed to meet the legislative objective are rationally connected to it; and (iii) the means used to impair the right or freedom are no more than is necessary to accomplish the objective.116

The English courts recognise that the right to restitution of unlawfully exacted taxes is reinforced by Article 1 of the First Protocol.117 However, the protection afforded by Article 1 is weak and qualified.118 Several factors contribute to this deficiency. First, this weak 108 R (Child Poverty Action Group) v Secretary of State for Work and Pensions [2009] EWCA Civ 1058, [2010] 1 WLR 1886 [34]; Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2010] EWCA Civ 103, [2010] STC 1251 [228]. See, also, Monro v HM Revenue & Customs [2008] EWCA Civ 306, [2009] Ch 69 [30]–[33]. Compare Longmore LJ’s comments in the same case at [41]. 109 James v United Kingdom (1986) 8 EHRR 123 [37]; Pressos Compania Naviera v Belgium (1996) 21 EHRR 301 [33]. See also MW Janis, RS Kay and AW Bradley, European Human Rights Law, 3rd edn (Oxford, OUP, 2008) 519. 110 HRA, s 3(1). 111 HRA, s 2(1). 112 HRA, s 2(1). 113 See, for example, Andreas Soteriou v Ultrachem Ltd [2004] EWHC 983 (QB); Child Poverty Action Group (n 108). 114 James (n 109) [50]; National & Provincial Building Society v United Kingdom (1997) 25 EHRR 127 [80] and [105]. 115 R (SRM Global Master Fund LP) v Treasury Commissioner [2009] EWCA Civ 788, [2009] UKHRR 1219 [43]. 116 de Freitas v Permanent Secretary of Ministry of Agriculture, Fisheries, Lands and Housing [1999] 1 AC 69 (PC) 80; R (Daly) v Secretary of State for the Home Department [2001] 2 AC 532, 547. 117 Child Poverty Action Group (n 108) [34]. See also Hoffman (n 107). 118 See Beatson J’s comments in R (Burnley Borough Council) v First Secretary of State [2006] EWHC 798 (Admin) [48].

142 Niamh Cleary protection is directed by the wording of Article 1 of the First Protocol itself. In R (Trailer and Marina (Leven) Ltd) v Secretary of State for the Environment, Food and Rural Affairs, Neuberger LJ (as he then was) observed that Article 1, unlike other Convention provisions, does not require measures adopted by Contracting States to be objectively necessary in a democratic society. Instead, it is sufficient that the Contracting State deemed the measure to be necessary.119 Secondly, the right to property is undermined by the weak formulation of the proportionality test adopted by the ECtHR, which requires that ‘the means used to impair the right or freedom are no more than is necessary to accomplish the objective’.120 This meansends test – which requires that the means adopted are proportionate to the objective sought to be achieved – offers considerably less protection than the means-effects proportionality test adopted in other jurisdictions,121 which requires a court to consider the impact that a particular measure has on individual rights. Moreover, although ECHR jurisprudence requires that the means adopted by the legislature are proportionate to the aim of the measure, this does not mean that the legislature is obliged to adopt the least restrictive means available.122 Consistently with Convention jurisprudence, the English courts have held that the proportionality test is a means-ends test.123 However, they have also held that the proportionality doctrine requires, not only that a justification is offered, but that the measure adopted bears some degree of proportionality to the aim, even if it is not the least restrictive option; the measure must be reasonably necessary, even if it is not absolutely necessary.124 Thirdly, the right to property under the ECHR is heavily qualified by the wide margin of appreciation afforded to states when adopting measures aimed at facilitating the payment of taxes125 and remedying social injustice.126 The ECtHR is particularly reluctant to intervene in matters pertaining to taxation and has taken a particularly deferential approach when determining whether measures restricting restitution of unlawfully exacted taxes violate Article 1 of the First Protocol.127 In National & Provincial Building Society v UK, the applicant building society challenged legislation, passed in the wake of the Woolwich decision, that retrospectively validated payments of taxes levied ultra vires. The applicants claimed that this retrospective removal of their cause of action in restitution infringed their property rights under the ECHR. The 119 R (Trailer and Marina (Leven) Ltd) v Secretary of State for the Environment, Food and Rural Affairs [2004] EWCA Civ 1580, [2005] 1 WLR 1267 [50]–[51]. 120 de Freitas (n 116) 80; Daly (n 116) 547. 121 Heaney v Ireland [1994] 3 IR 593 (Irish Sup Ct); Cox v Ireland [1992] 2 IR 503 (Irish High Ct); Chaulk v R [1990] 3 SCR 1303; R v Heywood [1994] 3 SCR 761. 122 Wilson v First County Trust Ltd (No 2) [2003] UKHL 40, [2004] 1 AC 816 [70]; R (Fisher) v English Nature [2003] EWHC 1599 (Admin) [2004] 1 WLR 503 [46]; Trailer and Marina (Leven) Ltd (n 119); Pascoe v First Secretary of State [2006] EWHC 2356 (Admin), [2007] 1 WLR 885. See also Bäck v Finland (2004) 40 EHRR 1184 [52]–[55]. However, this principle depends upon the context in which the measure was adopted: Smith v Secretary of State for Trade and Industry [2007] EWHC 1013 (Admin) [2008] 1 WLR 394 [42]. 123 SRM Global Master Fund LP (n 115) [46]. However, some cases have adopted a means-effects analysis: Beaulane Properties Ltd v Palmer [2005] EWHC 817 (Ch), [2006] Ch 79 [119]. 124 R (Clays Lane Housing Co-Operative Ltd) v The Housing Corporation [2004] EWCA Civ 1658, [2005] 1 WLR 2229 [25]. 125 National & Provincial (n 114). 126 James v United Kingdom (1986) 8 EHRR 123 [46]; Earl Cadogan v Sportelli [2008] UKHL 71, [2010] 1 AC 226 [48]. 127 Harris et al have described the power of the state to enforce the payment of taxes as ‘practically unlimited’: DJ Harris, M O’Boyle and C Warbrick, Law of the European Convention on Human Rights, 2nd edn (Oxford, OUP, 2009) 692.

Property, Proportionality, etc 143

ECtHR reaffirmed that Contracting States enjoyed a wide margin of appreciation in the context of taxation. The Court would not interfere with a national legislature’s assessment unless it was ‘devoid of reasonable foundation’.128 Although the measure had the effect of extinguishing the applicants’ claims for restitution, it did not appear that ‘the ultimate aim of the measure was without reasonable foundation having regard to the public-interest considerations which underpinned the proposal to legislate with retroactive effect and Parliament’s endorsement of that proposal’.129 The approach of the English courts to proportionality in taxation cases is almost as qualified as that of the ECtHR. In Wilson v First County Trust Ltd (No 2),130 Lord Nicholls held that the courts would interfere with the legislature’s choice of means where the legislature had attached insufficient importance to a person’s Convention rights, but that the ‘more the legislation concerns matters of broad social policy, the less ready will be a court to intervene’.131 In particular, the English courts have confirmed that the state enjoys a very wide margin of appreciation when imposing taxes,132 and that this margin of appreciation extends to the recovery of, as well as the levying of, taxes.133 However, this approach is questionable for two reasons. First, the role of the margin of appreciation is to take account of divergence in practice between different Contracting States to the ECHR. It has no relevance in a domestic context. In R v DPP ex parte Kebilene, Lord Hope confirmed that the margin of appreciation ‘is not available to the national courts when they are considering Convention issues arising within their own countries’.134 Thus, it is not clear why the state is entitled to rely on the margin of appreciation before the English courts. Secondly, even if the text of Article 1 of the First Protocol mandates a deferential approach to measures required to secure the payment of taxes, it is not clear why this approach should extend to the retention of unlawfully exacted taxes.135 Therefore, it is inappropriate for the English courts to afford a wide margin of appreciation to the legislature where recovery of unlawfully exacted taxes is sought, still less to take such a broad approach to the margin of appreciation that measures relating to taxation are effectively placed beyond the scope of Article 1 of the First Protocol. It is therefore open to the English courts to take a more rigorous view of the proportionality doctrine where its domestic jurisprudence so demands.136 In light of the acknowledged importance of the principle against unjust enrichment of the state through unlawful taxation, there are compelling reasons why the English courts should take a more rigorous approach to proportionality in the context of restitution of overpaid tax. In particular, the courts should consider the proportionality of measures that permit the retention of unlawfully exacted taxes. This approach should apply irrespective of whether the measure takes the form of a legislative provision or a common law doctrine. If the right to property affects the initial, unlawful, levying of the tax, it should also affect the ex post retention of money exacted unlawfully. Therefore, where the state seeks to rely on change of position to resist National & Provincial (n 114) [80]. ibid [81]. 130 Wilson (n 122). 131 ibid [70]. 132 Thorpe v HM Revenue & Customs [2009] EWHC 611 (Ch), [2009] STC 2107 [45]. 133 Monro (n 108) [33]. See also Hoffman (n 107) 70. 134 R v Director of Public Prosecutions ex parte Kebilene [2000] 2 AC 326 (HL) 380. 135 Lee (n 4). Compare Monro (n 108) [33], where Arden LJ held that the wide margin of discretion afforded to national authorities extended to ‘the determination of policy in relation to claims to recover tax’. 136 Clays Lane Housing Co-Operative Ltd (n 124) [18]. 128 129

144 Niamh Cleary repayment of unlawfully exacted taxes, it should be required to demonstrate that its retention of the money is proportionate to an objectively identifiable aim that justifies expropriating the taxpayer’s property.

(2) Change of Position and the Right to Property It is clear that the current formulation of the change of position defence does not adequately protect taxpayers’ property rights. As discussed in Part B, the current approach to change of position presumes that the state has incurred expenditure in reliance on its receipt of the tax. It also presumes that this expenditure has dissipated the whole of the enrichment. Considering this practice in light of the first limb of the proportionality test, the presumption in favour of expenditure means that the state does not have to justify its retention of the money by pointing to any pressing social objective. Even if the state was obliged to identify such an objective, it is not clear what that objective would be. The general public benefit has never been considered a sufficient justification for the outright expropriation of an individual’s property.137 Therefore, in the absence of exceptional circumstances, the cost to the public purse of making restitution should not constitute a pressing social objective that satisfies the proportionality principle. Thus, it cannot be said that the current permissive approach to change of position in FII and Bloomsbury requires the state to identify a legislative objective that is ‘sufficiently important to justify limiting a fundamental right’. Therefore, it fails the first limb of the proportionality test. The failure to identify a suitable objective means that the second and third limbs of the proportionality test fall along with the first. If the effect of the presumption of expenditure is that there is no need to identify any objective for retaining the money, it cannot be said that the measure in question – the change of position defence and the ensuing retention of the money – is rationally connected to any objective. Therefore, the change of position defence fails the second limb of the proportionality test. Similarly, if the state does not have to justify its expenditure, for example by pointing to any objective that would justify retention of the unlawfully exacted payments, there is no means of ascertaining that the means used to impair the taxpayer’s right to property ‘are no more than is necessary to accomplish the objective’, under the third limb of the proportionality test. The third limb of the test should not be satisfied where the state simply says that restitution of the taxes would disrupt the public finances. Instead, a severe potential disruption of the public finances should be required before such a justification would be considered acceptable. It is unlikely that the change of position defence can be modified to take account of property rights. Therefore, change of position should be replaced with a defence that considers whether allowing the state to retain the overpaid tax constitutes a disproportionate interference with the taxpayer’s property rights. To retain unlawfully exacted taxes, the state should be required to demonstrate that restitution would have serious adverse consequences for the public finances, that those consequences are sufficiently serious to justify retention of unlawfully expropriated taxes, and that no other revenue raising measure would be reasonably practical and effective to achieve the same result. This defence should 137 This principle predates the enactment of the HRA: London and North Western Railway Co v Evans [1893] 1 Ch 16 (CA) 28; Re De Keyser’s Royal Hotel Ltd [1919] 2 Ch 197 (CA) 225–26; Burmah Oil Co (Burma Trading) Ltd v Lord Advocate [1965] AC 75 (HL) 156.

Property, Proportionality, etc 145

be confined to claims for restitution of taxes or charges of a similar nature. However, within that category of cases, the defence should be available in respect of all claims where difficulties of proof preclude the state from relying on the change of position defence, irrespective of the cause of action relied upon by the state. After all, the same difficulties of proof attend a change of position defence by the state, irrespective of whether the claim is framed as a mistake claim or as a claim under the Woolwich principle. Moreover, the same public policy concerns underpin all claims for restitution of overpaid tax, irrespective of the specific cause of action relied upon by the taxpayer.138 This new, proportionality-oriented defence is one that is wholly based on public policy. It is also highly discretionary, insofar as judges are invited to assess whether or not the risk of fiscal disruption to the state is sufficient to justify interfering with the claimant taxpayer’s right to property. However, there are four reasons why these objections can, and should, be overcome. First, Part B of this chapter established that the traditional ‘expenditure and causation’ approach to change of position cannot be applied to the state. As a result, the unprincipled and excessively generous evidentiary approach adopted by the courts where change of position is relied on by the state strongly resembles a policy-motivated ‘fiscal chaos’ defence, without requiring what might be thought to be the fundamental requirement of ‘fiscal disruption’. On the other hand, framing a policy-motivated defence in terms of property rights would allow the courts to focus on the real concerns at stake in restitution claims against the state. When we speak of the state’s entitlement to rely on the security of its receipts, we are not concerned with causation or with specific items of expenditure. Instead, we are concerned with fiscal stability and the threat of fiscal disruption. Reconceptualising the defence in terms of interference with property rights allows us to reconcile the concern to prevent fiscal disruption against the taxpayer’s right to property. Secondly, an approach that focuses on the claimant’s property rights is consistent with the rationale for the change of position defence. Change of position is founded on a principle of justice. In claims between two private parties, change of position determines which of two innocent parties should bear a loss. However, in claims for restitution of overpaid tax, change of position seeks to establish whether a private individual, rather than the state, should bear the loss. Thus, any defence to claims for restitution of overpaid tax should consider whether individual taxpayers who have been taxed unlawfully should bear the loss or whether that loss should be borne by the Exchequer and, therefore, indirectly, by taxpayers as a whole. The English courts generally do not accept that burdens should be imposed upon individuals where they are more appropriately borne by society as a whole.139 It is not clear why this should be any different where the claimant’s claim is brought in unjust enrichment. The state should therefore be obliged to show compelling reasons why the loss should be borne by the individual taxpayer rather than by the Exchequer. This loss is often substantial: in FII, the sum sought to be recovered by the claimant was £5 billion. Therefore, if change of position is concerned with preventing injustice, the question of whether the impact on the claimant’s rights is proportionate to the social objective sought to be achieved by the state in retaining the money becomes a relevant consideration for any policy-motivated defence that seeks to replace change of position in claims against the state. See also, Bant (n 42) 141. R v Abbott [1897] 2 IR 362, 405; Cannon Brewery Company Ltd v Central Control Board (Liquor Traffic) [1918] 2 Ch 101 (CA) 120; Burmah Oil (n 137) 156. 138 139

146 Niamh Cleary Thirdly, unlike the traditional fiscal chaos defence, a proportionality analysis provides a principled framework for the operation of the defence. Instead of trying to determine whether permitting restitution would lead to ‘fiscal chaos’, a proportionality-based response is more sophisticated. Not only does it require the state to establish a threat of fiscal disruption, or other compelling policy objective for denying restitution, it also requires the state to demonstrate that this threat is sufficiently serious to justify expropriation of the claimant’s property. This latter requirement of the defence can be informed and supplemented by existing principles and cases concerning proportionate interference with the right to property, so that this restitutionary defence builds upon, and is informed by, existing property rights jurisprudence. Fourthly, it is clear that the courts are already willing to consider what constitutes a real and present threat of fiscal disruption, and whether that threat is sufficient to justify a denial of restitution. This approach is clearest in Irish restitutionary jurisprudence. In Murphy v Attorney General, the Irish Supreme Court adopted a generous change of position defence to protect against the perceived threat of fiscal disruption posed by potential restitution claims by ‘tens of thousands of married couples’140 who had been taxed at extremely high rates of tax over a considerable period.141 Conversely, in Re Article 26 and the Health (Amendment) (No 2) Bill 2004,142 the Supreme Court held that the state could not rely on fiscal chaos considerations to justify legislative measures that purported to retrospectively validate unlawfully levied health charges. It observed: [W]hile it is the opinion of the court that the financial burden on the State of making the relevant repayments is a substantial one, it is by no means clear that it can be described as anything like catastrophic or indeed that it is beyond the means of the State to make provision for this liability within the scope of normal budgetary management.143

Although this approach is less developed in England, in Waikato Regional Airport Limited v Attorney General, the Privy Council rejected a special defence to overpaid charges, holding that ‘the amount at stake, although significant for the appellants, cannot possibly be regarded as such as to threaten the disruption of public finances in New Zealand’.144 Each of these three cases can be seen as an example of the proportionality approach in action. In each case, the state claimed that restriction of the claimant’s right to recovery was necessary to meet a stated objective: to prevent disruption of the public finances. In each case, the court considered whether or not there was a real risk of fiscal disruption and whether this threat justified denying restitution to the claimant. Thus, the courts can, and do, assess the risk of fiscal disruption in claims for restitution against the state and consider whether the risk is sufficiently serious to deny redress to the claimant. There are compelling reasons to introduce a policy-motivated defence that considers the right to restitution in terms of the taxpayer’s right to property. However, a major obstacle to the approach discussed in this Part is the weak protection accorded to property rights under the ECHR. This protection is so weak that there is a significant risk that a defence that turns on property rights analysis will not give any meaningful protection to property rights or to the taxpayer’s interest in restitution. Therefore, the English courts’ approach to Murphy (n 50) 317. Y Scannell, ‘The Taxation of Married Women. Murphy v Attorney General ’ in E O’Dell (ed), Leading Cases of the Twentieth Century (Dublin, Round Hall Sweet & Maxwell, 2000). 142 Re Article 26 (n 96). 143 ibid [128]. 144 Waikato Regional Airport Limited v Attorney General [2003] UKPC 50, [2004] 3 NZLR 1 [82]. 140 141

Property, Proportionality, etc 147

protection of property rights under the ECHR may need to be fortified before this defence can offer any meaningful protection to taxpayers’ property rights.

E. CONCLUSIONS

The change of position defence was developed to do justice between private parties. However, recent experience suggests that it cannot do justice in claims where restitution is sought from the state. Part B demonstrated that the change of position defence, which relies on expenditure and causation, cannot be applied where the state seeks to defend itself against claims for restitution of overpaid tax. The generous approach to causation adopted by the courts where the state seeks to rely on change of position undermines the taxpayer’s interest in restitution, as well as the public interest in ensuring that the state does not retain unlawfully exacted taxes. However, these interests must be balanced against the equally important concern that the law of restitution should not cause serious disruption to state finances. Therefore, in some circumstances it will be necessary to protect the state’s security of its receipts. A policy-motivated defence is necessary to replace traditional defences that turn on causation and expenditure. However, none of the policy-oriented defences that have been proposed to date is capable of balancing the state’s security of receipt against the individual taxpayers’ right to restitution and the public interest in ensuring that the state is not enriched by unlawful receipts. Instead, any policy-motivated defence that seeks to replace the change of position defence should take account of the claimant’s right to property. Such a defence should consider whether the state’s retention of the money would constitute a disproportionate interference with the taxpayer’s property rights under Article 1 of the First Protocol to the ECHR. This approach allows the courts to strike a balance between preventing disruption of the state’s finances and ensuring that individual taxpayers are not required to bear the burden of unlawfully exacted taxes, where there are no compelling reasons for them to do so. Thus, a proportionality analysis allows the courts to focus on what is truly at stake in claims for restitution of overpaid tax. Finally, there are several issues that have not been touched on by this chapter, and may require further consideration. First, in addition to property rights derived from the ECHR, the right to property is also protected by Article 17 of the EU Charter on Fundamental Rights. Where EU law is engaged, the courts are obliged to act in a manner that is compatible with the state’s obligations imposed by the EU Charter. In such circumstances, any interference with the right to property must comply with the proportionality principle set out in Article 52 of the Charter. The right to property and the proportionality principle are formulated differently under the EU Charter and under the ECHR. Thus, the proportionality-based defence proposed in this chapter may require modification where EU law is engaged. However, in light of the significance accorded to ECHR jurisprudence in EU law,145 such differences are unlikely to be significant. Secondly, this chapter has focused on the spending decisions of central government. There may be smaller state bodies – for example local authorities and state regulatory bodies – whose budgets are tightly constrained and whose spending decisions are closely See Article 6 TEU; Opinion 2/94 on Accession by the Community Courts to the ECHR [1996] ECR I-1759.

145

148 Niamh Cleary related to their receipts. Thus, there may be circumstances in which there is no difficulty in establishing that a state body has changed its position in reliance upon specific receipts. These cases may therefore be amenable to determination pursuant to the traditional change of position defence. However, such an approach is undesirable for two reasons. First, a bifurcated approach that depends both on the nature the public body, and on the nature of spending decisions that it makes, is likely to be difficult to apply in practice. Secondly, the traditional approach is undesirable in principle, as it fails to have regard to the taxpayer’s right to property. However, the constrained nature of a public body’s finances may also entail that the threat of fiscal disruption is greater for small state bodies than for central government, making it easier for such bodies to justify retention of overpaid tax by reference to proportionality principles. Irrespective of the precise parameters of the defence that the courts choose to adopt, one consideration remains paramount: claims for restitution of unlawfully exacted tax concern the recovery of money to which the state is not, and was never, entitled. Therefore, there is a strong interest in ensuring restitution of those sums. Any policy-oriented defence that is adopted by the courts should be particularly sensitive to this consideration, and, in the absence of any risk of severe disruption of the public purse, should prioritise it over competing policy concerns that favour retention of the money by the state for reasons of administrative or fiscal efficiency.

8 Undoing Transactions for Tax Purposes: The Hastings-Bass Principle MONICA BHANDARI

A. INTRODUCTION

It is generally appreciated that some form of tax might be payable when a substantial gift is made. For example, a gift of company shares from a mother to a child might result in a capital gains tax (CGT) charge,1 and perhaps also an inheritance tax (IHT) charge.2 Suppose, though, that the mother mistakenly thought that a tax charge would not arise, for example, because she did not realise that tax could be levied on the gift, or received bad tax advice, or knew of the tax charge, and sought to minimise it, but failed to follow tax advice correctly. The mother might then wish to escape the charge by arguing that she would not have transferred the shares had she known that the tax would be payable. But it is too late. The tax liability crystallises at the time of the transfer, and once this has happened, the transfer cannot be ‘undone’ for tax purposes. The child can transfer the shares back to the mother, but the CGT liability will not disappear.3 This does not seem to be controversial.4 What is controversial, however, is that a trustee who makes a similar mistake when he transfers trust property to a recipient may be able to say exactly what the mother cannot say: that he mistook the tax consequences of the transfer, that he would not have transferred the property if he had understood these correctly, and that the transfer should therefore be ‘undone’ for tax purposes, ie the parties should be treated as though the transfer had never taken place and the tax liability had never arisen. This is the controversial principle in Re Hastings-Bass (deceased),5 first established in England, and since applied elsewhere.6 Taxation of Chargeable Gains Act 1992 (hereafter ‘TCGA 1992’), ss 286, 18 and 17. Such a gift would be liable to IHT only if the donor died within 7 years of the gift: Inheritance Tax Act 1984 (hereafter ‘IHTA 1984’), ss 2, 3, 3A and 7. 3 In fact, if the assets have increased in value between the time of the transfer to the child and the transfer back to the mother, then another CGT charge may arise. 4 Consider the hypothetical situation posited by counsel, and accepted by Henderson J, in Test Claimants in the Franked Investment Income Litigation v HM Revenue & Customs [2008] EWHC 2893 (Ch), [2009] STC 254 [257]: ‘if inheritance tax is received by the Revenue due to a failure to make an election in a deed of variation pursuant to IHTA 1984, s 142, the taxpayer cannot recover from the Revenue even if the failure to make the election was due to a mistake’. 5 Re Hastings-Bass (deceased) [1975] Ch 25. Differing judicial views have been expressed as to whether this case lends its name to a ‘principle’ or a ‘rule’. This chapter refers to the Hastings-Bass ‘principle’, to mark the fact that the courts have some discretion as to when it should be applied. 6 eg Ireland: Irish Pensions Trust Ltd v Central Remedial Clinic [2005] IEHC 87, [2006] 2 IR 126; Boliden Tara Mines Ltd v Cosgrove [2007] IEHC 60. Jersey: Re Green GLG Trust 2002 JLR 571; Friedman and Asiatrust Ltd [2006] 1 2

150 Monica Bhandari The Hastings-Bass principle puts trustees and their beneficiaries in a somewhat hallowed position, since it enables them to undo transactions to escape adverse tax consequences that they did not foresee. The principle has recently been reshaped and limited by Pitt v Holt.7 Yet the question remains as to whether the law should allow trustees (and their beneficiaries) to escape from the adverse tax consequences of transactions in circumstances where other taxpayers cannot. The cases which prompt this question have a different fact-pattern from the cases discussed in the other chapters of this volume. The other chapters all concern the situation where a claimant makes a payment to a tax authority in the mistaken belief that the money is owed as tax, and then makes a statutory or common law claim for restitution.8 This chapter concerns the situation where a trustee mistakenly transfers property to another party, the transfer triggers a tax liability, and the trustee then argues that the transfer should be ‘undone’, with the result that the basis of the tax liability falls away. The effect of a successful argument to this effect will vary according to whether the tax has been paid by the time of the proceedings: if it has, then the taxpayer can recover his money; if not, then he will be relieved of his obligation to pay.

B. THE ORIGINS OF THE HASTINGS-BASS PRINCIPLE

Although Hastings-Bass was decided as long ago as 1975, the principle to which the case gives its name is still said to be ‘emerging’ and ‘developing’.9 In fact, the Hastings-Bass case did not even apply the principle in the form in which it was applied in many later cases,10 the appearance of which dates from Mettoy Pension Trustees Ltd v Evans.11 Moreover, while the principle was successfully used to avoid tax liabilities in later cases, Hastings-Bass itself concerned an unsuccessful claim by the Revenue to have a transaction set aside. Thus the Revenue were instrumental in the creation of a principle that later came back to haunt them.12 The Hastings-Bass case concerned a tax avoidance scheme that went wrong. HastingsBass was a trust beneficiary with a life interest under a settlement made in 1947. In 1957 his sister made a settlement under which his son took a life interest, with trusts of capital in JRC 187; Re Howe Family Number 1 Trust [2007] JRC 248, 2007 JLR 660; Re Vistra Trust Co (Jersey) Ltd [2008] JRC 111; Re Seaton Trustees Ltd [2009] JRC 050; Re V Settlement [2011] JRC 046. Cayman Islands: A v Rothschild Trust Cayman Ltd [2006] WTLR 1129; Barclays Private Bank & Trust (Cayman) Ltd v Chamberlain [2007] WTLR 1697. However, the principle is not used in Scotland: D Francis, ‘Hastings-Bass and His Scottish Friends’ 2008 Scots Law Times 161. 7 Pitt v Holt [2011] EWCA Civ 197, [2012] Ch 132. This was a combined appeal from two first instance decisions: Pitt v Holt [2010] EWHC 45 (Ch), [2010] 1 WLR 1199, and Futter v Futter [2010] EWHC 449 (Ch), [2010] STC 982. 8 For statutory recovery regimes, see eg the Taxes Management Act 1970, ss 33 and 42, and the Value Added Tax Act 1994, s 80. There is some controversy over the interaction between the common law and statutory regimes where the legislation does not specifically exclude a common law claim. See J Beatson, ‘Restitution of Taxes, Levies and Other Imposts’ (1993) 109 LQR 405, 420; M Bhandari and C Mitchell, ‘Lessons of the Metallgesellschaft Litigation’ [2008] Restitution Law Review 1, 15–17. For legislation which specifically excludes a common law claim see eg the Value Added Tax Act 1994, s 80(7). 9 Breadner v Granville-Grossman [2001] Ch 523 [46] and [58]; Sieff v Fox [2005] EWHC 1312 (Ch), [2005] 1 WLR 3811 [118]; Gallaher Ltd v Gallaher Pensions Ltd [2005] EWHC 42 (Ch), [2005] Pens LR 103 [163]. 10 eg Breadner (n 9) [59]; AMP (UK) plc v Barker [2001] Pens LR 77 [85]; Sieff (n 9) [114]; Burrell v Burrell [2005] EWHC 245 (Ch), [2005] STC 569 [15] and [16]. 11 Mettoy Pension Trustees Ltd v Evans [1990] 1 WLR 1587, 1621. 12 M Gunn, ‘Turning the Clock Back’ (2002) 148 Taxation 634.

The Hastings-Bass Principle 151

favour of the son’s children and other trusts in default. In 1958 the trustees of the 1947 settlement advanced a fund valued at £50,000 to the trustees of the 1957 settlement, to be held for the son’s benefit on the 1957 trusts. The point of this was to remove the fund from the 1947 settlement so that there would be no estate duty payable on it when Hastings-Bass died, trust property in which he had a life interest being treated as passing on his death for estate duty purposes. The scheme would have worked but for the House of Lords’ decision in Re Pilkington’s Will Trusts.13 The effect of this decision on the arrangements in Hastings-Bass was that all the interests in the sub-settlement of the fund, other than the son’s life interest, were void for perpetuity. In 1964 Hastings-Bass then died and the Revenue argued that there had been no valid exercise of the trustees’ power of advancement because their purported exercise of this power had produced a result that was different from the one they had intended. It followed that Hastings-Bass’s life interest in the fund had continued until his death and hence that estate duty was payable. The trustees of the 1947 and the 1957 settlements brought proceedings against the Revenue to determine whether this was correct. Plowman J held that the transfer had not been effective, on the basis that the actual transaction in 1958 was substantially or essentially different from what the transferring trustees had intended. He reached this conclusion in line with the principle set out in Re Abrahams’ WT.14 However, his decision was overturned by the Court of Appeal, which considered that Hastings-Bass could be distinguished from Abrahams. In the latter case, the actual transaction had been substantially different from the intended transaction because it had not benefited the intended recipient. In contrast, in Hastings-Bass, the son had actually received a life interest in the fund, albeit that the other interests under the sub-settlement had not taken effect. The court held that it would not interfere with the exercise of a trustee’s discretion, even if the transaction did not fully achieve its intended effect, unless:15 it is clear that [the trustee] would not have acted as he did (a) had he not taken into account considerations which he should not have taken into account, or (b) had he not failed to take into account considerations which he ought to have taken into account.

This statement formed the starting point for later development of the Hastings-Bass principle. It makes it clear that there is generally no reason for the court to interfere in a trustee’s decisions where he has acted within the scope of his powers, and it holds that a court cannot do so unless the stated conditions are satisfied. This rendering of the HastingsBass principle takes a negative form,16 and in this respect it differs from Warner J’s later statement of the principle in Mettoy, where he said that:17 where a trustee acts under a discretion given to him by the terms of the trust, the court will interfere with his action if it is clear that he would not have acted as he did had he not failed to take into account considerations which he ought to have taken into account. 13 Re Pilkington’s Will Trust, Pilkington v Inland Revenue Commissioners [1964] AC 612 (HL), reversing [1964] AC 612 (CA). 14 Re Abrahams’ Will Trust, Caplan v Abrahams [1969] 1 Ch 463. 15 Hastings-Bass (n 5) 41. 16 As observed in Sieff (n 9) [46]; Burrell (n 10) [16]; Breadner (n 9) [59]; AMP (n 10) [84]; Mettoy (n 11) 1621. 17 Mettoy (n 11) 1621.

152 Monica Bhandari Many courts and commentators understood Warner J’s formulation to be nothing more than a positive version of the negative principle stated in Hastings-Bass, but in fact it marked a new departure for the law. In Hastings-Bass the court merely stated when it would not interfere with a trustee’s decisions. It did not say that there was a positive obligation on the court to intervene in cases where the stipulated conditions were met.18 Thus, Mettoy constituted a move away from Hastings-Bass, and took the law in an undesirable direction. There is no reason for the court to interfere in every case where a trustee takes irrelevant considerations into account or fails to take relevant considerations into account. Often no harm is caused by the trustee’s failure and such cases do not come to the courts’ attention, and even when a case comes to court, that does not mean that the court should always interfere, even if none of the parties objects to the transaction being set aside, especially where the only harm complained of is a tax liability. In the tax cases it is common for all the parties before the court to ask for the transaction to be set aside, since they will benefit from a reduced tax bill, and there is no party with an interest in opposing the application. An order setting the transaction aside obviously works to the detriment of the Revenue, but until recently, the Revenue did not appear in HastingsBass cases.19 In many cases, an offer was made to them to be joined as a party, but they always refused,20 although they sometimes asked for authorities to be brought to the court’s attention.21 It may be that the Revenue calculated that they would not be bound by the court’s orders if they refrained from participation, and hence that they would not have to refund (or waive a right to) tax if the court found that a transaction should be undone.22 Other reasons for their inaction may have been that they do not have the resources to fight every claim, and also that they were reluctant to have binding authority against them, which other parties might then use as a precedent without having to go to court. Nonetheless, as a result of the Revenue’s standoffishness, a series of cases were heard in which no-one seriously tried to dissuade the court from reversing transactions, and the scope of the Hastings-Bass principle steadily expanded. A striking illustration of the problems caused by a lack of Revenue participation is provided by Re Griffiths (deceased),23 although this was a mistake case rather than a case about the Hastings-Bass principle. The deceased made transfers of property with a view to minimising IHT, but this depended on his surviving for seven years thereafter. In the event, he died after two years because he had an aggressive form of cancer, a fact of which he was unaware when he made the transfers. If the transfers had not been made, the estate would have passed to the deceased’s wife tax free,24 but as matters turned out more than £1 mil18 A point made in ‘HMRC and the Hastings-Bass Principle’, Revenue Tax Bulletin 83 (June 2006), also released as Revenue Interpretation 278. 19 Other cases have used the principle in a different context, for example, in relation to pensions, which are outside the scope of this article. For more detail see for example Kerr v British Leyland (Staff) Trustees Ltd [2001] WTLR 1071 and Stannard v Fisons Pension Trust Ltd [1991] PLR 224. 20 eg Sieff (n 9) [29]; Burrell (n 10) [13]; Abacus Trust Company (Isle of Man) Ltd v Barr [2003] EWHC 114 (Ch) [12]; Abacus Trust Co (Isle of Man) Ltd v National Society for the Prevention of Cruelty to Children [2001] STC 1344 [2]. 21 eg Burrell (n 10) [13]. The same has been done in relation to mistake and rectification cases. See eg Re Griffiths (deceased) [2008] EWHC 118 (Ch), [2009] Ch 162 [6]; Farmer v Sloan [2004] EWHC 606 (Ch), [2005] WTLR 521 [8]. 22 Gunn (n 12) 636. See also NSPCC (n 20) [2] where the Revenue refused to agree to be bound by the decision of the court. 23 Griffiths (n 21). 24 IHTA 1984, s 18.

The Hastings-Bass Principle 153

lion was payable as IHT.25 Lewison J held that the transfers should be undone, on the basis that the deceased had made a mistake as to the present state of his health. However, the judge also said that it was difficult to be sure whether the deceased had had cancer at the relevant date, and that if this fact had been contested, it would have been difficult to make a finding to this effect.26 These comments suggest that if the Revenue had participated in the proceedings,27 adversarial argument about the facts might have made all the difference to the outcome.28 The Revenue attracted some criticism for their reluctance to become involved,29 and in 2006 they announced that they were more likely to participate in future litigation because they believed the Hastings-Bass principle to have been formulated too widely.30 Subsequently they involved themselves in two cases on the scope of the Hastings-Bass principle, Futter v Futter and Pitt v Holt, which have now been heard and decided together by the Court of Appeal.31 The court has substantially limited the principle, halting the runaway development which characterised the previous case law, and at the time of writing an appeal is pending to the Supreme Court.

C. THE DEVELOPMENT OF THE HASTINGS-BASS PRINCIPLE

(1) The Case Law Prior to Pitt v Holt One example of the Hastings-Bass principle having been successfully applied in the tax context is Green v Cobham.32 This was one of the first significant tax cases in which trustees were able to set aside a transaction, and it opened the doors for other trustees to escape from unwanted tax consequences. The case concerned a will trust in the British Virgin Islands. Its assets included shares in a holding company, which itself held shares in companies founded by the testator. The beneficiaries included the testator’s grandchildren. In 1990, the holding company had a large reserve of retained profits and the trustees wished to distribute these to the grandchildren. However, as three of them were minors, their shares were distributed via two accumulation and maintenance (A&M) trusts. The will trust and the A&M trusts were treated as a single settlement for CGT purposes. To prevent CGT being charged in the UK, the trust had to be non-resident in the UK. To achieve this, a certain number of the trustees had to be non-resident. When the will trust and the A&M trusts were set up, this requirement was satisfied because some of the trustees were non-resident by virtue of the rule established by the TCGA 1992, section 69(2). This Griffiths (n 21) [5]. Had the scheme worked correctly a great deal more tax would have been saved. ibid [18]. 27 ibid [6]. 28 Note that in rectification cases the Revenue has also been reluctant to participate, eg Farmer (n 21) [8]; Racal Group Services Ltd v Ashmore [1995] STC 1151, 1154; Ashcroft v Barnsdale [2010] EWHC 1948 (Ch), [2010] STC 2544. 29 eg Sieff (n 9) [83], though Lloyd LJ recognised there might be policy reasons for the Revenue’s stance. See too C Mitchell, ‘Reining in the Rule in Re Hastings-Bass’ (2006) 122 LQR 35, 36; Breadner (n 9) [61]; R Walker, ‘The Limits of the Principle in Re Hastings-Bass’ (2002) 13 King’s College Law Journal 173, 183. 30 Revenue Tax Bulletin 83 (n 18). 31 Pitt (n 7). See also Gresh v RBC Trust Co (Guernsey) Ltd 2009–2010 GLR 216. 32 Green v Cobham [2002] STC 820 (decided in January 2000). 25 26

154 Monica Bhandari provided that if a trustee carried on the business of managing trusts, and was the trustee of a will trust, he would be treated as non-resident wherever he actually resided.33 The A&M trusts were declared in November 1990, and in December 1990 one of the trustees of the A&M trust retired from practice. As a result he no longer fulfilled the section 69(2) condition and became a UK resident trustee. Hence there ceased to be sufficient non-resident trustees for the settlement to be treated as non-UK resident, UK tax rules applied to the settlement, and CGT became payable on all disposals by both the holding company and the will trust. This had significant implications for the tax treatment of the trusts, as unrealised gains of around £35 million were held within the settlement. The trustees asked the court to set aside the declaration of the A&M trusts, with the result that the retirement of the trustee would not affect the residency of the will trust, and no CGT would be payable. The trustees argued that they had failed to take the CGT consequences of the declaration into account, and would not have made the declaration if they had realised what these would be. They had not thought about them at all,34 nor had the lawyers drawn them to their attention. This was essentially because no one had appreciated that the A&M trusts would be treated as part of the same settlement as the will trust. Both parties to the case asked the court to set aside the declaration, and though the defendants advanced the opposing arguments, in the absence of any opposing party, this was hardly the same as the Revenue putting forward their own arguments. Jonathan Parker J accepted that the CGT consequences were matters that the trustees should have taken into account, and that if they had taken them into account they would not have made the declaration.35 He held that the CGT liability had arisen as a consequence of the declaration, albeit that it did not flow immediately from it, and said that the trustees’ error lay in making the residence of the will trust depend on the make-up of the trustees of the A&M trusts, something which was out of the will trust trustees’ control. This reasoning is problematic. The declaration of the A&M trusts only became ‘defective’ due to the A&M trustee’s retirement from practice, an event that was unrelated to the declaration itself. This may have fallen within the strict letter of the Hastings-Bass principle as it was understood in 2000, but it is doubtful that this was a deserving case. Jonathan Parker J said that if the declaration were effective, the CGT effects would be ‘catastrophic’.36 However, while it is true that a large tax liability would then have ensued, it is questionable whether ‘catastrophe’ is an apt description of a situation where someone shares in a £35 million gain, albeit one that triggers a CGT liability.37 Another striking feature of the tax cases prior to Pitt is that it seems to have made no difference what type of mistake the trustees made: the Hastings-Bass principle was able to rescue them from every error. Various mistakes can be at issue in cases where a taxpayer wishes to undo a transaction. The taxpayer may simply be unaware of a tax charge at the time of entering the transaction, he may receive incorrect tax advice relating to the transaction, or he may implement tax advice incorrectly. In the first case, where the taxpayer is simply unaware of the tax charge, this should not be considered a sufficient reason to undo the transaction. Many tax consequences can attach to transactions, and taxpayers have an 33 This legislation was changed by Finance Act 2006, s 88 and Sch 12, when the reference to residence related to managing trusts abroad was removed. 34 Green (n 32) 824. 35 ibid 828. 36 ibid 824. 37 Walker (n 29) 178.

The Hastings-Bass Principle 155

obligation to inform themselves of the tax liabilities which will result from their arrangements. Hence the fact that the charge was ‘unknown’ at the time of entering into the transaction cannot be a sufficient reason for undoing the transaction.38 Nevertheless the courts set aside transactions entered by trustees who made such a mistake. An example is Burrell v Burrell,39 a case about IHT liability. The settlor wished to transfer some company shares to members of his family. These were of the type that attracted Business Property Relief for IHT purposes,40 which should have meant that they were IHTexempt on transfer. An A&M trust of the shares was established in favour of the settlor’s son.41 However, when he reached the age of majority, the settlor thought that the dividends on the shares were too high for someone of such a young age to receive. So the trustees decided to end the interest in possession and create two new trusts. The shares were transferred into a discretionary trust. Although the shares were eligible for business property relief, which relieves a lifetime charge to IHT on entry into the trust, they had to be owned for two years before the transfer could attract the relief.42 The interest in the shares for IHT purposes only vested in the son when he gained an interest in possession of the trust,43 which was at the age of 18. However, the transfer actually took place when he was 19, and it followed that he had not owned the shares for the requisite period. Thus the transfer into the trust attracted an IHT liability of up to £1.47 million. Mann J said that this would be ‘a very serious loss to the trust estate’,44 and said that trustees must consider the tax consequences of their decisions and that failure to do so can trigger the Hastings-Bass principle.45 The trustees had thought about the tax consequences of their actions, but they had failed to give these proper attention, and it followed that the transfer should be set aside.46 Another way of looking at situations of this sort is to ask whether the trustees have failed to do something which it is their duty to do. They should take into account the tax consequences of a transaction and if they fail to do this, they have committed a breach of duty, and a negligence claim can be made against them, but there is no reason to undo the transaction to make the tax liability disappear. In fact, in Burrell it can be said that the trustee gave the tax issues as much thought as he was capable of giving them. He then relied on legal advisers to give him proper advice about the tax consequences. The solicitors were negligent in failing to give the trustees the full picture in relation to tax consequences.47 So Mann J granted the declaration of invalidity despite the fact that both the legal advisers and the trustees were negligent. As a result they escaped liability for their negligence and were

38 In Barclays (n 6) the trustees did not take into account a change in the law relating to CGT and this was sufficient to set aside the transaction on the basis that a relevant consideration was not taken into account. If such a rule is allowed to stand, there is no motivation for trustees to inform themselves of the law and keep up to date with it. As a general rule, knowledge of the law and changes to the law are imputed to individuals, yet trustees appear to be sheltered from this rule. 39 Burrell (n 10). 40 IHTA 1984, ss 105(1)(bb) and 122. 41 IHTA 1984, s 71. The rules relating to the beneficial IHT treatment of these trusts have changed since the Finance Act 2006, which undertook a major overhaul of the IHT treatment of trusts. 42 IHTA 1984, s 106. 43 IHTA 1984, s 49. 44 Burrell (n 10) [10]. 45 ibid [19]. 46 ibid [20]. 47 ibid [11].

156 Monica Bhandari able to avoid the transaction.48 Furthermore, the trustees had put in place another deed of appointment on the basis that the original creation of the discretionary trust was invalid, suggesting that a declaration of invalidity was insufficient to solve the trustees’ problems, and that this was really a rectification case.49 The foregoing discussion leads us on to other types of mistake. Where tax advice has been received but the advice is incorrect or flawed in some way, giving rise to a tax liability, this is also an error which should be insufficient to set aside the transaction. Nevertheless it was held to be sufficient in Hastings-Bass cases.50 For example, in Sieff v Fox,51 the assets of a settlement included Woburn Abbey and a number of chattels of high value. There was a discretionary trust of the chattels. On the 10-year anniversary of this trust in 2001, there was a very high 10-year IHT charge.52 The trustees sought a way to minimise this charge in the future. One option suggested by their advisers was to transfer the property from the original settlement to Lord Howland, the primary beneficiary, contingent on his being alive on a future date. At that time, Lord Howland would resettle the property on a more flexible trust. The trustees were advised that a CGT charge would arise on the transfer of the assets from one trust to the other,53 but that hold-over relief54 would be available and so there would be no tax payable at that time. Furthermore, no IHT exit charge55 would be payable on the transfer out of the discretionary trust, provided that the transfer occurred within three months of the 10-year charge.56 This advice was incorrect because this type of hold-over relief for CGT can only operate where there is an IHT charge. Here, there was no such charge, as the transfer was made soon after the 10-year charge. Thus no hold-over relief was available and the CGT charge of approximately £1 million was triggered.57 The other problem with the transfer to the second trust arose in relation to IHT. Lord Howland, as had been planned for some time, moved into an apartment at Woburn Abbey, where the chattels were kept. When the chattels were transferred, an IHT charge was triggered as it was an assignment of a contingent interest. However, once Lord Howland was enjoying some benefit from his gift, this would be seen as a gift with reservation of benefit,58 and so the chattels would be treated as part of Lord Howland’s estate on his death unless he paid market value for the use of the assets. This was approximately £40,000 a year.59 Thus, as Lloyd LJ60 pointed out, there would either be a high IHT charge on his death, or a high annual sum to be paid from taxed income.61 Quite why it is offensive for a taxpayer to have to pay a charge that anyone else in similar circumstances would have to pay from taxed 48 Hilliard suggests that this might in fact be a positive outcome, since it enables the long term relationship between the trustee and the beneficiary can be maintained: J Hilliard, ‘Limiting Re Hastings-Bass’ [2004] Conveyancer & Property Lawyer 208, 212. 49 Rectification was not mentioned in this case, but note Smithson v Hamilton [2007] EWHC 2900 (Ch), [2008] 1 WLR 1453 [60]–[80], where Park J suggested that rectification should not be allowed through the back door by using a Hastings-Bass claim. 50 Burrell (n 10) [104]. 51 Sieff (n 9). 52 IHTA 1984, s 64. 53 TCGA 1992, s 71(1). 54 TCGA 1992, s 260. 55 IHTA 1994, s 65. 56 ibid. 57 Sieff (n 9) [25]. 58 Finance Act 1986, s 102. 59 Sieff (n 9) [225]. 60 Lloyd J heard the case in the High Court but by the time he gave judgment, he had become Lloyd LJ. 61 Sieff (n 9) [25].

The Hastings-Bass Principle 157

income is unclear. Lloyd LJ also found that if the trustees had appreciated the tax consequences of the appointment, they would not have made it. Yet the trustees did not overlook the tax consequences. This was a case where the legal advisers gave erroneous advice. Despite this, the Revenue lost the tax which had become payable as a result of the transaction, rather than the erring legal adviser bearing the burden of the losses that had flowed directly from the incorrect advice that was given. If tax advice is taken and it is incorrect, then this is not a reason to set the transaction aside. Rather it is a reason for saying that remedies should be sought against the tax adviser.62 The transaction itself has achieved in legal terms what it was supposed to achieve, and the only problem is that it did not have the tax consequences that were hoped for. Where tax advice is sought, there is always a danger that it might not be correct, and the law has mechanisms to deal with such a problem, namely an action for negligence.63 There is no reason to give added protection to certain groups of people from the effects of such advice. If the advice were correct, then the benefit would have been achieved from the tax avoidance. If it is not, and a loss results, then that is simply the risk/reward balance that has to be taken into account when seeking professional advice. Where tax advice is taken and is correctly given, but the taxpayer does not effect the transaction in accordance with the advice, then once again, this is not an error that warrants the transaction being set aside. Yet, once again, the Hastings-Bass principle was used to rescue trustees in this situation. An example is Abacus Trust Co (Isle of Man) Ltd v NSPCC, which led to other cases brought by the same trust company.64 In NSPCC, a CGT liability of approximately £1.2 million was successfully avoided. The trustees entered a scheme to avoid CGT when a loan note matured. For the tax avoidance scheme to succeed, two trusts had to be set up in the tax year ending 5 April 1998 and an appointment to charity had to be undertaken after 6 April 1998. It was essential that the transactions had to be in different tax years. Counsel advised the trustee company of this fact, but the lawyer drafting the documents suggested that the appointment should be made on 3 April and the director of the trustee company agreed without referring to his notes from his meeting with counsel. Patten J held that the trustees had an obligation to consider if an appointment would result in a significant tax charge for the beneficiaries or the fund, and failure to do so could bring the Hastings-Bass principle into play.65 The problem here was that the trustees received good tax advice which they did not follow. Yet they were able to escape from the tax consequences and any liability. This seems at odds with the way one would expect the law to treat those who follow accurate advice in an incompetent manner. Since the fault lay with the trustees alone, why should the burden of their error have been borne by anyone else?66

62 See Tang Hang Wu, ‘Rationalising Re Hastings-Bass: A Duty to Act on Proper Bases’ [2007] Trust Law International 62. See too Revenue Tax Bulletin 83 (n 18) [6]. 63 In his discussion of Sieff (n 9), Mitchell (n 29) 36 remarks that there must have been a sigh of relief from the legal advisers when the claim was permitted. 64 Abacus v NSPCC and Abacus v Barr (n 20). 65 NSPCC (n 20) [16]. 66 See also Tang Hang Wu (n 62) 76.

158 Monica Bhandari (2) Pitt v Holt In Pitt v Holt it was recognised that the principle in Hastings-Bass had been abused. Lloyd LJ went back to the case itself and concluded that the subsequent case law had ignored the actual ratio of Hastings-Bass.67 He took this to be that where part of the purpose of a transaction fails, that does not mean that the whole transaction automatically fails. Thus, in a tax case where the trustees have sought to limit or avoid tax, but have failed to do so, that does not automatically mean that the transaction should be set aside. Rather, it is necessary to look at the trustees’ actions and to consider whether or not they have acted within their powers. Where they have acted outside their powers, the transaction is void and must be set aside. Where they have acted within their powers, matters are a little more complicated. In this situation there are two possibilities. The trustees may have committed a breach of duty, in which case the transaction must be set aside. Or they may have committed no breach of duty, in which case the transaction must stand. Hence, in cases where the trustees make an error respecting the tax consequences of a transaction, the transaction may still be set aside, but only if the trustees’ error is the result of a breach of duty. If they have taken advice, and the advice is wrong, then this does not constitute sufficient reason in itself to undo the transaction: only if the trustees committed a breach of duty when taking and acting on the advice will the transaction be set aside, and in most cases it can be anticipated that trustees will have acted reasonably, and will not have breached their duties, when employing advisers and following the advice they are given.68 This reinterpretation of Hastings-Bass means that the principle should no longer be understood in terms of trustees failing to take into account relevant considerations, and should now be seen instead in terms of trustees committing a breach of duty. As a result, the principle has been narrowed, and as Lloyd LJ acknowledged, many previous cases would have been decided differently if his reformulation of the Hastings-Bass principle had been applied.69 For the reasons that have already been given, this must be regarded as a positive development for the law. Furthermore, it can be anticipated that future cases seeking to invoke the principle are unlikely to be those where there is agreement between the parties as to the court order sought.70 This means that the rule will now develop with proper involvement of adversarial parties, rather than those who have a common goal. This is more likely to lead to sensible development and will avoid unnecessary and undesirable broadening of the principle. In these respects, Pitt really does clip the wings of the HastingsBass principle. However, in what remains of the principle, it is still problematic.

D. HASTINGS-BASS AS A REMEDY FOR BREACH OF TRUST

Pitt leaves the Hastings-Bass principle intact in situations where there has been a breach of trust. So, for example, in a case where the trustees have negligently failed to take advice in Pitt (n 7) [32]. ibid [127]. 69 ibid [129]. 70 ibid [130], where Lloyd LJ confirmed that this was a likely consequence of his judgment. 67 68

The Hastings-Bass Principle 159

relation to the tax consequences of a transaction,71 the beneficiaries can still seek a court order to have the transaction reversed. Whilst this reduces the availability of Hastings-Bass as a basis for a claim it still means that the reversal of a transaction, along with its tax consequences, is a possible remedy for breach of trust. To explain why this is problematic, it is necessary to consider the remedies that are available to comparable parties – namely individuals and the principals of fiduciaries other than trustees who enter transactions that turn out to have undesirable tax consequences.

(1) Rescission and Rectification Individuals who enter transactions by mistake may be entitled to rescission or rectification.72 Rescission undoes the transaction and the situation is treated as if the transaction never occurred. The rationale for this is that the mistake negates the claimant’s consent to the transaction. Rectification, on the other hand, is awarded in cases where the effect of the mistake is to prevent the transferor’s intention being carried through. Hence the documentation is rectified so that it accurately reflects the transferor’s intention. A full consideration of these two remedies is beyond the scope of this article,73 but two observations are worth making. First, in cases involving tax, these remedies are rarely available,74 and they are not usually available at all where the only mistake relates to tax consequences. For example, the courts generally follow Millett J’s finding in Gibbon v Mitchell,75 that rescission for mistake is awarded only where the mistake relates to the legal effects of the transaction, and is not awarded where the mistake relates only to the consequences of, or advantages to be gained by, entering the transaction. This distinction between the effects and the consequences of the transaction has been used to deny relief in cases where the mistake relates only to the tax consequences of the mistake. This distinction is not without its critics,76 but it is still the key test used in the case law. Secondly, and more importantly for the current purposes, it is notable that the HastingsBass principle has generally been used in cases where rescission for mistake and rectification are not available as remedies. Either these remedies have been expressly refused or it has been assumed that they are not available. It is clear, therefore, that although the law provides remedies to individuals who enter transactions by mistake, it does not usually enable them to escape from the tax consequences of such transactions. Furthermore, the Hastings-Bass principle allows trustees to undo transactions in order to escape from tax consequences in situations where the law would not permit an individual to do so.

ibid [128]. Gibbon v Mitchell [1990] 1 WLR 1304, 1307. See also B Häcker, ‘Mistakes in the Execution of Documents: Recent Cases on Rectification and Related Doctrines’ (2008) 19 King’s Law Journal 293, 320. 73 For more detail see J Hilliard, ‘Gibbon v Mitchell Reconsidered: Mistakes as to Effects and Mistakes as to Consequences: Parts 1 and 2’ [2004] Private Client Business 357 and [2005] Private Client Business 31; Häcker (n 72); M Bhandari, ‘Unravelling Transactions for Tax Purposes: Rescission and Rectification’ [2011] British Tax Review 82. 74 See eg Racal (n 28); Whiteside v Whiteside [1950] Ch 56; Allnutt v Wilding [2007] EWCA Civ 412, [2007] WTLR 941; Pitt (n 7). 75 Gibbon (n 72) 1309. 76 eg Hilliard (Conv) (n 48) 217. 71 72

160 Monica Bhandari (2) Remedies for Breach of Fiduciary Duty By using the Hastings-Bass principle to furnish trust beneficiaries with a remedy for breach of trust, the law gives them greater protection than it gives to individuals or the principals of other types of fiduciary. It can be argued that beneficiaries should be given special protection because the trust structure makes it hard for them to look after their own interests. Legal ownership of the trust property is vested in the trustee, who has the power to dispose of it, and the beneficiaries are vulnerable to negligent or self-serving abuses of this power. The trustee is therefore placed under a proscriptive obligation to act for the benefit of the beneficiaries, and if this includes a duty to take the tax consequences of a transaction into account, then the beneficiary must be provided with an adequate remedy where this duty is breached. Hence it can be said that the Hastings-Bass principle simply provides an adequate remedy to the beneficiary who should not have the bear the burden of a tax liability that results from the trustee’s breach of duty. The question arises, though, whether the special nature of the trustee-beneficiary relationship is sufficient reason to put a beneficiary in a better position than an individual acting for his own benefit? In response, it might be said that comparison with an individual is inapt. An individual who acts on his own behalf has the right to arrange his affairs so as to minimise the tax that he will have to pay.77 If he does not exercise that right, or does so incorrectly, so that a larger than expected tax liability arises, the blame can rest only with that individual and there can be no remedy for him.78 A beneficiary on the other hand has no choice whether to enter the transaction, and cannot minimise his tax liability in relation to the trust property, as it is only the trustee who is able to arrange the trust affairs. In this situation, therefore, it is appropriate to transform the trustee’s right to minimise the tax liability into a duty: to protect the beneficiary’s right that the least amount of tax paid should be paid either by the beneficiary or on his behalf, a duty must be imposed on the trustees, which will be breached in the event that they fail to take tax considerations properly into account when entering transactions. Should this breach of duty be treated differently to other breaches of fiduciary duty? In the case of other fiduciary relationships, where the tax consequences of a transaction are not taken into account, or not properly taken into account, the remedies available to the principal would appear to be limited to an action for breach of duty against the fiduciary. This suggests that in the trusts context, too, the beneficiaries should be restricted to an action for breach of trust against the trustees, and that any crystallised tax liability should be left undisturbed. It is important, therefore, to consider whether the remedy granted in the Hastings-Bass cases is significantly different from the remedies awarded in respect of other breaches of fiduciary duty. Breaches of fiduciary duty can give rise to a number of remedies, including orders that the fiduciary must pay equitable compensation or disgorge the profits of his wrongdoing; remedies can also be awarded against third party recipients of property transferred in breach of fiduciary duty.79 Personal remedies awarded against wrongdoing fiduciaries are Inland Revenue Commissioners v Duke of Westminster [1936] AC 1; Ensign Tankers v Stokes [1992] STC 617. Outside of the remedies of rescission and rectification where there is a problem with consent to the transaction. 79 See generally D Hayton, P Matthews, and C Mitchell, Underhill and Hayton: Law Relating to Trusts and Trustees, 18th edn (London, LexisNexis Butterworths, 2010) chs 9, 22 and 24. 77 78

The Hastings-Bass Principle 161

quite different from the remedy available under Hastings-Bass and are not concerned with ‘undoing’ the transaction. Proprietary remedies awarded against third party recipients are closer to the remedy available under Hastings-Bass, in terms of trying to ‘undo’ the transaction, but they are not available against a bona fide purchaser for value without notice.80 We can compare the Revenue to a bona fide purchaser. Whilst they have not ‘purchased’ tax payments in the traditional sense, a legal and legitimate liability has crystallised to the Revenue that should be honoured. There are also policy reasons for treating the Revenue as bona fide purchasers. The benefit is to the public purse and therefore in this sense, there can be no justification for stripping the benefit where the receipt was innocent and to be used for the public good. Even if this is not the case, it could be argued that the Revenue should not have to repay sums paid as tax as a consequence of a breach of fiduciary duty in cases where they have changed their position in good faith as a result of receiving the money. This is a difficult argument to make, because it will generally be hard to show that the Revenue were caused to change their position by the receipt of the claimant’s money.81 Nevertheless in Test Claimants in the FII Group Litigation v HMRC,82 Henderson J held that change of position is a defence which is available to the Revenue. A detailed discussion of this point is beyond the scope of this chapter, but what should be clear is that the remedy provided in HastingsBass cases does not sit comfortably with the other remedies available against third party recipients of property transferred in breach of fiduciary duty.

(3) Justifications for the Hastings-Bass Principle As it is clear that the Hastings-Bass principle, even in its more limited form since Pitt, leads to trust beneficiaries being treated differently from comparable parties, the question arises whether this different treatment can be justified. Equity between taxpayers is a cornerstone of tax policy which should be maintained.83 To consider justifications for differential treatment, it is essential to understand who actually benefits from the application of the principle. In most tax cases the obvious answer to this might appear to be beneficiary: he will benefit from the transaction being set aside because this will alleviate a tax burden that would otherwise be borne by the beneficiary personally or by the trust estate. If the beneficiary’s own tax burden is relieved, the benefit to the beneficiary is easy to see. If the trust estate is relieved of a tax burden, the beneficiary still benefits because there will then be more assets in the trust in which he has an interest.84 On closer inspection, however, we can see that it is not the beneficiary but the trustee who really benefits from application of the Hastings-Bass principle following the decision ibid, paras 99.14 ff. For which reason Niamh Cleary argues that the defence should not be available in chapter 7 of this volume. 82 Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2008] EWHC 2893 (Ch), [2009] STC 254. In Ireland, there have also been some statements in favour of change of position being available for the Revenue, for example in Murphy v AG [1982] IR 241. 83 The principle that taxpayers in a similar position should pay a similar amount of tax was recognised in A Smith, Wealth of Nations, Book V ch II Pt II ‘Of Taxes’ and is widely known as the concept of horizontal equity in the tax system. 84 cf NSPCC (n 20) where undoing the transaction would result in the removal of a gift to a charity but this was not contested by the charity, nor was it by the Attorney General on behalf of the charity, who was invited to make representations in the case. 80 81

162 Monica Bhandari in Pitt that the principle only applies where the trustee has committed a breach of duty. In these circumstances, the beneficiary does not need to undo the transaction to recover his loss, since he can sue the trustee for compensation, and it is the trustee who benefits from application of the principle since this enables him to escape his compensatory liability, at the expense of the public on whose behalf tax is collected by the Revenue.85 Hilliard has argued that the trustee is not in fact protected by the principle. One reason he gives is that in practice trustees are protected by a wealth of exemption clauses in the trust documents and so it will be very difficult for beneficiaries to pursue them.86 When limiting the Hastings-Bass principle in Pitt, Lloyd LJ also recognised this problem.87 However, as Tang Hang Wu points out88 this is an issue better dealt with by considering the rules relating to exemption clauses. If exemption clauses are available in such a wide range of situations that beneficiaries have no protection against trustees, perhaps the rules for exemption clauses need to be reconsidered. This is not a reason to provide the beneficiaries with another remedy that shifts their loss onto another person altogether, an outcome that is particularly repugnant when this other person is the Revenue, representing the public at large. Alternatively, if it is considered that there are sound reasons for allowing trustees to take the benefit of exemption clauses, because the ‘deal’ between the settlor and trustees means that the beneficiaries never acquire enforceable rights against the trustees when they act negligently, then the correct conclusion to be drawn is that the beneficiaries have no remedies at all in cases where the trustees’ negligence exposes them to an unforeseen tax liability. Hilliard also argues that the Hastings-Bass principle protects beneficiaries because it relieves them of the need to become entangled in a hostile negligence claim against their trustees which may entail the expenditure of their own money on legal costs.89 However, this argument is also weak because in other areas of life the recipients of negligent advice and services must enter into hostile litigation and incur expenditure on legal costs, and there is no reason why a trust beneficiary should be sheltered from these difficulties.90 It has also been argued that beneficiaries should be entitled to special treatment to the extent that no third party loses out.91 However, in tax cases there is always a third party who loses out – the Revenue.92 In all of the Hastings-Bass tax cases, a legitimate tax liability arises, and the Revenue acquires a legitimate corresponding right to payment.93 The fact that the parties did not foresee that this right would come into existence, and wish that it had not, cannot be a sufficient reason to undo the transaction, just as this would be an insufficient reason for an individual to escape the tax consequences of a transaction into which he has entered. The benefit of society as a whole should be put before the protection of a trustee.94 85 Tang Hang Wu (n 62) 76. See also I Dawson, ‘The Effect of an Unthinking Trustees’ Action’ [2002] Conveyancer & Property Lawyer 67, 71–72; R Nolan and M Conaglen, ‘Hastings-Bass and Third Parties’ (2006) 65 CLJ 499; Donaldson v Smith [2006] EWHC B9 (Ch), [2007] WTLR 421. 86 See Hilliard (Conv) (n 48) 212. See also Breadner (n 9) [57]. 87 Pitt (n 7) [98]. 88 Tang Hang Wu (n 62) 69–70 and n 48. 89 Hilliard (Conv) (n 48) 207 and 212–13. cf Dawson (n 85) 76. 90 See Tang Hang Wu (n 62) 69–70. 91 Hilliard (Conv) (n 48) 213. 92 Walker (n 29) 240 considers that this might be an option, but that the question is open to debate. 93 It could be argued that the Revenue is merely a ‘volunteer’ and therefore should not receive protection. However, the better view is that the Revenue’s right is triggered by legislation and as such is at least as strong as that of a purchaser and so the right should be protected as a purchaser’s right would be. 94 It is possible in some cases there might be different public policy issues at play which justify the operation of the Hastings-Bass principle, for example in the case of pensions, where the beneficiaries have purchased an interest

The Hastings-Bass Principle 163

In Sieff v Fox Lloyd LJ recognised the difference in treatment between individuals and trustees,95 and said that such a difference was justified for two reasons: first because trustees are dealing with property which is not their own and secondly because trust taxation is more complex than the taxation of individuals. In Re Griffith (deceased) Lewison J echoed the first reason by saying that a higher test should apply when an individual disposes of his own property.96 However, in neither case were these ‘justifications’ explained and, with respect, it is doubtful that these factors actually justify the differential treatment.97 So far as the first explanation goes, it is true that the trustee holds property for the benefit of others, but it does not follow from this that the trustee should be treated differently from any other legal owner of property when he seeks to escape the unforeseen tax consequences of transactions. It is also true that the beneficiaries should be protected from the negligence of the trustee, but that is a separate issue from the question whether the trustee as legal owner should have a privileged position vis-à-vis the Revenue. The complexity of trust taxation is not a sound reason for the difference either. Taxation can be complicated at any level and whilst the affairs of some individuals are straightforward, those of others are more complicated. As between individual taxpayers, no question arises of more favourable treatment being afforded to those who are subject to complex tax rules than is afforded to those whose tax affairs are simple. Furthermore, whilst taxation of trusts can be complex, the parties have voluntarily put themselves within this complex regime. It is their choice to have the taxation regime of trusts apply to them from the outset, and so they are in no position to complain about the complexity of the regime into which they voluntarily entered. Parties often use trusts in order to avoid tax.98 Where that is their motivation for choosing a more complicated taxation regime, they are therefore hoping to derive a benefit from their choice. If their plan fails and they find themselves obliged to pay as much as, or even more than, they would have done, had they not entered into a particular transaction, then it can be said that their choice turned out badly, but the risk of this happening was one that they took voluntarily. A connected point is that a taxpayer owes no obligation to set up his affairs so as to incur the most amount of tax possible,99 but he has no right to pay the minimum amount of tax possible. If he fails to take advantage of a scheme that could reduce his tax when he arranges his affairs, or improperly implements such a scheme, his tax liability crystallises, and no question can arise of this constituting a ‘windfall’ in the hands of the Revenue.100 It may be true that he would have avoided the tax if he had properly implemented the scheme, but that is by the by. If his tax liability were really a ‘windfall’ to the Revenue, then that would suggest that whenever a taxpayer arranged his affairs in a way that exposed him to a greater tax liability than the minimum he could have achieved with more careful planning, the and where the policy motivation might be quite different from cases in which the trust has a less meaningful aim – such as tax avoidance. There, the issue is not whether or not a tax charge is triggered, but rather whether other decisions taken by the trustees should stand. It has been recognised that pension cases are different, for example by the fact that it only needs to be established that trustees might have acted differently had they taken into account the relevant considerations. See Kerr (n 19) and Stannard (n 19). 95 Sieff (n 9) [85]. See also M Thomas and B Dowrick, ‘The Odd Couple? Hastings Bass and Mistake’ [2006] Conveyancer and Property Lawyer 91, 110. 96 Griffiths (n 21) [27]. 97 Mitchell (n 29) 41. 98 Although it should be noted that the use of trusts in tax planning has become less attractive in recent years as a consequence of the more punitive taxation of domestic trusts and those with offshore features. 99 IRC v Duke of Westminster [1936] AC 1; Ensign Tankers v Stokes [1992] STC 617. 100 As stated in eg Re Slocock [1979] 1 All ER 358, 363.

164 Monica Bhandari Revenue should not be allowed to keep the difference.101 If that were the rule then administration of the tax system would be significantly more burdensome and expensive, taxpayers would have no incentive to set out their affairs in a sensible manner from the outset, and tax advisers would effectively be given the benefit of hindsight and carte blanche to change transactions so as to minimise tax liability.102 Clearly the purpose of setting aside transactions with unforeseen tax consequences is to avoid a tax charge, but it is also interesting to consider that the transaction itself is often part of a tax avoidance scheme. The approach of the courts in dealing with such cases can be contrasted with their general approach to tax avoidance. One of the starting points in such cases is the fact that there is no obligation to pay the most amount of tax possible. However, the courts have acknowledged that if transactions have no commercial purpose other than to avoid tax, such transactions can be ignored for tax purposes and thus the transaction can be taxed as if artificial steps were not included.103 Whilst there can be legitimate tax avoidance, and tax avoidance schemes have not always been struck down,104 the courts are wary of tax avoidance and recognise the need to separate transactions with a true commercial nature from those with the sole aim of avoiding tax.105 Here we see a contrast with the approach in many Hastings-Bass cases, where the courts have turned a blind eye to the fact that the transaction was related to tax avoidance and have allowed trustees to escape from the tax consequences flowing from a tax avoidance scheme which has been improperly implemented. This is in particular contrast to the tax avoidance cases where the transaction entered into is extremely artificial.106 In fact, had the Revenue participated in such cases it would have been surprising if they had not sought to oppose the setting aside of the transaction on the ground that it formed part of a tax avoidance scheme. If courts can ignore artificial transactions where they have a tax avoidance purpose, then there is no reason to set aside a transaction which does not achieve its avoidance purpose. This is because, even if the transaction were put in place in the proper manner, the courts would be able to see through the transaction. The courts need not strike down every tax avoidance scheme, but there is a vast difference between this and aiding taxpayers in their desire to escape tax by helping them to set aside the unwanted consequences of the transaction. Even though tax avoidance can be legitimate, this is not to say it should be actively encouraged by the courts and if the courts are to maintain control over tax avoidance, they cannot send conflicting messages. Thus, it is important that the courts continue to treat tax cases separately, as they do for individuals, and prevent parties from altering or undoing transactions simply due to unwanted tax consequences flowing from transactions. 101 Note that this is quite different from cases where money paid as tax was never legally due and can therefore truly be described as a windfall that the Revenue must return. 102 cf Hilliard ([2005] PCB) (n 73) 36 and n 19. 103 Furniss v Dawson [1984] AC 474; WT Ramsay Ltd v IRC [1982] AC 300; Inland Revenue Commissioners v Burmah Oil Co Ltd [1982] STC 30. 104 eg Barclays Mercantile Business Finance Ltd v Mawson [2004] UKHL 51. 105 Also worthy of note is the emerging principle of ‘abus de droit’ in the European context, which uses this distinction. See Case C-255/02 Halifax plc v Commissioners of Customs & Excise [2006] ECR I-1609; WHA Ltd v HM Revenue & Customs [2007] EWCA Civ 728, [2007] STC 1695; Case C-103/09 HM Revenue & Customs v Weald Leasing Ltd [2011] STC 596. See too HL McCarthy, ‘Abuse of Rights: The Effect of the Doctrine on VAT Planning’ [2007] British Tax Review 160; R de la Feria, ‘Prohibition of Abuse of (Community) Law: The Creation of a New Principle of EC Law through Tax’ (2008) 45 Common Market Law Review 395; M Ridsdale, ‘Abuse of Rights, Fiscal Neutrality and VAT’ (2005) 14 European Community Tax Review 82. 106 Walker (n 29) 177–78.

The Hastings-Bass Principle 165

There is also a pragmatic reason for preventing the unravelling of transactions which have tax consequences in that the potential tax consequences of unravelling a transaction are extremely difficult. A simple example would be that of a discretionary trust which makes an appointment of assets to a beneficiary. At the time of the appointment, there will be a CGT charge on the difference between the market value at the time of the appointment and the previous acquisition cost. Then the beneficiary will be directly taxed on any income which is received from the assets.107 If the trustees, for some reason, realise that there are consequences of the transaction which are unpalatable, and wish to have the transaction set aside, the tax implications are difficult. The tax that was paid when the assets left the trust should not have been paid. If there was an exit charge for IHT purposes, that should not have been paid. Further, the tax paid on income arising from the assets was paid by the wrong taxpayer – if the appointment is set aside, the trustees owned the assets and so should have paid tax on the income arising from them. It is quite possible that the taxpayer in fact paid less tax than the trustee would have to pay,108 although it is unlikely that the trustees would seek to have the transaction set aside in these circumstances. In Hastings-Bass itself, the court pointed out that the income tax consequences of setting the transaction aside would be difficult, as the previous income tax and CGT consequences would have been different if the transaction were not effective.109 The point is that there are many tax issues to consider even when setting aside just one transaction, particularly after some considerable time period has elapsed. Whilst this is not a strong reason to deny the relief on its own, this is of course a factor and added to the more forceful arguments set out above, provides an extra incentive to deny a remedy where the issue pertains to tax consequences.

107 Income tax will be calculated and payable according to income deriving from the type of asset (eg interest on cash in a bank account, dividends from shares, profits from a business, etc) and so either under the Income Tax (Earnings and Pensions) Act 2003, the Income (Trading and Other Income) Act 2005, or the Income Tax Act 2007. 108 Rates of CGT used to vary as between trustees and individuals, from 20% to 40%. Now the lowest rate an individual can pay is 18% and the highest is 28% – trustees will always pay a CGT rate of 28%. 109 Hastings-Bass (n 5) 38.

9 Judicial Techniques in Relation to Remedies for Overpaid Tax CATHERINE BARNARD AND JULIAN GHOSH QC

A. INTRODUCTION

Orthodoxy requires taxpayers who wish to establish the scope and substance of European Union (EU) rights, or to enforce such rights, to rely on ‘direct effect’, and failing which to seek to have the national courts construe the relevant domestic legislation sympathetically so as to be Union-compliant. If sympathetic construction is not available, the taxpayers complaining of a breach of Union rights seek to have non-compliant provisions disapplied against them, failing which they must fall back onto either restitutionary remedies or a claim for damages (alleging non-contractual liability on the part of the offending Member State).1 The limitations of claims in restitution and damages (recovery in restitution limited to monies paid as an ‘inevitable consequence’ of the wrongly levied tax and the need for a ‘sufficiently serious breach’ for damages claims) are explored elsewhere.2 We merely observe that the limitations in question, being themselves notions of Union law, are at least presumptively consistent with general Union law principles, including effectiveness. This chapter considers the notion of sympathetic construction and disapplication of non-EU compliant provisions in the context of statutory remedies to recover overpaid tax. We conclude that the principles adopted by the English courts (all of the decisions so far have been passed down in England) differ in material ways from the principles set out by the European Court of Justice. Furthermore, it seems that the Court of Justice (CJEU, formerly ECJ) has recently ventilated a new and far more robust principle which imposes a general duty on the national courts to provide a remedy for the breach of Union rights in all circumstances, which goes well beyond the notions and limitations of either sympathetic construction or disapplication of non-EU compliant domestic provisions.

1 This seems to be the order of remedial action as a matter of the approach of the Court of Justice of the European Union (hereafter ‘CJEU’): see Case C-91/92 Faccini Dori v Recreb Srl [1994] ECR I-3325. 2 See especially Maximilian Schlote’s contribution in chapter 10 of the present volume (at pp 204–05).

170 Catherine Barnard and Julian Ghosh QC

B. SYMPATHETIC CONSTRUCTION

(1) The Approach of the Court of Justice So far as the Court of Justice is concerned, the notion of ‘sympathetic construction’ has been held to have its source in, variously, the ‘loyalty clause’,3 the general principle of effectiveness4 and the presumption of compliance of Member States with their Union obligations5. The duty is a strong one, requiring the national court to construe domestic provisions ‘so far as possible in such a way that they are applied in conformity with the objective of [EU law]’,6 with the national court being required to do ‘whatever lies within its jurisdiction’ in the exercise.7 Put shortly, ‘the national court is bound to interpret national law, so far as possible, in the light of the wording and the purpose of [the relevant Union legislation] concerned in order to achieve the result sought by [that legislation]’.8 This obligation applies to all domestic legislation, whether specifically implementing Union law or not, (and whether that domestic legislation predates particular Union law by reference to which it is construed or not9). The Court of Justice is quite clear that the national court, in its exercise of construing sympathetically, must consider the ‘whole national law’ and it is that ‘national law as a whole’ which must not produce a result contrary to the objectives of Union law.10 The jurisdiction to construe sympathetically is, of course, that of the national court. The Court of Justice has no jurisdiction to construe Member States’ national laws.11 Despite the powerful rhetoric of the Court of Justice in imposing this duty on national courts, it is tolerably clear that the Court expects the national courts of the respective Member States to adopt only the judicial techniques of construction appropriate to their particular jurisdiction.12 In other words, each national court’s duty is to apply the interpretative techniques which have evolved within the jurisdiction of the Member State, the legal system of which that particular national court forms part, albeit that Pfeiffer indicates that national courts must use all the tools at their disposal to do this.13 It follows that the exer3 Requiring ‘Member States [to] take all appropriate measures . . . to ensure fulfilment of the obligations arising out of [the EC Treaty]’: Article 10 EC, now Article 4(3) TEU. cf Case C-106/89 Marleasing SA v La Comercial Internacional de Alimentacion SA [1991] ECR I-4135 (hereafter ‘Marleasing’) para 8. 4 See Joined Cases C-397/01 to C-403/01 Pfeiffer v Deutsches Rotes Kreuz, Kreisverband Waldshut eV [2004] ECR I-8835 (hereafter ‘Pfeiffer’) paras 110–11, 114. 5 See Pfeiffer (n 4) para 112. 6 ibid para 117. 7 ibid para 119. 8 ibid para 113. In fact, Pfeiffer refers specifically to Directives, but as is made clear below, the duty to construe sympathetically extends to construing the whole of any Member State’s national law sympathetically by reference to all Union legislation. 9 See Marleasing (n 3) para 8. The position in the English courts is the same, although there are certain observations to the contrary. 10 Case C-105/03 Criminal proceedings against Maria Pupino [2005] ECR I-5285 (hereafter ‘Pupino’) para 47; Pfeiffer (n 4) para 115. 11 See HM Revenue & Customs v IDT Card Services Ireland Ltd [2006] STC 1252 (CA) (hereafter ‘IDT’) [68], [81] (Arden LJ). 12 Including anti-avoidance case law: see the AG’s Opinion in Case C-196/04 Cadbury Schweppes plc v Inland Revenue Commissioners [2006] ECR I-7995 (hereafter ‘Cadbury Schweppes’) fn 68, which expressly observes that UK national courts may apply the so-called ‘Ramsay doctrine’ in construing Treaty provisions in an abuse case. 13 See Pfeiffer (n 4) para 116.

Judicial Techniques 171

cise of construing sympathetically is subject to limits. First, a sympathetic construction of domestic legislation cannot offend legal certainty, so as to disturb either consistency or predictability of outcome.14 Thus domestic provisions cannot be construed contra legem: ‘[W]ords which do not require interpretation because they are perfectly clear, should not be distorted under pretence of interpretation’.15 The Court is quite clear that a point may be reached ‘when [national law] cannot receive an application which would lead to a result compatible with that envisaged by [a Member State’s Union obligations]’.16 Secondly, sympathetic construction should not lead to retroactivity. Importantly, the Court seems to have set its face against retroactivity generally (and not simply a retroactive aggravation of criminal liabilities or penalties).17 This is a slightly odd limitation on the notion of sympathetic construction since the very notion of construction (whether or not it leads to surprising results) is retrospective as a matter of principle: the construction of any provision reveals its meaning in a Platonic sense, and the fact that the construction of a domestic provision produces a backwards looking effect is, it is thought, neither here nor there.

(2) The Approach of the English Courts The principles of sympathetic construction have been often summarised in a deceptively neat package in the English case law.18 However, an unpacking of the English approach betrays a technique materially different from that articulated by the Court of Justice, at least prior to the decision of the Supreme Court in the Franked Investment Income (FII) Group Litigation.19 The approach of the English Courts, prior to FII (Supreme Court), went far beyond the principles articulated by the CJEU and indeed beyond previous guidance given by the House of Lords in Ghaidan v Godin-Mendoza.20 FII (Supreme Court) has made a welcome return to a principled application of the canons of sympathetic construction, but the English domestic case-law prior to FII (Supreme Court) is highly instructive in illustrating the dangers of an over-expansive approach to such an exercise. In English law, the duty to construe sympathetically as a function of the principle of effectiveness has been recognised,21 as has the source of the presumption of Member State 14 Pupino (n 10) para 44. The notion of legal certainty as a synonym for consistency and predictability of outcome is trite: see only IDT (n 11) [110]. 15 Case C-136/96 Scotch Whisky Association v Compagnie financière européenne de Financière Européenne de Prises de Participation (Cofepp)and Centrale d’Achats et de Services Alimentaires SARL (Casal) [1998] ECR I-4571 (hereafter ‘Scotch Whisky Association’), para 18 of the AG’s Opinion, endorsed by the Court at para 33. 16 Pupino (n 10) para 47. 17 Pupino (n 10) paras 44–45, where the Court seems to expressly condemn retroactive sympathetic construction generally and retroactive aggravation of criminal penalties only as an (albeit express) afterthought. 18 See eg Vodafone 2 v HM Revenue & Customs (No 2) [2009] STC 1480 (CA) (hereafter ‘Vodafone 2’) [37]–[38], which sets out the principles as imposing a requirement to go beyond conventional canons of construction, demanding sympathetic construction whether or not the particular domestic provisions under scrutiny are ambiguous, accepting that the exercise is not a semantic one, going beyond a literal construction, permitting the reading in of words required to make the particular provisions in question Union-compliant, without needing a precise linguistic formulation of the words to be read in, with the limitations that the process cannot go against the grain of the legislation or impute to the courts a duty for which they are not equipped. 19 Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2012] UKSC 19, [2012] 2 AC 337 (hereafter ‘FII (Supreme Court)’ or ‘FII (SC)’). 20 Ghaidan v Godin-Mendoza [2004] UKHL 30, [2004] 2 AC 557 (hereafter ‘Ghaidan’) [30]–[33] (Lord Nicholls), set out in the text accompanying the nn 28–29. 21 John Wilkins (Motor Engineers) Ltd v HM Revenue & Customs [2009] UKUT 175 (TCC), [2009] STC 2485 (hereafter ‘Wilkins’) [114].

172 Catherine Barnard and Julian Ghosh QC compliance with Union obligations.22 However, critically, the United Kingdom introduced a legislative basis upon which to import Union obligations into domestic law, by way of the European Communities Act 1972, Section 2(4), which provides that ‘any enactment passed or to be passed . . . shall be construed and have effect subject to’ the United Kingdom’s Treaty obligations incorporated into UK law by Section 2(1) and any implementing legislation made under Section 2(2).23 Thus all United Kingdom statutory provisions (whenever enacted) are made subject to a legislative trumping that they are to have effect ‘without prejudice to the directly enforceable Union rights of nationals of any member state of [the Union]’.24 Put another way, Parliamentary intention is legislatively (and retrospectively) modified so as to accommodate a respect for Union rights. Thus judicial observations that Union-compliant constructions of domestic provisions may well not encapsulate Parliamentary intention at the time of their enactment are unsurprising.25 Furthermore, section 2(4), by introducing a legislative mechanism to accommodate Union rights into United Kingdom law,26 leads to an immediate divergence from the Court of Justice’s approach to the reconciliation of domestic law and Union obligations due to its retrospective nature,27 but the writers do not consider this to be a vice, since had the courts been left to construe national legislation without an express modification of Parliamentary intention, that very process of construction should by its nature be retrospective (see above). It is worth setting out the locus classicus of the English approach to sympathetic construction laid down in Ghaidan v Godin-Mendoza, albeit in the context of section 3 Human Rights Act 1998, of which section 2(4) ECA 1972 is an analogue:28 [30] From this it follows that the interpretative obligation decreed by [Human Rights Act 1998, section 3] is of an unusual and far-reaching character. Section 3 may require a court to depart from the unambiguous meaning the legislation would otherwise bear. In the ordinary course the interpretation of legislation involves seeking the intention reasonably to be attributed to Parliament in using the language in question. Section 3 may require the court to depart from this legislative intention, that is, depart from the intention of the Parliament which enacted the legislation. The question of difficulty is how far, and in what circumstances, Section 3 requires a court to depart from the intention of the enacting Parliament. The answer to this question depends on the intention reasonably to be attributed to Parliament in enacting Section 3. [31] On this first point to be considered is how far, when enacting Section 3, Parliament intended that the actual language of the statute, as distinct from the concept expressed in that statute, should be determinative. Since Section 3 relates to the ‘interpretation’ of legislation, it is natural to focus attention initially on the language used in the legislative provision being considered. But once it is accepted that Section 3 may require legislation to bear a meaning which departs from the unambiguous meaning the legislation would otherwise bear, it becomes impossible to sup22 IDT (n 11) [261]; Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2010] EWCA Civ 103 (hereafter ‘FII (Court of Appeal)’ or ‘FII (CA)’) [108], which proposition was not at all disturbed by the Supreme Court in FII (SC) (n 19). 23 cf also Wilkins (n 21) [100]; IDT (n 11). 24 R v Secretary of State for Transport, ex parte Factortame Ltd [1990] 2 AC 85 (HL) 140 (Lord Bridge). 25 See Vodafone 2 (n 18) [38]; FII (CA) (n 22) [108]. 26 The European Communities Act 1972 recognised the duty of sympathetic construction some 20 years before the obligation was articulated in Pfeiffer (n 4). 27 Recognised in Vodafone 2 (n 18) [56]. 28 Section 3 of the Human Rights Act 1998 provides that primary and subordinate legislation must be read and given effect in a way which is compatible with Convention rights, which is clearly analogous to section 2(4) of the 1972 Act – and recognised to be such, so the same principles of construction apply: see eg IDT (n 11) [92] (Arden LJ); FII (SC) (n 19) [118] (Lord Walker).

Judicial Techniques 173 pose that Parliament intended that the operation of Section 3 should depend critically upon the particular form of words adopted by the Parliamentary draftsman in the statutory provision under consideration. That would make the application of Section 3 something of a semantic lottery. If the draftsman chose to express the concept being enacted in one form of words, Section 3 would be available to achieve convention-compliance. If he chose a different form of words, Section 3 would be impotent. [32] From this the conclusion which seems inescapable is that the mere fact the language under consideration is inconsistent with the convention-compliant meaning does not of itself make a convention-compliant interpretation under Section 3 impossible. Section 3 enables language to be interpreted restrictively or expansively. But Section 3 goes further than this. It is also apt to require a court to read in words which change the meaning of the enacted legislation, so as to make it convention-compliant. In other words, the intention of Parliament in enacting Section 3 was that, to an extent bounded only by what is ‘possible’, a court can modify the meaning, and hence the effect, of primary and secondary legislation. [33] Parliament, however, cannot have intended that in the discharge of this extended interpretative function the courts should adopt a meaning inconsistent with the fundamental feature of the legislation. That would be to cross the constitutional boundary Section 3 seeks to demarcate and preserve. Parliament has retained the right to enact the legislation in terms which are not convention-compliant. The meaning imported by application of Section 3 must be compatible with the underlying thrust of the legislation being construed. Words implied must, in the phrase of my noble and learned friend, Lord Roger of Earlsferry, ‘go with the grain of the legislation’. Nor can Parliament have intended that Section 3 should require the courts to make decisions for which they are not equipped. There may be several ways of making a provision convention-compliant, and the choice may involve issues calling for legislative deliberation.29

The terms of section 2(4) make it wholly unsurprising for the process of applying domestic provisions in the context of Union rights to be described as ‘not constrained by conventional rules of construction’,30 and the exercise of construing sympathetically as being not a ‘natural’ exercise in statutory construction.31 Thus section 2(4) requires an exercise of ‘requiring the court[s] to go beyond what could be done by way of statutory interpretation where no question of [EU] law or human rights’ arises.32 In the latter (ie non-EU, nonhuman rights) case, ‘[t]he courts exercise considerable caution before adding or omitting or substituting words’.33 In Inco Europe (a non-EU, non-human rights case), Lord Nicholls said that before interpreting a statute in this way the court must be abundantly sure of three matters: (1) the intended purpose of the statute or provision in question; (2) that by inadvertence the draftsman and Parliament had failed to give effect to that purpose and the provision in question; and (3) the substance of the provision Parliament would have made, although not necessarily the precise words Parliament would have used, had the error in the Bill been noticed.34

29 Ghaidan (n 20) [30]–[33] (Lord Nicholls). See also R v A (No 2) [2002] 1 AC 45, where the House of Lords adopts the same approach. 30 Vodafone 2 (n 18) [37]. 31 IDT (n 11) [82]. 32 FII (CA) (n 22) [260]. 33 Inco Europe Ltd v First Choice Distribution (a firm) [2000] 1 WLR 586 (HL) (hereafter ‘Inco Europe’) 592 (Lord Nicholls). Similar limitations were expressed to apply by the Court of Appeal in the context of fiscal legislation: O’Rourke (HM Inspector of Taxes) v Binks [1992] STC 710, 713 ( Scott LJ), reading the word ‘small’ to qualify the term ‘distribution’. 34 Inco Europe (n 33) 592.

174 Catherine Barnard and Julian Ghosh QC Lord Nicholls continued that sometimes, even when these conditions are met, ‘the court may find itself inhibited from interpreting the statutory provision in accordance with what it is satisfied was the underlying intention of Parliament. The alteration in language may be too far-reaching’.35 He noted that in Western Bank Ltd v Schindler 36 Scarman LJ ‘observed that the insertion must not be too big, or too much in variance with the language used by the legislature’.37 Judicial observations that a Union-compliant construction of domestic provisions permits a departure from the literal meaning of the words of the provision38 and furthermore permits a gloss to be put on the language of domestic provisions even though they may not be ambiguous39 are not earth-shattering.40 Neither is it surprising that the English courts have adopted and approved the techniques of construction such as reading words restrictively, expansively or reading words in. So a provision which restricted tax credits to dividends paid by ‘a company resident in the United Kingdom’41 to a UK resident person (company or non-corporate) could be sympathetically construed so as to confer tax credits to ‘a [recipient] company resident in the United Kingdom [and also to other persons entitled to claim a credit by Union law to the extent that they are so entitled]’.42 And a provision which prohibited relief for an error or mistake in a taxpayer’s tax return if the return was made in accordance with generally prevailing practice at the time when the return was made,43 could be read as subject to the limitation that the restriction on relief by reference to generally prevailing practice applies ‘if and only to the extent that the United Kingdom can impose such a restriction consistent with its [Union] obligations’.44 The words to be read in do not need to be precisely formulated.45 The technique adopted by the Court of Appeal in FII in sympathetically construing section 231(1) of the Income and Corporation Taxes Act 1988 and section 33(2A) of the Taxes Management Act 1970 certainly went beyond the conventional limitations imposed on the courts in reading words into a statute when Union rights or human rights are not involved. In FII, a tax credit was available under the Income and Corporation Taxes Act 1988, section 231(1), to a UK resident company which received a dividend from a UK resident paying company. The paying company had to pay Advance Corporation Tax (‘ACT’) on paying the dividend. The recipient company would, without the benefit of section 231(1), have to pay ACT on any onwards dividend of the amount paid to it. Section 231(1) provided a credit to the recipient company which acknowledged that because the distributing ibid. See Western Bank Ltd v Schindler [1977] Ch 1 (CA) 18. 37 Inco Europe (n 33) 592. 38 Ghaidan (n 20) [30] (Lord Nicholls). 39 IDT (n 11) [97]; Vodafone 2 (n 18) [37]; Ghaidan (n 20) [32] (Lord Nicholls). The authority for such a provision is now some 20 years old: Pickstone v Freemans plc [1988] 2 All ER 803 (HL) (hereafter ‘Pickstone’) 817; Litster v Forth Dry Dock and Engineering Company Ltd [1989] 1 All ER 1134 (HL) 1138 (Lord Oliver). 40 Indeed, similar principles apply in relation to the construction of domestic provisions even where no Union rights are in play. This observation hardly needs authority in the tax world, but see eg Jenks v Dickinson [1997] STC 853. 41 Income and Corporation Taxes Act 1988, s 231(1). 42 FII (CA) (n 22) [107]. 43 Taxes Management Act 1970, s 33(2A). A return made in compliance with UK tax law but which reflected a liability to tax in breach of Union law would be made in accordance with generally prevailing practice and thus without more could therefore not be modified and attract relief under s 33. 44 FII (CA) (n 22) [261] (Arden LJ). 45 Vodafone 2 (n 18) [37]; Pickstone (n 39) 807 (Lord Keith); Ghaidan (n 20) [122] (Lord Rodger); IDT (n 11) [114] (Arden LJ). 35 36

Judicial Techniques 175

company had paid ACT on its dividend (to the UK fisc) the recipient company need not pay ACT on any onwards dividend of the amount it had received. The UK had received ACT once and it was equitable for the UK to forgo ACT on that further onwards dividend. But of course if the distributing company was outside the UK corporation tax charge (because it was not UK resident), then the rationale for giving the UK-resident recipient a tax credit against its prima face liability to ACT on an onwards dividend disappears (since the UK would never have had an ACT payment). The Court of Justice had held that the restriction on tax credits to distributions paid by UK resident companies interfered with the freedom of establishment and the free movement of capital.46 The Court observed that in circumstances where the distributing company had paid corporation tax in its own jurisdiction within the Union (albeit not ACT as such), it was an unjustified impediment to the internal market to restrict tax credits to distributions paid by UK tax resident companies. That is unexceptionable as a matter of Union law: an example of the free movement principles enshrined in the TFEU trumping conventional fiscal thinking as a function of the supremacy of European Union Law. But for the aggrieved applicant to persuade the Court of Appeal that the statutory modification to be made to section 231(1) so as to extend the class of recipients entitled to a tax credit to those receiving distributions from companies tax resident not in the UK but elsewhere in the Union, required an extension which went well beyond the limits of what a court is permitted to read into a statutory provision as expressed in Inco Europe in the context of domestic legislation where Union rights are not relevant.47 Parliament intended to limit tax credits to distributions received from UK tax resident companies. The intention (and policy rationale) of the draftsman could hardly have been clearer. However, that clear intention as expressed on the face of section 231(1) was modified by section 2(4) of the 1972 Act and thus made it possible, according to the Court of Appeal, to read in the words they did. For the present, we merely observe that FII (Court of Appeal)48 is an illustration of the duty to construe sympathetically going well beyond conventional rules of construction. Similar observations apply to the words read into section 33(2A) of the 1970 Act which was also at issue in FII (Court of Appeal).49 Parliament clearly intended that relief for overpaid tax on a mistaken return ought not to be available where the overpayment was in line with generally prevailing practice (including practice biased on clear UK tax law, albeit in circumstances where the latter breached Union law). The substance of section 33(2A) was fundamentally changed in this case in a manner well outside the confines of the Inco Europe approach. This analysis of FII (Court of Appeal) shows that the terms of section 2(4) of the 1972 Act drive the English courts to conduct carpentry upon UK national law well beyond that envisaged by the Court of Justice, which only required domestic techniques of construction (albeit as many of them as possible) to be applied. The Supreme Court in FII 50 did not deal with the issue of the Union-compliant construction of section 231, as this issue was deferred until the Supreme Court had received guidance from the CJEU on certain other issues which were made the subject of a second 46 Case C-446/04 Test Claimants in the Franked Investment Income Group Litigation v Inland Revenue Commissioners [2006] ECR I-11753 (hereafter ‘FII (CJEU)’) paras 79 –98. 47 In fact Inco Europe (n 33) concerned the Arbitration Act 1980. 48 FII (CA) (n 22). 49 ibid. 50 FII (SC) (n 19).

176 Catherine Barnard and Julian Ghosh QC reference. However, there was no express endorsement of the Court of Appeal’s approach; merely recognition that this approach had to be assumed to be correct until the CJEU had delivered its further guidance and the construction of section 231 re-emerged as an issue to be considered by the Supreme Court.51 However, as we observe below, the Supreme Court made material (and, in our view, welcome) qualifications to the expansive approach of the Court of Appeal in FII in relation to section 33(2A). We consider the Supreme Court’s approach in FII and the extent to which it overturns the Court of Appeal’s approach (and indeed the approach in other pre-FII case law), in detail below. So far, we have shown that the terms of section 2(4) of the 1972 Act have provided for a more expansive approach by the English courts, at least prior to FII (Supreme Court), to sympathetic construction than that required by the Court of Justice in Pfeiffer. Moreover, the limitations imposed on sympathetic construction by the Court of Justice – by reference to legal certainty and contra legem which are based on the limitations of language – simply did not apply in the pre-FII English approach. As we have already seen, Ghaidan sets out the judicial techniques available to a national court to achieve Union-compliant construction of a domestic legislative provision.52 As it happens, even the Ghaidan limitation (that the construction must not be against the ‘grain of the legislation [being construed]’)53 is conceptually very different from the linguistic contra legem limitation described by the Court of Justice in Pupino.54 While the limit in Pupino is predicated on the notion of legal certainty, the limitation set out in Ghaidan is a constitutional one: one which recognises the limits of judicial intervention. The terms of Section 3 of the Human Rights Act 1998 (and indeed Section 2(4) of the European Communities Act 1972) makes this approach wholly legitimate. Parliament has intended that all domestic legislation is ‘subject to’ the United Kingdom’s European Union obligations (and the European Convention of Human Rights). But it is clear that Parliament may choose to breach either its Union obligations or its ECHR obligations and enact statutes which expressly do exactly that. Thus, the application of section 2(4) by the courts, which makes all statutes ‘subject to’ the United Kingdom’s European Union obligations, is constrained by the courts’ obligation not to interfere with the ‘cardinal features’ of the legislation under scrutiny. Thus the trumping process, which must be administered by the courts, must be justiciable. Furthermore, the application of section 2(4) by the courts must be justiciable. So in Fleming t/a Bodycraft the House of Lords, in striking down a provision which sought to reduce a limitation period in recovering overpaid VAT without an adequate transitional period (breaching the legitimate expectations of those VAT traders who had relied on the old, longer period), refused to ‘construe’ into the relevant provisions a replacement limitation period.55 This was because, in relation to justiciability, the courts cannot arrogate to themselves a function which they cannot fulfil. If the courts are faced with depriving a legislative provision of fundamental effect, the courts, according to Lord Nicholls and Lord Roger in Ghaidan, cross an impermissible constitutional boundary and undertake tasks for which they are not equipped.56 That is a very different and more distant limitation than one of linguistic contra legem. ibid [83] (Lord Walker), [175] (Lord Sumption). See Ghaidan (n 20) [30]–[32], set out in the text accompanying nn 28 –29. 53 ibid [33]. 54 Pupino (n 10) paras 44 – 47. 55 Fleming (trading as Bodycraft) v HM Revenue & Customs [2008] STC 324 (hereafter ‘Fleming’) [10] (Lord Hope), [21] (Lord Scott). 56 Ghaidan (n 20) [33] (Lord Nicholls), [108]–[116] (Lord Rodger]. 51 52

Judicial Techniques 177

C. THREE CASE STUDIES: FII, WILKINS AND IDT

It is instructive to consider further the decision of the Court of Appeal in FII 57, together with another important decision, that of the Upper Tribunal (Tax Chamber) in Wilkins,58 to test how the principles of interpretation have been applied in practice prior to the Supreme Court’s decision in FII in the field of remedies for the recovery of overpaid tax, and then to compare and contrast the approach of the Supreme Court in FII 59 to discover the extent to which the pre-FII (Supreme Court) approach is good in law as a matter of principle and authority.

(1) FII (Court of Appeal) As we have already observed, in FII the technique adopted by the Court of Appeal illustrates how much further the court went in making section 231(1) of the Income and Corporation Taxes Act 1988 Union-compliant, than would have been permitted by Lord Nicholls in Inco Europe (since, if nothing else, the words read in were larger than those envisaged by Scarman LJ).60 However a more fundamental point arises. We have already said that it made perfect sense, as a matter of tax theory, to restrict the tax credit (whether the distribution was received by a UK resident company or non-corporate person) to circumstances where the paying company was itself UK resident. By contrast, the Court of Justice found there to be a breach of Union law which involved different considerations, namely that it was precisely this fundamental feature of UK tax law which only gave a tax credit to a UK tax resident recipient of a dividend paid by a UK tax resident company which interfered with Union trade law (the freedom of establishment and the free movement of capital), since it discouraged the intra-EU cross-border establishment of subsidiaries and the acquisition of shares in non-UK resident companies as part of an investment portfolio.61 The decision of the Court of Justice does not take away from the fact that it is a fundamental feature (‘the grain of the legislation’) of section 231(1) that the distributing company must be UK tax resident in order for the recipient to receive a tax credit. It is difficult to see why the exercise of sympathetic construction should permit the reading in of the words which the Court of Appeal felt able to do. In other words, it is difficult to see how the ‘grain of the legislation’ did not apply to prevent a sympathetic construction of section 231(1) in FII (Court of Appeal). The Court of Appeal observed that the interpretation which read in the words set out above ‘does not cut across “the grain” or cardinal features of Section 231’62 without addressing this point. It is not as if the absence of an ability to sympathetically construe section 231 would have left the appellant without a remedy (subject to limitation periods etc). The appellant would, of course, have had the right to bring an action in damages, albeit subject FII (CA) (n 22). Wilkins (21). 59 FII (SC) (n19). 60 See the text accompanying n 47. 61 FII (CJEU) (n 46). 62 FII (CA) (n 22) [108] (Arden LJ). 57 58

178 Catherine Barnard and Julian Ghosh QC to establishing a ‘sufficiently serious’ breach by the United Kingdom (which the Court of Justice had heavily indicated was not present in FII).63 As we have already observed, the Supreme Court has deferred consideration of the construction of section 231 until the CJEU gives guidance on the second reference.64 In relation to Section 33(2A)(a) of the Taxes Management Act 1970, the Court of Appeal observed that to deny relief on the basis that the taxpayer had made an error or mistake in a tax return (because the UK domestic tax provisions could not comply with the United Kingdom’s Union obligations, which, in turn, meant that any return would, of course, have been based on ‘generally prevailing practice’) could be read as subject to the limitation that it applies only if and to the extent that the United Kingdom can impose a restriction consistently with its Treaty obligations.65 Again, the extension and substance of the disapplication of Section 33(2A) goes far beyond the approach permitted by Inco Europe 66 for domestic construction. More relevantly, again, it is difficult to see why this approach does not ‘cut across the grain’ of section 33(2A). Indeed, to construe section 33(2A) so as to make it Union-compliant with such radical linguistic carpentry meant that the appellant in FII was denied a remedy (on the basis that the tax code contained in the Taxes Management Act 1970 did in fact permit a remedy which, in turn, meant that the appellant could not go outside that exclusive tax code to find another remedy).67 Here the Supreme Court was clear in its rejection of the expansive approach of the Court of Appeal. Lord Walker observed that I have grave doubts as to whether that interpretation does not go against the grain of the legislation, since the ‘practice generally prevailing’ condition is of long standing and has always been regarded as an important safeguard for the public revenue. I am inclined to think that Mr Aaronson was right . . . to call it a ‘cardinal feature’ of the legislation.68

Lord Sumption also critcised ‘such radical surgery’ by the Court of Appeal.69 And, as for the effect of the Court of Appeal’s decision on the remedies available to the taxpayer, Lord Sumption observed: [I]t seems, with respect, eccentric to imply an ambit for section 33 which is inconsistent with EU Law and then to torture the express provisions so as to deal with anomalies that, but for the implication, would never have arisen.70

Although the analysis is compressed, it seems clear that the Ghaidan ‘against the grain’ limitation is being applied expressly more strictly than by the Court of Appeal and by refer63 FII (CJEU) (n 46) paras 212–17. As to whether that would have left the applicants in a satisfactory position with adequate protection of their Union rights is – of course – a separate question. 64 See n 51 and accompanying text. 65 FII (CA) (n 22) [261] (Arden LJ). 66 Inco Europe (n 33). 67 The Court of Appeal also sought to sympathetically construe the non-statutory right to reclaim wrongly paid tax under the principle in Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] AC 70 (HL), which exercise deserves separate analysis in another paper. We merely note here that non-statutory remedies may well be outside the scope of s 2(4) of the European Communities Act 1972 (not being ‘enactments’), which would subject such non-statutory remedies to the approach of the Court of Justice in Pfeiffer (n 4), unqualified by s 2(4). 68 FII (SC) (n 19) [119], having acknowledged (at [118]) that Ghaidan (n 20) provided the appropriate test. In fact, Lord Walker provided an elegant alternative Marleasing (cf n 3) Union-compliant construction of s 33, which was that s 33 could not be considered to be impliedly an exclusive remedy. 69 FII (SC) (n 19) [205]. 70 ibid.

Judicial Techniques 179

ence to the text and purpose of the particular domestic statute under scrutiny, ie section 33(2A). Thus, the Supreme Court’s approach restores the contra legem limitation laid down by the CJEU, at least to an extent, and reinforces the ‘against the grain’ Ghaidan limitation, which in turn promotes legal certainty.

(2) Wilkins The Upper Tribunal’s approach in Wilkins 71 can be criticised for different reasons: that it did not comply with either the Court of Justice’s approach in Pfeiffer nor with the United Kingdom’s approach in Ghaidan. In Wilkins the question was whether the term ‘interest’ in section 78 of the Value Added Tax Act 1994 could be construed as providing for compound interest. Section 78(1) provided that ‘interest’ was to be paid in circumstances in which output VAT had been overpaid to the UK Customs & Excise Commissioners. Section 78(3) provided that ‘interest . . . shall be payable at the rate applicable under Section 197 of the Finance Act 1996’, which in turn provided for the rate of interest to be specified by subordinate legislation. The Upper Tribunal observed that as a matter of conventional construction, the term ‘interest’ for section 78 purposes meant simple, not compound, interest, since compound interest depended on interest being compounded over a number of periods and no part of the relevant legislative code mentioned the frequency of rests.72 Wilkins echoed the observation in Pfeiffer that it is the application ‘of interpretative methods recognised by national law’ which the Court of Justice required to be used.73 However, the Upper Tier Tribunal observed that the doctrine of sympathetic construction ‘has nothing to say about the interpretation of the national legislative provision outside the context of the implementation of [EU] legislation’.74 In other words, the Tribunal suggested that the duty to conform sympathetically to Union obligations applied only to domestic provisions which specifically implemented Union legislation. It is submitted that that is simply wrong. First, the observations of the Court refer, as has already been mentioned, to ‘national law as a whole’.75 Indeed, since the Court of Justice has expressly observed that the duty to conform sympathetically applies to domestic legislation which pre-dated a Directive,76 which pre-dating legislation could not possibly have anything to do with a subsequently enacted Directive that nevertheless affects the construction of the prior domestic legislation. It is also inconsistent with the uncontroversial proposition that a taxpayer is entitled to the most favourable of available domestic remedies to enforce Union rights, which was expressly re-articulated and approved by the Supreme Court in in FII.77 Where a taxpayer is entitled to the most favourable of all available domestic remedies, all of those available domestic remedies must, if they are to enforce Union rights, themselves be Union-compliant (and be construed, in so far as possible, to be so). It follows that this observation by the Tribunal in Wilkins is unsustainable, certainly in the Wilkins (n 21). ibid [99]. 73 ibid [116], echoing the observations of the Court of Justice in Pfeiffer (n 4) para 116 (in fact citing Arden LJ in IDT (n 11) [116]). 74 Wilkins (n 21) [104], so that the question of the extent of any duty to construe sympathetically to nonimplementing national provisions was ‘an open question’: ibid [105]. 75 See the observations made in Pupino (n 10) para 47; Pfeiffer (n 4) para 115. 76 Marleasing (n 3) para 8. 77 FII (SC) (n 19) [13], [20]–[23] (Lord Hope), [41] (Lord Walker). 71 72

180 Catherine Barnard and Julian Ghosh QC light of FII (Supreme Court). Indeed, the Court of Appeal in Vodafone 2 78 afforded the controlled foreign companies provisions in the Income and Corporation Taxes Act 1988 a sympathetic construction by reference to the freedom of establishment provisions in the TFEU (rather than a Directive), where the controlled foreign company provisions were enacted with no thought to the United Kingdom’s Union obligations at all.79 Further, the Upper Tier Tribunal observed that to confer compound interest would go against the grain of the relevant legislation.80 The statutory scheme, said the Tribunal, was one of simplicity, which specified the relevant rates of interest payable (laid down in subordinate legislation). No ‘rests’ were identified by which compound interest could be calculated. The Tribunal held that to construe section 78 to allow for compound interest would arrogate a function and jurisdiction which the Tribunal did not have. The Tribunal in Wilkins also made a second, separate, observation on the nature of the duty to construe sympathetically. The Tribunal observed that the duty had no narrow focus on domestic implementing legislation . . . rather than national law as a whole. Accordingly, if domestic law makes provision somewhere for that which is required by the directive, it does not matter that it cannot be found in the particular piece of domestic legislation which is enacted to implement the directive.81

The Tribunal’s rejection of a ‘narrow focus’ on domestic implementing legislation is not at odds with its conclusion that the duty to construe sympathetically may not extend to national non-implementing legislation (albeit that we disagree with that conclusion). Rather the Tribunal here makes a separate point that the principle of effectiveness may not be breached even if an adequate remedy is not contained in national legislation which implements Union obligations (here under a Directive), if an adequate remedy is contained elsewhere in national law. It follows, according to the Tribunal, that in such a case where the principle of effectiveness is not breached (because of an effective remedy in national law somewhere, albeit not within the implementing legislation) the duty to construe sympathetically is not engaged at all.82 This analysis is – we consider – also contrary to principle and authority. In principle, the very notion of statutory construction of national provisions, in particular implementing provisions, assumes that such provisions have an inherent meaning which is consistent with Member States’ Union obligations. That Union-compliant meaning is not to be switched on and off depending on whether a Member States’ Union obligations are saved from breach by reason of other national law provisions. Put another way, the duty to construe sympathetically is not exclusively predicated on a potential breach of effectiveness. We have already observed that this duty arises from a general presumption of Member State compliance with Union obligations (from which follows a presumption that all national provisions and implementing provisions in particular have a Union-compliant Vodafone 2 (n 18). cf Cadbury Schweppes (n 12). The same is true of all of the case law involving the United Kingdoms’ direct tax law. Indeed the Upper Tier Tribunal refers to the Court of Appeal’s decision in Vodafone 2 (see Wilkins (n 21) [117]) but distinguishes that decision on the basis that, without a conforming construction, Vodafone 2 would have been left without an adequate remedy. The Court of Appeal in IDT (n 11) [91] considered this question to be open, but after Vodafone 2 the question seems now to be firmly resolved as a matter of English law. 80 Wilkins (n 21) [120]. 81 ibid [105]. 82 ibid [106], which makes it clear that the Tribunal founded its conclusion that the duty of sympathetic construction was not engaged in Wilkins for both of these reasons. 78 79

Judicial Techniques 181

meaning). Furthermore, this presumption finds specific expression in the loyalty clause. The breach or otherwise of the principle of effectiveness has nothing to say about the duty to construe sympathetically. As for authority, the rationale of the duty of sympathetic construction has been expressly explained in terms of this general presumption of compliance with Union obligations and the loyalty clause: nowhere has the Court of Justice suggested that the duty is dependent on a breach of the principle of effectiveness.83 Furthermore, this analysis of the Tribunal is also inconsistent with the proposition that a taxpayer may select the most favourable of all available domestic remedies to enforce Union rights. As it happens, so far as the specific issue in Wilkins is concerned, we have considerable sympathy for the proposition that, where rates are specified, it ‘goes against the grain’ of the legislation to specify a different rate. But as a matter of the degree of the distortion of language inflicted on the provision is concerned, it does not go against the grain any more than the extension of section 231 of the Income and Corporation Taxes Act 1988 to distributions from non-UK resident companies. Nor does it go against the grain any more than disapplying the exception in section 33(2A) of Taxes Management Act 1970 to circumstances where they would be otherwise in breach of Union law. Words such as ‘or alternatively compound interest where there has been a breach of Union law’ could, it is submitted, be perfectly intelligibly read into section 78. Be that as it may, FII (Supreme Court) confirms that all domestic provisions, not just those which implement Union legislation, are subject to a duty to construe sympathetically and that he availability of a remedy to enforce Union rights in one part of domestic law does not absolve national courts to construe all other parts of domestic law sympathetically.

(3) IDT This is especially the case when one considers the approach of the Court of Appeal in IDT.84 The Court of Appeal construed a ‘disregard’ provision in section 78 the Value Added Tax Act 1994, which – read literally – would result in a transaction which the draftsman of the Sixth Directive clearly considered should be a taxable supply. The Court of Appeal read words into the relevant provisions so that it the disregard did not apply ‘where the disregard would result in the non-taxation, contrary to the objectives of the Sixth Directive . . . of a taxable transaction of goods or services in the United Kingdom’.85 The formulation of the words to be read into section 78 is, it is submitted, no more against the grain than the formulation of words read into the domestic provisions considered in IDT. What this analysis shows is that the ‘against the grain’ limitation adopted by the English courts can be troublesome in practice. It led the Court of Appeal in both FII and IDT to make radical modifications to otherwise plain provisions. It is also worth observing that in Ghaidan the conclusion of the House of Lords was that the notion of ‘husband’ or ‘wife’ extended to same-sex partners.86 While this would clearly be viewed as a radical conclusion, say, 20 years ago, as a matter of language the House of Lords’ approach undertook linguistic carpentry of a far less major nature of the relevant statutory concepts than the See esp Pfeiffer (n 4) paras 110–14. IDT (n 11). 85 ibid [114]. 86 cf Ghaidan (n 20). 83 84

182 Catherine Barnard and Julian Ghosh QC Court of Appeal in FII and IDT. It may well be that the Court of Appeal’s approach in both of these cases went beyond what the House of Lords envisaged in Ghaidan.87 And lest it be thought that this more elastic limitation to sympathetic construction is a better safeguard for Union rights than the linguistic limitation of contra legem adopted by the Court of Justice, it is worth recollecting that the expansive approach to sympathetic construction of the English Courts has produced the paradoxical result that it has actually cut down the enforcement of Union rights.88 So the approach of the Court of Appeal in IDT, which construed domestic VAT legislation by reference to the Sixth Directive, resulted in a transaction which was otherwise outside the scope of VAT (on the face of the domestic UK legislation) becoming taxable. Thus, the approach of the Court in IDT weakens, almost to vanishing point, the ‘no reverse direct effect’ doctrine, whereby a Member State cannot rely on having either failed to implement or mis-implemented a Directive. And in FII (Court of Appeal), the ‘sympathetic construction’ of section 33(2A) resulted in the applicants being theoretically afforded a statutory remedy (thereby excluding non-statutory remedies), but one which was unenforceable because the limitation period for making an ‘error or mistake’ claim under section 33 had expired. Indeed, the approach encroaches on the observations in Kolpinghuis89 that sympathetic construction should not be used to achieve what cannot be achieved through reverse direct effect. The troublesome nature of an over-robust application of the ‘grain of the legislation’ limitation is also demonstrated by its encroachment on the application of the principle of legal certainty. In IDT the Court of Appeal observed that the reading of the relevant words into domestic legislation did not offend legal certainty because, firstly, the principle of sympathetic construction was well known and, secondly, the terms of the Sixth Directive made it clear that the transactions which were the subject of scrutiny ought to be taxable.90 But the prospective application of the techniques of sympathetic construction is a separate issue to what are the limits of those techniques. And as the writers have already observed, the result of the Court of Appeal’s approach in IDT has emasculated another well-established principle, that of ‘no reverse direct effect’, which might of itself offend legal certainty. However, to the extent that the approach of the Supreme Court in FII is applied to IDT, so that the carpentry (or ‘surgery’, to use Lord Sumption’s phrase) used by the Court of Appeal in IDT was properly treated as too radical to be permissible, by reference to the text and purpose of the relevant domestic provisions in the UK VAT code, the prospect of reverse direct effect and the offence against legal certainty disappear, at least to that extent.

(4) FII (Supreme Court): The Reining in of the Prior Case Law and a Welcome Approach to Sympathetic Construction The Supreme Court in FII91 has re-established sound principles of sympathetic UnionCompliant construction. The rejection by the Supreme Court of the expansive application 87 In that the words read into the relevant provisions in each case may well be said to lead to a ‘change in substance of a provision completely, to change a provision from one where Parliament says x is to happen to one saying x is not to happen’: ibid [110] (Lord Rodger). 88 See esp the text accompanying n 67. 89 Case 80/86 Criminal proceedings against Kolpinghuis Nijmegen BV [1987] ECR 3969. 90 IDT (n 11) [110] (Arden LJ). 91 FII (SC) (n 19).

Judicial Techniques 183

of sympathetic construction by the Court of Appeal is very welcome. The Court of Appeal’s approach in FII cast doubt on the contra legem limitation laid down by the CJEU, offended the ‘no reverse direct effect principle’ as it applied to Directives, potentially offended legal certainty and went beyond the limitations laid down in English case law itself. Certainly in its approach to the construction of section 33(2A), these concerns have been allayed by the Supreme Court. And the acknowledgment in FII by the Supreme Court that a taxpayer is entitled to the most favourable domestic remedy in enforcing Union rights should make it clear that the approach of the Upper Tier Tribunal in Wilkins, in using the availability of one domestic remedy as a licence to fail to construe another statutory remedy sympathetically, is wrong in principle and cannot stand in the light of FII. All of which is good news.

D. DISAPPLICATION

(1) The Problem As we have already seen, the Court of Justice has no jurisdiction to construe domestic legislation, let alone to disapply it.92 However, of course national courts may employ disapplication as a technique if a claimant is ‘so circumstanced that the offending provisions must not be invoked against them, either in particular cases or at all’.93 Disapplication makes perfect sense if the application of the relevant domestic provision is what leads to a breach of Union rights (or impedes their enforcement). So the offending provision which illegitimately curtailed the limitation period to recover overpaid VAT94 is simply to be disapplied as against those who have a legitimate expectation by reference to the original (longer) period.95 However the notion of disapplication makes no sense in encapsulating Union rights where the claimant seeks advantage of credits, reliefs etc, or where he wishes to take advantage of a particular enforcement provision to give him an effective remedy and is frustrated because of a non-compliant qualifying condition (such as a requirement of UK tax residence of either himself or a counterparty), as in FII (CJEU) and FII (Court of Appeal) where the tax credit was dependent on a distribution being paid by a UK tax resident company. In such a case, it is submitted that the notion of a ‘partial disapplication’ (so as to apply the provision excised of the words ‘UK resident’) is not really an intelligible notion. Rather, it is a question of sympathetic construction.96 In other words, disapplication does not really make any sense where a taxpayer wishes to take advantage of a provision the application of which is confined to a class of persons to which he does not belong.

Pfeiffer (n 4) para 115; Vodafone 2 (n 18) [34]. Fleming (n 55) [49] (Lord Walker). See also Vodafone 2 (n 18) [62]. 94 Regulation 29(1A) of the VAT Regulations 1995 did not provide for an adequate transitional period. 95 Fleming (n 55). 96 So, for example, in FII (CA) (n 22) the entitlement to a tax credit which was restricted to dividends received from ‘a UK tax resident company’ was seemingly considered to be a question of ‘removing the requirement of UK residence’ (at [100]), but in fact the precise judicial technique used to make s 231 Union-compliant was to read words in (at [107]). 92 93

184 Catherine Barnard and Julian Ghosh QC

(2) A New Approach by the Court of Justice? However, a new and potentially seismic approach has been adopted by the Court of Justice in Mangold v Helm97 and Kücükdeveci v Swedex GmbH & Co KG 98. The Court has exhibited a broad approach in its decisions which leads, to all intents and purposes, to direct horizontal effect of Directives and possibly giving national courts a general duty to grant a remedy in all circumstances. If this is the case, this would fundamentally alter the result in both FII (Court of Appeal) and Wilkins. In Mangold, Directive 2000/78 prohibiting age discrimination99 had been promulgated, but Germany was still within the grace period for implementing it. A private law action was brought by the applicant against his employer in relation to certain German provisions relating to fixed-term contracts which the applicant claimed contravened the provisions of Directive 2000/78. The Court held that Directive 2000/78, which would, once implemented, make the fixed-term contract contrary to German employment law, was but an illustration of the general principle of non-discrimination. The principle of nondiscrimination applied at all times (including the period leading up to the expiry of the grace period to implement the Directive), which meant, in turn, that Germany’s domestic employment law provisions which prospectively infringed Directive 2000/78 (once implemented) also immediately and presently infringed the general principle of non-discrimination. In other words, the general principle of non-discrimination engaged Union law and gave the applicant Union law rights, the precise content of which were to be found in the Directive, even though the grace period to implement it was as yet unexpired.100 This made the grace period, for all practical purposes, irrelevant.101 Thus Germany’s obligations to implement the Directive were accelerated in a private law action such that it was the applicant’s employer who bore the burden of this accelerated obligation. Even more tellingly, the remedy, according to the Court of Justice, was for Germany (not the employer) to ‘set aside’ the offending German domestic employment law provisions.102 Thus the Court of Justice engaged the general principle of non-discrimination as a surrogate to accelerate specific obligations addressed to Member States to implement Directives in the context of a private law action (so that the burden of that accelerated obligation was to fall on a private law employer who had no control over the implementation or otherwise of the relevant Directive). And in the context of remedies, the duty of the national court was to look at the national law as a whole and (quite simply) ‘set aside’ any provision which prevented the applicant from relying on his Union rights. Despite the extensive criticism of the judgment in Mangold, the approach was reinforced by the Court of Justice in Kücükdeveci.103 Here the relevant age discrimination Directive (the same Directive as that in issue in Mangold) was not yet implemented, but the grace period for implementation had expired. Certain German provisions (relating to employers Case C-144/04 Mangold v Helm [2005] ECR I-9981 (hereafter ‘Mangold ’). Case C-555/07 Kücükdeveci v Swedex GmbH & Co KG [2010] ECR I-365 (hereafter ‘Kücükdeveci’). 99 Council Directive 2000/78/EC of 27 November 2000 establishing a general framework for equal treatment in employment and occupation [2000] OJ L303/16. 100 Mangold (n 98) para 74. 101 See ibid para 66. 102 ibid para 77. 103 Kücükdeveci (n 99). 97 98

Judicial Techniques 185

being permitted to ignore periods of employment of an employee where the employee was under 25 years old) were considered by the Court of Justice to be directly age discriminatory. However, since the case involved two private parties there could be no horizontal direct effect of Directive 2000/78. Moreover, and significantly for our purposes, the referring national court considered sympathetic construction to be impossible because of the clarity and precision of the offending German provisions.104 As in Mangold, the Court of Justice observed that Directive 2000/78 merely gives expression to, but does not lay down, the principle of equal treatment in employment and occupation, and that the principle and non-discrimination on grounds of age is a general principle of European Union law in that it constitutes a specific application of the general principle of equal treatment (see, to that effect, Mangold, paragraphs 74–76).105

The CJEU continued that [i]n those circumstances, it is for the national court, hearing a dispute involving the principle upon discrimination of grounds of age . . . to provide, within the limits of its jurisdiction, the legal protection which individuals derive from European Union law and to ensure the full effectiveness of that law, disapplying if need be any provision of national legislation contrary to that principle (see, to that effect, Mangold, paragraph 77).106

The Court of Justice seems to be arrogating to the national court a general duty to ensure the encapsulation of Union rights (in this context ensuring that there is an adequate remedy for the breach of Union rights which resulted in age discrimination). Although the Court provides lip service to the ‘limits of [the national court’s] jurisdiction’,107 it directs the disapplication of any provision of national legislation which contravenes the general principle of equal treatment. Applying, by analogy, a duty on the national court to disapply any provision which contravenes the principle of effectiveness, it may well be that the notion of sympathetic construction has been made redundant. And if national courts indeed have a general and absolute duty to confer a remedy in all circumstances (which notion has been expressly aired before: the need for an effective remedy has been described as being of itself a ‘general principle’108), so that a Union right is tangibly enforced, by every applicant in every case, the result in both FII (Court of Appeal) (no remedy because the limitation period for making an error or mistake claim has expired and non-statutory remedies are not available because an in-time error or mistake claim ousts all such non-statutory remedies) and Wilkins (‘interest’ means the specific rate of interest expressed in subordinate legislation) must be reversed. In both cases, the English courts would be bound to provide a judicial remedy. Indeed this approach may be a resurrection of the approach of the Advocate ibid para 49. ibid para 50. 106 ibid para 51. 107 ibid. For an unconvincing account of a limited reading of the scope of Mangold, see Case C-427/06 Bartsch v Bosch und Siemens Hausgeräte (BSH) Altersfürsorge GmbH [2008] ECR I-7245, which suggested that the Court of Justice in Mangold was addressing a breach of the principle of equal treatment not in itself, but rather as a principle which was engaged because the offending German provision had implemented a previous Directive (1999/70) and it was this previous Directive which engaged Union Law: see paras 17–18, 23–24 of the judgment of the Court. But the Court of Justice in Mangold made it quite clear that there was no breach of the old Directive 1999/70 and it was a breach of the principle of equal treatment which carried with it the specific rights contained in the new Directive which engaged Union Law and conferred Union rights: see paras 26–27, 65–66, 70, 74 of the judgment of the Court. Kücükdeveci (n 99) now puts the position beyond doubt. 108 Case 68/88 European Commission v Greece [1989] ECR 2965 paras 23–24. 104 105

186 Catherine Barnard and Julian Ghosh QC General in FII (CJEU),109 where he made the observation that a breach of Union rights demanded a remedy and the principle of effectiveness was not satisfied if the only remedy was in damages, in relation to which the ‘sufficiently serious breach’ condition could not be demonstrated.110 This was rejected by the Court of Appeal in FII. As indicated above, we consider this rejection to have been well-founded. The ‘sufficiently serious breach’ condition is itself a notion of Union Law. It would be odd if it breached the principle of effectiveness. Nevertheless, the adoption of the Mangold approach by the Court of Justice may well presage the triumph of the Advocate-General’s thinking in FII (CJEU). Of course, it may be that the Court of Justice adopts an independent line of cases relating to tax law, so that the Mangold/Kücükdeveci approach will not apply to tax remedies.111 However, if this is not the case, it has significant implications for applicants seeking remedies for overpaid tax.

FII (CJEU) para 134. See also AG Bot in Kücükdeveci (n99) para 69. 111 As the Court of Justice has done in relation to limitation of benefit provisions, which were held to breach the principle of non-discrimination in non-tax cases (cf Case C-466/98 European Commission v United Kingdom (‘Open Skies’) [2002] ECR I-9427), but were held to be consistent with Member States’ obligations in tax cases, when contained in Double Tax Treaties in conventional form (cf Case C-374/04 Test Claimants in Class IV of the ACT Group Litigation [2006] ECR I-11673, paras 88–90). 109 110

10 The Principle of Effectiveness and Restitution of Overpaid Tax MAXIMILIAN SCHLOTE

A. INTRODUCTION

The principle of effectiveness is a well-established principle in the jurisprudence of the Court of Justice of the European Union (CJEU).1 In its most basic form, it requires a Member State to repay taxes or charges that were levied in contravention of EU law. The rationale underpinning the right to repayment is neither claimant-focused, nor defendantfocused. Indeed, it is not, or not primarily, concerned with principles of private law, such as unjust enrichment. Instead, its raison d’être is the reversal of the economic effects of the breach of EU law, which, as the CJEU reasons, ensures the effective enforcement of the primary EU law provision. This is why the right under EU law to the repayment of money charged as tax in breach of EU law extends to all sums that are directly related to the principal sum. It also explains why, in certain circumstances, individuals other than the taxpayer are given a right to recovery. In explaining and expanding the scope of the right to recovery, the CJEU has reduced the area of discretion available to domestic courts in giving effect to an individual’s EU law right to repayment. Pursuant to the principle of national procedural autonomy, it is for national law to lay down detailed rules of recovery. However, the guiding principle even at the national level is the principle of effectiveness, and not the principles that would normally govern the particular domestic cause of action. Hence, while there is a three-way relationship between the principle of effectiveness, the right to recovery, and the domestic remedy, the EU aspects of that relationship are dominant. It is therefore essential to have a clear understanding of the principle of effectiveness and its rationale before one tries to identify the rights of a claimant who seeks to recover money paid as tax charged in breach of EU law.

References to the CJEU include the European Court of Justice, ECJ.

1

188 Maximilian Schlote

B. EFFECTIVENESS, SAN GIORGIO AND THE RETURN OF OVERPAID TAX

(1) The Origins of and Justification for the Right to the Repayment of Unlawful Tax The leading case on the principle of effectiveness is Amministrazione delle Finanze dello Stato v SpA San Giorgio.2 The case concerned health inspection charges levied on the entry of animals or animal products into Italy. These were held to be charges equivalent to customs duties and therefore to be in breach of the provisions on the free movement of goods, now contained in Article 28 of the Treaty on the Functioning of the European Union.3 Under national law provisions, recovery of the illegal charge was practically impossible. Against this backdrop, the CJEU made its seminal statement of the principle of effectiveness: [E]ntitlement to the repayment of charges levied by a member state contrary to the rules of Community law is a consequence of, and an adjunct to, the rights conferred on individuals by the Community provisions prohibiting charges having an effect equivalent to customs duties.4

This principle has since been restated in many judgments of the CJEU, including Metallgesellschaft Ltd v IRC and Hoechst AG v IRC,5 Test Claimants in the FII Group Litigation v IRC,6 and BP Supergas Anonimos Etairia Geniki Emporiki-Viomichaniki kai Antiprossopeion v Greece.7 A Member State is therefore ‘required in principle to repay charges levied in breach of Community law’.8 In the extract from San Giorgio that has been quoted, the CJEU explained that the right to repayment is ‘a consequence of, and an adjunct to’ the rights which made the charge illegal in the first place.9 In Metallgesellschaft, the CJEU expressed the same idea slightly differently: the San Giorgio right to repayment is the ‘consequence and complement’ of the breached rights.10 The right to repayment of illegal charges exists because to allow repayment is to give effect to the claimant’s directly enforceable EU rights. For example, in Metallgesellschaft, a breach of Article 49 of the Treaty was committed when the UK subsidiaries of foreign parent companies had to pay tax earlier than the UK subsidiaries of domestic parents, and the award was said to be ‘essential in restoring the equal treatment guaranteed by Article [49]’.11 In Danfoss A/S and Sauer-Danfoss ApS v Skatteministeriet,12 the CJEU explained further how the right to repayment supports the effective enforcement of EU rights: Case C-199/82 Amministrazione delle Finanze dello Stato v SpA San Giorgio [1983] ECR 3595. European Union, Consolidated Version of the Treaty on the Functioning of the European Union, 13 December 2007, [2008] OJ C115/01 (hereafter ‘the Treaty’). 4 San Giorgio (n 2) [12]. 5 Joined Cases C-397/98 and C-410/98 Metallgesellschaft Ltd v Inland Revenue Commissioners; Hoechst AG v Inland Revenue Commissioners [2001] Ch 620 [84]–[86]. 6 Case C-446/04 Test Claimants in the Franked Investment Income Group Litigation v Inland Revenue Commissioners [2006] ECR I-11753, [2007] 1 CMLR 35 [201]–[208] (hereafter ‘FII (CJEU)’). 7 Case C-62/93 BP Supergas Anonimos Etairia Geniki Emporiki-Viomichaniki kai Antiprossopeion v Greece [1995] STC 805 [42]. 8 FII (CJEU) (n 6) [202]. 9 San Giorgio (n 2) [12]. 10 Metallgesellschaft (n 5) [84]. See also FII (CJEU) (n 6) [202]. 11 ibid [87]. 12 Case C-94/10 Danfoss A/S and Sauer-Danfoss ApS v Skatteministeriet, CJEU, 20 October 2011. 2 3

The Principle of Effectiveness 189 [T]he right to the recovery of sums unduly paid helps to offset the consequences of the duty’s incompatibility with EU law by neutralising the economic burden which that duty has unduly imposed on the [payor].13

As Henderson J recognised in Investment Trust Companies (in liq) v HMRC,14 the scope of the right must be interpreted in the light of this purpose. In other words, there is a reflective relationship between the right to repayment and its purpose, viz the neutralising of the economic burden imposed on the taxpayer. This ensures that the claimant has a right to restitution wherever this is required by the principle of effectiveness. (2) National Procedural Autonomy and Unjust Enrichment The claimant’s rights, generated by the EU law principle of effectiveness, are enforced in national law. Pursuant to the principle of national procedural autonomy, it is ‘for the domestic legal system of each Member State to . . . lay down the detailed procedural rules [for recovery]’.15 However, this principle should not be understood as giving the Member State any real discretion.16 The Member State has to make a remedy available to the full extent of the claimant’s San Giorgio right. Moreover, the Member State’s procedural rules must comply with the twin principles of equivalence and effectiveness, so that the rules applicable to an EU law claim are no less favourable than those governing similar domestic actions; nor may these rules render ‘virtually impossible or excessively difficult the exercise of rights conferred by Community law’.17 EU law does not require particular remedies to be applied or created to enforce the right to repayment,18 but it does require a Member State to make a remedy available that is effective to ensure that sums paid to the government in contravention of EU law are recoverable.19 A line of European and English court decisions shows that in English law a claimant can make a claim in unjust enrichment to recover unlawful charges.20 For example, in Metallgesellschaft, the CJEU found that there had been a breach of EU law when the claimants had been obliged to pay Advance Corporation Tax (ACT) when they should have been permitted to pay Mainstream Corporation Tax (MCT) at a later date. Pursuant to San Giorgio, they were therefore entitled to a sum of interest reflecting the use value of the money during the period between the premature payment date and the date when the tax should have been payable. In two cases, Sempra Metals Ltd v IRC21 and Deutsche Morgan Grenfell Group plc v IRC,22 the House of Lords held that this was recoverable in a claim for ibid [23]. Investment Trust Companies (in liq) v HM Revenue & Customs [2012] EWHC 458 (Ch), [2012] STC 1150 [118] (hereafter ‘ITC ’). 15 FII (CJEU) (n 6) [203]. See too Case 33/76 Rewe-Zentralfinanz eG and Rewe-Zentral AG v Landwirtschaftskammer für das Saarland [1976] ECR 1989 [5]. 16 Case C-591/10 Littlewoods Retail Ltd v HM Revenue & Customs AG’s Opinion, 12 January 2012 [23] (hereafter ‘Littlewoods (AGO)’). 17 FII (CJEU) (n 6) [203]; Rewe-Zentralfinanz (n 15) [5]. 18 As Lord Sumption noted in Test Claimants in the FII Group Litigation v HM Revenue & Customs [2012] UKSC 19, [2012] 2 AC 337 (hereafter ‘FII (SC)’). 19 Littlewoods (AGO) (n 16) [23]. 20 For the elements of an unjust enrichment claim, see Banque Financière de la Cité SA v Parc (Battersea) Ltd [1999] 1 AC 221 (HL) 227 (Lord Steyn). 21 Sempra Metals Ltd v Inland Revenue Commissioners [2007] UKHL 34, [2008] 1 AC 561. 22 Deutsche Morgan Grenfell Group plc v Inland Revenue Commissioners [2006] UKHL 49, [2007] 1 AC 558 (hereafter ‘DMG’). 13 14

190 Maximilian Schlote unjust enrichment. The use value of the money during the period of prematurity had enriched the government at the claimants’ expense, and this enrichment had been unjust because the claimants had mistakenly believed that the tax had been due at a date when it had not been, as a result of the Metallgesellschaft decision.23 This was an effective remedy for the San Giorgio claim arising out of Metallgesellschaft because the measure of enrichment in English law was the rate of interest at which the government would have had to borrow an equivalent sum (ie compound interest), and this corresponded to the use value of the money to which the claimants were entitled as a matter of EU law. In general, an action in unjust enrichment provides an effective remedy for a claimant’s San Giorgio claim. However, it must be remembered that the governing principle in this area is the principle of effectiveness and not domestic law principles. This means that if domestic law provisions prevent a claimant from recovering to the full extent required by EU law, then they must be disapplied.24 The principle of effectiveness dictates the scope of the claimant’s right to recovery although his claim lies in the law of unjust enrichment under English domestic law.

(3) The Principle of Effectiveness and Defences to a San Giorgio Claim The interaction between the principle of effectiveness and the domestic law of unjust enrichment, perhaps more appropriately termed the primacy of the former over the latter, can usefully be illustrated by a comparison of the defences available to an unjust enrichment claim and the defences available to a San Giorgio claim. In the English law of unjust enrichment, a ‘defence is available to a person whose position has so changed that it would be inequitable in all the circumstances to require him to make restitution’.25 For example, in Test Claimants in the FII Group Litigation v IRC, a case involving an unlawful charge of corporation tax, the government’s argument that it had changed its position in response to its enrichment was accepted by Henderson J, who considered that government expenditure would not have taken place at the same level but for the government’s enrichment.26

23 For more detailed discussion of how unjust enrichment operates to give effect to the San Giorgio claim, see the text around n 106. 24 See eg Test Claimants in the Franked Investment Income Group Litigation v Inland Revenue Commissioners [2010] EWCA Civ 103, [2010] STC 1251 [174] (hereafter ‘FII (CA)’): if a claim brought under the rule in Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] AC 70 required it to be shown that HMRC had actively demanded the unlawful tax, then this requirement would have to be disapplied because it would be contrary to EU law. The primacy of EU law also frequently raises questions as to whether limitation periods must be disapplied: see the text accompanying nn 59–63. 25 Lipkin Gorman (a firm) v Karpnale Ltd [1991] 2 AC 548, 580 (Lord Goff). For elaboration of this principle, see eg C Mitchell, P Mitchell and S Watterson (eds), Goff & Jones: The Law of Unjust Enrichment, 8th edn (London, Sweet & Maxwell, 2011) ch 27. 26 Test Claimants in the Franked Investment Income Group Litigation v Inland Revenue Commissioners [2008] EWHC 2893 (Ch), [2009] STC 254 [344] (hereafter ‘FII (Ch)’). Henderson J’s findings in relation to the ‘change of position’ defence were not considered on appeal in the FII case. However, his statement of the circumstances in which the defence is available was very wide and it is not clear whether it should be or, indeed, is available in such a wide set of circumstances. In another case, Vos J found that it was not enough that the Revenue could ‘show that they either spent or used the overpayment’: Littlewoods Retail Limited v Inland Revenue Commissioners [2010] EWHC 1071 (Ch), [2010] STC 2072 [124] (hereafter ‘Littlewoods (Ch)’). For further discussion of this point, see chapter 7 of the present volume by Niamh Cleary, at pp 128–32.

The Principle of Effectiveness 191

The justification for the change of position defence is a matter of some academic debate.27 However, it can be said with confidence that the defence is fundamentally concerned with the injustice that would be caused to the defendant if restitution were ordered in circumstances where he was not to blame for the unjust enrichment and where he changed his position in response to it. In Lipkin Gorman Lord Goff said that ordering restitution in such circumstances would be ‘inequitable’,28 and in FII Henderson J said that it would be ‘inequitable to require the Revenue to make restitution for the tax which was paid by mistake, because the money in question has long ago been spent in the public interest’.29 This defence, however, cannot be available in response to an unjust enrichment claim made by a claimant who seeks restitution of money paid as tax levied in breach of EU law.30 The defence is inconsistent with the principle of effectiveness because this principle denies restitution purely on the basis of the inequity caused to the defendant. The CJEU said as much in Weber’s Wine World Handels-GmbH v Abgabenberufungskommission Wien,31 and again in Lady & Kid A/S v Skatteministeriet.32 The CJEU stated that the ‘only one’ or ‘sole’ exception to the EU law requirement that overpaid tax must be repaid arises in cases where the claimant has passed on his loss to a third party.33 The defence of passing on is consistent with the principle of effectiveness for the reason that if the taxpayer has not suffered the economic burden of the unlawful tax because he has passed on his loss to a customer, then it would do nothing to reverse the economic consequences of the breach of EU law to make restitution to him.34 As the CJEU said in Danfoss, allowing a claim by the taxpayer would ‘be tantamount to paying him twice over, which may be described as unjust enrichment, whilst in no way remedying the consequences for the purchaser of the illegality of the charge’.35 Passing on, driven, as it is, by the principle of effectiveness, does not excuse the Member State from making restitution; it simply acknowledges that the correct claimant is not the taxpayer but the customer to whom the tax has been passed on. Although Vos J held otherwise in Littlewoods Retail Ltd v IRC, the passing on defence does not look primarily at the ‘justice (or injustice)’ of the reimbursement.36 In the EU context, the question whether the defence should be available turns instead on the principle of effectiveness and on whether restitution would reverse the economic consequences of the breach. In contrast, the change of position defence is inconsistent with the principle of effectiveness precisely because it does look at the injustice which restitution would cause the defendant. The fact that the government has changed its position does not affect its ability to reverse the economic consequences of the breach of EU law. Instead, like the ‘settled practice’ defence proposed by the Danish government in Fantask A/S v Industriministriet A Burrows, The Law of Unjust Enrichment, 3rd edn (Oxford, OUP, 2011) 526 –32. Lipkin Gorman (n 25) 580 (Lord Goff). 29 FII (Ch) (n 26) [344]. 30 This is the conclusion reached by Henderson J in FII (Ch) (n 26) and FJ Chalke Ltd v Inland Revenue Commissioners [2009] EWHC 952 (Ch), [2009] STC 2027. In Littlewoods (Ch) (n 26) Vos J disagreed and held that the defence might be consistent with EU law. 31 Case C-147/01 Weber’s Wine World Handels-GmbH v Abgabenberufungskommission Wien [2003] ECR I-11365. 32 Case C-398/09 Lady & Kid A/S v Skatteministeriet [2012] All ER (EC) 410, noted by R Williams [2011] British Tax Review 631. 33 Weber’s Wine (n 31) [94]; Lady & Kid (n 32) [20]. Thus, Lord Sumption in FII (SC) (n 18) [147] described the obligation to return overpaid tax as ‘absolute’. 34 Danfoss (n 12) [22]. 35 ibid [22]. 36 Littlewoods (Ch) (n 26) [137]. 27 28

192 Maximilian Schlote (Erhvervministeriet), allowing the government to rely on its change of position would ‘have the effect of encouraging infringements of Community law which have been committed over a long period’.37 Vos J, who held that the change of position defence may be consistent with the principle of effectiveness, also pointed to the permissibility of reasonable limitation periods to San Giorgio claims as proof that there is scope for new exceptions to the requirement to make restitution.38 The lawfulness of limitation periods is discussed below,39 but it can be said at this stage that in so far as they are lawful, they are lawful because the CJEU recognises that the principle of effectiveness is qualified by a further principle of EU law, namely the principle of legal certainty. There is no equivalent countervailing principle of EU law that would be given effect by the change of position defence. Consequently, in a case where a claimant seeks to enforce his San Giorgio right by bringing an action against the UK government under the English law of unjust enrichment, the government could not rely on the change of position defence even if it were able to establish such a defence as a matter of English domestic law.

C. BEYOND THE TAXPAYER: RECOVERY BY INDIRECT TAXPAYERS

(1) Recovery by Those Who Suffered the Economic Burden of the Tax The principle of effectiveness does not prevent defendants from relying on the passing on defence because the focus of this principle is on the reversal of the economic consequences of the breach. The corollary of this is that where a claimant has passed on his loss to a third party, the Member State may be required to make restitution to the third party, on the basis that he is the person who actually bore the economic burden of the unlawful tax. For example, in the Danfoss case, Denmark imposed a tax on the supply of lubricating oil in violation of EU Directive 92/81/EEC. A supplier (the direct taxpayer) owed and paid the tax to the national authorities but, as is the case with most indirect taxes, the economic burden of the tax was passed on to customers (the indirect taxpayers) by way of an element of tax on the invoice. The CJEU confirmed that in such circumstances allowing the direct taxpayer a claim in restitution would not serve the principle of effectiveness and that the Member State may therefore deny recovery to the direct taxpayer without infringing the San Giorgio principle.40 However, if the Member State denies a claim to the direct taxpayer on the ground that the economic cost of the tax has been borne by the indirect taxpayer, then the principle of effectiveness requires the Member State to repay the unlawfully levied charge directly to the indirect taxpayer.41 This does not mean that, in a case like Danfoss, the Member State must allow an indirect taxpayer a direct claim against the national authorities in all circumstances.42 Under the Case C-188/95 Fantask A/S ea v Industriministeriet (Erhvervministeriet) [1997] ECR I-6783 [40]. Littlewoods (Ch) (n 26) [137] and [138]. 39 In the text accompanying nn 59–63. 40 Danfoss (n 12) [21] and [22]. See too Lady & Kid (n 32) [18]–[20]. 41 Danfoss (n 12) [26]–[29]. 42 ibid [26]–[29] 37 38

The Principle of Effectiveness 193

principle of national procedural autonomy, the Member State can instead require the indirect taxpayer to seek restitution from the direct taxpayer. In those circumstances, however, the principle of effectiveness requires that the direct taxpayer must be able to recover from the national authorities in turn.43 In this way, the economic consequences of the unlawful charge are fully reversed. Furthermore, if the Member State denies the indirect taxpayer a claim against the national authorities, and requires the indirect taxpayer to pursue the direct taxpayer for the amount of the overpaid charge, as Denmark did in Danfoss, recovery by the indirect taxpayer from the direct taxpayer must comply with the principle of effectiveness: [I]f reimbursement by the taxable person were to prove impossible or excessively difficult – in particular, in the case of the insolvency of that person – the principle of effectiveness requires that the purchaser be able to bring his claim for reimbursement against the tax authorities directly.44

It follows that an indirect taxpayer can only be denied a direct claim against the national authorities if he is able to ‘obtain reimbursement through the taxable person of the amount of the charge passed on to him’.45 If it is ‘impossible or excessively difficult’ for him to recover this amount from the direct taxpayer, then he must be given his own claim against the national authorities.46 This principle is straightforward to apply in the situation identified by the CJEU in Danfoss.47 If the indirect taxpayer is generally required to pursue the direct taxpayer and forbidden to pursue the national authorities directly, but the direct taxpayer becomes insolvent, then it is highly unlikely that the indirect taxpayer will recover the whole of the unlawful charge passed on to him from the direct taxpayer. It follows that the indirect taxpayer must be given his own direct claim against the national authorities. However, insolvency is, in a sense, an easy example because it concerns the practical possibility of recovery. The test is harder to apply when the direct taxpayer cannot recover from the national authorities because domestic law does not provide him with a claim, with the knock-on result that domestic law makes it impossible for the indirect taxpayer to recover from the direct taxpayer. This was the situation in the ITC case.48

(2) Investment Trust Companies (in liq) v HMRC The ITC case concerned the supply of investment management services by investment trust managers (‘the managers’) to the claimant investment trust companies (‘the companies’). The companies paid the managers a sum in respect of VAT charged on those services and the managers accounted for this money to HMRC as output tax. When it transpired that the services supplied by the managers to the companies should have been VAT exempt under the Sixth Directive, the managers recovered some of the overpaid VAT from HMRC and passed this on to the companies. However, the managers were unable to 43 ibid [26]. See too Case C-192/95 Société Comateb v Directeur Général des Douanes et Droits Indirects [1997] ECR I-165 [24]; Case C-35/05 Reemtsma Cigarettenfabriken GmbH v Ministero delle Finanze [2007] ECR I-2425 [39]. 44 Danfoss (n 12) [28]. 45 ibid [26]. 46 ibid [28]. 47 ibid [28]. See too Reemtsma (n 43) [41]. 48 ITC (n 14). The English law aspects of this case are more fully examined by Charles Mitchell in chapter 6 of the present volume.

194 Maximilian Schlote recover from HMRC the whole of the VAT that the companies had paid to the managers. This allowed the managers to claim a change of position defence against a potential claim by the companies in respect of the irrecoverable part. As a matter of domestic law, the companies therefore found it impossible to recover the whole of the VAT from the managers. Applying the test stated above, they should therefore prima facie have been given a direct claim against HMRC. Before reaching this conclusion, however, it is necessary to understand the nature of the irrecoverable parts of the overpaid VAT. There were two reasons why the managers could not recover the whole of the VAT for which they had accounted to HMRC. First, the managers had incurred input tax on services provided to them by suppliers (‘the suppliers’). The suppliers had paid this VAT to HMRC and there was no dispute as to the legality of this VAT charge. Adopting the numerical example used by Henderson J in ITC,49 the managers paid the suppliers £25 in respect of lawful input tax. The services received from suppliers were in respect of the services provided by the managers to the companies; services which were thought to be subject to VAT. Therefore, at the time when the managers accounted to HMRC for VAT, it was believed that the £25 of VAT paid to the suppliers was ‘attributable to . . . taxable supplies’.50 It was therefore assumed that under VATA, section 25(2), the managers were entitled to deduct the £25 from any output tax that they owed on the provision of their investment services to the companies. Again using Henderson J’s numbers, the net payment which the managers therefore made to HMRC was £100 output tax in respect of the services provided to the companies, less £25 input tax, ie £75. Following the discovery that these services should have been exempt, HMRC repaid only £75 to the managers, this being the cash amount which they had received from the managers. The second reason why there was a shortfall in the amounts recovered by the managers from HMRC is that the whole of the managers’ claims were excluded by limitation periods between 4 December 1996 and 2001 (referred to as the ‘dead period’ in the judgment). There was no argument that those limitation periods were contrary to EU law, at least in so far as they only operated as against the managers. The upshot was that the companies had paid £100 in respect of an unlawful VAT charge but could, through the managers, recover only £75, and only for periods outside the dead period. Hence they pursued HMRC directly in unjust enrichment for recovery of the difference between the £100 they had paid to the managers and the £75 they had recovered, and for recovery of the full £100 in respect of the dead period. Henderson J held as a matter of domestic law that these claims were excluded by VATA, section 80(7).51 But he also held that if the companies had an EU right to recovery against HMRC, section 80(7) would have to be disapplied and the companies’ unjust enrichment claim would be permitted.

(3) The Requirement that Recovery of the Unlawful Charge Must Be Impossible As a matter of EU law, HMRC were within their rights to deny the companies a direct claim in respect of the overpaid tax, to the extent that the companies could recover from ITC (n 14) [15]. Value Added Tax Act 1994, s 26(1) and (2)(a) (hereafter ‘VATA’). 51 ITC (n 14) [88]–[110]. For an explanation of why VATA, s 80(7) excludes common law claims, see R Mullan and H Brown, ‘Littlewoods Ltd v HMRC: Compound Interest on Overpaid Interest on Overpaid VAT – Where Are We Now?’ [2010] British Tax Review 475. 49 50

The Principle of Effectiveness 195

the managers. However, the right to recover from the managers had to comply with the principle of effectiveness, and as a matter of domestic law it was impossible for the companies to recover £25 of their £100 from the managers.52 That in itself was not enough to justify giving the companies their own direct claim against HMRC: they also needed to show that the £25 represented part of an unlawful charge which had been imposed on the managers and then passed on to the companies. At first sight, this was difficult to show. After all, the very reason the managers could not recover the £25 was that the managers had only ever paid HMRC £75. It seems strange that the companies should have been entitled to recover more than the managers could recover; more, even, than the managers had paid. Nonetheless, that is what Henderson J found.53 He was correct to do so for the following reason. VATA, section 25(1) requires a taxable person to ‘account for and pay VAT’. The managers accounted for £100 of unlawful VAT by paying £75 cash, as permitted by VATA, section 25(2).54 As the VAT and Duties Tribunal said in Times Right Marketing Limited (in liq) v HMRC, at least in the context of a claim for bad debt relief under VATA 1994, section 36, a taxable person can also be said to have ‘accounted for and paid [VAT] . . . by way of input tax credit’.55 It follows that despite only paying £75 in cash, the unlawful charge accounted for by the managers was £100. This is perhaps easiest to appreciate when seen from the perspective of the companies. The companies paid the managers £100 as VAT, and the whole of that sum was paid in respect of the unlawful charge for which the managers subsequently accounted to HMRC. In those circumstances, it was only by giving the companies a direct right of recovery from HMRC, it being impossible or excessively difficult to recover from the managers, ‘that the claimants could obtain full reimbursement of the tax . . . which they [had] paid in full’.56 However, whilst Henderson J’s analysis is broadly consistent with this, the outcome of his analysis, namely recovery of £25 by the companies from HMRC, presents a problem. This can be seen if one considers the appropriate counterfactual, which is the situation where the services provided by the managers to the companies were always understood to be tax exempt. In that situation, the managers would have suffered £25 of input tax on the services received from the suppliers. The managers would have supplied VAT exempt services to the companies, and it would have been understood to follow from VATA, sections 25(2) and (3) and 26(1) and (2) that the managers were not entitled to a refund of the £25: in essence, the managers would have been treated as final consumers of the services provided by the suppliers. The evidence in ITC suggests that in this situation, the managers would in fact have borne this £25 loss, ie they would not have sought to pass the £25 on to the companies by charging them higher prices.57 It follows that in the counterfactual, the outcome would have been that HMRC received and kept £25 of lawful VAT, the companies paid no unlawful tax, and the managers bore the burden of paying £25 of lawful VAT. This outcome is consistent with the actual outcome of ITC, as analysed in this chapter, only in respect of the companies: for them, the economic consequences of the breach of EU law were reversed by the finding that they could recover £25 from HMRC. See the passage quoted from Danfoss (n 12) in the text to n 44, and ITC (n 14) [118]. ITC (n 14) [145]. 54 See the text accompanying n 49. 55 Times Right Marketing Limited (in liq) v HM Revenue & Customs London VAT and Duties Tribunal, 13 March 2008 [30]. 56 ITC (n 14) [145]. 57 ibid [84] and [137]. 52 53

196 Maximilian Schlote Does it follow, then, that in ITC the companies should have been unable to claim directly against HMRC? Arguably not. It must be kept in mind that the reversal of the economic consequences of the breach is at the heart of San Giorgio and Danfoss. HMRC suffered a shortfall of £25 in ITC not in consequence of the breach, but in consequence of mistakenly having allowed the managers an input tax credit. Expressed differently, the £25 that HMRC was correctly found liable to pay to the companies was not identical with the £25 that HMRC should have been able to recover from the managers, on the ground that it had made a mistake when it had allowed them an input tax credit. The £25 owed by HMRC to the companies was owed in respect of the unlawful charge of £100. The £25 that HMRC should have been able to recover from the managers was owed in respect of the input tax suffered on the supply of services to the managers by the suppliers, which input tax should not have been deductible. In other words, the correct outcome was that HMRC should have repaid the whole of the £100 of unlawful output VAT, charged on services provided by the managers to the companies, and that HMRC should then have raised a separate assessment against the managers for the recovery of £25 of input VAT which had turned out not to be deductible by virtue of the same event that had rendered the £100 output VAT recoverable, ie the discovery that the true status of the supplies by the managers to the companies was exempt rather than taxable. This analysis also applies vis-à-vis HMRC and the managers. HMRC should have repaid the full amount of VAT for which the managers had wrongly accounted and, separately, should have raised an assessment against the managers for the recovery of any related input tax. If this is correct, then the managers should have pursued HMRC for the £25. The companies could then in national law have pursued the managers for this £25, since the managers’ change of position defence would have fallen away. In a sense, this analysis, which has been used here to support the conclusion that the companies should have had a direct claim against HMRC, somewhat undermines itself. It will be recalled that the companies should only have been entitled to a direct claim against HMRC if recovery from the managers was impossible or excessively difficult. The problem arises because the same analysis which explains why the companies should have been able to recover £25 from HMRC, also explains why they should have been able to recover the same sum from the managers – in which case they should not have been entitled to recover directly from HMRC after all. However, such a claim against the managers would have been complex: the managers would have had to pursue HMRC, arguing that HMRC’s practice of reimbursing the net amount was inconsistent with the San Giorgio right to restitution. This would then have been followed by a claim by the companies against the managers. To justify giving the companies a direct claim against HMRC it must be shown that an indirect claim through the managers is impossible or excessively difficult and, as Henderson J said in ITC, this requirement of exhaustion of remedies should not be interpreted too strictly in light of the overarching principle of effectiveness.58 The key concern is that the whole of the unlawful charge is returned to the person who bore the burden of it and that aim has been met in ITC. (4) Effectiveness and National Limitation Periods So far, this chapter has focused on the expansive reach of the principle of effectiveness and the requirement that restitution should achieve a reversal of the economic consequences of ibid [140].

58

The Principle of Effectiveness 197

the breach of EU law. However, it will now be demonstrated by analysis of the other irrecoverable sums in ITC, the sums in respect of the dead period (‘dead period VAT’), that the principle of effectiveness is neither unlimited in its reach, nor is it the only principle applicable in this area. The managers could not recover unlawful VAT paid to HMRC during the dead period. As a result, the managers would have had a change of position defence in reply to any claim by the companies in respect of that VAT. The companies therefore argued, and Henderson J accepted, that it was impossible for them to recover this dead period VAT from the managers.59 However, that impossibility was not enough, in and of itself, to justify giving the companies a direct claim against HMRC. For the purposes of EU law, the impossibility must relate to recovery of an amount in respect of which restitution is required by the principle of effectiveness, and Henderson J did not accept that the companies had made this out. The main reason for this was that the limitation periods which caused the dead period for the managers were themselves fully compliant with EU law. Henderson J therefore concluded that these limitation periods should be extended to apply to any direct claims by the companies. To assess the merits of this analysis, it is necessary to briefly survey the impact of the principle of effectiveness on national limitation periods, and then to assess the significance of this for indirect taxpayers who make a direct claim against HMRC where it is impossible for them to recover from direct taxpayers, as a result of a lawful limitation period applying to a claim by the direct taxpayers against HMRC. The principle of effectiveness, and EU law more generally, have no general objection to limitation periods being imposed by Member States. As the CJEU said in the ReweZentralfinanz case, ‘the laying down of such time-limits with regard to actions of a fiscal nature is an application of the fundamental principle of legal certainty protecting both the taxpayer and the administration concerned’.60 All that EU law requires is that limitation periods are reasonable and comply with the principles of effectiveness, equivalence and legitimate expectations. Periods of five years and even three years have been held to be acceptable.61 These principles also apply in relation to limitation periods that are introduced in order to curtail a pre-existing limitation period. In Marks & Spencer plc v Customs and Excise Commissioners, the CJEU said that such curtailment of existing limitation periods should not be ‘intended specifically to limit the consequences of a judgment of the Court’.62 Furthermore, there is not only a condition that the new limitation period is reasonable but also that the new legislation includes transitional arrangements allowing an adequate period after the enactment of the legislation for lodging the claims for repayment which persons were entitled to submit under the original legislation.63

This shows that the principle of effectiveness does not exist in a vacuum. There are other principles in EU law, including the principle of legal certainty, that may also be in play. Reasonable limitation periods can, therefore, curtail the San Giorgio right to recovery even ibid [135]. Rewe-Zentralfinanz (n 15) [5]. 61 Case C-90/94 Haahr Petroleum Ltd v Å´benrå Havn [1997] ECR I-4085; Case C-231/96 Edilizia Industriale Siderurgica v Ministero delle Finanze [1998] ECR I-4951. 62 Case C-62/00 Marks & Spencer plc v Customs and Excise Commissioners [2003] QB 866 [36] (hereafter ‘M & S’). 63 ibid [38]. 59 60

198 Maximilian Schlote where this makes a reversal of all the economic consequences of the breach of EU law impossible. However, EU law requires Member States to respect accrued rights. Where a limitation period is curtailed, the principle of effectiveness requires that transitional arrangements must be put in place to enable those with accrued San Giorgio claims to exercise their rights. The dead period in ICT arose as a result of the application of these principles. Section 47 of the Finance Act 1997 reduced the six-year limitation period that had previously applied to claims for the recovery of VAT to three years without allowing for a transitional period. Following the M & S case, HMRC accepted that this violated EU law and allowed the claimants to claim for the six-year period ending before 4 December 1996, the date when section 47 came into force. The three-year period applied, lawfully, for the claims made in 2004. Thus, the dead period ran from 1996 to 2001. Importantly, for present purposes, the fact that the managers could not claim in respect of the dead period was consistent with EU law because the three-year period introduced by section 47 was clearly reasonable and, therefore, entirely in compliance with EU law. The lack of a transitional period was remedied by allowing the managers to claim in respect of the pre-1996 period. It follows from this, and from the principle of legal certainty, that although VAT had been unlawfully charged during the dead period, there could have been no right of recovery under the San Giorgio principle. The principle of effectiveness did not require it, because this principle was outweighed by the principle of legal certainty.

(5) Does a Lawful Limitation Period Applicable to a Direct Taxpayer Prevent a Claim Against the National Authorities by the Indirect Taxpayer? The limitation rule creating the dead period only applied to claims by the managers, but its effect was to make it impossible for the claimants to recover from the managers in respect of payments made during the period, because the managers’ inability to recover from HMRC gave the managers a ‘change of position’ defence in respect of claims made against them by the companies. Henderson J accepted this, but held that because the limitation period was EU law-compliant in relation to the managers’ claims against HMRC, its effect should be extended to the companies’ claims against HMRC as well: Since HMRC have an EU law-compliant limitation defence against the taxable persons themselves, it would in my judgment be anomalous and unreasonable if, in the name of effectiveness, EU law were to require the national court to make available to a person who has borne the economic burden of the tax a claim which would not be subject to the same defence.64

In other words, Henderson J accepted, correctly, that the companies could not be ‘placed in a more favourable position for limitation purposes than the Managers’.65 At first sight, this looks strange. After all, if the companies could pursue their own claim against HMRC, then under the English law of unjust enrichment this would take the form of a common law claim founded on mistake to which a different, longer limitation period would apply. If a shorter limitation period were applied to their claim, which domestic law would have imposed only on a claim by the managers against HMRC, then it might be ITC (n 14) [142]. ibid [142].

64 65

The Principle of Effectiveness 199

asked whether this would violate the principles of EU law that forbid curtailing a limitation period without putting transitional arrangements in place. However, this is not the correct way of approaching the issue. It should be remembered that EU law allows a Member State to deny a direct taxpayer a San Giorgio claim in respect of unlawful taxes passed on to indirect taxpayers. In that case, the indirect taxpayer would have his own San Giorgio-based claim against the national authorities and it would clearly be wrong to argue that any limitation periods which apply to the direct taxpayer should, by extension or analogy, apply to the indirect taxpayer, even if EU law-compliant. The principles of effectiveness, equivalence and, in particular, legitimate expectations have to be applied to the issue of limitation periods from the perspective of the indirect taxpayer. The indirect taxpayer’s claim is wholly independent from the position of the direct taxpayer. The same cannot be said, however, if a Member State in principle denies an indirect taxpayer a direct claim against the national authorities. It will be remembered that the Member State is free to do this, provided that it lets the indirect taxpayer recover from the direct taxpayer who, in turn, can recover from the national authorities. The claim belongs to the direct taxpayer and the indirect taxpayer cannot, under this system, recover more than the direct taxpayer. The indirect taxpayer may be entitled to pursue the national authorities directly if recovery from the direct taxpayer turns out to be impossible, but only to the extent of the sums which the direct taxpayer should have been able to recover pursuant to San Giorgio. This is why the example used by the CJEU in Danfoss was well chosen: it concerned insolvency of the direct taxpayer, ie the practical impossibility of recovering from the direct taxpayer. This boundary is harder to police in cases of legal impossibility. But ITC illustrates the point well. The companies had a direct claim in respect of the £25 because this constituted part of the unlawful VAT which had to be returned under the San Giorgio principle. The fact that HMRC did not return that sum to the managers made recovery from the managers impossible, but it did not change the fact that there had to be restitution in relation to the £25. In other words, HMRC’s refusal to return the £25 to the managers was itself contrary to the San Giorgio principle. The same cannot be said in respect of the dead period VAT. Non-recovery of the dead-period VAT was justified by the principle of legal certainty, and so the principle of effectiveness did not require restitution of this money. Expressed differently, in a system where an indirect taxpayer normally pursues his own claim against the national authorities, and in a system where an indirect taxpayer normally pursues such a claim through the direct taxpayer who, in turn, recovers from the national authorities, the amount of the unlawful charge is identical and, to that extent, both systems will always lead to the same result. The difference arises in the application of limitation periods: in the former system, only the limitation periods applying to the indirect taxpayer are relevant. In the latter, the indirect taxpayer can never recover more than the whole of the unlawful charge to the extent that the direct taxpayer could have recovered; thus, rules limiting recovery by the direct taxpayer also limit recovery by the indirect taxpayer. This difference is justifiable because in the latter system, the Member State would not have in place limitation periods specifically directed at recovery by the indirect taxpayer, for the simple reason that recovery by the indirect taxpayer is usually denied.

200 Maximilian Schlote

D. BEYOND OVERPAID TAX

(1) Interest So far this chapter has emphasised that the principle of effectiveness requires domestic law to disapply defences and, where appropriate, to allow indirect taxpayers a claim to enable the recovery of overpaid tax. However, in many cases, recovery of the principal sum paid as tax will not suffice to reverse the economic consequences of the breach of EU law. In particular, in the period after payment and prior to restitution, the government benefits from having received this sum because its borrowing needs are correspondingly reduced, saving it the expense of having to pay interest. This is referred to as the ‘use value’ or the ‘time value’ of the money.66 In effect, the taxpayer provides an ‘interest free loan’ to the Member State.67 This was recognised as a matter of English law by the House of Lords in Sempra. There the claimants made premature payments of corporation tax that had become lawfully due by the time the claims were brought, so that the claimants could not recover the principal amount of the tax. Nevertheless, they could recover the use value of the money during the period of prematurity. The benefit accruing to the government as a result of the premature payment was measured with reference to its borrowing costs over the relevant period. Since the government could only borrow on compound interest terms, the claimants were entitled to an award of compound interest.68 It follows that in many cases where a claimant brings an action in unjust enrichment to recover money paid as tax that is not due there is no need to invoke EU law with a view to establishing a right to compound interest, as this right already exists as a matter of English domestic law. However, in some cases, in particular in cases involving the recovery of overpaid VAT, statute excludes the common law claim and confines claimants to the recovery of principal and simple interest.69 In such cases, if the tax were imposed in contravention of EU law, the claimants would be entitled to recovery as a matter of EU law. If this right incorporated a right to the recovery of interest, and more specifically, of compound interest, then such English law limitations would have to be disapplied. The position seems to be that EU law does not, per se, require the claimant to receive interest qua interest on the overpaid tax. This is an ancillary matter for national law.70 However, what EU law does require is that the claimant should receive the amount of the use value of money paid as tax levied in breach of EU law.71 In some ways, this is a distinction without a difference because the measure of the use value of money is interest and, in that way, there is an EU right to interest. Nevertheless, the distinction is important because See eg Littlewoods (Ch) (n 26) [67] (Vos J); Sempra (n 21) [103] (Lord Nicholls). Sempra (n 21) [102] (Lord Nicholls). 68 ibid [50] (Lord Hope), [131] (Lord Nicholls) and [188] (Lord Walker). 69 VATA, ss 78 and 80. See Chalke (n 30) [57]–[75]; Littlewoods (Ch) (n 26) [62]; ITC (n 14) [88]–[110]; Mullan and Brown (n 51). 70 Metallgesellschaft (n 5) [86]. 71 M Schlote, ‘Littlewoods v HMRC: Compound Interest as a Matter of EU Law’ [2012] British Tax Review 144; M Schlote, ‘Test Claimants in the FII Group Litigations v IRC and Littlewoods Retails Limited v HMRC: Concurrency of Remedies and Compound Interest in Recovering Overpaid Tax’ [2012] British Tax Review 500. These are respectively notes on Littlewoods (AGO) (n 16) and Case C-591/10 Littlewoods Retail Ltd v HM Revenue & Customs, 19 July 2012, [2012] ECR I-000, [2012] STC 1714 (hereafter ‘Littlewoods (CJEU)’) and FII (SC) (n 18). 66 67

The Principle of Effectiveness 201

it defines the scope of the procedural autonomy of the Member States. If the right was a right to interest, there would be more scope for Member States to argue that any interest award would discharge their restitutionary obligations. In contrast, if the obligation is to return the use value of money, the Member State has to make restitution to the full extent of that use value. In Metallgesellschaft, the claimants were UK subsidiaries of foreign parent companies. They were compelled to pay Advance Corporation Tax (ACT) in circumstances in which the UK subsidiaries of UK parents were exempt from doing so. This fell foul of Article 49 of the Treaty but, by the time the case came to the courts, the illegal ACT had already been set off against lawful Mainstream Corporation Tax (MCT). The CJEU recognised that in those circumstances the claimants were entitled to a remedy for the ‘cost of loss of the use of the sums’,72 or, expressed differently, ‘the amount of interest which would have been generated by the sum, use of which was lost as a result of the premature levy’, this being essential in restoring the equal treatment guaranteed by Article 49.73 The claim for interest was not ancillary and a matter for national procedural autonomy, but amounted to ‘the very objective sought by [the claimants]’.74 In FII, another case concerning the recovery of unlawfully charged tax, the CJEU later confirmed that it had intended to lay down a general principle in Metallgesellschaft: [W]here a Member State has levied charges in breach of the rules of Community law, individuals are entitled to reimbursement not only of the tax unduly levied but also of the amounts paid to that State or retained by it which relate directly to that tax. As the court held in . . . Metallgesellschaft, that also includes losses constituted by the unavailability of sums of money as a result of tax being levied prematurely.75

In its recent decision in Littlewoods, the CJEU referred to both Metallgesellschaft and FII when concluding that a ‘Member State [must] repay with interest amounts of tax levied in breach of EU law’.76 It is important to understand that these cases establish that the claimants have a right to recover a sum of interest which reflects the use value of the unlawful tax and that that right arises in recognition of the fact that Member States benefit to the extent of the use value of money by holding on to it during the period between receipt and restitution of the unlawful tax. The amount of the use value of money in any given period is, in all cases, a question of fact. It is, therefore, to some extent a question of national procedural autonomy. This explains the CJEU’s statement in Littlewoods that ‘it is for the internal legal order of each Member State to lay down the conditions in which such interest must be paid, particularly the rate of that interest and its method of calculation (simple or “compound” interest)’.77 This does not give the Member State a discretion to make an award of simple interest in circumstances in which the use value of money is best measured in compound interest terms. It simply enables the Member State to determine whether the best measure is simple or compound interest, on the facts of the case. As the CJEU continued in Littlewoods, national rules ‘must comply with the principles of equivalence and effectiveness’,78 and in Metallgesellschaft (n 5) [87]. ibid [88]. 74 ibid [87]. 75 FII (CJEU) (n 6) [205]. 76 Littlewoods (CJEU) (n 71) [26]. 77 ibid [27]. 78 ibid [27]. 72 73

202 Maximilian Schlote the context of making an award for the time value of money, the principle of effectiveness ‘requires that the national rules referring in particular to the calculation of interest which may be due should not lead to depriving the taxpayer of an adequate indemnity for the loss occasioned through the undue payment of VAT’.79 On the facts of Littlewoods, the claimant taxpayer paid £204,774,763 as VAT over a period of 40 years. It then transpired that the VAT had been unlawfully charged. The claimant received restitution in respect of the principal sum, and simple interest amounting to £268,159,135. On these facts, Advocate General Trstenjak had concluded that the interest payment was not ‘so low that it largely deprived the interest claim stemming from EU law of substance’,80 and that because the interest payment ‘exceeds the principal sum by more than 25%’,81 the UK’s obligation to make restitution of interest was satisfied.82 The Advocate General, in essence, found that because the interest payment was large in absolute terms and relative to the principal, there was no breach of the effectiveness principle. The CJEU’s approach was different. Determination of whether the simple interest payment was adequate on the facts was left to the national court, in accordance with the principle of procedural autonomy. However, in giving guidance on applying the principle of effectiveness, the CJEU stated that the award of interest had to be an ‘adequate indemnity’ for the ‘loss occasioned through the undue payment of VAT’. Thus, the award made should not, as the Advocate General thought, be measured relative to the principal sum, but instead it should be compared to the size of the use value of the overpaid tax. As Vos J indicated in his judgment at first instance, the use value of the overpaid tax in Littlewoods is best measured in compound terms.83 If this is confirmed in evidence in the High Court, then it follows that the claimants in Littlewoods should be awarded the difference between the simple interest they have received so far and the compound interest to which they are entitled. In Littlewoods, the CJEU recognised the important link between the principle of effectiveness, the obligation to make an interest payment, and the fact that that payment is made in respect of the use value of the unlawful tax. This makes sense of the national court’s procedural autonomy in calculating the amount of interest. Rather than following the approach of the Advocate General, which was to give the national court discretion to make a claim available for any interest payment that was more than de minimis, the national court must allow a claim in respect of the sums lost by the claimant, or gained by the government,84 in consequence of the unlawful tax.

(2) Other Sums ‘Directly Related to the Overpaid Tax’ The principle of effectiveness requires restitution of the unlawful charge as well as of the time value of the amount of the charge. In addition, the Member State may benefit at the ibid [29]. Littlewoods (AGO) (n 16) [34]. 81 ibid [37] and [32]. 82 It should be noted that the CJEU also referred to the interest payment exceeding by ‘more than 23%’ the amount of the principal. However, in light of the remainder of the CJEU’s judgment, it is clear that the national court’s task is to assess whether the award of simple interest provides an ‘adequate indemnity’ in respect of the ‘loss occasioned through the undue payment of VAT’: Littlewoods (CJEU) (n 71) [30] and [28]. 83 Littlewoods (Ch) (n 26) [149] (x). 84 The CJEU referred to the sums lost by the claimant. In the text around n 99 below it is argued that the CJEU should have referred to the government’s gain. 79 80

The Principle of Effectiveness 203

expense of the claimant in additional ways as a result of imposing the unlawful charge. A full reversal of the economic consequences of the breach of EU law requires restitution in respect of those benefits as well. The CJEU recognised this in FII. In the passage quoted above,85 the court made it clear that the recovery of the use value of money was merely an example of a broader principle: the claimant is entitled to recovery of the overpaid tax as well as of ‘amounts paid to that state or retained by it which relate directly to that tax’.86 As in Metallgesellschaft, where the CJEU expressly said that Article 49 of the Treaty requires claimants to have a right to interest,87 the CJEU explained this in FII in terms of effectiveness: in order to provide an effective remedy for the Member State’s breach of EU law, the claimant’s remedy had to extend to all of the Member State’s gains.88 Advocate General Geelhoed in FII expressed the requirements of effectiveness in terms of the following ‘underlying principle’: ‘[T]he UK should not profit and companies (or groups of companies) which have been required to pay the unlawful charge must not suffer loss as a result of the imposition of the charge’.89 This reflects the CJEU’s statement in Metallgesellschaft that the claimants should have an effective legal remedy in order to obtain reimbursement . . . of the financial loss which they have sustained and from which the authorities of the Member State concerned have benefited as a result of the advance payment of tax.90

This also mirrors the CJEU’s approach in FII.91 To assess how this operates in practice, it is necessary to look at the facts of FII. The claimants were UK parent companies who received dividends from their foreign subsidiaries. In the first instance, the CJEU held that the charge to Case V corporation tax of those foreign dividends, in circumstances in which domestically-sourced dividends were exempt from corporation tax, may have violated Article 49 of the Treaty.92 Furthermore, the UK parents had to account for Advance Corporation Tax (ACT) on the onward distribution of foreign-sourced dividends when the payment of a domestically-sourced dividend did not attract ACT. This was held to violate Article 49 of the Treaty. Recovery of the unlawful Case V tax and the unlawful ACT, to the extent not set off against lawful Mainstream Corporation Tax, constitutes a straightforward San Giorgio claim for the recovery of overpaid tax. The claimants also sought recovery of tax reliefs, eg capital allowances, which the claimants disclaimed so as to be able to set off more of the unlawful ACT against lawful MCT (waiver claims). Furthermore, the claimants sought restitution in respect of tax reliefs, such as credit in respect of lawful ACT, group relief or management expenses, which they applied to reduce the amount of unlawful Case V MCT (relief claims). Lastly, some of the In the text to n 75 above. FII (CJEU) (n 6) [205]. 87 See the text to n 73. 88 FII (CJEU) (n 6) [204], [205] and ruling 6. 89 ibid [AG132]. 90 Metallgesellschaft (n 5) [96]. 91 FII (CJEU) (n 6) [204], [205] and ruling 6. 92 The Court of Appeal held that a further reference to the CJEU was required to resolve this issue. Foreignsourced dividends were subject to Case V tax but a tax credit was allowed in respect of foreign tax paid. The CJEU is being asked whether in comparing this system with the exemption system applicable to domestically-sourced dividends the Court should look to nominal or to effective tax rates. Advocate General Jääskinen’s Opinion, delivered on 19 July 2012, is inconclusive. He accepts HMRC’s interpretation of the earlier CJEU reference in FII so that the relevant rate is the nominal rate. However, he prefers the finding of Advocate General Geelhoed in FII to the effect that the discrimination between exemption system and credit system is itself in violation of EU law, opening the door to an effective reversal of that decision. 85 86

204 Maximilian Schlote claimants elected for the dividends they paid to their shareholders to be taxed under the foreign investment dividend (FID) regime. This allowed the claimants to recover any surplus ACT. However, it still disadvantaged the claimants compared to dividends paid wholly out of domestically-sourced funds because, first, the claimants had to pay ACT and then recover it, rather than the dividend simply being exempt from ACT and, secondly, because the FID did not carry a refundable tax credit in the hands of the shareholder, unlike a domestically-sourced dividend. That regime, therefore, violated Article 49 of the Treaty. The claimant companies, having adopted the FID regime, felt the need to increase the size of their dividends to take account of the fact that their dividends did not carry the refundable tax credit and the claimants sought to recover in respect of this enhancement (enhancement claims). Clearly, none of the waiver claims, relief claims or enhancement claims was made in respect of money paid pursuant to an unlawful charge, which would be recoverable under a straightforward San Giorgio claim. However, as the CJEU explained in FII, the UK would have to make restitution in respect of these claims if they concerned benefits that were sufficiently closely related to the overpaid tax. The question is, for example, whether the waiver claims are ‘amounts paid to [the UK] or retained by it which relate directly to [unlawful ACT]’, or whether the relief claims are amounts so related to unlawful Case V corporation tax? With one exception, the CJEU denied that any of the waiver claims, relief claims or enhancement claims could fall within the San Giorgio principle. They were insufficiently directly related to the UK’s breach of EU law and did not ‘constitute, on their part, an inevitable consequence’ thereof. Instead, the UK was benefited as a ‘result of decisions taken by [the claimants]’.93 It seems, therefore, that it is not enough that the government received a benefit caused by the unlawful tax. A benefit is only recoverable under the San Giorgio principle if it also constitutes an inevitable consequence of the breach of EU law. This test is, arguably, impossible to satisfy. Nothing is literally inevitable: both claimant and government will usually be able to point to some alternative outcome.94 It is also hard to accept, as the CJEU and the Court of Appeal both seem to have accepted in FII, that the waiver, relief and enhancement claims arose as the ‘result of discretionary tax planning or purely commercial decisions’,95 when the only reason that the waivers occurred, and the reliefs were claimed, was the existence of an unlawful tax charge. In addition, it is hard to draw a distinction between these rejected claims, and the one claim permitted by the CJEU, viz the claim for recovery in respect of lawful ACT set against unlawful MCT.96 The Court of Appeal explained that this part of the relief claims was recoverable under San Giorgio because EU law treats ‘ACT as an advance payment of MCT’.97 However, this does not explain why, in terms of remoteness, setting off a tax credit gained in respect of lawful ACT against unlawful MCT is different from setting off any other tax credit – for example one gained in respect of group relief – against the same unlawful corporation tax. Of course, the two credits are very different from a tax law perspective, but these differences cannot explain why the use of one in FII (CJEU) (n 6) [207]. See the criticism of this test in R Williams, Unjust Enrichment and Public Law (Oxford, Hart Publishing, 2010) 115. 95 FII (CA) (n 24) [149]. See too FII (CJEU) (n 6) [207]. 96 ibid [148]. 97 ibid [148]. 93 94

The Principle of Effectiveness 205

respect of the unlawful tax is sufficiently directly related to the unlawful tax, but not the use of the other. Not only is this test difficult to apply, it also seems to be at odds with the principle of effectiveness. But for the breach of EU law, the claimants would not have used or disclaimed their tax reliefs so that, but for the breach of EU law, the UK would have had to allow those reliefs in later years in accounting for lawful corporation tax. Far from reversing the consequences of a breach of EU law, the ‘inevitable consequence’ test therefore allows the Member State to hold on to benefits obtained in breach of EU law and, in the words of the CJEU in Fantask, it thereby encourages breaches of EU law.98 An approach that was more in line with the principle of effectiveness would focus more explicitly on reversing the benefits that a Member State receives at the claimants’ expense. The need for an abstract remoteness test in FII arose primarily because the CJEU was insufficiently clear about the scope of the San Giorgio remedy. The court frequently referred to the claimants’ ability to recover for their losses.99 For example, in respect of the enhancement claims, the CJEU considered that the ‘loss and damage’ caused by the FID regime was too remote. In light of the claimants’ apparent ability to recover their losses pursuant to a San Giorgio claim, it is not particularly surprising that the CJEU laid down strict remoteness tests. Instead of allowing the claimants to recover some of their losses pursuant to a San Giorgio claim, subject to strict remoteness tests, the CJEU would have done better to recognise that San Giorgio is, in the court’s own words, a claim for ‘restitution’,100 ie that it is limited to allowing the claimants to recover benefits transferred by them to the Member State. It might be thought that the requirement of the principle of effectiveness to reverse the economic consequences of the breach of EU law should extend to the claimant’s losses. However, it should be remembered that in order to recover losses caused by a breach of EU law, a claimant must show, under the principles of state liability, that the breach was ‘sufficiently serious’,101 a condition that is rarely satisfied in tax cases.102 As Arden LJ said in the Court of Appeal in FII, the CJEU ‘established the requirement of a sufficiently serious breach because it considered it just to do so’.103 It would, therefore, be wholly unprincipled to allow recovery of the claimants’ losses under the San Giorgio remedy, given that this remedy is not subject to the requirement that a breach of EU law be sufficiently serious. It follows that the claimants should have been unable to recover any of their losses, but should have been able to recover the whole of the benefit received by the government at their expense. As in the English law of unjust enrichment, the appropriate test to identify recoverable benefits in the government’s hands should have been a ‘but for’ causation test.104 The claimants’ right of recovery and the government’s corresponding liability Fantask (n 37). FII (CJEU) (n 6) [204]; also see Metallgesellschaft (n 5) [88]. 100 Metallgesellschaft (n 5) [82]. 101 Case C-6/90 Francovich v Italy [1991] ECR I-5357; Joined Cases C-46/93 and C-48/93 Brasserie du Pecheur SA v Germany; R v Secretary of State for Transport, ex parte Factortame Ltd [1996] ECR I-1029. 102 FII (CJEU) (n 6). Advocate General Geelhoed explained this on the basis that ‘boundaries of the application of the Treaty free movement provisions in the field of direct taxation [are] extremely complex and, in parts, in the process of development’ (AG Geelhoed’s opinion, para 137]); confirmed by the CJEU at [212]–[217]; confirmed by the Court of Appeal in FII (CA) [206]. 103 FII (CA) (n 24) [137]. 104 Goff & Jones (n 25) ch 6. The authors consider that the ‘but for’ test is well-established in two-party cases and advocate the adoption of the same test for three-party cases. 98 99

206 Maximilian Schlote should have been limited to the value of the government’s enrichment, and in that case, there would have been no need for the CJEU to worry about uncapped liability. Applying this reasoning to the waiver, relief, and enhancement claims, it is clear that the first two claims should have fallen within the San Giorgio principle. They can be analysed together. But for the imposition of the unlawful charge, the reliefs would still have been in the claimants’ hands. They could then have been set against other lawful tax liabilities, either immediately or in later years. HMRC was, therefore, enriched to the extent that it received more tax. Although Arden LJ held otherwise in FII,105 the legality of the tax received in later years cannot have been a bar to the claimants’ recovery, because recovery would not have been in respect of that later, lawful tax, but in respect of the waivers and reliefs. When the claimants disclaimed and used reliefs, they absolved HMRC from liability to allow a credit against lawfully due tax. The value of that enrichment depended on what was lawfully due in later years, but the lawfulness of that later tax should not have affected the right to the restitution in respect of the earlier enrichment. Furthermore, DMG shows that the legality of the government’s receipt of a tax payment is not itself an answer to a San Giorgio claim when there is an earlier, causative illegality. In DMG, the illegality related to the fact that the claimants were denied the opportunity to make a group income election which would have exempted them from ACT. Since the election was never made, the claimants’ ACT payment was lawfully due, but this payment could nonetheless be recovered. The CJEU’s refusal to include the enhancement claims in the San Giorgio remedy is more defensible. The FID regime violated EU law because of the denial to make a tax credit available to the shareholders. This refusal caused the shareholders a loss which they suffered regardless of the size of the dividend. The argument that the enhancement of the dividend should be viewed as saving HMRC a necessary expense because it saved HMRC from having to provide the tax credit to the shareholders ignores the fact that the shareholders continued to have their own claim against HMRC in respect of HMRC’s discriminatory failure to make available a tax credit on the dividends, however enhanced. The enhancement of the dividend did not, therefore, constitute a benefit to HMRC; at most, it amounted to a loss in the hands of the claimants. Giving the claimants a right of restitution in respect of all the gains which accrued to the Member State at their expense in consequence of the charging of an unlawful tax, and applying a simple ‘but for’ test of causation in establishing this, fully reverses the Member State’s enrichment. So interpreted, the San Giorgio principle adequately addresses the economic consequences of the breach, thus satisfying the principle of effectiveness, without violating the distinction between the San Giorgio principle and the recovery of losses pursuant to the principle of state liability.

E. CONCURRENCY OF REMEDIES

This chapter begun by stating that while the principle of effectiveness requires Member States to make restitution of tax charged in contravention of EU law, it leaves enforcement of that right to the legal systems of Member States. The foregoing discussion has shown the FII (CA) (n 24) [178]–[184].

105

The Principle of Effectiveness 207

limited scope of that procedural autonomy in respect of setting the parameters of the right of recovery, in relation to whether the claimant can recover, who can recover, and how much the claimant can recover. Before concluding, this Part will examine the effects of European law on the kind of remedy that a Member State must make available to the claimant in a San Giorgio action. Claimants which pursue the UK government for recovery of tax under the San Giorgio principle typically frame their claim as a claim in unjust enrichment.106 The tax payment is at the expense of the claimant and it enriches the UK government. The claimant then has to establish the ‘unjustness’ of the payment by reference to an unjust factor. The relevant unjust factors in an overpaid tax case are mistake of law107 and a claim under the rule in Woolwich Equitable Building Society v IRC.108 In FII, for example, the claimants mistakenly believed that the ACT they had to pay on the onwards distribution of foreign-sourced dividends was lawfully due. Hence they could make a claim founded on mistake. At the same time, since the tax was not lawfully due, the government had received the claimants’ payments without statutory authority. Hence, as the Court of Appeal and the Supreme Court confirmed, the claimants could also make a claim founded on the Woolwich principle. Similarly, in Littlewoods, both claims were available. In ITC, the claimants did not pursue a Woolwich claim but relied solely on the mistake claim. A claimant pursuing a claim in unjust enrichment grounded purely in domestic law, ie a claim that does not seek to enforce a San Giorgio right, may choose to pursue either a mistake claim or a Woolwich claim, or may bring both.109 This is so even in circumstances in which the mistake claim is subject to a more favourable limitation period, such as that established by the Limitation Act 1980, section 32(1)(c). It follows from an application of the principles of effectiveness and equivalence that a similar choice is available to the claimants in FII, Littlewoods and ITC. As noted earlier, it is for each Member State to lay down the procedural rules governing actions for safeguarding rights which individuals derive from EU law, including rights to recover taxes charged in violation of EU law. Such procedural autonomy is, however, subject to the principles of equivalence and effectiveness. According to the principle of equivalence, the procedural rules governing EU law claims must not be less favourable than those governing similar domestic actions. A claim for the recovery of taxes levied by a Member State in breach of EU law is similar to a claim for restitution of taxes levied contrary to domestic law.110 In many cases, the recovery of taxes is governed by the common law,111 and, as a matter of the common law of unjust enrichment, a claimant pursuing a claim for the recovery of taxes charged unlawfully under English law can choose freely between the mistake claim and the Woolwich claim. As Lord Hope said in DMG and FII, this concurrency of remedies creates an advantage for the claimant: ‘In fairness, the claimant ought to be free to choose the remedy that best suits his case’.112 When choosing between a mistake claim and a Woolwich claim, many claimants 106 Examples are Littlewoods (Ch) (n 26), FII (SC) (n 18), and ITC (n 14). For the elements of an unjust enrichment claim, see Banque Financière (n 20) 227. 107 Recognised in English law since Kleinwort Benson Ltd v Birmingham City Council [1997] QB 380 (CA). 108 Woolwich (n 24). 109 DMG (n 22). See too FII (SC) (n 18) [21] (Lord Hope). 110 FII (SC) (n 18) [219]. 111 Some English statutes also provide for recovery, eg the Taxes Management Act 1970, s 33, but none of these applied in FII: FII (SC) (n 18) [219]. 112 FII (SC) (n 18) [21]; DMG (n 22) [51].

208 Maximilian Schlote wish to choose the mistake claim in order to take advantage of a longer limitation period that applies to claims of this kind. It follows, therefore, from the principle of equivalence that ‘the same grounds of action, and the same freedom of choice, must equally be available in analogous circumstances to a person seeking to recover taxes which have been levied in breach of EU law’.113 Once it is admitted that a choice of remedies is available to claimants pursuing a claim for the recovery of taxes levied unlawfully under domestic law, and that that choice of remedies confers an advantage on the claimants, that choice of remedies must be made available to claimants pursuing a San Giorgio claim.114 As the CJEU said in Rewe-Handelsgesellschaft Nord mbH v Hauptzollamt Kiel, ‘it must be possible for every type of action provided for by national law to be available for the purposes of ensuring observance of Community provisions’.115 There might not be an EU law principle of concurrency of remedies,116 but the principle of equivalence has the effect of importing the freedom of choice given to domestic claimants by DMG into European law. In FII, the claimants were pursuing a mistake claim and a Woolwich claim. The former claim was subject to the extended limitation period in section 32(1)(c). However, without making transitional arrangements, the government curtailed that limitation period by the Finance Act 2004, section 320, and the Finance Act 2007, section 107. In agreement with the argument presented above, five of their Lordships came to the provisional conclusion that the mistake claim had to be available to the claimants as a matter of EU law and that the lack of transitional period meant that the curtailment of the limitation period in sections 320 and 107 violated the principle of effectiveness.117 However, Lord Sumption118 found that the mistake claim did not have to be available as a matter of EU law and that, therefore, no question as to its effectiveness arose.119 In light of this disagreement, the issue was referred to the CJEU. Lord Sumption found that because the Woolwich claim provided the claimants with an effective remedy in respect of their San Giorgio claim, there was no need for a mistake claim, or any other claim, to be made available as a matter of EU law.120 However, this merely addresses the question of whether English law would have satisfied the requirements of the principle of effectiveness if ‘English law had evolved differently, and the ground of action based on mistake had not been available’.121 The effectiveness of the Woolwich claim cannot affect the availability of the mistake claim pursuant to the equivalence principle. A similar error undermines Vos J’s reasoning in Littlewoods. In Littlewoods VATA, sections 78 and 80 had the effect of excluding the mistake and the Woolwich claim for the recovery of compound interest.122 If, as argued in this chapter, EU law requires compound interest to be made available, EU law also requires that the effect of sections 78 and 80 be FII (SC) (n 18) [220]. Littlewoods (AGO) (n 16) [58]. 115 Case 158/80 Rewe-Handelsgesellschaft Nord mbH v Hauptzollamt Kiel [1981] ECR 1805 [44]. 116 FII (SC) (n 18) [196]. 117 ibid [22] (Lord Hope), [120] (Lord Walker), [126]–[139] (Lord Clarke), [140] (Lord Dyson), and [246] (Lord Reed). 118 With whom Lord Brown agreed: ibid [121]–[125]. 119 ibid [142]. 120 ibid [142]. 121 ibid [224]. 122 The claimants in ITC are facing a similar issue. 113 114

The Principle of Effectiveness 209

lifted from both claims. It is not enough, pace Vos J,123 that the restriction be lifted from the Woolwich claim. Even if the Woolwich claim alone can provide the claimants with an effective remedy, such an approach would violate the principle of equivalence.124 The first step is to identify a comparable remedy. The claimants have, on the argument presented in this chapter, a right to compound interest. Compound interest as a reflection of the use value of money of overpaid tax is, in English law, recovered in a claim for unjust enrichment.125 As we have seen, the English law of unjust enrichment gives claimants a free choice between a mistake claim and a Woolwich claim and that choice must be extended to the claimants in Littlewoods. Support for this approach can be found in Advocate General Trstenjak’s opinion and the CJEU’s decision in Littlewoods.126 The CJEU referred to the principle of equivalence127 before adding that a national court ‘when faced with a rule of law that is incompatible with directly applicable EU law . . . is required to disapply that national rule’.128 Ergo, sections 78 and 80 – which prevent the claimant from exercising their EU right to compound interest – must be disapplied. Of course, the obligation to disapply does not ‘restrict the power of the competent national courts to apply, amongst the various procedures of the internal order, those which are appropriate to safeguard the individual rights conferred by EU law’.129 This does not, however, give the national court a power to exclude remedies which have to be available to the claimants as a matter of the principle of equivalence. The CJEU in Littlewoods referred to Ministero delle Finanze v IN.CO.GE.’90 Srl,130 where the court explicitly stated that the national court, in exercising this procedural autonomy to apply the ‘appropriate’ domestic remedy, remained subject to the principles of effectiveness and equivalence.131 In IN.CO.GE.’90 the claimants argued that the fact that a tax violated EU law meant that the tax loses its fiscal character and that the usual rules applicable to the recovery of taxes should not apply. These facts illustrate the scope of the national court’s power to identify an ‘appropriate’ remedy. It was acceptable for the Italian courts to apply the special tax recovery rules to the exclusion of the general unjust enrichment claim because even for claimants pursuing a claim for the recovery of taxes unlawfully levied under Italian law, the general law of unjust enrichment was not available to recover the tax. In those circumstances, the principle of equivalence does not require the general claim to be made available. In other words, IN.CO.GE.’90 would apply to FII and Littlewoods if the mistake claim was also not available to a claimant pursuing a claim for the recovery of tax or compound interest which arose as a result of tax unlawfully levied under domestic law. The reasoning of Vos J in Littlewoods, the Court of Appeal in FII, and the minority of the Supreme Court in FII depends on the idea that because a Woolwich claim is effective, a mistake claim does not have to be made available or tested against the rules of equivalence and effectiveness. It seems that the courts think of Woolwich as a domestic equivalent of the San Giorgio principle, the Woolwich claim being described as more accurately and Littlewoods (Ch) (n 26) [92] (ii). Advocate General Trstenjak confirmed the applicability of the principle of equivalence to this issue: Littlewoods (AGO) (n 16) [58]. 125 Sempra (n 21). 126 Littlewoods (AGO) (n 16) [58]. 127 Littlewoods (CJEU) (n 71) [31]. 128 ibid [33]. 129 ibid [33]. 130 ibid [33]. See n 131 below. 131 Joined Cases C-10/97 to C-22/97 Ministero delle Finanze v IN.CO.GE.’90 Srl and Others [1998] ECR I-6307 [26] and [29]. 123 124

210 Maximilian Schlote comprehensively giving effect to the San Giorgio principle than a mistake claim.132 The Court of Appeal went so far as to say that even if a Woolwich claim were not an effective remedy because it applied only if HMRC had actively demanded the unlawful tax, the court would remove the requirement for a demand in the San Giorgio context, rather than allow the claimants to pursue a mistake claim that would otherwise have to be made available.133 This approach was wholly misguided. The principle of effectiveness is not a limiting principle. The idea is not that the CJEU determines what precisely the claimant can recover. Instead, it lays down the minimum requirements of the principle of effectiveness, a principle which finds no equivalent in domestic law. The CJEU determines simply what any claimant in any Member State must recover – at a minimum. But it does not describe the extent of recovery beyond that. The whole point of the principle of equivalence is that if a comparable domestic claimant can recover more than what is required at a European level by the principle of effectiveness, then the San Giorgio claimant must recover to that extent, not to the minimum extent guaranteed by EU law. It is, therefore, contrary to EU law to attempt to ‘match’ a particular remedy to San Giorgio, to try to create a domestic version of San Giorgio. Whilst the Advocate General’s and the CJEU’s decision in Littlewoods offer support for this position, however, final resolution of the concurrency of remedies issue will have to wait until the CJEU has given judgment in the reference by the Supreme Court in FII.

F. CONCLUSION

The overpaying taxpayer has contributed significantly to the development of the EU law principle of effectiveness in recent years. However, despite or perhaps because of these developments, the CJEU has not yet achieved complete conceptual clarity and significant uncertainty remains in relation to almost all of the issues discussed in this article. This article has sought to provide an answer to many of these issues by employing a concept of the EU principle of effectiveness which focuses on reversing the economic consequences of a breach of EU law whilst respecting the boundary between the principle of state liability and the right to repayment pursuant to San Giorgio. In light of the principle of national procedural autonomy, the identification of a coherent underlying principle of the right of repayment goes far beyond mere academic interest. The decision of the Supreme Court in FII to refer the concurrency of remedies issue to the CJEU illustrates this well: despite a five to two majority in favour of the outcome advocated in this chapter, their Lordships felt the need to make a reference to the CJEU. To avoid such uncertainty, the CJEU should more clearly enunciate the principles guiding the right to repayment. It will have the opportunity to do so in FII.

Littlewoods (Ch) (n 26) [90]. FII (CA) (n 24) [174].

132 133

11 Absence of Basis: A German Perspective ANNE SANDERS*

A. INTRODUCTION

Since the House of Lords’ decision in Woolwich Equitable Building Society v IRC 1 was handed down, questions have been raised as to how the case should be analysed. Was it an unjust factor which rendered the enrichment of the Revenue unjust, or was it the lack of any legal justification for the payment? Today, this question is not only at the heart of claims for restitution of unlawfully demanded taxes, as for example in the 2012 Supreme Court decision in Test Claimants in the FII Group Litigation v HMRC,2 but it is of major importance for the whole of the law of unjust enrichment. The swaps litigation convinced the late Professor Birks3 to reinterpret the English law of unjust enrichment as being based on an absence of basis approach, and courts4 and academics5 alike are questioning whether this new approach should be followed. Cases such as Woolwich or FII could be decided under a general ‘absence of basis’ approach. However, this chapter asks whether such an interpretation is desirable. It takes Woolwich and FII as a starting point, and tries to explain the origins and difficulties of the * I wish to thank Miss Eleanor Campbell and Mr William Swadling for their helpful comments on previous drafts of this chapter. I also wish to thank Dr Birke Häcker, Dr Steven Elliott, Mr Richard Hart and Professor Charles Mitchell for making this book possible. 1 Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] AC 70 (HL) (hereafter ‘Woolwich’). 2 Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2012] UKSC 19, [2012] 2 AC 337 (hereafter ‘FII ’). 3 P Birks, Unjust Enrichment, 2nd edn (Oxford, OUP, 2005) 108–17. 4 Deutsche Morgan Grenfell Group plc v Inland Revenue Commissioners [2006] UKHL 49, [2007] 1 AC 558 (hereafter ‘DMG’) [21]–[22], [28] (Lord Hope), [150]–[158] (Lord Walker); FII (n 2) [81] (Lord Walker), [162], [173], [188]–[190] (Lord Sumption); Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2010] EWCA Civ 103, [2010] STC 1251, [156]. 5 T Krebs, ‘In Defence of Unjust Factors’ in D Johnson and R Zimmermann (eds), Unjustified Enrichment − Key Issues in Comparative Perspective (Cambridge, CUP, 2002) 76; A Burrows, ‘Absence of Basis: The New Birksian Scheme’ in A Burrows and Lord Rodger of Earlsferry (eds), Mapping the Law: Essays in Memory of Peter Birks (Oxford, OUP, 2006) 33; G Dannemann, ‘Unjust Enrichment as Absence of Basis: Can English Law Cope?’ in A Burrows and Lord Rodger of Earlsferry (eds), Mapping the Law: Essays in Memory of Peter Birks (Oxford, OUP, 2006) 363; B Häcker, ‘Still at the Crossroads’ (2007) 123 LQR 177, 182; S Meier ‘No Basis: A Comparative View’ in A Burrows and Lord Rodger of Earlsferry (eds), Mapping the Law: Essays in Memory of Peter Birks (Oxford, OUP, 2006) 343; H Scott, ‘Restitution of Extra-Contractual Transfers: Limits of the Absence of Legal Ground Analysis’ [2006] Restitution Law Review 93; R Stevens, ‘Is there a Law of Unjust Enrichment’ in S Degeling and J Edelman (eds), Unjust Enrichment in Commercial Law (Sydney, Thomson Reuters, 2009) 11; T Baloch, ‘The Unjust Enrichment Pyramid’ (2007) 123 LQR 636; Canadian perspectives: M McInnes, ‘The Test of Unjust Enrichment in Canada’ (2007) 123 LQR 34, 37; L Smith, ‘The Mystery of Juristic Reason’ (2000) 12 Supreme Court Law Review

214 Anne Sanders absence of basis approach as it is found in Germany. The chapter focuses on the German law of unjust enrichment because German law is often cited as an example of an absence of basis model.6 Moreover, the German law of unjust enrichment was of great interest to Professor Birks, who was inspired by comparative work on the German law of unjust enrichment by academics such as Professor Zimmermann,7 Professor Meier,8 Professor Dannemann,9 and Dr Krebs.10 This chapter aims to explain some of the problems an absence of basis approach faces, both in theory and practice, and concludes that its adoption would introduce more problems into English law than it would solve.

B. WOOLWICH AND FII – ABSENCE OF BASIS CASES?

In Woolwich, the House of Lords held that tax paid in accordance with an ultra vires statutory instrument was recoverable. The question what (if any) unjust factor was applied has been debated ever since.11 The possible ground for restitution discussed in this chapter was given by Lord Browne-Wilkinson. He said that a ‘stream of authority is founded on the concept that money paid under an ultra vires demand for tax or other impost has been paid without consideration’.12 The term ‘no consideration’ was used again in Westdeutsche Landesbank v Islington LBC13 to justify restitution under a closed swap agreement. Neither Lord Browne-Wilkinson nor Hobhouse J explained, however, what they meant by ‘consideration’. The term consideration was certainly not used in its contractual sense. Even if one were to apply the notions of John Locke and Jean Jacques Rousseau, according to whom the state has its origins in an agreement concluded among its subjects, it would be difficult to explain even lawfully demanded taxes as payments made pursuant to a bargain. Moreover, there was certainly consideration in the contractual sense, in that there was a promise in exchange for a promise in the swaps cases. In the law of unjust enrichment, the term ‘consideration’ has long been used as the English translation of the Latin term ‘causa’.14 Causa is one of the vaguest terms of Roman

(2d) 211; L Smith, ‘Demystifying Juristic Reasons’ (2007) 45 Canadian Business Law Review 281; C Hunt, ‘Unjust Enrichment Understood as Absence of Basis: a Critical Evaluation with Lessons from Canada’ (2009) 6 Oxford University Comparative Law Forum: ouclf.iuscomp.org/articles/hunt.shtml. 6 Scott (n 5); Hunt (n 5) II (i); J Edelman and E Bant, Unjust Enrichment in Australia (Oxford, OUP, 2006) 159–61; G Dannemann, The German Law of Unjustified Enrichment and Restitution (Oxford, OUP, 2009). 7 R Zimmermann, ‘Unjustified Enrichment: The Modern Civilian Approach’ (1995) 15 Oxford Journal of Legal Studies 403. 8 S Meier, ‘Unjust Factors and Legal Grounds’ in D Johnson and R Zimmemann (eds), Unjustified Enrichment – Key Issues in Comparative Perspective (Cambridge, CUP, 2002) 37; S Meier, Irrtum und Zweckverfehlung (Tübingen, Mohr Siebeck, 1999). 9 Dannemann (n 5). 10 T Krebs, Restitution at the Crossroads (London, Cavendish Press, 2001); Krebs (n 5). 11 E McKendrick, ‘Restitution for Unlawfully Demanded Tax’ [1993] Lloyds Maritime and Commercial Law Quarterly 88, 93–98. For a further contribution to this debate see the discussion by Charlie Webb in chapter 5 of the present volume. 12 Woolwich (n 1) 197. 13 Westdeutsche Landesbank Girozentrale v Islington London Borough Council [1994] 1 WLR 938 (CA). 14 Birks (n 3) 117.

Absence of Basis: A German Perspective 215

jurisprudence,15 and means reason, purpose or motive. In the realm of the law of unjust enrichment, the term ‘consideration’ is regularly found in the context of the unjust factor ‘failure of consideration’. This unjust factor also has its origins in Roman law, in the condictio causa data causa non secuta.16 Lord Birkenhead translated this as an ‘action to recover something given for a consideration which has failed’.17 By ‘consideration’, Lord Birkenhead meant the purpose, basis or condition upon which the thing had been given.18 Thus, the unjust factor of failure of consideration is applicable when the basis upon which a payment was made subsequently fails. However, this was not the issue in the Woolwich case. In such a case, ‘no consideration’ can only mean that a benefit was transferred which was not due. Professor McKendrick has observed that the acceptance of this meaning of ‘no consideration’ would be tantamount to the introduction of a condictio indebiti.19 Professor Dannemann has gone one step further, arguing that English law embraces an absence of basis approach for ultra vires public authority claims.20 There is certainly something to say in favour of such an analysis. The speeches of Lords Goff, Browne-Wilkinson and Slynn seem to turn on the fact that the tax was not due. However, their Lordships did not state that all payments not due had to be refunded. On the contrary, Lord Goff said that the English law of unjust enrichment had not developed a condictio indebiti. As Lord Sumption stated in FII,21 the case has also not been understood subsequently to provide authority for a general absence of basis approach. English case law would look much different if this had been the case. The Woolwich principle can be understood as an unjust factor developed for a special policy reason.22 According to Lord Goff, restitution is justified by the fundamental idea expressed in the Bill of Rights that no public authority may extract money without a lawful right to do so.23 The Canadian Supreme Court subsequently held in Kingstreet Investments v New Brunswick24 that such claims belong not in unjust enrichment but in the realm of public law. In England, however, a completely different route was taken, with the House of Lords holding in Deutsche Morgan Grenfell v IRC 25 that Woolwich claims could be brought in private law, alongside claims under mistake of law. Nevertheless, it must be conceded that a claim under the Woolwich principle has noticeable public law elements.26 On the interpretation of the Supreme Court in FII v HMRC,27 a Woolwich-claim does not even require that taxes were paid pursuant to a demand. It is sufficient that tax was A Berger, Encyclopedic Dictionary of Roman Law (Clark NJ, Lawbook Exchange Ltd, 1953 [reprint 2004]) 382. See Sir Walter Evans, ‘Essay on the Action for Money had and Received’ (1802), reprinted [1998] Restitution Law Review 9–11; Birks (n 3) 118. 17 Cantiere San Rocco SA v Clyde Shipbuilding and Engineering Co [1924] AC 226 (HL) 235. 18 Edelman and Bant (n 6) 242. 19 McKendrick (n 11) 95. 20 Dannemann (n 6) 36, 202. 21 FII (n 2) [162] (Lord Sumption). 22 Burrows (n 5) 33, 43. 23 P Birks, ‘Restitution from the Executive: A Tercentenary Footnote to the Bill of Rights’ in P Finn (ed) Essays on Restitution (London, Sweet & Maxwell, 1990) 164; see also J Beatson ‘Restitution of Taxes, Levies and Other Imposts’ (1993) 109 LQR 401, 411–12. 24 Kingstreet Investments Ltd v New Brunswick (Department of Finance) 2007 SCC 1, [2007] 1 SCR 3 (hereafter ‘Kingstreet’); compare L Smith, ‘Public Justice and Private Justice: Restitution after Kingstreet’ (2008) 46 Canadian Business Law Journal 11; see also J Alder ‘Restitution in Public Law: Bearing the Cost of Unlawful State Action’ [2002] Legal Studies 165. See also Robert Chambers’ contribution in chapter 14 of the present volume. 25 DMG (n 4). 26 R Williams, Unjust Enrichment and Public Law (Oxford, Hart Publishing, 2010); see also C Mitchell and P Oliver, ‘Unjust Enrichment and the Idea of Public Law’ in R Chambers, C Mitchell and J Penner (eds), Philosophical Foundations of the Law of Unjust Enrichment (Oxford, OUP, 2009) 394. 27 FII (n 2). 15 16

216 Anne Sanders received which was not lawfully due.28 This focus on the absence of a legal claim in FII’s interpretation of Woolwich could indeed be understood as establishing an absence of basis approach in relation to claims for restitution of overpaid tax. It is difficult to fully understand the Woolwich principle as established in FII, however. Lord Sumption, speaking of ‘unlawful exaction’ rather than ‘unlawful demand’,29 stated that the Woolwich principle was ‘probably not available’ where there was ‘no unlawful exaction but the taxpayer has simply paid in error: eg he has miscalculated his liability under a self-assessed tax or accidentally paid twice’.30 This statement suggests that Lord Sumption assumed that the Woolwich claim was available only if the taxpayer paid in order to avoid an enforcement of the tax by public authorities and not just because the state had no lawful claim to the taxes paid. If an absence of basis approach had been adopted in relation to overpaid tax, however, it should be irrelevant whether the taxpayer had paid twice mistakenly or because he was afraid of the revenue authorities. Even if FII could be interpreted as adopting an absence of basis approach in relation to overpaid taxes, however, this would say nothing about a general switch of the English law of unjust enrichment to an absence of basis approach. FII seems to lead in a different direction. Lord Sumption remarked with respect Lord Goff’s speech in Woolwich that English law still had not developed a condictio indebiti.31 Lord Walker, who expressed his hopes in Deutsche Morgan Grenfell v IRC 32 that English law might develop along the lines suggested by Professor Birks, did not even comment on a possible development towards absence of basis in FII, even though Mr Rabinowitz QC had brought up the point.33 Thus, it seems as if a switch to a general absence of basis approach is more unlikely than it was in 2006 when Deutsche Morgan Grenfell v IRC 34 was decided. Nevertheless, the debate is not over yet. The late Professor Peter Birks argued that if an absence of basis approach had been applied in the first place, the current struggle to define the grounds and scope of the Woolwich principle could have been avoided.35 Moreover, there are powerful voices advocating a switch to an absence of basis approach. In addition to Professor Peter Birks, Professor Robert Stevens36 has argued that absence of basis, though no sufficient ground for recovery,37 is a necessary element of every claim and forms the ‘golden thread’ which links together all claims in unjust enrichment. He argues that it would make English law more coherent if it embraced an absence of basis approach which could provide a single explanation for all unjust enrichment claims rather than a patchwork of different unjust factors. The purpose of this chapter is to discuss some issues in relation to the question whether Professors Birks and Stevens are right, and whether their Lordships should therefore have adopted a general absence of basis analysis in Woolwich and in FII.

FII (n 4) [157]. FII (n 2) [173], [186]. 30 ibid [186]. Many thanks to Dr Birke Häcker for discussing this issue with me. 31 ibid. 32 DMG (n 4) [158]. 33 FII (n 2) [81]. 34 DMG (n 4). 35 Birks (n 3) 134. 36 Stevens (n 5) 11; Baloch (n 5). 37 Stevens (n 5) 12–13, 30–33. 28 29

Absence of Basis: A German Perspective 217

C. THE ORIGINS OF ABSENCE OF BASIS

The absence of a legally enforceable claim might be said to be at the heart of the Woolwich principle, because the Bill of Rights provides that no citizen may be taxed without Parliamentary consent.38 But this explanation does not hold true for claims against private defendants. It must therefore be asked why the notion of absence of legal basis became so important in Germany and why enrichments should require bases at all to be safe from claims for restitution. The answer to these questions is more difficult than one might expect and requires a brief excursus into legal history.

(1) Roman Condictiones In Roman law, there were a number of condictiones, actions by which a certain thing or a sum of money could be claimed. The formula of the claim did not state any cause or ground, but simply alleged that the defendant was under a duty to convey to the claimant a certain sum of money or a thing. Among the applications of the condictio were the enforcement of written contracts, the notion of an enforceable claim arising from all purely oral agreements having not yet developed. Another role for the condictiones was to enable restitutionary relief. Such condictiones included the condictio indebiti, a condictio to recover a payment made in the mistaken belief that it was owing.39 There was also the condictio causa data causa non secuta40 for the recovery of property transferred, where the purpose for the transfer had failed,41 and the condictio furtiva, which granted a restitutionary claim against a thief. When the notion of enforceable agreements outside a restricted number of special contracts later took hold, the sum of the condictiones was summarised by Justinian and later in the ius commune under the heading of ‘quasi-contract’, the reason being that they granted the same kind of claim as a contract but without the conclusion of a binding agreement.42 From the beginning, there were attempts to find a common denominator for all the enrichment condictiones. One such attempt was made by Pomponius, who said ‘it is natural justice that nobody can enrich himself at the expense of another’. As a less abstract denominator, Papinian said that the common root of all the enrichment condictiones was that the enrichment was ‘sine causa’.43 Justinian included the condictio sine causa in the Digest under a special head, which might have served to be used to solve cases which did not fall under the other condictiones. It should be recalled, however, that the word causa is one of the vaguest terms of Roman law,44 which makes it doubtful that a precise idea of legal basis existed. In the ius commune – Roman law applied in Europe before the great codifications – the condictio indebiti grew in importance, while the position of the condictio sine causa remained uncertain.45 See Birks (n 23). R Zimmermann, The Law of Obligations (Oxford, OUP, 1996) 848–49. 40 Digest. 12.4.0. 41 B Nicholas, An Introduction to Roman Law (Oxford, Clarendon Press, 1962) 229–30. 42 Zimmermann (n 39) 837. 43 Zimmermann (n 39) 854–55. 44 Berger (n 15) 382. 45 Zimmermann (n 39) 866–67. 38 39

218 Anne Sanders (2) Grotius and Savigny The natural lawyer Hugo Grotius was among the first to develop the abstract notion of enrichment and the idea that enrichments obtained without a ‘legal title’ had to be given up.46 However, apart from stating that there would be no legal basis where there was no antecedent gift or contract, he did not analyse the concept of legal title further.47 In the early nineteenth century, in a not yet unified Germany, Roman law in the form of the ius commune was applied in many areas and was the core subject of academic research. Professor Friedrich Carl von Savigny was convinced that there should be no codification of German private law before the existing law was systematised.48 From the untidy mess of Roman law, he sought to extract the terms and structure that had developed organically over time without the detrimental intervention of a legislator. Consequently, he wished to detect the one rationale unifying all the condictiones and to emancipate this area of law from quasi-contracts and the notion of fairness expressed by Pomponius. Looking at the condictio sine causa, he became convinced that the unifying characteristic of the quasicontracts was an absence of contracts. The common feature of all condictiones, he argued, was the purposeful enlargement of the assets of one party by way of a diminution of the assets of another, leading to a state of unjustified having: unjustified because a legal basis, a causa, the purpose for this shift of assets, had either not existed in the first place or had subsequently fallen away.49 Professor Savigny was aware that the notion of legal basis was crucial for the development of his system. However, he deferred working on this difficult problem until his next book, which he unfortunately never finished.50 It can be deduced from his work, however, that he saw at the heart of the condictio indebiti the transfer of an enrichment in the mistaken belief that a certain purpose could be achieved by it. This still left the problem of how the condictiones, which justified restitution in cases where the defendants had received the enrichment without a wilful act of the claimant, for example the condictio furtiva, might be included in the picture. The outcome is a beautiful though still incomplete picture, as the notion of legal basis was not fully developed by either Grotius nor Savigny. The starting point of absence of basis seemed not to have been the normative force of the argument that every enrichment required a basis, because at that time there was no complete notion of what a legal basis was. Its aim was rather to detect a unifying denominator of the Roman condictiones. However, two possible approaches can be detected from the work of these two pioneers. The first is an approach which was alluded to by Grotious’ remark that gifts or contracts provide legal bases. This first, ‘objective’ approach assumes that there are certain legal bases, for example contractual obligations or gifts, which are the same for every person. The other ‘subjective’ approach was foreshadowed by Savigny, who spoke of the individual purpose to be achieved by the transferor as the legal basis of the enrichment.51 The vague F Schäfer, Bereicherungsrecht in Europa (Berlin, Duncker & Humblot, 2001) 93–94. Scott (n 5) 93, 95. 48 F Carl von Savigny, ‘Vom Beruf unserer Zeit für Gesetzgebung und Rechtswissenschaft’, reprinted in Hans Hattenhauer (ed), Thibaut und Savigny − Ihre Programmatischen Schriften (München, Vahlen, 1973) 125–26. 49 Zimmermann (n 39) 872–73; Schäfer (n 46) 174–78. 50 Schäfer (n 46) 151. 51 A discussion of both approaches will be undertaken below. 46 47

Absence of Basis: A German Perspective 219

term causa could both point to the individual purpose of a person making a payment and to the existence of an obligation that demanded a certain payment be made.

(3) Absence of Basis in the German BGB It was left to the next generation of academics to develop Savigny’s ideas further and to replace the ius commune with the German Civil Code, the Bürgerliche Gesetzbuch (BGB). Professor Bernhard Windscheid, a highly influential scholar of Roman law and draftsman of the BGB, introduced the term ‘unjustified enrichment’ and said that enrichment and absence of legal ground were at the heart of this area of law. He argued that there was no one formula to detect the legal basis of a transfer. A distinction had to be drawn between cases where the enrichment was transferred deliberately from the claimant to the defendant and other forms of enrichments, for example, where the defendant enriched himself by the unauthorised use of the claimant’s property.52 The starting point of the BGB was the Roman condictiones. During the drafting process, more and more of the traditional condictiones were summarised in § 812 I 1 BGB, thereby combining the performance-based condictiones such as the condictio indebiti with condictiones such as the conditio furtiva, where the enrichment was obtained without a deliberate act of the claimant.53 § 812 I 1 BGB is often cited when an absence of basis model is discussed. It reads, ‘[a] person who obtains something by performance by another person [first alternative] or in another way [second alternative] at the expense of this person without legal ground is bound to give it up to him’.54 However, the elegant abstraction of this absence of basis claim conflicts with a piecemeal approach in some of the provisions immediately following § 812 I 1 BGB. §§ 812–822 BGB still contain some Roman condictiones which could be described as applying an unjust factor and not an absence of basis approach. Thus, § 812 I 2 BGB provides a right to restitution where performance was aimed at a result which did not occur; the old condictio ob causa data causa non secuta. This section is closer to the English unjust factor of failure of consideration than absence of basis. § 816 BGB gives a right to recover the purchase price a defendant obtains by selling the claimant’s property to a bona fide purchaser. The section also allows the recovery of the property itself in case an unauthorised person made a gratuitous disposition to a recipient in good faith.55 Moreover, § 817 1st sentence BGB provides that, ‘[i]f the purpose of a performance was defined in such a way that by his acceptance the recipient violated a statutory prohibition or offended good morals, the recipient is obliged to provide restitution’.56 Since many of the cases where § 817 1st sentence BGB could be applied are also covered by § 812 I 1 BGB, its scope is doubtful today.57 However, it cannot be denied that the section does not refer to

52 B Windscheid, Pandektenrecht, 9th edn with revisions by T Kipp (Aalen, Scientia, 1906 [reprint 1963]) vol 2, § 422, 871. 53 Schäfer (n 46) 309–10. 54 Translation by Dannemann (n 6) 308, indications for the first and second alternatives added. 55 Thus, title can be obtained in good faith even if no consideration is given. Only the action in unjust enrichment, a right in personam, not in rem, allows recovery. 56 Translation by Dannemann (n 6) 309–10. 57 ibid 79.

220 Anne Sanders any legal basis, but rather requires a positive ground for restitution, making use of an unjust factor comparable to the disputed58 English factor of illegality.59 Outside the specific provisions on unjustified enrichment, the BGB also contains six more instances of restitution, for example rights after termination of contracts,60 rights arising because of a negotiorium gestio,61 restitutionary rights connected with tort, cessio legis,62 and rights arising from the owner/possessor relationship.63 These models are not discussed in this chapter, except to say that they too are not based on an absence of basis model. If English law wanted to justify such rights in this way, it would derive little assistance from the German model.

(4) Initial Problems As Professor Zimmermann has pointed out, the German experience with its general action of unjust enrichment in § 812 I 1 BGB was not a happy one at first.64 Scholars and courts realised that this broad rule needed to be limited, especially because it seemed impossible to develop a single notion of legal basis for both performance-based enrichments and enrichments received without a deliberate act of the claimant. It was acknowledged that the section expressed a great legal vision whose implementation was, however, frustrated by the limitations of practical reality.65 Though there have always been attempts to interpret §§ 812–822 BGB as a unified system governed by a single idea, the absence of legal bases or, as courts sometimes claim, the idea of fairness or corrective justice,66 the majority of scholars and courts today stress that understanding the distinctions between different ways in which the enrichment was received and why it has to be given up are crucial to achieving both legal certainty and clarity.67 Professor Lieb argued that if unjust enrichment claims stepped in whenever wealth changed hands, the law of restitution would interfere with the responsibility of peo-

58 See for a critical discussion of this unjust factor, W Swadling, ‘The Role of Illegality in the English Law of Unjust Enrichment’ in D Johnson and R Zimmermann (eds), Unjustified Enrichment – Key Issues in Comparative Perspective (Cambridge, CUP, 2002) 289. 59 Compare A Burrows, The Law of Restitution, 3rd edn (Oxford, OUP, 2011) 488–97; Edelman and Bant (n 6) 294–300. 60 The relationship between the law of termination and the law of unjustified enrichment has proved to be a highly complex area of German law, see P Hellwege, ‘Unwinding Mutual Contracts: Restitutio in Integrum v the Defence of Change of Position’ in D Johnston and R Zimmermann (eds), Unjustified Enrichment − Key Issues in Comparative Perspective (Cambridge, CUP, 2002) 243; Meier (n 5) 345; Dannemann (n 6) 60–74. 61 For a comparative introduction see J Kortmann, Altruism in Private Law (Oxford, OUP, 2005) 106–10. 62 An assignment that operates by virtue of law, for example if the guarantor pays instead of the principal debtor the BGB transfers the creditor’s claim onto the guarantor who can enforce it against the debtor: Dannemann (n 6) 17. 63 This area of German property law concerns restitutionary claims the owner might bring against the processor who held his property without a right to do so: ibid 89–91. 64 Zimmermann (n 39) 887–91; Meier (n 5) 361. 65 Schäfer (n 46) 312. 66 BGHZ 36, 232, 234–35; 55, 128, 134; critical M Lieb in Münchner Kommentar, 4th edn (München, Beck, 2004) § 812 [22]. 67 M Lieb, Die Ehegattenmitarbeit (Tübingen, Mohr Siebeck, 1970) 61–64; D Reuter and M Martinek, Unrechtsfertigte Bereicherung, Handbuch des Schuldrechts (Tübingen, Mohr Siebeck, 1983) 52–62; K Larenz and C Canaris, Schuldrecht II 2, 13th edn (München, Beck, 1994) 129–31; H Westermann in Ermann BGB, 10th edn (Köln, Otto Schmidt, 2000) § 812 [1], [44].

Absence of Basis: A German Perspective 221

ple to organise their life by means of contract.68 The limits of such a wide law of unjust enrichment would have to be left to vague notions of fairness and judicial discretion.69 The first step in limiting this broad notion of unjustified enrichment was a restrictive interpretation of the requirement of ‘at the expense of this person’. Later, Professors Wilburg70 and von Caemmerer71 proposed a distinction between performance-based (Leistungskondiktion) and non-performance-based (Nichtleistungskondiktion) enrichments.72 Only the former were prima facie recoverable if obtained without a legal basis. A highly sophisticated definition of performance (Leistung) was developed to narrow the field of automatic restitution.73 In non-performance cases, for example, when money was obtained without the consent of the defendant, restitution should only be allowed in certain restricted cases, for example when the defendant had obtained the enrichment by violating the defendant’s rights. Thus, § 812 I 1 alternative 2 BGB today includes claims that could be classified in English law under the unjust factor of ignorance and restitution for wrongs. The Wilburg and von Caemmerer approach solved the problem of the potentially explosive nature of § 812 I 1 BGB by breaking up the notion of ‘enrichment in another way’ into a number of specific claims. This focus on the reasons why and how the defendant unlawfully received the enrichment shows some similarities with the unjust factor approach applied by scholars who embrace the unjust factor of ignorance. Though Professor Dannemann advocates English law embracing an absence of basis approach, he acknowledges that it would have to follow the same path to make an absence of basis system work.74 It might be argued that restitution for wrongs and the disputed unjust factor of ignorance75 already follow different rules than other claims in unjust enrichment. However, as Professor Scott has forcefully argued, within German law, the distinction sacrifices much of the generality and abstraction its absence of basis system seems to provide at first glance.76

D. WHAT IS A BASIS?

Even though German law follows a mixed and not a pure absence of basis approach, it cannot be denied that ‘absence of legal ground or basis’, as provided for in § 812 I 1 BGB, is of central importance to its law of unjustified enrichment. For that reason, attention must turn to the notion of ‘legal basis’. The question what it is will be asked in the context of performance cases. Those in which the enrichment was received without the claimant’s Lieb (n 67) 99–100. ibid 65. 70 W Wilburg, Die Lehre von der Ungerechfertigten Bereicherung im Österreichischem und Deutschem Recht (Graz, Leuscher & Lubensky, 1934). 71 E von Caemmerer, ‘Bereicherung und Unerlaubte Handlung’ in H Dölle, M Rheinstein and K Zweigert (eds), Feschrift für Rabel I (Tübingen, Mohr Siebeck, 1954) 333, 337–38. 72 Reuter and Martinek (n 67) 232. See in English, Krebs (n 10) 207–17; T Krebs, ‘The Fallacy of “Restitution for Wrongs”’ in A Burrows and Lord Rodger or Earlsferry (eds), Mapping the Law: Essays in Memory of Peter Birks (Oxford, OUP, 2006) 379, 380–88. 73 For a critical discussion of this notion of ‘Leistung’ see Lieb (n 66) § 812 [26]–[31]. 74 Dannemann (n 5) 376–77. 75 P Birks, Introduction to the Law of Restitution (Oxford, Clarendon Press, 1989) 140–46; Burrows (n 59) 403– 35; Edelman and Bant (n 6) 271–82; for a critical discussion W Swadling, ‘Ignorance and Unjust Enrichment: The Problem of Title’ (2008) 28 Oxford Journal of Legal Studies 627. 76 Scott (n 5) 93, 96. 68 69

222 Anne Sanders consent are, as was pointed out above, analysed in a way which is difficult to describe as a pure absence of basis approach.77 Professor Birks began his analysis of absence of basis with an explanation of the notion of legal basis from a civilian standpoint. He described it as ‘an explanation known to the law’, and went on to say: ‘An enrichment which turns out to have no such explanation was inexplicable and could not be retained’.78 But though this notion of basis describes its function, it does not state the preconditions such explanations must fulfil. Theoretically, my receiving money by way of a mistaken payment ‘explains’ my enrichment at the defendant’s expense from a legal perspective just as much as saying it was money which was due. Professor Birks’ notion does not explain why in the first case I must give up the enrichment whereas in the second I need not. It is evident that a legal basis must provide some normative reason justifying the retention of the enrichment. When that is so, however, is difficult to say. Canadian courts, which have embraced an absence of basis approach, have struggled with defining the notion of ‘juristic reason’. In Garland v Consumers’ Gas Co,79 the Canadian Supreme Court held that the established categories that constitute juristic reasons include a contract, a disposition of law, a donative intent, and other valid common law, equitable or statutory obligations.80 However, by giving ample consideration to factors such as the reasonable expectations of the parties and public policy, courts could even develop a new category of juristic reason.81 It is not surprising, therefore, that the system was described as ‘very complex’ by Bastarache J in Kingstreet Investments v New Brunswick.82 In the later case of BMP Global Distribution Inc v Bank of Nova Scotia,83 the Supreme Court did not even allude to Garland or the notion of ‘juristic reason’, but solved the case applying the unjust factor of mistake and Barclay’s Bank v Simms.84 In Kerr v Baranow, a case concerning financial rights of unmarried cohabitants after the breakdown of their relationship, the Supreme Court of Canada used the absence of juristic reason approach again.85 It is thus unclear whether Canada has now abandoned its absence of basis approach or is applying two different systems concurrently,86 as German law does. Whatever the case may be, it can be assumed that the difficulties in developing an applicable notion of juristic reasons have caused the system to lose some of its charm. Professor Smith argued that the juristic reason approach still required some fine-tuning. The result would not be a civilian approach, however, but probably a trans-systemic approach, taking lessons from both systems.87 Dr Hunt, who has pointed to the problems inherent in drawing up a list of legal bases or juristic reasons, has argued that the notion is clear in codified legal systems.88 However, he For discussion of such cases see Krebs (n 72) 379. Birks (n 3) 102, 127. 79 Garland v Consumers’ Gas Co 2004 SCC 25, [2004] 1 SCR 629 [44]–[46]. The case is discussed by Robert Chambers in chapter 14 of the present volume. 80 ibid [44]. 81 ibid [46]; Hunt (n 5) III (i). 82 Kingstreet (n 24) [38]. 83 BMP Global Distribution Inc v Bank of Nova Scotia 2009 SCC 15, [2009] 1 SCR 504. 84 Barclay’s Bank Ltd v WJ Simms Son & Cooke (Southern) Ltd [1980] QB 677 (QB). 85 Kerr v Baranow 2011 SCC 10, [2011] 1 SCR 269, [36]–[45]; compare M Lower, ‘Case Comment – The Constructive Trust: From Common Intention to Relationship? Kerr v. Baranow’ [2011] Conveyancer & Property Lawyer 515; M McInnes, ‘Cohabitation, Trusts and Unjust Enrichment in the Supreme Court of Canada’ (2011) 127 LQR 339. 86 Compare Smith, ‘Demystifying Juristic Reasons’ (n 5). 87 ibid 304. 88 Hunt (n 5) II (iv), III (i), (iii). 77 78

Absence of Basis: A German Perspective 223

might give too much credit to continental lawyers. One might think that every German lawyer can easily give an answer to the question of what is a legal basis. There is no definition in the BGB, however, and most lawyers will just give examples when asked for a definition, approaching the problem in an ‘I don’t know what it is, but I know it when I see it’ kind of way. Even amongst German academics, the question how to define a legal basis is far from clear. Professors Lieb and Schwab describe a legal basis as simply meaning the legal reason why an enriched person may keep the enrichment,89 an explanation used for both participatory and non-participatory enrichments. However, as with Professor Birks, this only defines the function of the basis; it gives no help in deciding when such a basis exists. In the same way, a definition of unjust factor which provided that it is the ‘thing which justifies restitution’ would be in need of further development. If absence of basis is to offer an elegant guide to which enrichments have to be given up, the legal explanation required must be something that provides a normative justification in all cases where an enrichment should be retained. In the section which follows, different cases are analysed in an attempt to discover what kind of legal bases were developed by the BGB, courts, and scholars and whether they provide a normative basis for the retention of enrichments.

(1) Enforceable Obligations Both German and Canadian law agree that enforceable obligations are legal bases.90 That was also the view of Professor Birks, who said that if a payment was made to discharge a valid obligation, the enrichment had an explanatory basis and was not unjust.91 As far as such obligations are concerned, an absence of basis approach even has some normative force. This can be explained by reference to Woolwich and FII themselves. The state must not keep money it could not lawfully demand from its citizens. Likewise, if a private defendant could not have claimed a particular enrichment by process of law, he too should now give it up as an unjustified enrichment. Professor Meier used the same reasoning when arguing that the questions ‘do I have to perform?’ and ‘can I recover what I have already performed?’ were inter-connected.92 For a German lawyer, most cases of legal basis involve an obligation created by a contract or a response to a wrong. For them, an obligation is seen as having two functions. It creates both the right to claim performance and the right to retain performance when received. The first function can be described as the ‘claiming function’ of the obligation. A contract for the sale of a car, for example, creates obligations to transfer the title of the car, on the one hand, and to pay the agreed price, on the other. Once performed, under a German analysis, the obligations in the contract of sale turn into legal grounds, justifying the retention of the title to the car and the money.93 This second function could be described as the ‘keeping function’ of the obligation. If a contract is void ab initio or is subsequently 89 M Schwab in Münchner Kommentar zum BGB, 5th edn (München, Beck, 2009) § 812 [338]; Lieb (n 66) § 812 [171]–[174]; cf Windscheid (n 52) § 423, 880, who stated that an enrichment was unjustified when the law denied its approval. 90 See Garland (n 79) [44]. 91 Birks (n 3) 129. 92 Meier (n 5) 343, 347. German law does not make this connection in all cases: see Part E(2)(b) below. 93 Schwab (n 89) § 812 [337]–[338].

224 Anne Sanders avoided because of the rules in contract law on mistake or duress, the legal basis disappears and the defendant has to give back what he received.94 Since most unjust enrichment cases heard in German courts concern contractual obligations, this part of the system works relatively well in practice. There are a number of problems that might arise if this approach was applied in England which will be discussed later. For the moment, however, it can be said that as far as this goes, the absence of basis approach seems elegant and convincing.95 On a more philosophical level, academics have debated how an obligation which is fulfilled and is thus extinguished can still form a basis. They have turned to slightly metaphysical words to explain this event. Thus, Professor Gernhuber states: ‘As a legal basis, a fulfilled obligation is nothing more than a memory of an obligation which once existed and was extinguished by performance’.96 Others have declared that it is the legal relationship, for example a contract of sale between the parties, which provides the legal basis, not the obligation as such.97 This explanation, however, takes away some of the normative force of the argument developed with reference to the Woolwich principle described above. The two approaches which focus either on the contractual relationship or on the obligation, however, share an important characteristic. They assume that there are certain legal bases, for example a contractual obligation or gift, which are the same for every person. Those bases can be absent because a contract was void from the beginning or avoided later. This theory is called the ‘objective’ approach in Germany and was, as we have already seen, already foreshadowed by Hugo Grotius’ first attempts to define legal basis. The majority of German scholars have, however, rejected this approach. They argue that a basis is created by the achievement of the purpose which the transferor intended.98 This approach is called the ‘subjective’ approach because it focuses on the claimant’s state of mind, the individual purpose he pursued, when transferring the enrichment. Once the purpose he pursued fails, an absence of basis is assumed. Viewed from a historical perspective, these scholars are in excellent company, as Professors Savigny and Windscheid99 placed stress on the transferor’s purpose as the legal explanation or basis which justified enrichments. It can be said in favour of this approach that without knowing the purpose a claimant pursued, it is impossible to link a payment to a certain (possibly non-existing) obligation. According to a widely accepted German definition, a performance requires a purposeful enrichment of the defendant. The ‘subjective’ approach takes this a step further, defining the purpose to be achieved as the legal basis itself.100 In most cases, the ‘objective’ and ‘subjective’ approaches reach the same results, however. If a payment is made in order to discharge a non-existing debt, there is no valid obligation to provide a legal basis. Moreover, the purpose to discharge the debt cannot be achieved. It is difficult to tell which approach Professor Birks took, since, as we have seen, he only described a legal basis as an explanation known to the law.101 The Birksian scheme requires This is an approximate description. Avoidance shall be discussed in part E below. Zimmermann (n 7); Meier (n 8). 96 J Gernhuber, Die Erfüllung und ihre Surrogate (Tübingen, Mohr Siebeck, 1983) 99–100. 97 B Kupisch, ‘Zum Rechtsgrund i. S. des § 812 BGB bei Erfüllung’ [1985] Neue Juristische Wochenschrift 2370, 2374–75. 98 Reuter and Martinek (n 67) 106 –10; Westermann (n 67) § 812 [1], [44]. 99 Windscheid (n 52) § 423, 884. 100 From a German perspective, it can be argued that it would be surprising if the legal basis in § 812 I 1 BGB had the same meaning as the purpose in § 812 I 2 BGB. In that case, the Leistungskondiktion would be superfluous, see Schwab (n 89) § 812 [337]. 101 Birks (n 3) 102. 94 95

Absence of Basis: A German Perspective 225

inquiries into purposes when a voluntary enrichment is not made in order to discharge a debt. In the case of a payment made to discharge a debt, however, Professor Birks focuses on the existence or absence of obligations. The ‘subjective’ approach, which focuses on the purpose pursued by the payer, was subjected to convincing criticism by Professor Scott.102 It is questionable because it lacks the ‘surgical simplicity’ Professor Birks claimed for an absence of basis model, because there is no direct connection between the absence of a valid obligation and the absence of legal basis. In many cases when restitution should follow, a consequent application of the ‘subjective’ approach would bar recovery. In the Woolwich case, Woolwich paid in order to avoid bad press. Following a purpose-focused analysis, the purpose of the payment was achieved and Woolwich should not recover the money. Only if the purpose is strictly defined as fulfilling some (objective) obligation can this approach solve the case. It might even be argued that this approach is not a proper absence of basis approach at all. The approach does not focus on the actual absence of a valid obligation or gift as justification for the retention of the enrichment, but rather on a positive reason for recovery − the failure of a purpose − which is in truth not very different from the unjust factor of failure of consideration.

(2) Unenforceable Promises – Gifts The previous part explained that the notion of legal basis is not clear-cut, even when it comes to enforceable obligations. The meaning of legal basis is, however, even more difficult outside the realm of enforceable obligations.103 (a) Contractual Approaches – Natural Obligations A device used in Germany to explain how there could be a legal basis despite the fact that there was no enforceable obligation to transfer the enrichment was to broaden the realm of contracts and obligations to include gratuitous legal acts. Gifts, gratuitous services (mandatum) and bailments are all considered contracts in German law. Professor Dannemann suggests that this approach should be adopted in England. He recommends a re-conceptualisation of all non-contractual agreements to develop an absence of basis approach for the English legal system.104 However, this would require another fundamental change in English law, which in itself makes the approach questionable. Moreover, the German experience shows that there are problems with the notion of legal basis in respect to gratuitous enrichments. In Germany, a promise to make a gift is only enforceable if it is done in writing and signed in the presence of a notary. This limits the ‘claiming function’ of the promise. However, once it is performed, even an unenforceable promise to make a gift provides a legal basis, just like a contract of sale. The ‘keeping function’ therefore also applies in cases where a claim could not have been raised. The explanation of the Woolwich principle that ‘you should only retain what you could have asked for in court’, does not hold true in cases Scott (n 5) 93, 100 –101. ibid 93, 99. 104 Dannemann (n 6) 189–96. 102 103

226 Anne Sanders without an enforceable obligation, but one might be content that at least there is some natural obligation,105 as difficult as these are to explain. Most presents, however, are not promised but just handed over. In case of a spontaneous or surprising gift, the donor never makes any promises. How then could there be said to be a legal basis? This was the problem which faced by the German Imperial Supreme Court in a case in 1925.106 A husband had made a spontaneous gift of a house to his mistress. After the man had died, his widow, who was not as happy about her husband’s generosity, sued the mistress to recover the house. The Court of Appeal of Hamburg and the Imperial Supreme Court both agreed that the gift had been immoral and ruled in favour of the widow, though for different reasons. The Court of Appeal held that a spontaneous gift does not provide the requisite legal basis. However, apparently to avoid the apparently absurd result that most donees would then be unjustly enriched, the court immediately went on to hold that such gifts were outwith § 812 I 1 alternative 1 BGB (Leistungskondiktion) and its absence of basis approach. The widow instead won through an application of § 817 1st sentence BGB, which provides that an enrichment received violating public policy has to be given back. § 817 1st sentence BGB, it should be recalled, is not concerned with any legal basis. Its approach is instead closer to the English approach of unjust factors. The Imperial Supreme Court held that the Court of Appeal was mistaken in denying the application of § 812 I 1 alternative 1 BGB to gifts. A legal basis, the court held, could arise even though there was no obligation, not even an unenforceable promise. The agreement of the husband and his mistress that the house was given gratuitously created a contract the only function of which was to provide a legal basis for the transfer.107 In the instant case, the gift to the mistress was immoral and the gift-contract thereby rendered void. There was consequently no legal basis and the house could be recovered because of an absence of legal basis. This decision had the important effect of applying the rules of contract law on mistake, illegality, immorality and duress to spontaneous gifts, with the result that the unjust factor approach of § 817 1st sentence BGB was no longer needed. Was the Imperial Supreme Court right? A common lawyer would probably consider it odd to apply contract law where there was no bargain. However, German law, as we have seen, embraces a broader notion of contract. An important reason this is done is that such contracts can provide legal bases. If the Imperial Supreme Court had upheld the decision of the Hamburg Court of Appeal, German law might well have developed an unjust factors approach to gifts and an absence of basis approach to enforceable obligations. By applying the absence of basis approach to spontaneous gifts, the decision strengthened the coherence of the German law of unjust enrichment. However, one cannot but wonder whether the notion of contract was stretched in order to allow the finding of a legal basis where the court wanted to find one. However, the expansion of contract in this way did not resolve all problems. As in English law, there is a requirement in German contract law that contracting parties must have an intention to create legal relations.108 German lawyers agree, however, that people often act for other people’s benefit without such intentions. Take, for example, the walking of someone’s dog. If there is no intention to create legal relations in such a case, there can D Sheehan, ‘Natural Obligations in English Law’ [2004] Lloyds Maritime and Commercial Law Quarterly

105

188. RGZ 111, 151. RGZ 111, 151, 152–53. 108 ‘Rechtsbindungswille’: BGHZ 21, 102, 106. 106 107

Absence of Basis: A German Perspective 227

be no contract and thus no legal basis for the dog walker’s services. This means that there are cases where benefits are provided without a legal basis in German law in situations where no-one would argue that restitution should follow. However, this problem has never been properly addressed in German law, let alone resolved. Without an answer to this question, the German law of unjust enrichment remains incomplete. If it were ever brought to court, such a case might be solved by application of the defence in § 814 BGB. (b) The New Birksian Approach Professor Peter Birks avoided these problems by assuming that the achievement of a certain purpose formed the explanatory basis of all voluntary enrichments and that a gift could be made without the consent of its recipient.109 This approach, however, fails to offer any normative basis for the retention of the enrichment, apart from the fact that if the transferor wanted the enriched person to have the enrichment, he should not be entitled to demand it back. The weaknesses of an approach focusing on the purpose pursued have already been pointed out. It is extremely difficult to decide which purposes may form a legal basis and which do not. Professor Birks, for example, argued that in these cases the legal basis must be defined with an eye to risk-taking.110 If the claimant took the risk that the purpose pursued would not materialise, his claim should be rejected. However, the decision as to which purposes could form a legal basis, and whether the claimant took the risk that the purpose of the enrichment fails, leaves much to the discretion of the courts and may thus cause uncertainty. Moreover, the Birksian approach makes use of an extremely broad notion of gift, which has come in for criticism.111 It attempts to provide a legal basis for all kinds of accidental benefits, including heat rising through a ceiling which warms a neighbour’s flat. As we have seen, the German approach of the conclusion of a contract requires an agreement between two people, thus limiting the scope of the law of unjust enrichment. Moreover, accidental benefits only trigger a claim in unjust enrichment under very limited conditions, for example, when the enrichment is obtained through an infringement of the claimant’s rights. (c) The Invention of Legal Bases Commenting on the Birksian notion that a person turning up the heat in his flat was making a gift to his neighbour, Professor Watts writes: ‘Surely there is a gift here only because the law has already determined that there should be no claim. The better explanation is that enrichment per se does not need justification’.112 It is submitted that if the notion of absence of basis is introduced into English law, numerous legal grounds or defences will have to be developed to avoid restitution being available in too many cases. The problem that this might bring back a law of unjust enrichment based on equity and policy considerations113 can be seen in the Canadian case of Garland v Consumers’ Gas Co.114 As we have 109 Whether the donee’s consent is necessary for a gift is not a question with a straightforward answer, however, see J Hill, ‘The Role of the Donee’s Consent in the Law of Gift’ (2001) 117 LQR 177. 110 Birks (n 3) 144. 111 Burrows (n 5) 46. 112 P Watts, ‘Birks’ Unjust Enrichment’ (2005) 121 LQR 163, 166; see also Burrows (n 5) 46. 113 Dannemann (n 6) 195–96. 114 Garland (n 79) [44]–[46].

228 Anne Sanders already seen, Dr Hunt assumed that in a codified legal system, the development of new bases might not be necessary.115 This is not entirely true, however. In Germany, courts and scholars have developed legal grounds to prevent restitution where it did not seem desirable, the Imperial Supreme Court case discussed above being an example.116 One more example may be usefully given in this context. For many years, contributions to the purchase of a family home registered in the name of one partner or spouse alone were considered to be gifts and thus to have bases.117 When this approach was abandoned, it became clear to German courts and scholars that such enrichments had to have another basis during the time of the relationship or else the value put into the purchase of the family house could be claimed back at any moment. This, however, would have raised the question of limitation periods and the problem that creditors of the claimant might claim the right to restitution. German scholars thus developed the idea that contributions to the purchase of the family home were not gifts but contributions made under a ‘familyspecific contract’ concluded for the organisation of the couple’s partnership. This sui generis contract then provided the legal basis.118 In contrast to Australia,119 Canada120 and Scotland,121 English courts do not follow an unjust enrichment approach in family home cases. However, both Professor Gardner122 and Sir Terence Etherton, writing extra-judicially,123 have suggested unjust enrichment as a new doctrinal basis for family home cases. Etherton argues that it would be easy to grant a claim under an absence of basis approach since contributions to the family home are clearly provided without basis. However, he does not explain the legal character of such contributions and thus does not discuss what bases such contributions could have. This would be necessary, not only in this case, but in many more if English law decided to switch sides.

E. VOIDNESS, DEFENCES AND VOIDABILITY

There still are problems with an absence of basis approach even in those cases where there is agreement on the legal basis. In Germany, the wide principle that absence of basis triggers restitution is kept in check by a technical notion of invalidity, and by a number of defences in the German Civil Code. The English system would need to spend much time and effort developing such necessary safeguards. Moreover, an absence of basis approach cannot explain powers to avoid a contract as being triggered by unjust enrichment.

Hunt (n 5) III (i), (ii). RGZ 111, 151, 152–153. 117 Family home cases are not a problem only common law courts have to deal with. While courts used constructive and resulting trusts in England, German courts assumed implied partnerships and family cooperation agreements. These approaches have been attacked in both countries by academics with surprisingly similar arguments, see A Sanders, ‘Vermögensausgleich bei Solidargemeinschaften: Trust, (Ehegatten) Innengesellschaft und Bereicherungsrecht in Deutschland und England’ [2011] Zeitschrift für Europäisches Privatrecht 65. 118 The contribution to the house, however, can be asked back under the principle of frustration (§ 313 BGB) in case the relationship fails: BGHZ 84, 361, 367; 115, 261, 265; 127, 48; 142, 137, 149. 119 Baumgartner v Baumgartner (1987) 164 CLR 137; Muschinski v Dodds (1985) 160 CLR 583. 120 Pettkus v Becker [1980] 2 SCR 834; Peter v Beblow [1993] 1 SCR 980. 121 Satchwell v McIntosh 2006 SLT (Sh Ct) 117. 122 S Gardner, ‘Family Property Today’ (2008) 124 LQR 422. 123 T Etherton, ‘Constructive Trusts: A New Model for Equity and Unjust Enrichment’ (2008) 67 CLJ 265, 280. 115 116

Absence of Basis: A German Perspective 229

(1) Void Contracts Even if there is no clear cut definition of a basis, under certain circumstances the German lawyer knows that there has to be a basis and that enrichments have to be returned if that basis turns out to be void. Thus, if the contract of sale by means of which someone buys a car is void, restitution will be granted under § 812 I 1 alternative 1 BGB. The reasons why the contract is void can be many. The seller of the car might have forced the buyer to buy the car and thus created a right to avoid the contract. Or the contract might be void because of some import regulation law. None of these are questions for the law of unjust enrichment but for the law of contract regulated in other parts of the law. To the German lawyer, this seems a neat and elegant approach. To the English lawyer, who sees the law of unjust enrichment as a unity, this approach means that difficult questions are pushed out of sight into other areas of law.124 Professor Zimmermann has argued that the German approach prevents the same question being discussed twice, once in the realm of contract law and again in the law of unjust enrichment.125 It must be conceded, however, that the question of when a contract is void does not become any easier by giving it an importance for both the law of contract and unjust enrichment.126 German law works with a technical notion of invalidity, meaning that it is only found where restitution is to follow, as Dr Krebs has pointed out.127 By adopting an absence of basis approach, the German legal system made the choice that voidness in contract law should trigger restitution in the law of unjust enrichment. The automatic connection between invalidity and restitution would have made the swaps cases easy to solve for a German lawyer.128 In other legal systems, like the English, however, invalidity might be triggered by a number of policy grounds which do not justify restitution.129 In such a system, merely describing a contract as void says nothing about whether restitution should follow. Invalidity itself is just a consequence of a particular policy, and one has to go back to that policy to see whether it justifies restitution. Professor Meier concedes that English law might not be based on a coherent notion of voidness and suggests that ‘void contract’ need not necessarily mean ‘no basis’.130 This might give courts more freedom to decide where restitution should be allowed. At the same time, however, it would take away much of the beauty of the system that comes from the automatic connection between voidness and restitution. Moreover, as Professor Meier concedes, it would make the notion of basis even more difficult to define. Thus, rather than providing an elegant common law absence of basis model, with absence of basis at the top and the traditional unjust factors at the base, the Birksian pyramid131 has been described as resembling an iceberg, with 90 per cent of the problems lurking beneath the surface.132 Burrows (n 5) 47; Edelman and Bant (n 6) 159–65; Hunt (n 5) II (i). Zimmermann (n 7) 416. 126 M Chen-Wishart, ‘In Defense of Unjust Factors: A Study of Rescission for Duress, Fraud and Exploitation’ in D Johnson and R Zimmermann (eds), Unjustified Enrichment − Key Issues in Comparative Perspective (Cambridge, CUP, 2002) 163. 127 Krebs (n 10) 88, 179, 247–48. 128 Birks (n 3) 108–13. 129 Krebs (n 5) 92–98. 130 Meier (n 5) 349. 131 Birks (n 3) 116. 132 G Virgo, ‘Demolishing the Pyramid: The Presence of Basis and Risk-Taking in the Law of Unjust Enrichment’ in A Robertson and Tang Hang Wu (eds), The Goals of Private Law (Oxford, Hart Publishing, 2009) 477. 124 125

230 Anne Sanders German law has developed a number of safeguards to prevent too much restitution. In some cases, the BGB itself prevents restitution by declaring void contracts binding after performance.133 For example, a contract of sale of land requires notarisation and is void if this formality is not complied with. However, once the buyer is registered, the contract becomes binding (§ 313b II BGB). Dr Krebs has forcefully argued that German law created such ‘just factors’ to prevent restitution in cases where it is not desirable.134 Other safeguards have been developed by the courts. German judges will only declare contracts void when they are prepared to grant restitution.135 To give an example, according to § 134 BGB, a contract that violates a statute is void. Since the sale of drugs is prohibited, a contract for the sale of heroin is void. However, it is not in every case that the courts have been prepared to translate a prohibition into voidness.136 In such cases, voidness may not be assumed when the policy reasons behind the statute do not justify restitution. Another instance is that of contracts creating long-term relationships, where courts have for decades denied restitution even though the underlying contract turns out to be void. If, for example, the statutes of a company or partnership137 or employment contracts138 are void, the parties cannot claim restitution of the services and payments provided. This is because courts have assumed that it would be too difficult to unravel such long-term arrangements retrospectively.139

(2) Defences Having praised the coherence of German law and its synchronisation of invalidity and restitution, it must be admitted that invalidity and voidness, absence of basis and restitution, do not always go hand in hand, even in German law. This is because the wide absence of basis principle is kept in check, not only by the courts’ use of voidness, but also by a number of defences which go beyond change of position. Indeed, Professor Birks stressed the importance of such defences when explaining his new scheme.140 It has been rightly pointed out that this is another aspect which takes away some of the beauty of the absence of basis system.141 (a) Mistake − § 814 BGB Mistake is probably the most important unjust factor in the English law of unjust enrichment. In German law, a mistaken payment is usually recoverable because there will have §§ 313b, 766, 518 II BGB. Krebs (n 5) 76, 88–89, natural obligations 90. 135 ibid 86–88. 136 The most popular practitioner’s commentary to the BGB, the Palandt, takes five narrowly-printed pages just to list the most important cases in which § 134 BGB does or does not render contracts void: see J Ellenberger, Palandt, 71st edn (München, Beck, 2012) § 134 [14]–[25]. 137 Articles and decisions on this subject fill libraries, see only BGH ZIP – Zeitschrift für Wirtschaftsrecht 2005, 753; W Goette, ‘Fehlerhafte Personengesellschaftsverhältnisse in der Jüngeren Rechtsprechung des Bundesgerichtshofes’ (1996) Deutsches Steuerrecht 266; K Schmidt, Gesellschaftsrecht, 4th edn (Köln, Heymann, 2002) s 6; C Schäfer, Die Lehre vom fehlerhaften Verband (Tübingen, Mohr Siebeck, 2002). 138 R Richardi and H Buchner in Münchener Handbuch zum Arbeitsrecht, 3rd edn (München, Beck, 2009) § 34. 139 RGZ 78, 303, 305–6; RGZ 81, 303, 305–6; RGZ 165, 193; Goette (n 137) 267. 140 Birks (n 3) 263. 141 Burrows (n 5) 33, 47. 133 134

Absence of Basis: A German Perspective 231

been no basis for the transfer. However, mistake nevertheless plays an important role in German law. § 814 BGB provides that payments made in the knowledge that they are not due need not be given up, even if there was no legal basis for them. It is difficult to explain in a principled way why mistake should play a role at all if there is no legal basis.142 Since the BGB provides the defence, such reasoning is of secondary importance in Germany. In England, it would be different. Professor Dannemann argues that cases such as CTN Cash & Carry Ltd v Gallaher Ltd 143 and Nurdin & Peacock plc v DB Ramsden & Co Ltd,144 where the claimants made payments knowing that they were not due, demonstrate that a switch to an absence of basis model is desirable.145 Moreover, Professor Burrows states that CTN Cash & Carry might well have been decided differently under an absence of basis approach.146 However, under the Birksian scheme, this is not entirely clear. The question whether a payment was made in respect of a non-existing obligation or was a voluntary one can only be decided by reference to a person’s state of mind. Assume the claimant in CTN Cash & Carry knew perfectly well that the defendant had no right to receive the payment he made and was not forced to make it.147 In that situation, the payment must be considered a gift. According to the Birksian scheme, in that situation, restitution can only be allowed if the purpose for which the payment was made does not come to fruition. In CTN Cash & Carry, however, the purpose of the payment was to stop the defendant from discontinuing its business relationship with the claimant. That purpose, it seems, was actually achieved. As a consequence, restitution should not be allowed. Or perhaps the purpose was to persuade the defendant to carry on delivering cigarettes but to claim the payment back later? In that situation, restitution might be possible. It becomes apparent that restitution will depend on the purpose of the individual payer, which may not be easier to ascertain than a relevant mistake. Would CTN Cash & Carry have been decided differently under German law? Again, the answer is far from obvious. Though there was no legal ground for the payment, the claimant made the payment knowing it was not due. In that situation, the defence in § 814 BGB might prevent restitution. § 814 BGB, as we have seen, provides a defence when a claimant pays knowing the money is not due. Literally applied, German law would not give restitution in cases like Woolwich, CTN and Nurdin. Courts and academics, however, have interpreted the section as only preventing restitution in cases in which the claimant make a ‘voluntary’ payment.148 Here, however, the problem again arises how to decide when a payment is made voluntarily. The problems of submission to an honest claim and compulsion raise their heads. It shows, as Edelman and Bant have pointed out,149 that the same questions arise, irrespective of the label of unjust factor or legal basis. This fact can be explained by another example. In FII,150 Lord Sumption discussed the case of Reemtsma Cigarettenfabriken GmbH v Ministero delle Finanze151 as one where overpaid tax was considered recoverable because of an absence of a legal basis. In this case, a Scott (n 5) 100. CTN Cash & Carry Ltd v Gallaher Ltd [1994] 4 All ER 714 (CA). 144 Nurdin & Peacock plc v DB Ramsden & Co Ltd [1999] 1 WLR 1249 (Ch). 145 Dannemann (n 5) 205. 146 Burrows (n 5) 45, n 56. 147 If the amount of pressure imposed on the claimant had been unlawfully high, the case might have been decided differently under the unjust factor approach. 148 BGHZ 83, 278, 282; Dannemann (n 6) 77–79. 149 Edelman and Bant (n 6) 159–65. 150 FII (n 2) [188]–[190]. 151 Case C-35/05 Reemtsma Cigarettenfabriken GmbH v Ministero delle Finanze [2007] ECR I-2425. 142 143

232 Anne Sanders German company had mistakenly paid VAT which was not due to an Italian company. Though the underlying law did not violate European law, the ECJ held that an effective remedy either against the state or the Italian company was necessary. Under English law, the case would qualify as one of mistake, no other claim being available, Lord Sumption held. However, if Lord Sumption thought that English law could be considered ineffective in such a case, German law should be considered equally so. If there was no mistake about the obligation to pay VAT, a private law claim against the company would have been blocked under German law because of § 814. (b) Policy Defences − § 817 2nd sentence BGB Restitution is not the automatic result in cases where contracts are void, not even in Germany. Professor Birks conceded this when he emphasised that policy defences had an important role to play in countering automatic restitution in many cases.152 In Germany, the development of such a policy defence took some time. § 817 1st sentence BGB provides that an enrichment received in violation of public policy has to be returned. § 817 2nd sentence BGB provides a defence, however, where both parties have consensually violated public policy. Scholars soon argued that § 817 2nd sentence BGB contains a general principle applicable also to § 812 I 1 alternative 1 BGB.153 If a contract is void, § 817 2nd sentence BGB can be used to deny restitution if the court feels that the policy justifying invalidity does not justify restitution. Take the case where someone pays money to a villain, who in return severely injures a victim. If the person who encouraged this violent act seeks to recover the payment, both German and English law agree, that the claim must be disallowed.154 In English law, there is no unjust factor. In German law, this result is achieved by means of § 817 2nd sentence BGB. Another example is provided by a case where a married man gave the mother of his illegitimate child money to keep his identity secret. After the mother had revealed the father’s identity to the child, the father demanded his money back. The court held that the agreement was immoral and invalid but denied restitution. Mother and father had both behaved badly, but the court assumed the mother rather than the father should have the money.155 In other cases where the court felt that despite a consensual misdeed restitution should be allowed, the scope of § 817 2nd sentence was limited,156 for example, when premises to run a brothel were leased. Courts held that the enrichment was recoverable in order not to leave the brothel in the premises and so encourage prostitution.157 These examples illustrate that when applying § 817 2nd sentence BGB, courts inquire into the policy reasons which caused the enrichment’s basis to fail and decide whether or not they justify restitution. It is obvious that this approach further detracts from the elegant simplicity that supposedly recommends the absence of basis model.

Birks (n 3) 263; see also Burrows (n 5) 33, 39. BGHZ 35, 103, 107; BGHZ 50, 90, 91; BGH Neue Juristische Wochenschrift: Rechtsprechungs-Report 1993, 1457, 1458; for further references see Lieb (n 66) § 817 [10]. 154 Dannemann (n 6) 80–82. 155 Unreported, see A Sanders, ‘§ 817 BGB und der heimliche Vater, Anmerkung zum Urteil des LG Koblenz vom 12. Juli 2006 (15.0.389/05)’ [2007] Forum Familienrecht 158. 156 Schwab (n 89) § 817 [20]–[29]. 157 RGZ 71, 432; BGH Wertpapier-Mitteilungen 1969, 1083. 152 153

Absence of Basis: A German Perspective 233

(3) Avoided Contracts In German law, rescission is an important way to remove the basis of a contract. If a contract is rescinded because of mistake, duress, etc, § 142 BGB in the general part of the BGB provides that the contract is void ab initio. It follows that there was never a legal basis for an enrichment and § 812 I 1 alternative 1 BGB will apply.158 The power to rescind a contract is not, however, understood as being triggered by unjust enrichment, but is merely a tool provided by the BGB. Explaining his new model, Professor Birks argued that both terminable159 and voidable contracts provided no valid legal basis even before the power to rescind or terminate was exercised.160 This argument has been heavily criticised.161 Professor Burrows argues that the Birksian analysis of rescission and termination is unnatural and problematic.162 He explains Birks’ reasoning as the consequence of his desire to adhere to his long-held view that the right to rescind and to terminate a contract was a response triggered by unjust enrichment.163 Moreover, the Birksian system requires that in cases where only one party, for example the victim of a misrepresentation, has the right to rescind or to terminate, there is an absence of basis for that person only.164 This reasoning complicates the notion of legal basis further,165 because it assumes that there can be a legal basis and an absence of legal basis at the same time − a ‘theory of relativity’ for the law of unjust enrichment. Professor Stevens, however, argues that Professor Birks was right. He concedes that the basis of a payment made under a voidable contract only disappears once the contract is avoided. That must be right if one accepts that an enrichment requires a basis in English law and that a contract does provide such a basis. However, Professor Stevens goes on to argue that the obligations created by the conclusion of a voidable contract are an enrichment that has no basis and that therefore the power to rescind is triggered by an absence of basis. That seems hard to understand. What would be the basis of that contract if there had not been a mistake? One would assume that a contract is concluded in order to create obligations, which in turn provide legal bases once they are fulfilled. However, if those obligations themselves need legal bases, what could they possibly be? If it were other contracts, how would those be concluded? And if so, would not those contracts require still more obligations to provide more bases for them? In that case, the legal world would be filled with obligations designed purely to provide bases for other obligations. One could argue, however, that the conclusion of a contract which creates an obligation provides a voluntary enrichment under the Birksian scheme. According to that analysis, the basis of that voluntary enrichment would be the purpose the parties wished to achieve in concluding the contract. In order to create an absence of basis, this purpose would need 158 Not uncontentious: H Sprau, Palandt, 71st edn (München, Beck, 2012) § 812 [26] argues that a basis falls away the moment the contract is rescinded. 159 Termination will not be discussed in this chapter. 160 Birks (n 3) 125–26. 161 K Barker, ‘Responsibility for Gain: Unjust Factors or Absence of Legal Grounds? Starting Points in Unjust Enrichment Law’ in C Rickett and R Grantham (eds), Structure and Justification in Private Law (Oxford, Hart Publishing, 2008) 65; Hunt (n 5) II (iii). 162 Burrows (n 5) 39–41, 47. 163 Burrows (n 5) 40; Birks (n 3) 126. 164 Birks (n 3) 126. 165 Hunt (n 5) II (iii).

234 Anne Sanders to fail. This, however, would be a highly artificial notion that one could not possibly explain to the man on the Clapham omnibus. It shows again that a legal system requiring a legal basis for every kind of enrichment including the creation of contractual rights might run into problems. Moreover, it would be difficult to detect when the purpose behind a contract fails. The purpose of making a profit would probably not be enough, as this would have considerable effects on our economic system. It is suggested that a system which adopts an absence of basis approach faces difficulties defining a power to avoid a contract as being triggered by any absence of basis.166 Avoidance offers a tool to destroy an existing basis; if there is no longer a basis, avoidance is not needed. However, the unjust factor, mistake, duress, etc, has been there from the moment the transfer caused by mistake or duress was made. It does not require any intermediate step to allow restitution of a payment made under a voidable contract or to trigger a power to avoid a contract. Two possible consequences might be drawn from this. One would be, as Professor Meier has suggested, to streamline the Birksian system and understand the power to avoid as merely a tool to remove a basis.167 Another would be to stick to an unjust factor approach, being content that it can explain the origins of a power the BGB does not need to rationalise. Professors Meier168 and Stevens might argue that an absence of basis approach is already at the heart of the English system, because restitution is not allowed so long as an existing contract is not avoided or terminated. Stevens argues that the priority of the contractual claim over the restitutionary one cannot be explained unless an absence of basis approach is applied.169 However, this is not necessarily so. As long as the defendant has a valid claim to the enrichment, the claimant would have to give it back immediately.170 That this circle is broken in favour of the defendant’s contractual claim is not a question of priority, but because of the fact that even if the defendant did make restitution, this would have no effect on his contractual claim to receive an enrichment unaffected by any unjust factor unless the contract were avoided or terminated. It would be a waste of court resources to allow a restitutionary claim in this situation.

F. PRACTICAL ISSUES

Finally, the practical differences between the two approaches shall be discussed.

166 Powers in rem may, however, be triggered by an absence of basis once a power to avoid has been exercised. However, different explanations for different powers might not be the most elegant approach. Häcker, though stressing that a power in rem model could be applied under an absence of basis approach, argues that one of the advantages of the power model is that it is sensitive to the respective reason for restitution: B Häcker, ‘Proprietary Restitution After Impaired Consent Transfers; a Generalised Power Model’ (2009) 68 CLJ 324, fn 6. cf also B Häcker, Consequences of Impaired Consent Transfers (Tübingen, Mohr Siebeck, 2009) 171–74, discussing the conceptual basis of rescission from a comparative angle. 167 Meier (n 5) 349. 168 ibid 346. 169 Stevens (n 5) 23. 170 Burrows speaks of anti-circularity, see A Burrows, Understanding the Law of Obligations: Essays on Contract, Tort and Restitution (Oxford, Hart Publishing, 1998) 21.

Absence of Basis: A German Perspective 235

(1) Burden of Proof The position in English law at present seems to be that the claimant has to prove all the elements of his cause of action in unjust enrichment (enrichment of the defendant, at the claimant’s expense, and the unjust factor). The defendant then has to prove any defences, such as change of position. If the absence of basis approach were adopted, one would assume, as a young practitioner suggested,171 that the claimant would still have to prove an enrichment of the defendant at his expense, but that defendant would have to prove the basis for the payment. So, if the claimant paid money to the defendant as a consequence of a mistake as to the extent of his liability under a contract, one could suppose the defendant would plead and prove the contract. It would then, one would assume, be for the claimant to rebut the defence with evidence of his own mistake as to his legal liability under the contract, and the defendant would then be liable unless he could make out one of the defences. The practical consequence of this would be that rather than pleading the unjust factor in the particulars of claim, the claimant would need to do it in the reply to the defence. Thus, the claimant’s case would not be clear until after the defendant has filed his defence. Since most cases do not get to trial, it might be more cost-effective for a case to be set out in one pleading than to require the claimant to draft his or her particulars of claim and then also a reply to the defence. This approach, which seems logical from a theoretical perspective, would contradict the basic rule of German procedural law that the claimant has to bring substantial proof to substantiate his claim, however.172 In German law, the claimant has to prove that the defendant was enriched at his expense and also that there is an absence of basis. The claimant thus has to prove a negative fact,173 something which has been criticised by Professor Smith.174 In Germany, it seems to be insufficient to claim that there is an absence of basis even if the defendant never objected to that claim.175 A German judge has to be given the facts that constitute the preconditions of a successful claim. Absence of basis, however, is not in itself a fact but a legal opinion. To bring facts from which a judge could conclude that there was indeed an absence of basis, the claimant must explain how and why he made a payment and why he wants his money back. Or to put it differently, claimants have to point to some unjust factor when arguing for restitution. This means in practice that the German law of unjustified enrichment works much more like the English model.176

Many thanks to Eleanor Campbell of One Essex Court for discussing this problem with me. BGH Neue Juristische Wochenschrift 1999, 2887; L Rosenberg, Die Beweislast, 5th edn (München, Beck, 1981) 98; Meier (n 5) 350–51. 173 BGH Neue Juristische Wochenschrift: Rechtsprechungs-Report 1991, 574, 575; BGH Neue Juristische Wochenschrift 1995, 662–63, 727–28; Sprau (n 158) § 812 [77]. 174 Smith, ‘The Mystery of “Juristic Reason”’ (n 5) 228–29. 175 Many thanks for discussing this issue with me to the judges Dr Heike Dohrn, Richterin am Landgericht Koblenz; Anne Bollacher, Staatsanwältin, formerly Richterin working at the Amtsgericht Waiblingen; and Dr Christoph Syrbe, Richter am Oberlandesgericht Koblenz; all delegated in 2010 to the Federal Constitutional Court (Bundesverfassungsgericht) as judicial assistants. 176 For a discussion of South African experience see Scott (n 5) 103–4. 171 172

236 Anne Sanders However, in some cases, courts place a so-called ‘secondary burden of proof’ on the defendant.177 Courts do not require claimants to produce endless lists of possible legal grounds when it would be easy for the defendant to bring proof of a basis justifying the retention of the enrichment. In that situation, the defendant has to bring proof of a fact justifying the retention of the enrichment which the claimant then has to rebut.178 Though it might work in practice, this approach shows the tension between an absence of basis approach and an unjust factor approach which is generated by the basic requirement that a claimant has to substantiate his claim. It might help that German lawyers do not usually know what an unjust factor is and thus do not realise that they are in fact often applying something that comes close to it. In Canada, where we have seen a switch from one to the other being attempted,179 this tension might be felt more strongly. In Garland v Consumers’ Gas Co180 the Canadian Supreme Court introduced a two-stage procedure to show how a negative fact could be proved in court. First, the claimant must show that there is no juristic reason from an established category. If there was no juristic reason, the defendant can rebut the prima facie right to restitution by showing that there was another right justifying his retention of the enrichment. However, Bastarache J in Kingstreet Investments Ltd v New Brunswick 181 rightly described this as a very complex enterprise.

(2) Limitation Periods Another practical question is how limitation periods are calculated in an absence of basis model. Cases like Deutsche Morgan Grenfell v IRC 182 and FJ Chalke v HMRC 183 give ample proof of how important this issue is. The most natural thing would be to calculate limitation periods from the moment the basis has failed or the moment the relevant enrichment was received. That means that if there was no legal basis from the start, the limitation period would start running with the payment. German law works slightly differently. According to § 199 I BGB, the section applicable to claims in unjust enrichment,184 a limitation period of three years starts running at the end of the year in which the claimant obtained knowledge of the claim or could have obtained knowledge if he had used reasonable care. With respect to a claim in unjust enrichment, this would mean that a claimant has to obtain knowledge that he enriched the defendant without legal basis. That means that the claimant has to receive knowledge of a negative fact. In practice, of course, this means that limitation periods start running from the moment the claimant becomes aware or should have become aware of the facts that constitute his mistake, duress etc, in other words, when the claimant receives knowledge of the facts that constitute an unjust factor which would directly trigger restitution under English law. 177 BGH Neue Juristische Wochenschrift 1999, 2887; BGH Neue Juristische Wochenschrift 2003, 1449; BGH Neue Juristische Wochenschrift: Rechtsprechungs-Report 2004, 556. 178 BGH Neue Juristische Wochenschrift 1999, 2887–88. 179 After BMP (n 83) the fate of juristic reason is uncertain. 180 Garland (n 79) [44]–[46]. 181 Kingstreet (n 24) [38]. 182 DMG (n 4). 183 FJ Chalke Ltd v HM Revenue & Customs [2010] EWCA Civ 313, [2010] STC 1640. 184 BGH Neue Juristische Wochenschrift 2007, 1584; BGH Neue Juristische Wochenschrift: RechtsprechungsReport 2008, 1237; Sprau (n 158) § 812 [69].

Absence of Basis: A German Perspective 237

G. CONCLUSION

Woolwich was not decided upon an absence of basis approach, and it should not have been. English law has no concept of legal basis which is capable of explaining the normative force behind all deliberate transfers, voluntary and obligatory, and has thus no principled way of scrutinising whether there was an absence of basis. Moreover, it is doubtful if such a system could be introduced. German law has developed an approach that works in both theory and practice. This, however, came at the price of difficulties which are not realised in everyday work because the system has had more than 100 years to get used to them. It is not maintained that the English system is without problems. However, it is submitted that by switching to an absence of basis system, the English system would confront new problems in addition to old ones. German law follows, in fact, a mixed approach, making use not only of absence of basis but also of unjust factors. German law has developed a number of tools by which to prevent automatic restitution from following in every case in which enrichment was obtained without a legal basis. First, there is the distinction between performance-based and nonperformance-based enrichments. Only the former trigger automatic restitution when there is no legal basis. Together with the rules in § 812 I 2 and § 817 1st sentence BGB, it can thus be argued that German law provides claims based on an unjust factor approach alongside the absence of basis approach in § 812 I 1 BGB. Secondly, the development of new bases, the technical notion of invalidity in the BGB, and the creative application of such rules and defences ensure that restitution only follows when it is felt that policy reasons justify it. Thirdly, practical issues are often solved in Germany by the de facto application of an unjust factors system. Fourthly, many restitutionary rights, for example the power to avoid, are regulated outside the law of unjustified enrichment and not explained by an absence of basis model. All this shows that the elegant general rule of absence of basis loses much of its charm when inspected in more detail. Professor Meier suggests that the law be broken down into specific categories, scrutinising in every individual case what makes restitution desirable.185 This is certainly an approach which is easier to apply and which serves the needs of practice much better than a broad theory which has to be limited considerably to make it workable. It might be noted, however, that this is exactly what the unjust factor approach is doing. While the absence of basis approach may provide a useful cross-check for new cases,186 it is submitted that it should not be adopted by English law. English and German law aim to answer the same questions, namely what kind of events and policy considerations should trigger restitution.187 This question may be equally difficult to answer in both countries. English law, however, has the advantage of answering the question in a more straightforward way. Embracing an absence of basis approach would force English law to give up what is itself an approach with its own elegance and beauty188 in exchange for an approach that would make English law more untidy and decisions more difficult to predict. Meier (n 5) 361. Burrows (n 59) 114. 187 From a Canadian perspective, see Smith, ‘Demystifying Juristic Reasons’ (n 5) 302. 188 Burrows (n 5) 36, 46, n 61. 185 186

12 ‘Public Law Restitutionary Claims’: The German Perspective BIRKE HÄCKER*

A. INTRODUCTION

A recent spate of cases concerning the circumstances under which UK taxpayers can claim restitution from the Inland Revenue for taxes levied in breach of European Community (or since the Lisbon Treaty: European Union) law1 has given fresh impetus to a debate simmering amongst Common Lawyers for some time. In their wake, the repercussions of the landmark House of Lords decision in Woolwich Equitable Building Society v IRC 2 have come under close scrutiny. The rationale underpinning Woolwich and the consequences flowing from the ruling are currently being reassessed in the light of what Lord Hoffmann in Deutsche Morgan Grenfell Group plc v IRC 3 described as the ‘forensic fall-out’4 from the Metallgesellschaft litigation5 and what has now culminated in the Supreme Court’s ruling in the Franked Investment Income (FII) case.6 The proper relationship between the ordinary private law of unjust enrichment and public law principles has become a particular focus of attention. Rebecca Williams’ masterly study on the subject, arguing for a ‘hybrid’ approach, contains a detailed comparative analysis of the relevant French and European Community/Union jurisprudence.7 It is the purpose of this chapter to add to the discussion a few comments from the per spective of German law.8 In Germany, the so-called ‘public law restitutionary claim’ * I am grateful to the participants of the ‘Restitution of Overpaid Tax’ conference held in Oxford in July 2010 and to those attending the FII discussion meeting in London on 18 June 2012. All translations are my own. 1 This chapter still speaks of ‘European Community’ (EC) law in places because many of the cases mentioned were decided under the first pillar of the pre-Lisbon structure of the Union. From December 2009 onwards, however, the only point of reference is the ‘European Union’ (EU). To avoid ambiguities, ‘EC’ and ‘EU’ law are here used side by side where appropriate. 2 Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] AC 70 (HL) (hereafter ‘Woolwich’). 3 Deutsche Morgan Grenfell Group plc v Inland Revenue Commissioners [2006] UKHL 49, [2007] 1 AC 558 (hereafter ‘Deutsche Morgan Grenfell’). 4 ibid [2]. 5 Cases C-397/98 and C-410/98 Metallgesellschaft Ltd v Inland Revenue Commissioners and Hoechst AG v Inland Revenue Commissioners [2001] ECR I-1727; [2001] Ch 620 (ECJ) (hereafter ‘Metallgesellschaft’). 6 Test Claimants in the Franked Investment Income Group Litigation v HM Revenue Commissioners [2012] UKSC 19, [2012] 2 AC 337 (hereafter ‘FII (SC)’). 7 R Williams, Unjust Enrichment and Public Law: A Comparative Study of England, France and the EU (Oxford, Hart Publishing, 2010). See also Rebecca Williams’ contribution in chapter 2 of the present volume. 8 The reader may also be referred to the summary of the German law on ultra vires demands given by T Krebs, Restitution at the Crossroads: A Comparative Study (London, Cavendish, 2001) 192–97.

240 Birke Häcker (öffentlich-rechtlicher Erstattungsanspruch) is a well-established legal concept which differs in certain significant respects from its private law counterpart. Using the German experience as a point of contrast and comparison may – without being prescriptive in any way – help catalyse the English debate.9 The chapter will begin by tracing the development of the German ‘public law restitutionary claim’ and its emancipation from ordinary private law during the first quarter of the twentieth century. It will then address three particular aspects which have arisen – in different ways – on either side of the Channel: first, how to establish a restitutionary claim in the public law context; secondly, the availability of interest on such a claim; and thirdly, the operation of the change of position defence. For the sake of completeness, the discussion will take account both of situations where a claim is brought against the state and where it is brought by the state.10 What the present chapter will, on the other hand, not do is provide a detailed account of how the normal rules governing public law restitutionary claims in Germany have become overlaid by specific requirements of EC/EU law when Community/Union concerns are at stake. Where national rules need to be modified or even disapplied to comply with EC/EU law, this will be mentioned (only) in the footnotes.11

B. THE ORIGINS AND FOUNDATION OF PUBLIC LAW RESTITUTIONARY CLAIMS

Looking at the developments triggered by the Woolwich decision almost exactly 20 years ago12 and the historical origins of the public law restitutionary claim in Germany, one finds some striking similarities, but also a number of differences. In pre-Woolwich England, restitutionary claims by and against the executive were – by and large – perceived purely as matters of private law.13 In line with the Diceyan notion that ‘every man, whatever be his rank or condition, is subject to the ordinary law of the realm and amenable to the jurisdiction of the ordinary tribunals’,14 courts did their best to fit cases involving public bodies into the general framework of ‘restitution’ gradually emerging as a discrete category from beneath the old forms of action. Woolwich highlighted that this was not always readily feasible.15 It raised the still virulent question about the extent to 9 On the Common Law side, the main focus of this chapter is of English law, although it will sometimes throw side-glances at developments in other Common Law jurisdictions. 10 A clarification is apposite at this point. When the word ‘state’ is used, it is taken as a shorthand for various entities including not only the state in the narrow sense (ie central government as well as, in Germany, the Länder) but also local authorities and other public bodies with separate legal identities. The chapter will not specifically consider situations where the claim is both by and against the ‘state’ (ie, cases where one public body demands restitution from another). 11 cf especially nn 234, 239 and 240 as well as nn 212, 250. On the German position specifically as regards restitution of tax levied in breach of EC/EU law see U Lange, Der Anspruch auf Erstattung gemeinschaftsrechtswidrig erhobener Steuern (Berlin, Duncker & Humblot, 2008). 12 The decision in Woolwich (n 2) was handed down on 20 July 1992. 13 This is not to deny that certain aspects of (what we now call) the law of unjust enrichment had long been subtly influenced by public law considerations. The existence of the colore officii doctrine vividly illustrates this for payments made by individuals to public officials, and the case of Auckland Harbour Board v R [1924] AC 318 (PC) does the same for the reverse scenario, namely ultra vires receipt from the state. The fact that the latter was nominally subject to a traceability requirement (cf end of n 124) does not detract from its distinct public law flavour. 14 AV Dicey, Introduction to the Study of the Law of the Constitution, 8th edn (London, Macmillan, 1915) 189. 15 In Woolwich (n 2), none of the traditional unjust factors could be made out. For an analysis and discussion see Williams, Unjust Enrichment and Public Law (n 7) 23–30.

‘Public Law Restitutionary Claims’ 241

which public law principles influenced or superseded the ordinary private law of ‘unjust enrichment’, as the subject was by then coming to be called. On the one hand, the pure private law stance could still be maintained by lining up a new ‘policy-based’ unjust factor alongside the existing grounds of recovery such as mistake and duress.16 On the other end of the spectrum, some academics argued that Woolwich-type cases should be regarded as exclusively a matter of public law, being based on a ‘free-standing public law principle enforceable in the Administrative Court’.17 A third group of authors began to advocate an intermediate, ‘hybrid’ approach by means of which public law principles were to be more overtly ‘finessed’18 into private law causes of action, such that – as Rebecca Williams puts it: the reason for restitution, the ‘unjust factor’, is provided by the existence of a public law ultra vires event, leaving the definition of enrichment, the claimant’s expense and relevant defences to be worked out by the law of unjust enrichment in the usual manner.19

Woolwich itself is indeed compatible with this hybrid approach.20 Yet in Kingstreet Investments Ltd v New Brunswick (Department of Finance),21 the Supreme Court of Canada endorsed the pure public law explanation,22 while the House of Lords ruling in Deutsche Morgan Grenfell (to the effect that the Woolwich principle does not exclude an ordinary mistake-based claim)23 smacks more of the pure private law stance. The UK Supreme Court’s recent FII Group Litigation decision is in a similar vein to Deutsche Morgan Grenfell. Despite emphasising the public law backdrop of the Woolwich claim, their Lordships recognised potential competing grounds of recovery as being grounded in ordinary private law, thus confirming that public bodies are at any rate not exempt from the remit of the latter when performing their public functions.24 German law, too, started out treating actions for the recovery of money paid to (or by) the state as ordinary private law claims. Although public and private law had been thought of as separate since the early to mid-nineteenth century,25 responsibility for the 16 In this vein most recently A Burrows, The Law of Restitution, 3rd edn (Oxford, OUP, 2010) ch 20, 498–520. This is not to say that advocates of the pure private law stance fail to recognise the public law dimension of Woolwich-type cases. Their approach is simply to convert public policy into a new private law unjust factor. 17 J Alder, ‘Restitution in public law: bearing the cost of unlawful state action’ (2002) 22 Legal Studies 165. cf also Ho Hok Lai, ‘Beyond Restitution and into Public Law’ [1993] Singapore Journal of Legal Studies 582. 18 This term is borrowed from J Beatson, ‘Finessing substantive “public law” principles into “private law” relations’ [1997] Acta Juridica 1. Beatson discusses the Woolwich case at 16–19 and 23. 19 Williams, Unjust Enrichment and Public Law (n 7) 38. An overview of the academic debate with further references is provided by C Mitchell and P Oliver, ‘Unjust Enrichment and the Idea of Public Law’ in R Chambers, C Mitchell and J Penner (eds), Philosophical Foundations of the Law of Unjust Enrichment (Oxford, OUP, 2009) 394, 397–400. 20 In establishing what became commonly referred to as a ‘policy-based’ unjust factor, the majority of their Lordships did not regard themselves as stepping wholly outside the framework of the ordinary private law of restitution. They relied heavily on comparisons with the traditional grounds for recovery, especially compulsion, failure of consideration and the colore officii doctrine: Woolwich (n 2) 164–65, 172–73 (Lord Goff), 197–98 (Lord Browne-Wilkinson), 200–4 (Lord Slynn). On the other hand, however, the public law principles at stake loomed large in the speeches and were clearly of paramount importance in establishing the claim: Woolwich (n 2) especially 172 (Lord Goff), 198 (Lord Browne-Wilkinson), 203 (Lord Slynn). 21 Kingstreet Investments Ltd v New Brunswick (Department of Finance) 2007 SCC 1, [2007] SCR 3 (hereafter ‘Kingstreet’). 22 See especially the statement of Bastarache J (ibid [40]) which is set out in the text to n 60. On the Kingstreet case and the Canadian position generally see Robert Chambers’ contribution in chapter 14 of the present volume. 23 Discussed in the text to nn 129 ff. 24 cf discussion in the text to nn 136 ff. 25 For an account of the development with further references see eg H de Wall, Die Anwendbarkeit privatrechtlicher Vorschriften im Verwaltungsrecht (Tübingen, Mohr Siebeck, 1999) 9–13.

242 Birke Häcker enforcement of such claims lay with the ordinary courts rather than with the administrative tribunals that were coming to be created from the 1860s onwards.26 The ordinary courts naturally applied the general private law of ‘unjustified enrichment’, to be found – ever since the coming into force on 1 January 1900 of the Civil Code (Bürgerliches Gesetzbuch, abbreviated BGB) – in §§ 812 ff BGB. In practical terms, this arrangement served to ensure comprehensive judicial protection for aggrieved claimants, given that the jurisdiction of the administrative courts was at the time still limited to certain specifically listed causes of action.27 On a theoretical level, the application of pure private law principles could be justified by reference to the so-called ‘fisc theory’ (Fiskustheorie) of the state.28 According to this theory, the state was, on the one hand, the supreme public sovereign. As such, it could do no wrong. On the other hand, standing beside the sovereign, there was the state as ‘fisc’, conceived of as a separate legal entity and subject to the ordinary law of the land. The fisc was the public purse, answerable to citizens in the ordinary courts.29 It was thus that, when a case similar to Woolwich came to be decided in 1909, there was no question either about the jurisdiction30 of the Imperial Supreme Court (Reichsgericht, abbreviated RG: the highest court in civil and criminal matters) or about the direct31 applicability of §§ 812 ff BGB.32 Things began to change in 1920, when the Reichsgericht declined jurisdiction in a case concerning goods requisitioned during World War I and later released upon payment of an appropriate deposit.33 The claimant sought to recover the amount deposited on the basis, inter alia, that the original requisitioning had been illegal. It was held that the legality of this act of sovereign power by the state could not be challenged, either directly or indi26 On the history of the administrative courts in Germany see eg W Rüfner, ‘Die Entwicklung der Verwaltungsgerichtsbarkeit’ in KGA Jeserich, H Pohl and G-C von Unruh (eds), Deutsche Verwaltungsgeschichte, vol 3 (Stuttgart, Deutsche Verlags-Anstalt, 1984) 909–30; G Sydow, Die Verwaltungsgerichtsbarkeit des ausgehenden 19. Jahrhunderts (Heidelberg, Müller, 2000). 27 So called ‘enumerative principle’ (Enumerationsprinzip), roughly comparable to the old English system of forms of action. The assignment did not include what we would now identify as public law restitutionary claims. By contrast, the ordinary courts were competent to decide all ‘civil disputes’ (bürgerliche Rechtsstreitigkeiten) by virtue of a general clause in § 13 of the 1877 Judicature Act (Gerichtsverfassungsgesetz, abbreviated GVG). In modern German administrative law, the jurisdictional competence of the administrative courts is no longer limited in the same way. A general clause in § 40 of the Code of Procedure of the Administrative Courts (Verwaltungsgerichtsordnung, abbreviated VwGO) allocates to them ‘all public law disputes of a nonconstitutional nature’ to the extent that such disputes are not specifically delegated to another branch of the judiciary by legislation: see also n 49. 28 On the ‘fisc theory’ cf de Wall (n 25) 11–12 with further references. For a critical historical assessment see W Rüfner, Verwaltungsrechtsschutz in Preußen von 1749 bis 1842 (Bonn, Röhrscheid, 1962) 169–86. 29 The maxim ‘dulde und liquidiere’ implied that that a citizen had to ‘endure’ the primary measure, but could then seek financial redress for harm suffered as a result of it. This led to the idea of the fisc functioning as the sovereign state’s ‘scapegoat’ (Prügelknabe). 30 Contrast RG (20.1.1890), RGZ 25, 302, where the Imperial Supreme Court declined jurisdiction because it thought that – on the specific facts of the case – the dispute was one allocated to the administrative courts. This barred recourse to the ordinary courts because according to § 13 GVG, their jurisdiction was subsidiary. 31 §§ 812 ff BGB were also directly applied in RG (17.2.1903), RGZ 54, 24; RG (19.12.1905), RGZ 62, 238 (claim by the state); RG (12.5.1911), RGZ 76, 270; RG (14.12.1909), RGZ 83, 159. Later cases suggest an analogous application: RG (3.10.1913), RGZ 83, 161 (claim by the state); RG (12.6.1917), RGZ 90, 314. 32 RG (29.10.1909), RGZ 72, 152, concerning overpaid stamp duty. In Kleinwort Benson v Lincoln City Council [1999] 2 AC 349 (HL) 374–75, Lord Goff referred to this case, making the point that, as in Woolwich, it did not matter that the claimant may not have been mistaken when making the payment. § 814 BGB, the provision in the Civil Code that excludes recovery under the condictio indebiti in the absence of a liability mistake (cf text after n 108) does not in fact receive any mention in RGZ 72, 152, perhaps because the parties had confined their dispute to the question of interest (on this aspect of the case see the text to n 171). 33 RG (30.4.1920), RGZ 99, 41.

‘Public Law Restitutionary Claims’ 243

rectly, in the ordinary courts34 and that, in so far as the claimant was seeking to characterise the deposit as a private law agreement, it did not lie in the parties’ hands to alter the nature of their dispute by ‘hiding’ the real questions of public law ‘behind an ostensible private law cloak’.35 Yet this did not solve the question of what was to be done if the dispute fell clearly outside the (still limited) remit of the administrative courts.36 Here, the Reichsgericht retained jurisdiction notwithstanding any public law aspects of a given case.37 In an influential monograph published in 1921, a young academic by the name of Gerhard Lassar argued that the time had come to formally emancipate restitutionary claims by and against the state from the private law of unjustified enrichment and to openly recognise the existence of a general ‘public law restitutionary claim’.38 He maintained that §§ 812 ff BGB, being tailored to the relationship between private individuals, were neither intended nor suitable to govern cases in which the (supposed) ‘basis’ for an enrichment39 was rooted in public rather than private norms.40 Moreover, since the legislator of the early twentieth century had already enacted a number of specific provisions dealing with claims for the recovery of overpaid taxes and the like,41 it was – in Lassar’s view – possible to construct the general public law restitutionary claim by analogy with these.42 Lassar formulates the principle as follows: ‘A performance rendered under public law without cause must, if its nature is such that it has monetary value, be restored to the party at whose cost it was rendered’.43 Perhaps unsurprisingly in view of having just given up the fisc theory, the Reichsgericht soon followed suit.44 In 1923, it allowed a claim by the state against a former civil servant who had received up-front salary payments for the time in which – as it turned out – he was no longer in office.45 The court emphasised that this claim was not based on §§ 812 ff BGB but on public law.46 It held: 34 Instead, it fell to the administrative courts to decide whether the requisitioning had been lawful. To a certain extent, this argument (or one very similar to it) had already been anticipated by RGZ 25, 302: see n 30. 35 RGZ 99, 41, 45. 36 See comment in n 27. 37 cf already the observations in RG (12.3.1918), RGZ 92, 310, 313: ‘The notion of a civil dispute as presupposed by § 13 GVG [cf n 27] is not wholly co-extensive with that of a private law dispute and does not necessarily exclude claims which, from today’s perspective, are properly regarded as public law claims’. It was, however, not until RGZ 107, 189 (see text to n 45) in 1923 that restitutionary claims by and against the state were fully acknowledged as being – in appropriate circumstances – true ‘public law claims’. On the interpretation of the term ‘civil dispute’ as used in the then applicable § 13 GVG see de Wall (n 25) 12–13. 38 G Lassar, Der Erstattungsanspruch im Verwaltungs- und Finanzrecht (Berlin, Liebmann, 1921). 39 § 812 I BGB reads: ‘A person who, without legal cause, obtains something by way of another’s performance or in some other way at another’s expense is obliged to surrender to that person what he has so obtained. The same obligation exists if the legal cause subsequently lapses or if the result which a performance was intended to bring about . . . does not eventuate’. 40 Lassar, Erstattungsanspruch (n 38) 94–103. Lassar summarises his conclusion as follows (at 101): ‘The claim for the recovery of a transfer without causa falls under §§ 812 ff BGB only if the putative causa is based on a private rather than a public relationship’. 41 See Lassar, Erstattungsanspruch (n 38) 181–209. Most importantly, the first German Tax Code (Reichsab gabenordnung, abbreviated RAO), which had been passed in 1919, specifically provided for the recovery of overpaid taxes, administrative fees, penalties and the like (§§ 127 ff RAO). In terms of the jurisdictional competence to enforce this restitutionary claim, it expressly excluded recourse to the ordinary courts (§ 227 RAO). 42 Lassar, Erstattungsanspruch (n 38), especially at 225. This attempt to establish and justify the general public law restitutionary claim by way of analogy with specific codified instances has subsequently been discredited: cf the references in n 53. 43 Lassar, Erstattungsanspruch (n 38) 226: ‘Eine causa-lose öffentliche Leistung, deren Inhalt einen Vermögens wert hat, ist demjenigen zu erstatten, auf dessen Kosten sie bewirkt ist’. 44 Albeit without mentioning Lassar’s work. 45 RG (12.6.1923), RGZ 107, 189. 46 On the question of the court’s jurisdiction, which the Reichsgericht did not specifically address, see n 37 and accompanying text.

244 Birke Häcker The relationship between a civil servant and the state is a public relationship that needs to be assessed exclusively in accordance with public law even in so far as the financial consequences flowing from it are concerned. It is not subject to the principles and doctrines of private law. Nor can private law rules be applied by analogy to the relationship between a civil servant and the state. They can only be made serviceable in the present context to the extent that they express a general legal principle which applies equally to public [as to private] law and which must, in order to fill a lacuna in the existing law, therefore be recognised as a constituent part of the public rules governing the civil service.47

Today, the existence of a free-standing public law restitutionary claim is beyond any doubt. It is enforced (almost)48 exclusively in the administrative courts49 or – where appropriate – by the specialised fiscal and social security tribunals.50 Whenever a transfer made in accordance with public law principles has to be reversed, the re-transfer will also be governed by public law. The two transactions are seen as two sides of the same coin.51 This leaves only a very limited scope for genuine private law cases involving the state, such as where the sale of a car to or by a public body or the purchase of office stationary has to be unwound.52 It will be seen in the next part that, within the field of public law, a plethora of specific legislation has been passed since the 1920s, so that the default ‘general’ public law restitutionary claim has lost some of its practical importance. Nevertheless, the latter remains an integral part of the administrative law system, although its doctrinal underpinnings are still not finally settled. Lassar’s proposed analogy with the codified public law grounds of recovery has not found favour.53 For a while the analogy with §§ 812 ff BGB was popular,54 but this is now widely rejected as unnecessary for lack of a real lacuna in the law.55 RGZ 107, 189, 189–90. There is a relatively small group of cases in which the proper qualification of the claim – and therefore the question of jurisdictional competence – is a matter of dispute: see the overview given by S Lorenz in Staudinger, Kommentar zum Bürgerlichen Gesetzbuch, §§ 812–22 (Berlin/München, Sellier de Gruyter, 2007) Vorbemerkung zu §§ 812 ff [80]–[81]. In another very small group of cases, public law restitutionary claims are allocated to the ordinary courts for other reasons, eg where the overpayment was initially received by such a court: cf BayObLG (9.12.1998), Neue Juristische Wochenschrift 1999, 1194; OLG Hamm (19.10.2000), Neue Juristische Wochenschrift: Rechtsprechungs-Report 2001, 1440. § 71 II Nr 1 GVG, providing for claims by civil servants against the state as fisc to be heard by the ordinary courts, has in practice been superseded by § 126 I BRRG (see end of n 49). 49 The general clause contained in § 40 I VwGO (already mentioned in n 27) provides: ‘Recourse to the administrative courts may be had in all public law disputes of a non-constitutional nature, unless a federal law expressly allocates such disputes to a different court. Public law disputes concerning State [as opposed to federal] law may also be allocated to a different court by State law’. For disputes between a civil servant and the state concerning the civil service relationship, the jurisdictional competence of the administrative courts is expressly provided for in § 126 BRRG (cf § 40 II 2 VwGO). 50 See § 33 FGO (jurisdiction of fiscal courts for public law disputes concerning taxes and other charges) and § 51 SGG (jurisdiction of social security tribunals for public law disputes concerning social security law). 51 So-called Kehrseitentheorie or Gleichlaufprinzip. 52 Contractual and restitutionary claims arising from these kinds of transactions are enforceable (only) in the ordinary courts. This is because – despite the state’s involvement – they are pure private law claims. The state acts like any private individual when entering the transaction; it does not exercise its sovereign power. 53 cf M Wallerath, ‘Das System der öffentlich-rechtlichen Erstattungsanspüche’ [1972] Die Öffentliche Verwaltung 221, who puts this down to the great divergence between the various statutory provisions. See also F Schoch, ‘Der öffentlichrechtliche Erstattungsanspruch’ [1994] Juristische Ausbildung 82, 84 (text after fn 34); K Windhorst, ‘Staatshaftungsrecht’ [1996] Juristische Schulung 894, 895. 54 cf E Tiedau, ‘Die öffentlich-rechtliche Bereicherungsklage und ihre Geltendmachung im Rechtswege’ [1952] Monatsschrift für Deutsches Recht 330; OVG Münster (20.4.1966), Die Öffentliche Verwaltung 1967, 271, 272; and still BVerwG (1.2.1980), Neue Juristische Wochenschrift 1980, 2538. 55 See only Wallerath (n 53) 222; F Ossenbühl, Staatshaftungsrecht, 5th edn (München, Beck, 1998) 422–23 (the book chapter dealing with public law restitutionary claims may alternatively be found in [1991] Neue Zeitschrift für Verwaltungsrecht 513) – note that the 6th edition of Ossenbühl’s textbook, due to be published in the course of 2013, was not yet available when this chapter was finalised; OVG Münster (27.2.1992), Neue Juristische Wochenschrift 1992, 2245. 47 48

‘Public Law Restitutionary Claims’ 245

The two most popular justifications either invoke the rule of law constitutionally enshrined in Article 20 III of the German Basic Law (Grundgesetz, abbreviated GG) or simply regard the postulate that unjustified enrichments have to be reversed as a general principle of the law expressed both through §§ 812 ff BGB (for private law) and through the public law restitutionary claim.56 Which of these explanations ultimately prevails is nowadays thought not to matter any longer, given that the general public law restitutionary claim is so well established that it has already obtained the status of customary law.57 In 1985, the Federal Administrative Court (Bundesverwaltungsgericht, abbreviated BVerwG) thus stated in connection with a misdirected postal order: Performances rendered without a legal basis and other transfers of wealth lacking a legal basis have to be reversed. This legal principle, which flows directly from the postulates of corrective justice, has found its private law expression in the provisions of §§ 812 ff BGB concerning unjustified enrichments; in public law it is reflected within the various areas in a plethora of provisions governing restitution of what has been obtained without legal basis within the specific area in question. But even where – as here – there is no express legislative provision, transfers of wealth lacking a legal basis [in the public law context] have to be reversed. This is the realm of the public law restitutionary claim which has been recognised for so long that judges and academic writers are already speaking of it as customary law. . .58

In the light of its clear affiliation with public law, it is tempting to compare the modern public law restitutionary claim of German law with the ‘exclusively public’ view of Woolwich-type cases so distinctly espoused by the Supreme Court of Canada in the Kingstreet Investments decision.59 There are indeed remarkable similarities between the justifications advanced by the German authorities and the reasoning of Bastarache J in Kingstreet. He said: Restitution for ultra vires taxes does not fit squarely within either of the established categories of restitution. The better view is that it comprises a third category distinct from unjust enrichment. Actions for recovery of taxes collected without legal authority and actions of unjust enrichment both address concerns of restitutionary justice, but these remedies developed in our legal system along separate paths for distinct purposes. The action for recovery of taxes is firmly grounded, as a public law remedy in a constitutional principle stemming from democracy’s earliest attempts to circumscribe government’s power within the rule of law.60

On the other hand, however, it would be wrong to assume that the public law restitutionary claim can be developed and applied wholly without reference to the corresponding private law principles. In the above-mentioned postal order case,61 the Federal Administrative Court went on to observe that although the free-standing nature of the public law restitutionary claim was beyond doubt, this did not mean that is was impossible or inappropriate to determine its exact scope by drawing on the rules enshrined in §§ 812 ff BGB: ‘For insofar as the private law provisions are mere expressions of a general legal

56 See only Ossenbühl (n 55) 422; Windhorst (n 53) 895; BVerwG (12.3.1985), BVerwGE 71, 85, 87–88 (see text to n 58); BVerwG (15.6.2006), Neue Juristische Wochenschrift 2006, 3225, 3226. 57 cf Schoch (n 53) 84–85; Ossenbühl (n 55) 422–23. 58 BVerwG (12.3.1985), BVerwGE 71, 85, 87–88 = Neue Juristische Wochenschrift 1985, 2436. 59 See n 21 and accompanying text. 60 Kingstreet (n 21) [40]. 61 See n 58 and accompanying text.

246 Birke Häcker principle equally valid in the public law context, there can be no fundamental objections to applying them [ie the private law provisions] by analogy’.62 This sounds a little like the ‘hybrid’ approach propagated by Rebecca Williams, according to which public law determines the reason for restitution while other issues like the valuation of the relevant enrichment and the availability of defences are left to private law principles.63 In essence, the methodology is the same: once established, the public law restitutionary claim behaves like a private law restitutionary claim unless there is a particular reason why the private law rules ought to be modified so as to take account of the public law context.64 There are, indeed, authors in Germany who worry that courts may have lost sight of some of the more subtle parallels between private and public law restitutionary claims and whose aim is to re-establish the unity of the legal system by drawing attention to common features.65 The main theoretical difference between the German model and the ‘hybrid’ approach is, then, that the former employs analogies to private law in order to determine the scope of the claim,66 while the latter has direct recourse to private law. The main practical difference relates to jurisdictional competence. Whereas in German law the public law restitutionary claim has to be enforced via the administrative courts or similar tribunals,67 a postulate shared by advocates of the pure public law interpretation of Woolwich,68 proponents of the hybrid approach have a slight preference for the private law procedure.69 Too much should not be made of this distinction. Most Common Law countries never developed the stark dichotomy between the ordinary (civil and criminal) courts and a separate branch of administrative courts.70 For reasons of procedural expediency, public law restitutionary claims in Germany are best decided by the administrative courts,71 whose verdict on the ‘legal basis’ for the transfer will usually be indispensable in any event.72 62 BVerwGE 71, 85, 88–89 = Neue Juristische Wochenschrift 1985, 2436. The question at issue was whether the provisions of the BGB governing the circumstances under which a defendant can or cannot rely on disenrichment (§§ 818 III, IV, 819 I BGB) were applicable to the public law restitutionary claim. It was ultimately held that they were not: see discussion in the text to nn 226–31. 63 See the text to n 19. 64 Two particular issues are discussed below, namely the availability of interest (Part D) and the problem of dealing with a defendant-enrichee’s changes of position (Part E). 65 Most prominently, W Lorenz, ‘Verbindungslinien zwischen öffentlichem Erstattungsanspruch und zivilrechtlichem Bereicherungsausgleich’ in P Badura and R Schulz (eds), Wege und Verfahren des Verfassungslebens: Festschrift für Peter Lerche zum 65. Geburtstag (München, Beck, 1993) 929–48, writes (at 931): ‘The problem raised by the law of restitution does not . . . exhaust itself in the question about the scope of the duty to surrender [an unjustified enrichment]. A closer inspection of the relevant jurisprudence in the areas of administrative and social security law as well as tax law shows that case constellations recur with which private law has been grappling for a long time. Where public law and private law evince such common substantive problems, one will therefore have to ask whether we ought not to aspire to reach the same or similar solutions. If there are no morally significant differences between the two fields, this could gradually give birth to uniform legal concepts and thus boost the idea of the unity of the legal system’. cf also M Weber, ‘Anwendbarkeit bereicherungsrechtlicher Grundsätze auf den öffentlich-rechtlichen Erstattungsanspruch nach § 37 Abs. 2 der Abgabenordnung’ [1992] Betriebs-Berater 404. 66 The claim itself is said to be of a public nature – any analogy with §§ 812 ff BGB being firmly rejected (see n 55 and accompanying text). However, it is interesting to note that the Federal Fiscal Court (Bundesfinanzhof, abbreviated BFH) once actually resorted to the private law principles developed under §§ 812 ff BGB in order to determine against whom a particular claim ought to be brought, ie the proper defendant: BFH (19.10.1976), BFHE 120, 424. 67 See nn 48–50 and accompanying text. 68 See n 17 and accompanying text. 69 Williams, Unjust Enrichment and Public Law (n 7) 49–52. 70 In England, the Administrative Court is in fact formally a sub-division of the High Court. 71 Or – where appropriate – by the fiscal or social security tribunals: see n 50 and accompanying text. 72 As will be seen in Part C, the ‘legal basis’ for performances rendered in the public law context is usually constituted by a so-called ‘administrative act’ (Verwaltungsakt) which can only be jettisoned by the administrative courts or other specialised tribunals (see text to nn 79–88). Once the legal basis for the transfer is thus destroyed, restitution follows, and the court may make an ancillary order to this effect.

‘Public Law Restitutionary Claims’ 247

In England, where public law is frequently and competently applied by the ordinary civil courts as an incidental matter in private law proceedings, there is much less pressure to assign restitutionary claims by or against the state to specialised ‘public law’ tribunals.73

C. ESTABLISHING A CLAIM

The previous part has outlined how a growing consciousness of the public law perspective led to the emancipation of public law restitutionary claims in Germany and to the recognition of a special ‘policy-based’ unjust factor in the Woolwich case. To the extent that a full emancipation from private law has not or not yet occurred in the Common Law world, the Woolwich claim can indeed be regarded as a hybrid, with the ‘policy-based’ unjust factor providing a gateway into public law.74 According to Rebecca Williams’ approach, the fact that a payment made or received by a public body was ultra vires by public law standards should, in and of itself alone, be a sufficient reason for restitution.75 This would put Woolwich-type cases (where the claim was brought by a taxpayer against the state) in the same category as Auckland Harbour Board v R (involving a restitutionary claim by the government)76 and all of the swaps cases.77 On the other hand, it will be seen that recent English case law does not unequivocally support such a hybrid approach, but that it instead aligns the policy-based Woolwich principle with traditional private law unjust factors such as mistake.78 In order to assess this development, a closer look at the relevant German and English law governing the establishment of ‘public law restitutionary claims’ is warranted. It is well-known and has already been mentioned that German law does not adopt an ‘unjust factor’ approach to restitution, as English law is traditionally understood to do, but that – like other Civilian jurisdictions – it follows the ‘absence of basis’ model. Both § 812 I 73 Forceful arguments against requiring claimants to initiate judicial review proceedings under CPR Part 54 (formerly RSC Order 53) before bringing an action for restitution at common law are made by Williams, Unjust Enrichment and Public Law (n 7) 49–52, and by Burrows (n 16) 508–10. But contrast Jones v Powys Local Health Board [2008] EWHC 2562 (Admin) as well as Hemming (Trading as Simply Pleasure Ltd) v Westminster City Council [2012] EWHC 1260 (Admin), on which see Williams, Unjust Enrichment and Public Law (n 7) 50–52, and her contribution in chapter 2 of the present volume (at pp 33–34). 74 See nn 19–20 and accompanying text. 75 See especially Williams, Unjust Enrichment and Public Law (n 7) 31–39, 54–55, 56–69, as well as her contribution in chapter 2 of the present volume. 76 Auckland Harbour Board v R [1924] AC 318 (PC) (hereafter ‘Auckland Harbour Board’). The parallel between cases concerning ‘ultra vires exaction by a public authority’ and ‘ultra vires payments by a public authority’ is also emphasised by Burrows (n 16) ch 20, especially 518. cf furthermore the statement by Cranston J in Charles Terence Estates Ltd v Cornwall Council [2011] EWHC 2542 (QB) that rent paid by local authorities under lease agreements which subsequently turned out to be ultra vires and void would have been recoverable but for the change of position defence (appeal pending). Note, however, the Supreme Court’s ruling in R (on the application of Child Poverty Action Group) v Secretary of State for Work and Pensions [2010] UKSC 54, [2011] 2 AC 15, to the effect that the state’s common law right to seek restitution of overpaid social security benefits was excluded by the ‘exhaustive code’ for recovery provided by s 71 of the Social Security Administration Act 1992. On ultra vires payments by public bodies and the resulting statutory and common law claims see the discussion in C Mitchell, P Mitchell and S Watterson, Goff & Jones: The Law of Unjust Enrichment (London, Sweet & Maxwell, 2011) ch 23. 77 eg Hazell v Hammersmith and Fulham London Borough Council [1992] 2 AC 1 (HL); Westdeutsche Landesbank Girozentrale v Islington London Borough Council [1996] AC 669 (HL); Guinness Mahon & Co Ltd v Kensington and Chelsea Royal London Borough Council [1999] QB 215 (CA); Kleinwort Benson Ltd v Lincoln City Council [1999] 2 AC 349 (HL). On this approach, the difficult search for a traditional unjust factor in each of the swaps cases (‘failure’ or ‘absence’ of consideration, mistake of law, etc) was unnecessary. 78 See discussion in the text to nn 129 ff.

248 Birke Häcker BGB79 and the public law restitutionary claim80 make liability depend on whether one party’s enrichment at the other’s expense was without a sufficient legal ground (causa) or juristic reason. A brief excursion into German administrative law helps in understanding how this relates to the taxonomy of public law restitutionary claims as developed since the 1920s. The recognition of free-standing public law restitutionary claims is only one product of a far more general movement emphasising the distinctness of public from private law. One of its main protagonists, Otto Mayer, introduced the concept of the so-called ‘administrative act’ (Verwaltungsakt) into the developing doctrinal framework.81 The concept has become a cornerstone of modern German administrative law. An administrative act is defined as ‘every decree, decision or other sovereign measure passed by a public authority in order to govern an individual case in the field of public law which is intended to be immediately effective vis-à-vis the citizen’.82 The administrative act is thus a means of concretising and applying general public law norms to specific cases. Where, for example, a local authority is entitled to impose a levy on residents for the use of local infrastructure,83 it will pass an appropriate administrative act of demand to validate and enforce its claim. Conversely, in order to obtain social security benefits, eligible citizens have to apply for an administrative act affirming their entitlement and specifying the exact amount payable to them. As regards its recipient, the administrative act has the force of law. Only the most glaring substantive errors or grave procedural irregularities entail the nullity of an osten sible administrative act.84 In all other cases, an individual who believes that the law has been incorrectly applied by the public authority concerned is required to challenge the decision. This often involves a first step of remonstrating with the authority in question85 and then, if necessary, taking the matter to the administrative court or other designated tribunal.86 An administrative act remains valid unless and until it is either set aside by the authority which passed it87 or struck down in court.88 Given that judicial review proceed See the translation in n 39. See Lassar’s formulation in the text to n 43. 81 See O Mayer, Deutsches Verwaltungsrecht, vol 1 (Leipzig, Duncker & Humblot, 1895) 64–65, 95; 3rd edn (Berlin, Duncker & Humblot, 1924) 60–62, 92–103. Mayer was strongly influenced by French administrative law: cf R Schmidt-De Caluwe, Der Verwaltungsakt in der Lehre Otto Mayers (Tübingen, Mohr Siebeck, 1999). 82 This is the official definition provided by § 35 I of the Administrative Procedure Act (Verwaltungs verfahrensgesetz, abbreviated VwVfG): ‘Verwaltungsakt ist jede Verfügung, Entscheidung oder andere hoheitliche Maßnahme, die eine Behörde zur Regelung eines Einzelfalls auf dem Gebiet des öffentlichen Rechts trifft und die auf unmittelbare Rechtswirkung nach außen gerichtet ist’. An equivalent provision is contained in § 118 AO. 83 So-called Kommunalabgabe, including the equivalent to council tax and various other local rates and charges. 84 § 44 VwVfG defines the circumstances under which an administrative act is to be treated as null and void, and it lists some typical invalidating factors (eg physical impossibility of performing the required act). In the tax context, an equivalent provision is contained in § 125 AO. 85 So-called Widerspruchsverfahren (§§ 68 ff VwGO). Increasingly, the German Länder are making use of an opt-out clause. A number of States have recently abolished or severely restricted pre-trial opposition proceedings in cases concerning State (as opposed to federal) administrative law. In the tax context, the opposition proceedings are known as Einspruchsverfahren (§§ 347 ff AO). 86 There are stringent time limits for launching pre-trial opposition proceedings and for taking a matter to court. In most cases, the citizen affected has to act within one month of notification (§§ 70 I, 74 I VwGO). In Woolwich (n 2) 174, Lord Goff – having outlined the German position – stated that ‘such draconian time limits as these may be too strong medicine for our taste; but the example of a general right of recovery subject to strict time limits imposed as a matter of policy is instructive for us as we seek to solve the problem in the present case’. 87 Or by a supervisory body higher up in the hierarchy (so-called Widerspruchsbehörde). Note that, apart from revisiting its decision in the course of pre-trial opposition proceedings (see n 85), a public authority may also avoid or revoke administrative acts of its own accord later down the line if circumstances change or facts emerge which warrant such a response. Such avoidance or revocation of administrative acts is dealt with in n 96 and the accompanying text. 88 In a case where the citizen objects to an administrative act and challenges it in court (so-called 79 80

‘Public Law Restitutionary Claims’ 249

ings before an administrative or other specialised court are therefore usually an indispensable prerequisite for public law restitutionary claims in any event, it makes sense to allocate jurisdiction over such claims to the same courts.89 Since its recognition in the 1920s, the general public law restitutionary claim has been supplanted by a whole host of individually codified grounds of liability catering for specific situations and taking precedence over the general (default) doctrine. Various statutes concerned with social security law provide for the recovery of payments that were not due to their recipient.90 Restitution of salary overpayments to civil servants, historically the trigger for the turnaround in the Imperial Supreme Court’s jurisprudence,91 is now also governed by legislation.92 Another important area is tax law, where the successor provisions to the ones relied on by Lassar93 continue to supersede the general public law restitutionary claim. § 37 II of the General Tax Code (Abgabenordnung, abbreviated AO) reads: If a tax [or another performance based on the tax relationship] has been paid or repaid without legal ground, the party at whose cost the payment was made has a claim against the recipient to recover the amount paid or repaid. The same applies if the legal ground for the payment or repayment subsequently lapses. [. . .]94

Yet by far the most important and broadest statutory implementation of Lassar’s doctrine is to be found in the Administrative Procedure Act (Verwaltungsverfahrensgesetz, abbreviated VwVfG), which came into force in 1977.95 It contains provisions governing the circumstances under which administrative acts may be avoided or revoked by the authority Anfechtungsklage), the administrative act will be struck down if it is unlawful and therefore violates the claimant’s rights (§ 113 I 1 VwGO). In a case where the citizen seeks to have a specific administrative act passed in his favour (so-called Verpflichtungsklage), a successful action will result in a judicial declaration requiring the public authority concerned either to pass the desired act (§ 113 V 1 VwGO) or to revisit its decision-making process (§ 113 V 2 VwGO). Courts other than the designated ones have no competence to quash administrative decisions and therefore normally have to treat even unlawful administrative acts as fully effective. 89 See already the text to n 72. Where, for example, a ratepayer successfully challenges a local authority’s demand for rates and taxes (see n 83 and accompanying text), the administrative court will not only strike down the relevant administrative act (§ 113 I 1 VwGO), but it may also make an ancillary order for repayment of any money already paid to fulfil the demand (§ 113 I 2 VwGO). Such upfront payments by citizens are very common, given that the institution of judicial review proceedings will usually not bar the enforcement of financial claims by the state (see § 80 II Nr 1 VwGO, see cf also the parallel provisions in § 361 I AO, § 69 I FGO). 90 eg § 26 II SGB IV (claims by the citizen for the recovery of overpaid pension fund contributions); § 50 SGB X (claims by the state against the citizen: § 50 I concerns payments made pursuant to an administrative act later nullified; § 50 II concerns payments made without a preceding administrative act); see also §§ 102 ff SGB X (concerning cases where one public body claims restitution from another for rendering social security services or payments under circumstances where the financial burden should properly have been borne by the other). 91 See n 45 and accompanying text. 92 eg § 12 II 1 BBesG (applying to civil servants at the federal level – there are equivalent provisions for the civil service of each of the Länder). Salary payments to civil servants are, of course, not preceded by an administrative act. The right to recover overpayments therefore follows automatically and without more. Interestingly, § 12 II 1 BBesG states that such restitutionary claims are regulated by ‘the principles of the [BGB] governing the surrender of an unjustified enrichment’. On the correct interpretation of this reference, see n 114. 93 See n 41 and the accompanying text. 94 In the German original, § 37 II AO reads: ‘Ist eine Steuer, eine Steuervergütung, ein Haftungsbetrag oder eine steuerliche Nebenleistung ohne rechtlichen Grund gezahlt oder zurückgezahlt worden, so hat derjenige, auf dessen Rechnung die Zahlung bewirkt worden ist, an den Leistungsempfänger einen Anspruch auf Erstattung des gezahlten oder zurückgezahlten Betrags. Dies gilt auch dann, wenn der rechtliche Grund für die Zahlung oder Rückzahlung später wegfällt. [...]’. By virtue of § 37 I AO, the restitutionary claim set out in § 37 II AO forms part of the ‘tax liability relationship’ (Steuerschuldverhältnis) between the citizen and the fisc. 95 References throughout this chapter are to the Federal Code of Administrative Procedure (VwVfG). There are parallel provisions in each of the Länder. It should be noted that the VwVfG is a general statute governing the administrative process, applying only to the extent that no more specific provisions regulate a given case.

250 Birke Häcker which passed them (§§ 48, 49 VwVfG).96 Following these provisions, § 49a I 1 VwVfG decrees that ‘[i]n so far as an administrative act has been avoided or revoked with retroactive effect or become ineffective due to the materialisation of a resolutive condition, performances already rendered [in accordance with it] have to be refunded’.97 § 49a VwVfG thus establishes a legislative framework for the recovery of money paid or other enriching transfers effected by a public body to an individual citizen pursuant to an administrative act which – following avoidance or revocation – no longer justifies its retention.98 Unlike its 96 Under the circumstances set out in §§ 48, 49 VwVfG, avoidance or revocation is possible even against the wishes of the person to whom the administrative act was addressed (contrast text to nn 85–87). § 48 VwVfG governs the avoidance of administrative acts which later turn out to have been unlawful all along (Rücknahme rechtswidriger Verwaltungsakte), while § 49 VwVfG concerns lawful administrative acts being overtaken by new developments and therefore having to be revoked (Widerruf rechtmäßiger Verwaltungsakte). Interestingly for lawyers brought up in the Common Law tradition, one finds amongst the criteria justifying the avoidance or revocation in a given case some which bear a striking similarity with the traditional English ‘unjust factors’, namely fraud, duress and failure of consideration. So far as they are relevant, the provisions read: § 48 VwVfG Avoidance of Unlawful Administrative Act (1) An unlawful administrative act may, even after the limitation period for challenging it has expired, be avoided wholly or in part with prospective or retroactive effect. An administrative act which has bestowed or confirmed a right or a legally relevant advantage (beneficial administrative act) may only be avoided in accordance with the restrictions set out in sub-sections 2 to 4. (2) An unlawful administrative act which grants a single or recurring monetary benefit . . . may not be avoided in so far as the beneficiary has relied on the continued existence of the administrative act and [in so far as] his reliance is deserving of protection when weighed against the public interest in avoiding the [unlawful] administrative act. His reliance will generally be deserving of protection where the beneficiary has used up what he has received or where he has made a financial disposition which he is unable to reverse or at any rate not without incurring unacceptable disadvantages. The beneficiary cannot invoke reliance where 1. he has secured the administrative act by means of fraud, threat or bribery; 2. he has secured the administrative act by means of statements which were untrue or incomplete in some relevant respect; 3. he knew of the unlawfulness of the administrative act or did not know of it as a result of his own gross negligence. In cases falling under the third sentence [of this sub-section], the administrative act will normally be avoided with retroactive effect. (3)–(4) . . . § 49 VwVfG Revocation of Lawful Administrative Act (1) A lawful non-beneficial administrative act may, even after the limitation period for challenging it has expired, be revoked wholly or in part with prospective effect except where an administrative act with the same content would have to be passed again or where its revocation is impermissible for other reasons. (2) A lawful beneficial administrative act may, even after the limitation period for challenging it has expired, be revoked wholly or in part with prospective effect only 1. if its revocation is expressly permitted by law or reserved in the administrative act [itself]; 2. if the administrative act is connected with a condition [so-called ‘charge’] which the beneficiary has not fulfilled within a stipulated time limit; ... (3) A lawful administrative act which grants a single or repeated payment of money . . . in order to further a particular aim or which is a pre-condition [for receiving such a payment] may, even after the limitation period for challenging it has expired, also be revoked wholly or in part with retroactive effect 1. if the performance is not used [at all], [or if it is] not used soon after being rendered or [if it is] no longer used for the purpose stipulated in the administrative act; 2. if the administrative act is connected with a condition [so-called ‘charge’] which the beneficiary has not fulfilled within a stipulated time limit. ... (4)–(6) . . . 97 § 49a I 1 VwVfG is subsidiary to the more specific provisions contained in eg § 50 SGB X, § 12 II BBesG or § 37 II AO (cf nn 90, 92, 94 and accompanying text). 98 Interestingly, § 49a I 2 VwVfG provides: ‘The performance to be restored is to be assessed by means of a written administrative act’. This means – in effect – that under § 49a VwVfG, the restitutionary relationship (entailed

‘Public Law Restitutionary Claims’ 251

more specific counterpart in the tax law context,99 it does not work both ways.100 And although, in practice, a great number of restitutionary claims both by and against the executive will nowadays be caught by one or the other statute, the uncodified customary principle still applies to situations not governed by legislation.101 Taking a step back, and leaving aside for the time being the distinction between the statutory and non-statutory context, it is possible to sub-divide the public law restitutionary claims of German law into three broad categories. First, cases in which a performance has been rendered pursuant to an administrative act.102 In such cases, the administrative act provides the legal basis for the recipient’s enrichment, and the restitutionary claim depends on its being void or invalidated by a competent body.103 Secondly, situations in which a public authority has entered into a contractual relationship with the other party, and where this contractual relationship is of a public law nature.104 The existence of the restitutionary claim then turns on the validity of the public law (government) contract.105 In the third and last category of case, there is neither an administrative act nor a contractual relationship, but a simple performance rendered by one party to the other for any of a number of different reasons.106 Only in this last scenario will the courts determine the presence or absence of a legal basis for the enrichment directly by reference to the substantive norms of public law. An interesting question arises in connection with public law restitutionary claims when one compares the German ‘absence of basis’ approach with the traditional English ‘unjust factor’ approach.107 Its point of departure is the observation that, as a matter of private law, the two ways of proceeding are not as far apart as may appear at first sight. Just as the by the avoidance or revocation of the original administrative act) is intended to be governed by a second administrative act which will then provide a fresh legal basis for the repayment or other return performance. 99 Set out in the text to n 94. 100 Although its wording may suggest otherwise, the provision is not intended to cover cases where the performance to be reversed is one rendered by the citizen to the state: see only F Kopp and U Ramsauer, Verwaltungsverfahrensgesetz, 12th edn (München, Beck, 2011), § 49a [3]–[4] with further references. 101 The customary ‘general’ public law restitutionary claim thus becomes relevant in four main situations. First, where a citizen successfully challenges an ordinary administrative act under which performance has already been rendered (note that when an administrative act is quashed in judicial review proceedings, the court has the power to make an adjunct order for restitution under § 113 I 2 VwGO). Secondly, where an administrative act of demand is subsequently avoided or revoked in accordance with §§ 48, 49 VwVfG by the public authority which passed it (given that § 49a VwVfG does not apply to claims against the state: see n 100 immediately above). Thirdly, where a public law or government contract has to be unwound because it turns out to be invalid: see the text accompanying nn 104–5 immediately below. And finally, when a performance is rendered under other circumstances without any legal ground (eg a simple mistaken overpayment): see the text to n 106. 102 See the examples given in the text accompanying nn 83–84. 103 See the text to nn 84–88. 104 § 54 VwVfG characterises the notion of ‘public law contract’ (öffentlich-rechtlicher Vertrag) as a contract ‘establishing, modifying or cancelling a relationship governed by public law’, applicable in particular where the public authority concerned, instead of passing an administrative act, reaches an agreement with the individual to whom the administrative act would otherwise be directed. In such a situation, the public authority – having chosen to govern by way of consensual public law contract rather than by means of the exercise of sovereign power – is bound to see the relationship through on this basis. It cannot enforce its rights under the contract or those flowing from the contract’s invalidity by means of a unilateral administrative act. 105 Just as is the case with administrative acts (cf n 84), mere unlawfulness does not render a public law contract void. Instead, the grounds of invalidity are closely circumscribed (§ 59 VwVfG). Where a public law contract is prima facie valid, it may nevertheless be retroactively rescinded in accordance with the ordinary contractual principles (§ 62 VwVfG, §§ 119 ff, 142 ff BGB). 106 eg where the state makes salary payments to a civil servant. The salary entitlement is based directly on the relevant legislation, without the need for any transposing administrative act or contract (cf § 2 I BBesG). 107 cf in this context also the remark in n 96 (about ‘unjust factors’ within §§ 48, 49 VwVfG).

252 Birke Häcker presence of a valid obligation to make the transfer usually precludes an unjust enrichment claim in English law,108 so the absence of a liability mistake or the like ordinarily bars unjustified enrichment claims in German law.109 Thus, § 814 BGB provides that ‘[w]hat has been rendered in discharge of a legal obligation cannot be reclaimed if the performing party knew that he was not obliged to perform’. Yet how does this translate into the public law context? It will be recalled that the House of Lords in Woolwich denied the need for the taxpayer to have made the payment under any kind of mistake or pressure.110 This was because, in Lord Goff’s words, ‘the simple fact that the tax was exacted unlawfully should prima facie be enough to require its repayment’.111 In German law, the particular problem can hardly arise in the tax context, since a payment is due whenever a valid administrative act says that it is.112 It is thus up to the taxpayer to challenge the administrative act in question.113 However, there are several cases where German courts have had to face up to the issue of whether § 814 BGB (or the general principle underlying it) is applicable to public law restitutionary claims. They have consistently held that it is not. Sometimes this result was reached through an appropriate interpretation of the statute establishing the claim.114 As far as the general (uncodified) doctrine is concerned, claims by individuals against the state for recovery of amounts paid in the knowledge that they were not due are frequently justified as follows: The provision [ie § 814 BGB] rests on the idea that the party rendering the performance is at liberty to comply with an obligation which does not, in fact, exist. If he knows that he is not obliged to perform and nevertheless renders performance, then he would be acting inconsistently with his 108 In Barclays Bank Ltd v WJ Simms Son & Cooke (Southern) Ltd [1980] QB 677 (QBD Comm) 695, Robert Goff J – as he then was – stated that a restitutionary claim based on causative mistake of fact would fail inter alia if the payment which the claimant seeks to recover ‘is made for good consideration, in particular if the money is paid to discharge, and does discharge, a debt owed to the payee’. 109 This aspect is also discussed by Anne Sanders in chapter 11 of the present volume (at pp 230–31). 110 cf n 15 and accompanying text. 111 Woolwich (n 2) 173. 112 See the text to nn 102–3. Interestingly, the conclusiveness of national administrative acts in cases where a tax assessment contained therein contravened EC/EU law was recently questioned before the Federal Tax Court (Bundesfinanzhof, abbreviated BFH). The court held that European law did not require administrative acts which had become final according to national rules to be disregarded, even if the taxpayer had no way of knowing about the contravention before the time period for challenging the assessment had expired: BFH (16.9.2010), case refs V R 46/09, 48/09, 49/09, 51/09, 52/09, and 57/09. The Federal Constitutional Court (Bundesverfassungsgericht, abbreviated BVerfG) has now refused permission to challenge from a constitutional point of view the BFH’s decision not to make a preliminary reference to the CJEU in these cases: BVerfG (29.5.2012 and 6.6.2012), case refs 1 BvR 1234/11, 1390/11, 1395/11, and 1403/11. 113 In the tax context, this is done by launching a protest against the tax assessment (so-called Einspruch) and then, if necessary, by bringing an action in the fiscal court. Since this challenge does not suspend the taxpayer’s liability, the argument that § 814 BGB might subsequently exclude claims for the recovery of tax paid under protest fails from the outset. First, because the taxpayer’s situation seems more analogous to that of a claimant relying on the condictio ob causam finitam (§ 812 I 2 alt 1 BGB – where § 814 BGB does not apply) than to that of a claimant bringing a condictio indebiti (§ 812 I 1 alt 1 BGB – where § 814 BGB does apply): H Meier-Branecke, ‘Die Anwendbarkeit privatrechtlicher Normen im Verwaltungsrecht’ [1926] Archiv des öffentlichen Rechts 230, 258–59. And secondly, even if § 814 BGB were applicable in principle, the provision is generally understood not to encompass payments made under protest while at the same time expressly reserving the right to seek restitution. 114 Thus, the Federal Administrative Court ruled in one case that § 814 BGB does not apply to claims based on § 12 II BBesG (see n 92). It reasoned that the reference in § 12 II BBesG to the ‘principles of the [BGB] governing the surrender of an unjustified enrichment’ was to be understood as envisaging only the legal consequences of private law unjust enrichment claims (so-called Rechtsfolgenverweisung) and not also the conditions under which unjust enrichment claims are established in private law (so-called Rechtsgrundverweisung). In the opinion of the court, § 814 BGB contained a negative constituent element of the claim and thus went to the conditions of establishing liability in the first place: BVerwG (28.2.2002), BVerwGE 116, 74, 77 = Neue Zeitschrift für Verwaltungsrecht 2002, 854, 854–55.

‘Public Law Restitutionary Claims’ 253 own previous conduct if he were later to demand back the subject-matter of his performance. The provision is thus a special means of protecting reliance; the recipient of a performance deliberately rendered in order to comply with a non-existent obligation is entitled to rely on being able to retain the subject-matter of the performance . . . This principle, however, does not fit where a public law restitutionary claim by the citizen against the executive is concerned. The public purse is bound by the [constitutional] principle which subjects all public administration to the rule of law. Its interest must be directed at eliminating an enrichment that has occurred without legal basis and at re-establishing a lawful state of affairs. This [constitutional] principle of legality also applies in the case where the public purse itself has obtained something without legal basis. Its reliance on being able to defend an unlawful gain against the claim for recovery is therefore never deserving of protection.115

This passage subordinates the general principle that a person who consciously discharges a non-existent debt is debarred from recovering his performance to competing considerations of upholding the rule of law. It is very reminiscent of the reasoning in Woolwich, where one the main arguments in favour of allowing the claimant building society to reclaim the unlawfully levied tax – the broad ground – was the fundamental constitutional doctrine that there must be no taxation without proper Parliamentary authority.116 Yet the parallels do not end here. The second – slightly narrower – reason why their Lordships in Woolwich were content to waive any requirement of mistake or actual compulsion, namely the fact that demands by the state are implicitly backed by its coercive powers,117 is also reflected in the German case law. Another justification often advanced for not applying § 814 BGB to citizens’ restitutionary claims against the state is this: The Senate is of the opinion that there is at any rate no room for an analogous application of § 814 BGB to the public law restitutionary claim of the citizen where – as in the present case – one is concerned with a performance without legal basis rendered by the citizen within a relationship of subordination as regards the public authority. The disqualification of the performing party from a restitutionary claim in accordance with § 814 BGB where he knew that he was not obliged to perform . . . rests on the general idea that inconsistent conduct is not permissible[118] . . . Yet this principle does not support the disqualification . . . if, at the time, the debtor rendered the performance under pressure . . . If, therefore, the application of § 814 BGB presupposes a performance by the debtor free from external pressure, then there can be no room for an analogous application of the provision to the recovery of a performance rendered without legal basis by the citizen in the context of a relationship of [state] supremacy . . . If a public body has used its superiority resting on its sovereign powers in order to require the citizen to enter into a [public law] contract,[119] the 115 OVG Koblenz (28.11.1991), Neue Zeitschrift für Verwaltungsrecht 1992, 796, 798. The passage is repeated – almost verbatim – by VG Darmstadt (20.1.2005), case ref 7 E 1201/01. Similar: VG Lüneburg (28.8.2002), case ref 5 A 124/01. This aspect is also emphasised by Ossenbühl (n 55) 434. 116 In Woolwich (n 2) 172, Lord Goff said that ‘the retention by the state of taxes unlawfully exacted is particularly obnoxious, because it is one of the most fundamental principles of our law – enshrined in a famous constitutional document, the Bill of Rights 1688 – that taxes should not be levied without the authority of Parliament; and full effect can only be given to that principle if the return of taxes exacted under an unlawful demand can be enforced as a matter of right’. 117 Immediately following the passage set out in n 116, Lord Goff continued (ibid): ‘[W]hen the revenue makes a demand for tax, that demand is implicitly backed by the coercive powers of the state and may well entail (as in the present case) unpleasant economic and social consequences if the taxpayer does not pay’. 118 The impermissibility of inconsistent conduct (venire contra factum proprium) is broadly equivalent to the English principle of estoppel. Interestingly, the court accepted that the principle of good faith might in appropriate circumstances estop a claimant from enforcing his or her right to restitution (so-called Ausübungssperre), but held that – on the facts of the particular case – the claimant was not debarred from seeking recovery of his performance. 119 The case concerned a public law contract (cf n 105 and accompanying text) which was void because the defendant local authority had effectively purported to ‘sell’ a planning permission to the claimant. An account of the facts in English and a detailed discussion are provided by Krebs (n 8) 194.

254 Birke Häcker latter will – in consequence of the [public authority’s] supremacy – typically be under pressure to fulfil the contract since he would otherwise have to expect the [public authority’s] sovereign powers to be used to his detriment.120

What about cases in which the state is the party seeking to recover a payment or other performance made in the knowledge that it was not due? German courts have dealt with this issue, too. Very occasionally, § 814 BGB is mentioned as a possible reason for striking out a restitutionary claim.121 In the vast majority of cases, however, courts once again emphasise that the balance of interests underlying the provision cannot be transposed into the domain of public law and that the principle of legality requires all performances rendered without a proper legal basis to be returned.122 A public authority is not free to spend public money as it likes. It must not make unauthorised voluntary payments, nor may it waive claims for their recovery. This dictates that the public authority’s knowledge as such should be irrelevant to a public law restitutionary claim.123 The argument resembles that put forward by the Privy Council in Auckland Harbour Board v R, where a decisive factor militating in favour of recovery was also the ultra vires nature of the payment.124 Yet this is not necessarily the end of the matter in German law. Under certain special circumstances, the recipient’s reliance interest may trump the public interest in upholding the rule of law.125 The citizen is therefore relieved of his liability to make restitution ‘if the private interest in protecting [his] reliance outweighs the public interest in re-establishing a distribution of assets that accords with the law’.126 Doctrinally, this limitation of the public law restitutionary claim is founded on the concept of ‘good faith’, which is regarded as ‘a general principle of the law pervading the entire legal order’.127 It would seem that only the most glaring procedural errors will estop a public authority from reclaiming money paid out ultra vires. Looking at the case law, the relevant decisions tend to be concerned with grossly incompetent public officials on the one side and individual defendants claiming (but unable to prove) that they have irreversibly changed their position on the other.128 120 VGH Mannheim (18.10.1990), Neue Zeitschrift für Verwaltungsrecht 1991, 583, 587. The aspect of latent state pressure is also emphasised by Lorenz, Festschrift Lerche (n 65) 938–39 fn 38; Schoch (n 53) 89. 121 VG Berlin (21.10.2003), case ref 8 A 218.00, Zeitschrift für das Fürsorgewesen 2005, 162 (obiter). Note that this case concerned a restitutionary claim between two public authorities. 122 VGH Kassel (17.7.1990), Neue Juristische Wochenschrift 1991, 510, 512; OVG Weimar (22.10.2002), Neue Zeitschrift für Verwaltungsrecht: Rechtsprechungs-Report 2003, 830, 832; VG Lüneburg (24.6.2003), case ref 4 A 78/02; VG Gera (11.3.2004), case ref 4 K 277.02.GE, ThürVGRspr 2005, 196. 123 See Windhorst (n 53) 899. Interestingly, § 814 BGB has been held to be inapplicable even in cases where courts have classified the particular claims at issue as falling under the private law jurisdiction of the ordinary courts: OLG Karlsruhe (30.12.1987), Neue Juristische Wochenschrift 1988, 1920, and see also Lorenz, Festschrift Lerche (n 65) 938. 124 In Auckland Harbour Board (n 76) 326–27, Viscount Haldane said: ‘[I]t has been a principle of the British Constitution now for more than two centuries, a principle which their Lordships understand to have been inherited in the Constitution of New Zealand with the same stringency, that no money can be taken out of the consolidated Fund into which the revenues of the State have been paid, excepting under a distinct authorization from Parliament itself. The days are long gone by in which the Crown, or its servants, apart from Parliament, could give such an authorization or ratify an improper payment. Any payment out of the consolidated fund made without Parliamentary authority is simply illegal and ultra vires, and may be recovered by the Government if it can, as here, be traced’. The suggested traceability requirement is an odd feature about the case which ought arguably not to be taken too seriously: cf Burrows (n 16) 518; Goff & Jones (n 76) [23–25]. 125 cf also the closely related discussion on the ‘disenrichment’ defence (§ 818 III BGB) in the text to nn 226–39. 126 VGH Kassel (17.7.1990), Neue Juristische Wochenschrift 1991, 510, 512. Similar: VG Lüneburg (24.6.2003), case ref 4 A 78/02. 127 OVG Münster (27.2.1992), Neue Juristische Wochenschrift 1992, 2245. cf also OVG Weimar (18.11.2009), case ref 1 KO 693/07. Emphasising the exceptional nature of an estoppel based on good faith: OVG Weimar (22.10.2002), Neue Zeitschrift für Verwaltungsrecht: Rechtsprechungs-Report 2003, 830, 832. 128 cf especially the facts of OVG Münster (27.2.1992), Neue Juristische Wochenschrift 1992, 2245.

‘Public Law Restitutionary Claims’ 255

The most interesting aspect about English law from a comparative perspective is that ‘public law restitutionary claims’ by citizens against the state (if we may call them this for the time being) are not confined to the Woolwich principle but can – following the House of Lords in Deutsche Morgan Grenfell – also be brought under the traditional unjust factor umbrella of ‘mistake’.129 In Deutsche Morgan Grenfell itself, this enabled the claimants to circumvent the otherwise applicable limitation period.130 I have argued elsewhere that it was wise of their Lordships to refuse the invitation to move English law over to a wholesale ‘absence of basis’ model of unjust enrichment.131 That a generalised condictio indebiti is still not on the cards was recently confirmed by the Supreme Court in the FII Group Litigation.132 However, until Deutsche Morgan Grenfell and especially FII, it seemed perfectly possible to understand the Woolwich policy unjust factor as effectively denoting the lack of a legal basis in the public law sphere. The argument would have been that while private defendants were entitled to keep enrichments bestowed upon them in circumstances where the claimant’s consent was neither impaired nor qualified, the state was strictly bound to return anything which it had not been specifically empowered to receive or keep. All would have boiled down to the vires question so elegantly incorporated into Rebecca Williams’ ‘hybrid’ approach.133 Had this view of Woolwich won the day, a comparative side-glance at German law might have exposed the recognition of a concurrent mistake-based claim in Deutsche Morgan Grenfell as a fifth wheel on the coach.134 For if the distinguishing feature of the Woolwich case lay in dispensing with the need for a liability mistake in the light of considerations dictated by public policy (essentially the same considerations as persuaded the German courts to disapply § 814 BGB in the public law context), then it should have been unnecessary to posit a separate mistake-based claim as a competitor to Woolwich, because the former would inevitably have encompassed all of the elements of the latter.135 Deutsche Morgan Grenfell did not expressly advert to the extent of the overlap, but in FII the Supreme Court made it clear that mistake-based claims retained an independent field of application within the public law context which Woolwich could not cover. While rejecting the suggestion that a formal or official ‘demand’ was necessary in order to trigger the See the text accompanying n 23 above. They were able to take advantage of the ‘extension’ provided by section 32(1)(c) of the Limitation Act 1980, according to which the limitation period in actions ‘for relief from the consequences of a mistake’ does not begin to run until the claimant has discovered the mistake or could with reasonable diligence have discovered it. 131 B Häcker, ‘Still at the Crossroads’ (2007) 123 Law Quarterly Review 177, 182. The reader may also be referred to Anne Sanders’ contribution in chapter 11 of the present volume which explains the various follow-on problems that could result from such a shift. 132 FII (SC) (n 6) [162] (Lord Sumption). 133 See n 75 and accompanying text. 134 Rebecca Williams in her monograph similarly argued for a ‘hierarchy of reasons for restitution’ under which the public law (ultra vires) unjust factor established in Woolwich was to take precedence over private law unjust factors such as mistake or duress: Williams, Unjust Enrichment and Public Law (n 7) especially at 70–122 (ch 4), 275–83 (conclusion). 135 In order to establish the mistake-based claim allowed in Deutsche Morgan Grenfell (n 3), a claimant has to show two things. First, that he fell victim to an error of some sort which caused him to make the payment. In Deutsche Morgan Grenfell itself, identifying the claimant bank’s mistake was no easy task: cf C Mitchell, ‘Mistaken Tax Payments’ [2007] Restitution Law Review 123, 126–28; Williams, Unjust Enrichment and Public Law (n 7) 94–96. And secondly, as a preliminary or incidental matter, he has to demonstrate that the alleged or supposed tax liability did not in fact exist. This is necessary to establish a liability mistake or – further down the line – to avoid running into the bar to restitution which exists when a mistaken payment discharges a valid liability (cf n 108). It is a contentious question whether or not Deutsche Morgan Grenfell was correctly decided on the second count: see only the dissenting speech delivered by Lord Scott: Deutsche Morgan Grenfell (n 3) [81]–[89], discussed by Williams, Unjust Enrichment and Public Law (n 7) 78–86. On an ‘ultra vires’ or ‘public law absence of basis’ conception of Woolwich, only the second element has to be shown before a claim will be allowed. 129 130

256 Birke Häcker Woolwich principle,136 their Lordships nevertheless required something more than a mere absence of public law basis before a performance could be reclaimed from the state. Lord Walker spoke in terms of a ‘perceived obligation to pay’ and said that [w]e should restate the Woolwich principle so as to cover all sums paid to a public authority in response to (and sufficiently causally connected with) an apparent statutory requirement to pay tax which (in fact and in law) is not lawfully due.137

Lord Sumption was even more explicit, explaining the relationship between the two types of claim as follows: [T]he principle in Woolwich Equitable applies generally in all cases where tax has been charged unlawfully, whether by the legislature or by the tax authorities, whether by overt threats or demands or simply by the taxpayer’s appreciation of the consequences of not paying, and whether the taxpayer was mistaken or not. By comparison, an action for restitution on the ground of mistake is a more limited remedy, for the obvious reason that it is necessary to prove the mistake. That will not always be easy, as the facts of Woolwich Equitable itself demonstrate. On the face of it, the only case where the Woolwich Equitable cause of action is probably not available and where a claimant may therefore need a right of restitution for mistake, is the case where there is no unlawful exaction of tax but the taxpayer has simply paid in error: e.g. he has miscalculated his liability under a self-assessed tax or accidentally paid twice.138

These statements indicate that the unjust factor encapsulated by the Woolwich principle is not merely the policy that the state must not keep what it is not entitled to as a matter of public law – though it is hard to say whether the gist of the action lies primarily in the performing citizen’s state of mind (as the language of ‘perceived obligation’ might be taken to suggest) or in the fact that the whole problem somehow emanated from the state’s sphere of responsibility (as the language of ‘unlawful exaction’ strongly implies). The exact limits of the Woolwich claim post-FII thus remain to be worked out. For the time being, however, it would seem that the ordinary mistake-based claim is now indispensable in the public law context to complement the Woolwich principle, catering for situations in which the citizen falls victim to his own miscalculation or makes an accidental overpayment. Moreover, we must infer that a person who deliberately overpays the state without being required to do so (such as a taxpayer who hopes to receive a better return on his ‘investment’ than he could get on the market in times of historically low interest rates) would be precluded from recovering altogether. In this respect, English and German law now apparently diverge.139 The bifurcation of ‘public law restitutionary claims’ in England brought about by Deutsche Morgan Grenfell and now confirmed by FII is probably here to stay. What must be 136 FII (SC) (n 6) [64]–[83] (Lord Walker), [171]–[174] (Lord Sumption), upholding the Court of Appeal decision in this respect. 137 ibid [79]. At [81], his Lordship added that ‘to decide that an official demand is not a prerequisite for the recovery of tax paid when not due . . . would not be a decisive step towards a general “absence of basis” principle in place of the “unjust factors” approach that has prevailed in the past. It would merely be creating . . . a rather larger island of recovery in respect of undue tax’. 138 ibid [186]. There is an interesting parallel here with the decision of the Supreme Court of Canada in Alberta v Elder Advocates of Alberta Society 2011 SCC 24, [2011] 2 SCR 261, where a distinction was drawn (at [91]) between, on the one hand, ‘claims relating to restitution of taxes levied under an ultra vires statute’ and, on the other hand, ordinary restitutionary claims arising in the public law context: ‘Kingstreet [cf n 21 and accompanying text] leaves open the possibility of suing for restitution in other circumstances’. The passage is discussed by Robert Chambers in chapter 14 of the present volume (at pp 309–10). 139 On the position in German law see the text to n 115, stressing that the rule of law itself requires public bodies to return any enrichment received without a valid legal basis.

‘Public Law Restitutionary Claims’ 257

hoped for is that the incidents of claims based on the Woolwich principle and those based on a traditional unjust factor such as mistake will, in due course, be streamlined either by the courts themselves or by the legislature, at least in so far as divergences between them are not directly tied up with a special feature of the relevant unjust factor. To some of these incidents we must now turn.

D. THE AVAILABILITY OF INTEREST

In Sempra Metals Ltd v IRC,140 the House of Lords used the Metallgesellschaft litigation as a springboard to remedy what had long been perceived as a defect in English law and – by a majority – held that compound interest should be available as of right for restitutionary claims brought at common law.141 Previously it had been thought, following Westdeutsche Landesbank Girozentrale v Islington LBC,142 that while courts could make an award of simple interest under section 35A of the Supreme Court Act 1981, they did not have the power to order payment of compound interest unless a given case fell within equity’s exclusive jurisdiction.143 Lord Mance in Sempra considered the equitable route to recovery preferable and argued in favour of a non-exclusive, discretionary approach.144 The question of jurisdictional basis apart, their Lordships in Sempra only disagreed over what the claimant taxpayer had to establish in evidence in order to substantiate his claim. Did he have to positively prove that the Revenue had been enriched by actually using the money received, such as by investing it to earn interest, or was there a (rebuttable) presumption that the Revenue had at least saved itself the cost of borrowing an equivalent amount elsewhere? While two of their Lordships adopted the former approach,145 the majority preferred to rely on the presumption, but quantified the state’s enrichment by reference to the relatively low government rate of borrowing rather than the higher commercial rate.146 It will be seen that in Germany, there are special rules governing the payment of (simple) interest in tax cases.147 Yet apart from these, the prospects of German claimants recovering any interest when seeking restitution from the state are bleak. Claimants would therefore appear to be in a markedly worse position than litigants in England.148 This is so despite the 140 Sempra Metals Ltd (formerly Metallgesellschaft Ltd) v Inland Revenue Commissioners [2007] UKHL 34, [2008] 1 AC 561 (hereafter ‘Sempra’). A comparative case note on this decision, which will be of interest to the German reader, is by S Arnold, ‘Ansprüche auf Zinszahlung im englischen Schadensersatz- und Bereicherungsrecht’ [2008] Recht der internationalen Wirtschaft 264. 141 Sempra (n 140) [22], [26], [36] (Lord Hope), [112]–[113] (Lord Nicholls), [149]–[153] (Lord Scott). 142 Westdeutsche Landesbank Girozentrale v Islington London Borough Council [1996] AC 669 (HL) (hereafter ‘Westdeutsche’). 143 In Westdeutsche (n 142), it was held that a mistaken payment did not – or at least not without more – lead to the creation of a trust in the payer’s favour. While the majority thought that compound interest was therefore not available to the claimant seeking restitution, Lord Goff and Lord Woolf both dissented on the ground that an award of compound interest was in their view necessary to ensure full restitution at common law. 144 Sempra (n 140) [231]–[240]. Lord Walker’s approach is more compatible with Lord Mance’s than with that of the majority: cf ibid [184]–[187]. 145 ibid [145]–[148] (Lord Scott), [231]–[234] (Lord Mance). 146 ibid [46]–[49] (Lord Hope), [116]–[119] (Lord Nicholls), [188] (Lord Walker). 147 See n 161 below and accompanying text. 148 Even where the above-mentioned tax statute applies, a claimant can only recover simple interest (albeit at a currently favourable rate of 6% per annum). In England, too, there are statutory provisions concerning the availability of simple interest on tax repayments. Their exact relationship with Sempra is still being worked out: cf only FJ Chalke Ltd v Revenue and Customs Commissioners [2009] EWHC 952 (Ch), [2010] EWCA Civ 313; Littlewoods

258 Birke Häcker fact that the basic starting point of German law is the same as the post-Sempra English position. As a matter of private law, § 818 I BGB provides that the enrichment to be disgorged by the recipient under ordinary private law principles ‘extends to secondary benefits derived from the performance’.149 This has always been understood to refer to any interest actually earned.150 According to the more recent jurisprudence of the Federal Supreme Court (Bundesgerichtshof, abbreviated BGH), it also extends to expenditure saved as a result of using the money received to repay an existing loan earlier than would otherwise have been the case.151 Where it is established that the defendant has actually benefited from the ‘time value’ of the capital obtained from the claimant,152 there is a presumption that the extent of his benefit equates to the ordinary cost of market borrowing.153 All of these principles are applied almost as a matter of course to public law restitutionary claims by the state against private individuals.154 As a general rule, defendants are therefore required to disgorge not only the capital sum obtained without a sufficient public law basis, but also any interest earned or saved using it.155 If the defendant is a commercial enterprise and the money has become untraceably mixed with its assets, then the difficulties of establishing the use value actually realised are overcome by means of the presumption that a businessman does not leave capital lying idle.156 Sometimes a statute will even Retail Ltd v Revenue and Customs Commissioners [2010] EWHC 1071 (Ch); John Wilkins (Motor Engineers) Ltd v Revenue and Customs Commissioners [2010] EWCA Civ 923. In the preliminary reference made by Vos J in the Littewoods case, the Court of Justice of the European Union (CJEU) has now ruled that ‘[i]t is for national law to determine, in compliance with the principles of effectiveness and equivalence, whether the principal sum must bear “simple interest”, “compound interest” or another type of interest’: Case C-591/10 Littlewoods Retail Ltd v HM Revenue & Customs [2012] STC 1714. For a discussion of this question and especially the CJEU’s decision see Maximilian Schlote’s contribution in ch 10 of the present volume (at pp 200–2). 149 To the extent that the performance itself or the use drawn from it cannot be surrendered in specie (as will typically be the case with the latter), § 818 II BGB substitutes a duty to make restitution of the ‘value’, subject – of course – to any relevant disenrichment (§ 818 III BGB). 150 See only M Schwab in K Rebmann, FJ Säcker and R Rixecker (eds), Münchener Kommentar zum Bürgerlichen Gesetzbuch, vol 5: §§ 705–853, 5th edn (München, Beck, 2009) § 818 [11] with further references (and with an account of the academic debate surrounding the issue at [12]–[14]). Where compound interest has been earned, it follows that the full amount ought therefore in principle to be recoverable. 151 BGH (6.3.1998), BGHZ 138, 160 = Neue Juristische Wochenschrift 1998, 2354. See also Schwab in Münchener Kommentar (n 150) § 818 [15]. 152 In the case of banks and similar institutions, there is a presumption that any capital received has in fact been put to use, and in a line of decisions (all handed down before the most recent economic crisis left interest rates plummeting) the Federal Supreme Court held that such use can typically be valued at a rate of 5 percentage points above base rate: BGH (12.5.1998), Neue Juristische Wochenschrift 1998, 2529; BGH (24.4.2007), BGHZ 172, 147, 157 = Neue Juristische Wochenschrift 2007, 2401, 2404. 153 cf RG (27.4.1936), RGZ 151, 123, and see also Schwab in Münchener Kommentar (n 150) § 818 [22]–[24]. Note that there is a question about whether a distinction needs to be drawn between, on the one hand, cases where the use value is an ancillary matter in restitutionary proceedings concerning the primary object of the defendant’s enrichment and, on the other hand, cases where – as in Sempra (n 140) – the ‘enrichment’ itself consists in no more than having a particular asset at one’s disposal for a particular time. For an overview of the debate see Arnold (n 140) 272 with further references; Schwab in Münchener Kommentar (n 150) § 818 [20]–[30], who thinks that the abstract possibility of making use of an asset should never be sufficient to trigger a liability under §§ 812, 818 I BGB. As far as claims under §§ 819 I, 818 IV BGB are concerned (applying against a defendant who positively knew of his obligation to make restitution) cf nn 164–65 and accompanying text. 154 See only BVerwG (18.5.1973), Neue Juristische Wochenschrift 1973, 2122; BVerwG (24.11.1977), Deutsches Verwaltungsblatt 1978, 608; BVerwG (22.4.1982), case ref 3 C 71.81; recently VG Berlin (11.2.2010), case ref 16 K 117.09, [46] (obiter). 155 A detailed justification for the analogous application of § 818 I BGB in the context of public law restitutionary claims against the citizen is given by BVerwG (24.11.1977), Deutsches Verwaltungsblatt 1978, 608, 609. 156 VGH Mannheim (16.3.1977), Neue Juristische Wochenschrift 1978, 2050, 2052. Under the circumstances, an interest rate of 5% per annum was thought appropriate as a conservative estimate of what the defendant could be assumed to have made using the money. But contrast the more cautious statement by VG Berlin (11.2.2010), case ref 16 K 117.09, [46].

‘Public Law Restitutionary Claims’ 259

specify that a particular rate of interest is payable irrespective of the use to which a recipient has put the money. Thus, § 49a VwVfG, the provision dealing with restitution following the avoidance or revocation of administrative acts,157 decrees (in § 49a III 1 VwVfG) that ‘[t]he amount to be repaid carries interest at a rate of five percentage points above base rate from the moment at which the administrative act is rendered ineffectual’.158 However, the statute also enables the public authority concerned to waive the claim to interest (by virtue of § 49a III 2 VwVfG): It is possible [for a public authority] to abstain from enforcing the claim to interest, in particular where the recipient beneficiary [of the administrative act, ie the ‘enrichee’] is not responsible for the circumstances leading to the avoidance, revocation or ineffectiveness of the administrative act and [where he additionally] pays the relevant [principal] amount within the time frame stipulated by the public authority.159

A similar provision exists – in the context of a more specific statute – for the recovery of social security payments.160 When it comes to claims by the citizen against the state, an entirely different picture appears. Here interest payments are handled most restrictively. With the exception of tax cases, where the issue is specifically dealt with in the Tax Code (in §§ 233 ff AO),161 there are no practically significant legislative grounds for recovering interest on top of the capital sum.162 As far as the application of general principles is concerned (especially where the basis of liability is the non-statutory general public law restitutionary claim),163 courts are – on the whole – reluctant to grant interest for periods before the defendant authority either positively knew about its liability to make restitution or where an appropriate action had at least been launched by the claimant. That interest is available on monetary claims from the commencement of legal proceedings accords with ordinary private law principles formulated by the BGB and applying inter alia to the law of unjustified enrichment.164 Positive knowledge See the text to nn 97–98. Since § 49a III VwVfG governs (only) claims to restitution based on § 49a I VwVfG, which is itself concerned with retroactive invalidity of administrative acts, the point in time from which interest is payable will necessarily lie some time in the past. See also Kopp and Ramsauer (n 100) § 49a [4], [6], [8]. Note that the exact relationship between the claim to interest under § 49a III 1 VwVfG and the right to demand disgorgement of secondary benefits under §§ 49a II 1 VwVfG, 818 I BGB remains yet to be worked out: contrast Kopp and Ramsauer (n 100) § 49a [12] with OVG Berlin-Brandenburg (19.5.2011), case ref OVG 10 B 2.08, especially [39]. 159 On the legislative history of § 49a III VwVfG see F Schnekenburger, ‘Zinsverlust? Zur Neuregelung der Zinsbezugsgrößen auf öffentlich-rechtliche Erstattungsansprüche’ [2003] Neue Zeitschrift für Verwaltungsrecht 36, 36–37. 160 § 50 IIa SGB X. 161 § 233 AO sets out the principle that interest is only payable where this is specifically provided for by statute. §§ 234–237 AO proceed to list various circumstances under which interest falls due, including restitutionary claims against the state (§§ 233a, 236 AO). § 238 AO fixes the interest rate at 0.5% per month. Any doubts about the compatibility of this limitation to simple interest with European law may have been allayed following the recent ruling of the CJEU in the Littlewoods preliminary reference (n 148), but cf the argument made by Maximilian Schlote in chapter 10 of the present volume (at pp 201–2). 162 An isolated example is to be found in planning law (§ 133 III 4 BauGB), yet this provision is given a very restricted interpretation by the courts: cf BVerwG (13.12.1991), Neue Zeitschrift für Verwaltungsrecht 1992, 131. 163 But also where a public law statute simply directs an analogous application of the rules of the BGB concerning unjustified enrichments. 164 § 818 IV BGB referring to §§ 291, 292 BGB. Under §§ 291, 288 I 2, II BGB, simple interest is payable at a rate of five/eight percentage points above base rate from the point in time in which legal proceedings to enforce a monetary claim become pending. Although a claim to compound interest as such is not possible (§§ 291, 289 BGB), the claimant is free to argue that the defendant has culpably failed to invest the money profitably and that he ought therefore to compensate the claimant for any extra gain which could have been made by means of such an investment: §§ 292, 987 II BGB: cf OLG Karlsruhe (30.9.2004), Wertpapier-Mitteilungen 2005, 645 – albeit without mentioning the claim to simple interest under §§ 291, 288 I BGB. 157 158

260 Birke Häcker of a claim on the side of the defendant, though extremely hard to prove in practice, is equated with his being sued.165 While it has been held that these rules also apply to public law restitutionary claims,166 the principle embodied by § 818 I BGB (ie that full restitution requires the disgorgement of secondary benefits irrespective of the defendant’s knowledge) has not been transposed with the same stringency via-à-vis the state as vis-à-vis the citizen. On the basis of the existing case law, it seems almost impossible for a private individual to successfully argue that the state has obtained a valuable benefit from using the capital advanced sine causa. As the Federal Administrative Court reiterated in 2003: Although – in accordance with § 818 I BGB – the public law restitutionary claim [normally] extends to the disgorgement of interim secondary benefits actually derived from the performance obtained without legal cause, it is well-established in the judicature of the [Federal Administrative Court] that the payment of ‘interest’ on the basis of benefits actually derived [from the performance] is usually out of the question where a public law restitutionary claim against a public authority is concerned.167

To understand why this should be thought to be the case, and in order to evaluate the statement, it is necessary to take a look back in legal history. At the time when the ‘fisc theory’ of the state prevailed,168 it was not far-fetched to accord to the fisc the special privileges as regards interest which it had already enjoyed under Roman law.169 This tradition apparently remained alive even after it had become recognised that § 818 I BGB was in principle applicable to the state as much as to private defendants. In the above-mentioned case of 1909 in which §§ 812 ff BGB were still treated as directly applicable to claims for the recovery of overpaid tax,170 the Imperial Supreme Court (Reichsgericht) justified the denial of interest liability as follows: The claimant has not established that the defendant has derived interest or other secondary bene fits from the 15.600 M [Marks] unlawfully exacted as stamp duty; and it is legally irrelevant whether [the defendant] would have been able to derive such benefits. In the case of public taxes and other state revenues payable to the treasury, it will usually not be possible to trace the fate of amounts of money paid by the individual so as to discover how they were used by the state and especially whether they were invested at an interest. If, as the appeal judge assumed, the 15.600 M were held amongst the moneys kept readily disposable to meet [general] state expenditure, it would nevertheless not be warranted to conclude . . . that the defendant has in this way at least saved interest on a loan. This is because, in the light of the relatively small amount of 15.600 M when compared with the overall [financial] needs of the state, it cannot be assumed that its absence alone would have led the state to issue bonds for a greater amount than was actually the case.171 § 819 I BGB referring to § 818 IV BGB (on which see n 164 immediately above). See only BGH (11.6.1970), Neue Juristische Wochenschrift 1970, 1637; BVerwG (7.2.1985), BVerwGE 71, 48 = Neue Juristische Wochenschrift 1985, 2208. There are exceptions, sometimes based on statutory interpretation, sometimes for procedural reasons: cf especially BSG (25.11.1977), case ref 2 RU 127/75; BVerwG (4.5.1994), Neue Juristische Wochenschrift 1994, 3116. 167 BVerwG (30.4.2003), Neue Zeitschrift für Verwaltungsrecht 2003, 1385, 1387. 168 See nn 28–29 and accompanying text. 169 cf Digest 22.1.17.5 (Paul): ‘Fiscus ex suis contractibus usuras non dat, sed ipse accipit’ (The fisc does not give interest on its contracts/obligations, but itself receives interest). See also §§ 2 and 3 of the Prussian ‘Gesetz über die Rechte des Fiskus hinsichtlich der Zinsen’ (Law concerning the rights of the fisc as regards interest) of 7 July 1833 (No 1443), to be found in Gesetzes-Sammlung für die Königlichen Preußischen Staaten 1833 (Berlin, 1833) 79. On the reception and application of this statute see W Schön, ‘Bereicherungszinsen der öffentlichen Hand’ [1993] Neue Juristische Wochenschrift 3289, 3290. 170 See the text to nn 30–32. 171 RGZ 72, 152, 153. 165 166

‘Public Law Restitutionary Claims’ 261

Whether or not the Reichsgericht was right to hold that the state had not been causally enriched by having the use of the money pending repayment,172 its reasoning does at any rate seek to fit the fiscal privilege into the mould provided by § 818 I BGB. The same cannot necessarily be said for the argument since relied on by the Federal Administrative Court in connection with the now emancipated public law restitutionary claim:173 Whether a businessman, who is in the habit of putting money to his own use, is made liable to pay interest, or whether it is the state, is not . . . the same, because the public purse always disposes of the funds made available to it in the public interest . . . which will usually exclude profitable investments.

This assertion of the state not using its revenues to make money has been frequently reiterated by the administrative courts,174 and – interestingly – also by the Bundesgerichtshof as the highest ordinary court adjudicating on pure private law claims to restitution from the state.175 The only situation for which the Federal Administrative Court has so far allowed interest to be recovered in accordance with the principle set out in § 818 I BGB is where the defendant enrichee was not the state as such, but a mutual pension assurance association entrusted with executive powers.176 In the case of such an organisation, the court argued, there was a factual presumption that all revenues were put to a profitable use.177 Not surprisingly, the state’s far-reaching exemption from liability to pay interest on money received without a sufficient legal basis has come under attack.178 As pointed out by Wolfgang Schön, the fact that the state is not a profit-making organisation in the conventional sense does not mean that it refrains from realising the use value of money.179 The state is, indeed, under a public law duty to behave in an economically efficient way. Its budgetary rules and constraints are such that we can establish a causative link between increased liquidity and interest actually earned on state investments or saved on state borrowing.180 It is not necessary for this purpose to identify the traceable proceeds of a given payment by following its value through the labyrinth of various substitutions, as the Reichsgericht thought.181 Instead, all that is required is to show that the state had – for a certain amount of time – money at its disposal to which it was not entitled. cf the discussion in the text to nn 178–81 and also the text accompanying n 193. BVerwG (18.5.1973), Neue Juristische Wochenschrift 1973, 2122, relying on BVerwG (7.6.1962), BVerwGE 14, 222, 231. 174 BVerwG (18.3.1973), Neue Juristische Wochenschrift 1973, 1854, 1854 f; BVerwG (30.4.2003), Neue Zeitschrift für Verwaltungsrecht 2003, 1385, 1387; OVG Lüneburg (23.1.2004), case ref 11 LB 257/03, [31], confirmed by BVerwG (7.9.2004), case ref 3 B 35/04, [10]–[11]. But contrast the notable early exception constituted by OVG Hamburg (9.7.1930), Deutsche Juristen-Zeitung 1931, 174. 175 BGH (3.2.2004), BGHZ 158, 1, 9 = Neue Juristische Wochenschrift 2004, 1315, 1317; BGH (30.3.2004), case ref XI ZR 145/03, [32]. One therefore has to conclude that the question of recovering interest from the state is not one confined to public law restitutionary claims proper but arises whenever the state is in the role of a defendant enrichee. But cf now BGH (24.5.2012), Wertpapier-Mitteilungen 2012, 1208, discussed in the text to nn 190–93. 176 BVerwG (27.10.1998), BVerwGE 107, 304 = Neue Juristische Wochenschrift 1999, 1201. The defendant Pensions-Sicherungs-Verein was a private law organisation invested with a certain amount of sovereign state power for the purpose of collecting pension contributions (a mechanism known as Beleihung). 177 BVerwGE 107, 304, 310–11 = Neue Juristische Wochenschrift 1999, 1201, 1202–3. Under the circumstances, the court thought that an interest rate of 4% per annum was appropriate (being the minimum of what the defendant could be presumed to have made using the money). 178 What is perhaps surprising is that the criticism is not much more widespread. 179 W Schön, ‘Bereicherungszinsen der öffentlichen Hand’ [1993] Neue Juristische Wochenschrift 3289, 3291. 180 ibid 3291–93. 181 ibid 3291, referring to the passage set out in the text to n 171. cf now also BGH (24.5.2012), WertpapierMitteilungen 2012, 1208, 1209. 172 173

262 Birke Häcker For a while, it looked as if the academic criticism might bear fruit. Following the pension fund decision, where the Federal Administrative Court had expressly left open the question whether there was going to be a more general change of direction,182 a number of courts boldly awarded interest on restitutionary claims against the state.183 They either relied on the budgetary argument outlined above184 or sought to establish a rebuttable presumption of the kind applicable to private businesses.185 Yet, given that the orthodox position has since been reconfirmed by both the Federal Administrative Court and the Federal Supreme Court,186 the chances for a German Sempra decision have dwindled somewhat for the immediate future. The debate has, however, not died down completely. There is no system of binding precedent in Germany, and so lower courts may if they wish depart from the jurisprudence of courts higher up in the hierarchy. In 2008, for example, the Landgericht Potsdam came out with a forceful judgment in favour of awarding interest against state defendants irrespective of whether the restitutionary claim is of a private or public law nature.187 Observing that ‘the State of Brandenburg is – to the knowledge of the court – in debt’ and that this was already the case when it obtained the relevant payment in 1995, the court found that the enriching receipt had enabled the state to save money by either discharging existing loans or at least not having to increase its borrowing.188 The court continued: It follows from these observations that the . . . lack of a genuine business interest by the public purse or [reference to] the ‘public interest’ is not a suitable criterion for determining the scope of the public purse’s liability to make restitution for secondary benefits. All that is relevant is whether the public purse factually benefits from receiving the amounts paid sine causa. In view of the [current] budgetary position and the expenditure necessarily saved on interest payments, this is indeed the case. The law of unjustified enrichment does not consider the enrichee’s possible motives when using the revenues obtained sine causa. There are [therefore] no objective reasons whatsoever for privileging the public purse when compared with a private enrichment debtor who is without doubt under an obligation to make restitution of expenditure saved on interest payments.189

Interestingly, the Federal Supreme Court in a very recent ruling cited this decision of the Landgericht Potsdam with approval.190 Although it expressly refused to question the cor BVerwGE 107, 304, 308 = Neue Juristische Wochenschrift 1999, 1201, 1202. cf text to nn 176–77. The relevant cases were mostly decided by courts within the ordinary jurisdiction, not by administrative courts. They were either concerned with pure private law actions against the state (see n 52 and accompanying text) or with special circumstances under which a public law restitutionary claim exceptionally falls within the competence of the ordinary courts (see n 48). 184 BayObLG (9.12.1998), Neue Juristische Wochenschrift 1999, 1194, 1195; OLG Hamm (19.10.2000), Neue Juristische Wochenschrift: Rechtsprechungs-Report 2001, 1140; OLG Köln (22.12.2000), case ref 2 Wx 32/00, Rechtspfleger 2001, 203, [17]–[19]. 185 LG Göttingen (13.11.2003), case ref 6 O 16/03 and 25/03, [33]. Similar: OLG Karlsruhe (6.3.2003), case ref 11 U 21/01; LG Hannover (28.9.2004), case ref 18 O 379/03, [17]. 186 BVerwG (30.4.2003), Neue Zeitschrift für Verwaltungsrecht 2003, 1385, 1387; BVerwG (7.9.2004), case ref 3 B 35/04; BGH (3.2.2004), BGHZ 158, 1 = Neue Juristische Wochenschrift 2004, 1315; BGH (30.3.2004), case ref XI ZR 145/03, [32]. 187 LG Potsdam (22.2.2008), Neue Zeitschrift für Verwaltungsrecht: Rechtsprechungs-Report 2008, 513. The case concerned the estate of a deceased. The deceased had died in 1992, and since heirs could – at the time – not be found, the State of Brandenburg received a total of almost DM 130,000 in 1995. It later turned out that the estate should in fact have fallen to the claimants. In 2005, the principal sum was paid over to them (just over € 66,000), but without any interest. 188 ibid 514. 189 ibid. 190 BGH (24.5.2012), Wertpapier-Mitteilungen 2012, 1208 (at [9]). The case concerned the recovery of tax by a trustee in bankruptcy in circumstances not caught by §§ 37 II, 233a AO, but instead governed by the revocatory action in § 143 I 2 InsO. 182 183

‘Public Law Restitutionary Claims’ 263

rectness of its own previous jurisprudence privileging the fisc in the context of § 818 I BGB,191 and while emphasising that there could be no presumption of the state benefiting from the use value of capital to the same extent as a bank would,192 the court acknowledged that the reasoning of the Reichsgericht could not be maintained under modern budgetary conditions and that it was not necessary to follow or trace the amounts received in order to establish a causative link with saved expenditure.193 All in all, however, it is clear that German law is some way off the post-Sempra English position. Not only does compound interest not seem to be anywhere on the cards, but – with the exception of situations specifically governed by statute or involving semi-private organisations operating like commercial enterprises – the state appears to be largely immune from claims directed at recovering the use value of money, at least until positive knowledge supervenes or legal proceedings become pending. What is, nevertheless, noticeable (albeit perhaps not surprising) is that those courts willing to engage with the task of identifying and valuing the benefits accruing to the state from having a certain amount of money at its disposal are struggling with the same issues which divided the House of Lords in Sempra. The core problem lies in allocating the burden of proof and, if one treats it as reversed,194 in determining the extent of any use-based enrichment, ie the applicable interest rate. Remarkably, English and German courts are coming at these questions from exactly opposite ends of the map. Sempra itself was a case with strong public law elements, and it has the potential to overhaul the entire English law of unjust enrichment even in cases where both parties are private individuals. German law, by way of contrast, has long boasted a reasonably well-established regime of disgorging secondary benefits in the purely private context, and it is only in connection with restitutionary claims against the state that the use value of money still fails to get proper recognition.195

E. APPLYING THE CHANGE OF POSITION DEFENCE

Once it is established that a defendant has been unjustly or unjustifiably enriched at the claimant’s expense, and once the value of the enrichment received has been quantified,196 the question arises whether and to what extent the defendant may argue that subsequent events affecting him197 ought to be recognised as reducing or extinguishing his liability. The appropriate defence is commonly known as ‘change of position’ in English law. The German equivalent is called ‘disenrichment’ (Wegfall der Bereicherung). There are a whole 191 By holding that the restitutionary liability in the case at hand differed from the standard scenario, § 818 I BGB not being at issue: ibid [14]–[15]. 192 ibid [12]. 193 ibid [10]–[11]. 194 On the argument employed by Schön (n 179), it would seem that the state could never successfully rebut the presumption since it is not free to leave capital lying idle: [1993] Neue Juristische Wochenschrift 3189, 3292–93. 195 It has been suggested that a solution may lie in an application of § 818 II BGB for cases where the defendant’s enrichment lies in no more than having received a capital sum (which was in fact due) earlier than would have been required. Since the option of using the capital in the interim period cannot be surrendered in specie, § 818 I BGB gives way to § 818 II BGB (cf n 149), which allows for an objective valuation based on the market cost of borrowing: Arnold (n 140) 272–73. 196 This is not to deny that, clear as these doctrinal steps may be in theory, there can in practice be a substantial overlap between the factors informing the valuation exercise and those taken into account at the defence level: see Vos J in the text to n 220. 197 Lack of space forbids the discussion of claimant-related defences, such as ‘passing on’.

264 Birke Häcker host of extremely contentious issues surrounding the defence, especially in England, where change of position as a free-standing doctrine is of relatively recent pedigree.198 Since it is impossible within the confines of this chapter to examine all of these issues under a public law lens,199 the focus here will be on outlining the general relationship between the ordinary private law defence and public law restitutionary claims of the type already discussed.200 In England, the debate about the applicability and operation of the change of position defence in the context of ‘public law restitutionary claims’ has centred on actions against the state. Perhaps triggered by the attention (and widespread disapproval)201 with which the ‘disruption of public finances’ defence advocated by the Supreme Court of Canada in Air Canada v British Columbia202 had been met, many authors signalled that the change of position defence should either not be applied at all or at least be severely curtailed where public authorities were concerned.203 In Woolwich itself, the issue was not directly addressed, but it has once again become live following the House of Lords decision in Deutsche Morgan Grenfell. In the FII Group Litigation case, Henderson J at first instance distinguished between claims based on the Woolwich principle and those based on mistake.204 While the change of position defence was unavailable for the former, he thought it could in principle operate in respect of the latter.205 Henderson J justified the exclusion of the defence for Woolwich-type claims relying on what Lord Goff had said in Lipkin Gorman v Karpnale Ltd, namely that it was ‘not open to a wrongdoer’.206 Seeing that the gist of the Woolwich principle was the unlawful levying of tax, the state was debarred from relying on the defence.207 And even though the Court of Appeal – now upheld in this respect by the Supreme Court208 – subsequently clarified that an actual ‘demand’ was not necessary to bring the principle into play,209 Vos J in Littlewoods Retail Ltd v HMRC 210 reasoned that since ‘wrongdoing in 198 Although there are a number of earlier cases where the issue was relevant, the change of position defence as such was not openly recognised until Lipkin Gorman (a firm) v Karpnale Ltd [1991] 2 AC 548 (HL) (hereafter ‘Lipkin Gorman’). 199 To follow up these issues, the reader may be referred to the article by E Bant, ‘Change of Position as a Defence to Restitution of Unlawfully Exacted Tax’ [2012] Lloyds Maritime and Commercial Law Quarterly 122, and to the contribution by Niamh Cleary in chapter 7 of the present volume. 200 Many of the issues canvassed here are also dealt with by M Jewell, ‘The Boundaries of Change of Position – A Comparative Study’ [2000] Restitution Law Review 1, to which the reader may wish to turn for an in-depth discussion of the relevant German case law. Going beyond the present chapter, an excellent comparative analysis specifically concerning the problem of anticipatory reliance (in public and private law) is provided by M Jewell, ‘Change of Position’ in P Birks and F Rose (eds), Lessons from the Swaps Litigation (London, Mansfield Press, 2000) 273–90. 201 See the reaction by the Law Commission, Mistakes of Law and Ultra Vires Public Authority Receipts and Payments, Law Com No 227 (London, 1994) [11.6]. 202 Air Canada v British Columbia (Attorney General) [1989] 1 SCR 1161, 59 DLR (4th) 161. 203 J Beatson, ‘Mistaken Payments in the Law of Restitution’ in The Use and Abuse of Unjust Enrichment: Essays on the Law of Restitution (Oxford, Clarendon Press, 1991) 144–45; A Burrows, The Law of Restitution (London, Butterworths, 1993) 358; T Hill, ‘Restitution from Public Authorities and the Treasury’s Position: Woolwich Equitable Building Society v IRC’ (1993) 56 Modern Law Review 856, 864. 204 Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2008] EWHC 2893 (Ch) (hereafter ‘FII (Ch)’). 205 ibid [335]–[342]. 206 Lipkin Gorman (n 198) 580. 207 FII (Ch) (n 204) [339]–[340]. 208 See n 136 and accompanying text. 209 Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2010] EWCA Civ 103 (hereafter ‘FII (CA)’) [157]–[173]. 210 Littlewoods Retail Ltd v HM Revenue & Customs [2010] EWHC 1071 (Ch) (hereafter ‘Littlewoods’).

‘Public Law Restitutionary Claims’ 265

the form of unlawful exaction of taxes remains at the heart of the Woolwich claim’, the exclusion continued to apply.211 By contrast, the change of position defence is said to be available for mistake-based claims of the type recognised in Deutsche Morgan Grenfell.212 The argument is – broadly – that where a claimant chooses to take advantage of the more generous limitation periods associated with the ordinary mistake-based claim,213 he should incur the concomitant disadvantage of being (potentially) confronted with a change of position on the part of the state.214 Nevertheless, the case law to date indicates that the defence will be hard to make out. In Haugesund Kommune v Depfa ACS Bank,215 a swaps case, both Tomlinson J at first instance and the Court of Appeal held that the local authorities concerned216 could not rely on it because they had deliberately taken on the risks connected with the investments for which they decided to use the money received under ultra vires contracts.217 And although Vos J in Littlewoods accepted that change of position could in principle apply to a mistakebased claim,218 he ultimately found with respect to it: [T]he change of position defence cannot succeed, because at the end of the fiscal year, the government has prima facie received a benefit in that its borrowing has been reduced by the amount of the VAT overpayment in that year. That benefit may or may not be carried forward, but the change of position defence fails because the Commissioners have failed to prove, the burden being on them, that they increased their spending in any way on the basis of the expectation or the happening of the overpayment, even if they can show that they either spent or used the overpayment to reduce borrowing.219

What is interesting about the Littlewoods decision is that Vos J saw the potential overlap between the interest issue discussed above and the change of position argument in cases where the parties are fighting over the proper quantification of the ‘time value’ of money: I have some doubt as to whether this defence, as the Commissioners have advanced it in this case, is really a defence as such at all. It seems to me to be more an argument as to how the benefit that has accrued to the Commissioners should be calculated for the purposes of assessing the enrichment that is to be the subject of an order for restitution.220

211 ibid [108]. At [107], he had held that even after FII (CA) (n 209), the Woolwich restitutionary claim was ‘still based on unlawfulness in the form of unlawful exaction of taxes from the citizen’. 212 In the Littlewoods litigation, counsel for the claimants accepted this point only for the purposes of the specific hearing, reserving the right to raise the issue in a higher court: see Littlewoods (n 210) [103]. But cf n 250 concerning the special considerations and restrictions now applying in the EU context according to the Court of Justice’s recent jurisprudence, on which see Maximilian Schlote’s discussion in chapter 10 of the present volume (at p 191). 213 See n 130 and accompanying text. 214 FII (Ch) (n 204) [341]–[342]. But cf the critical comments by Niamh Cleary in chapter 7 of the present volume (at p 134). 215 Haugesund Kommune v Depfa ACS Bank [2009] EWHC 2227 (Comm), on appeal [2010] EWCA Civ 579 (hereafter ‘Haugesund’). 216 On the facts of the case these were Norwegian municipalities. 217 cf Haugesund (n 215) [162]–[169] (Ch), and [106]–[130] (CA). This clarification is to be welcomed, albeit that the outcome is not surprising. The recipient of a loan under a void contract cannot normally escape restitutionary liability by arguing that he has lost the capital in a bad investment. He knows that the loan is repayable and should – even if the contract turns out to be void – bear the risks he deliberately incurred when making the investment. The judgments in Haugesund rightly drew on the principles underlying Goss v Chilcott [1996] AC 788 (PC). 218 In Littlewoods (n 210) [109], he said that ‘in my judgment, a change of position defence is available in English law to a mistake-based restitutionary claim, but not to a Woolwich based claim’. 219 ibid [124]. 220 ibid [110]. cf also ibid [129].

266 Birke Häcker Under German law, the state can never rely on the disenrichment defence. Although § 818 III BGB provides for ordinary private claims that ‘[t]he obligation to make restitution . . . is excluded to the extent that the recipient is no longer enriched’,221 it is commonly agreed that this does not apply where the recipient is the public purse. The reason generally advanced to justify the exclusion of the defence is that the executive is strictly subject to the rule of law (Article 20 III GG)222 and that budgetary constraints therefore prevent it from indulging in the kind of extravagances which characterise a private individual’s changes of position.223 The state may simply not rely on the security of a receipt which lacks a proper public law basis, and it must avoid all unnecessary expenditure.224 Another consideration is this: the disenrichment defence is a way of shifting the risk of the enrichment being lost or reduced before it can be reversed from the defendant enrichee to the claimant enrichor. Where the enrichee is a public body rather than a private party, that risk ought not to be so shifted. It ought instead to be borne by the public as a whole.225 In the reverse case, where the state has a public law restitutionary claim against the individual, German law is not so categorical. While the disenrichment defence contained in § 818 III BGB is once again held to be inapplicable both directly and by analogy,226 the idea underlying it – namely the protection of the recipient’s reliance interest – is generally thought to be reflected in public law of its own right. The function performed by § 818 III BGB in the realm of private law is fulfilled by the doctrine of legitimate expectations (Vertrauensschutz) as far as public law is concerned. In the words of the Federal Administrative Court: To answer the question [of] when the citizen can defend himself against a restitutionary claim by the public purse by pointing to a disenrichment, it is necessary to resort to a balancing of competing interests: the citizen’s interest in protecting his reliance on the security of his financial position and the interest of the executive in enforcing the principle of the rule of law. The obligation to make restitution ceases if the private interest in the protection of reliance outweighs the public interest in restoring the financial state of affairs that accords with the law.227

How, then, is this balance struck? Under the private law regime of the BGB, the disenrichment defence is unavailable only if the defendant positively knew of his restitutionary liability.228 The public law regime chooses to reconcile the competing interests in a slightly 221 § 818 III BGB refers both to the obligation to surrender in specie what has been received (‘Herausgabe’ as mentioned in § 818 I BGB) and to the default obligation to make restitution in value terms (‘Wertersatz’ as mentioned in § 818 II BGB). 222 cf the text to n 56. See only Ossenbühl (n 55) 435; Schoch (n 53) 88; H Weber, ‘Der öffentlichrechtliche Erstattungsanspruch’ [1986] Juristische Schulung 29, 34; Windhorst (n 53) 899; BVerwG (17.9.1970), BVerwGE 36, 108, 113–14, concerning a restitutionary claim between two public bodies; BVerwG (12.3.1985), BVerwGE 71, 85, 89, where the court emphasised that the defence was denied irrespective of the fact that it would normally be impossible for a public body to establish a disenrichment on the evidence in any event. 223 Ossenbühl (n 55) 435; see also Schoch (n 53) 88 (there text to fn 99). This argument concerns expenditure which the claimant would not otherwise have had and which he incurs in reliance on the security of his receipt, such as going on a luxury cruise. 224 Fault is entirely irrelevant in this context. 225 cf Meier-Branecke (n 113) 259. 226 Seminal: BVerwG (12.3.1985), BVerwGE 71, 85, 89 ff. See Ossenbühl (n 55) 433 with further references and a brief overview of the pre-1985 debate. Note, however, that where the extent of a defendant’s restitutionary liability is determined by § 49a VwVfG (discussed further in the text to nn 236 –38), § 818 III BGB does apply, albeit with certain modifications. For the tax context cf the critical assessment by M Weber (n 65) 406–8. 227 BVerwGE 71, 85, 90. See also BVerwG (18.1.2001), BVerwGE 112, 351, 357. 228 §§ 819 I, 818 IV BGB. From the fact that a defendant with the relevant knowledge is liable according to the BGB’s ‘general provisions’ (cf nn 164–65 and accompanying text), it is inferred that such a defendant cannot rely on § 818 III BGB: RG (11.5.1920), RGZ 99, 161, 168; RG (30.10.1942), RGZ 170, 65, 68; BGH (7.1.1971), BGHZ 55, 128, 132; BGH (26.10.1978), BGHZ 72, 252, 254–55.

‘Public Law Restitutionary Claims’ 267

different way. On the one hand, a defendant’s gross negligence usually debars him from holding out any expectation as being ‘legitimate’.229 The disqualification threshold is therefore lower in public than in private law. On the other hand, however, it seems that public law will more readily recognise non-disenriching changes of position as relevant.230 As the Federal Administrative Court went on to observe in its leading decision: What can work to the advantage of the citizen is that the financial value remaining need not be disgorged under all circumstances. This is because [his] reliance may be deserving of protection even where what has been obtained without legal cause still exists, but where the citizen has already made dispositions with respect to it that cannot be reversed without incurring unacceptable disadvantages. [Yet] what can work to the disadvantage [of the citizen] is that his reliance in the security of his financial position is not necessarily always deserving of protection whenever [he] was ignorant of the fact that the shift of wealth lacked a legal basis.231

In defining these parameters, the Federal Administrative Court relied heavily on various legislative provisions dealing with those public law restitutionary claims that had already found their way into one or the other statute.232 What should be noted in this context is that the German concept of the ‘administrative act’233 will often mean that the protection of reliance is realised at a stage preliminary to the actual unwinding process. In particular, where a public authority has to decide whether to avoid or revoke an administrative act (eg the grant of a subsidy to an individual or company), it will already take into account the recipient’s legitimate expectations.234 Thus, for example, § 48 II VwVfG provides: An unlawful administrative act which grants a single or recurring monetary benefit . . . may not be avoided in so far as the beneficiary has relied on the continued existence of the administrative act and [in so far as] his reliance is deserving of protection when weighed against the public interest in avoiding the [unlawful] administrative act. His reliance will generally be deserving of protection where the beneficiary has used up what he has received or where he has made a financial disposition which he is unable to reverse or at any rate not without incurring unacceptable disadvantages. The beneficiary cannot invoke reliance where 1. he has secured the administrative act by means of fraud, threat or bribery; 2. he has secured the administrative act by means of statements which were untrue or incomplete in some relevant respect; 3. he knew of the unlawfulness of the administrative act or did not know of it as a result of his own gross negligence. In cases falling under the third sentence [of this sub-section], the administrative act will normally be avoided with retroactive effect.235 See the discussion by Ossenbühl (n 55) 433–44; Schoch (n 53) 88–89. Non-disenriching changes of position cannot be easily accommodated within the framework of § 818 III BGB, which is a defence based on disenrichment, not change of position. 231 BVerwGE 71, 85, 90. 232 Beside various now-repealed provisions in the field of civil service law, the court drew on the then applicable § 48 II 7 VwVfG, the predecessor of the current § 49a II 2 VwVfG (on which see the text after n 236). 233 cf the text to nn 81–88. 234 Note that the protection of reliance is much more restricted in the European context (especially where the grant of a particular national subsidy is found to contravene EU law). Here, the so-called principle of effectiveness (effet utile) requires § 48 VwVfG to be applied with substantial modifications. One of these is that the public authority concerned does not have the usual discretion in deciding whether or not to avoid the unlawful administrative act. It is bound to do so. Another is that the restrictions contained in § 48 II VwVfG are factually disapplied. Once the Commission has decided that the subsidy in question is incompatible with EC/EU law, the recipient’s legitimate expectations or his reliance on the security of his receipt are no bar to its avoidance. See also nn 239– 40 and cf Kopp and Ramsauer (n 100) § 48 [7]–[14], [100], [126]. 235 cf also § 49 II and III VwVfG, set out in n 96. 229 230

268 Birke Häcker If the performance was rendered by the state to the citizen on the basis of a valid administrative act, the recipient’s changes of position may hence be relevant to the question of whether the legal basis for the transfer is avoided at all. A public law restitutionary claim proper will only come into existence where the administrative act is so avoided or where it is otherwise revoked or invalidated. The provision applicable in such cases (§ 49a VwVfG) has already been mentioned.236 After stipulating that performances rendered in pursuance of an administrative act later avoided or revoked ab initio have to be returned, § 49a II continues: The scope of restitution – except as regards interest[237] – is to be determined by reference to the rules of the BGB concerning the surrender of an unjustified enrichment. The beneficiary may not plead disenrichment in so far as he knew of the circumstances leading to the avoidance, revocation or invalidity of the administrative act or [in so far as] he was ignorant of them as a result of his gross negligence.

In effect, therefore, § 49a II BGB refers to the private law rules on the reversal of unjustified enrichments (including § 818 III BGB), but modifies the disqualification threshold in line with the public law approach to legitimate expectations.238 This ensures that the change of position defence is as unavailable to grossly negligent defendants as to those with positive knowledge.239 Of course, the question of legitimate expectations can only arise where an administrative act is invalidated after the period during which it may be challenged has expired.240 If an administrative act granting a particular benefit (eg a subsidy) is opposed by a third party (eg a competitor) within the statutory time-limit241 and is avoided or revoked in the course of these opposition proceedings or set aside by a judicial decision, the beneficiary can never argue that he has already detrimentally relied on the security of his receipt.242 On the other hand, where there is nothing to avoid or revoke because the performance itself was not preceded by any administrative act, reliance will potentially be an issue right from the 236 cf the text to nn 97–98. Note that – for systematic reasons – § 49a VwVfG applies only to situations where a public authority has avoided or revoked its own administrative act ab initio under §§ 48, 49 VwVfG or where an administrative act is retroactively invalidated when a particular condition subsequent comes about. In other situations (eg where an administrative act is challenged by a third party within the statutory time limit and struck down in court), recourse must be had to the uncodified ‘general’ public law restitutionary claim. 237 It will be recalled that § 49a II VwVfG contains special rules on the payment of interest: see the text to nn 157–59. 238 There is a question about the relationship between § 48 II 3 Nr 3 VwVfG and § 49a II 2 VwVfG. While § 48 II 3 Nr 3 VwVfG requires a defendant to have known or been grossly negligent in not realising the actual unlawfulness of the administrative act in question, it is sufficient for a disqualification under § 49a II 2 VwVfG that the defendant knew of all the relevant circumstances on which the verdict of unlawfulness is based. The latter provision is therefore slightly wider: Kopp and Ramsauer (n 100) § 49a [15]. 239 In the European context, the disenrichment defence is not available at all (see above n 234). On the one hand, it is regarded as grossly negligent for the recipient of a national subsidy not to have reassured himself by checking that the subsidy is compatible with the requirements of EC/EU law. On the other, it has become established that – even where gross negligence is not made out – the principle of effectiveness precludes reliance on § 49a II 2 VwVfG: Kopp and Ramsauer (n 100) § 48 [126], § 49a [18], with further references. 240 Even then, administrative acts may usually only be avoided or revoked within one year of the public authority concerned learning about the circumstances which would warrant avoidance or revocation: §§ 48 IV, 49 II 2, III 2 VwVfG. Again, however, this limitation period is disapplied in the EC/EU context: Kopp and Ramsauer (n 100) § 48 [151]. 241 Recall that the time limits for challenging an administrative act are extremely short, usually one month: see n 86. 242 See § 50 VwVfG, providing that the normal restrictions on the avoidance or revocation of ‘beneficial’ administrative acts, being based on the addressee’s legitimate expectations, do not apply in such a case.

‘Public Law Restitutionary Claims’ 269

moment of receipt. It will be recalled that in such a case the existence of a legal basis for the performance is judged directly by reference to the substantive law,243 and the uncodified public law restitutionary claim applies. This was in fact the context in which the abovecited general statement of the Federal Administrative Court was made.244 Looked at from the outside, the discrepancy between the private law regime and the public law restitutionary claim is striking. But is it justified? Does the balance really need to be struck differently depending on whether an individual’s interest in the security of his receipt is set up against a private party’s interest in obtaining restitution (disqualification only in the case of positive knowledge) or against the public interest in maintaining the rule of law (disqualification also in the case of gross negligence)? Interestingly, there are almost no calls for the public law principles governing changes of position to be adjusted so as to align with §§ 818 III, IV, 819 I BGB.245 Quite the opposite is true. Those authors who would favour a harmonisation of the divergent rules generally regard the private law regime as misconceived. Fritz Ossenbühl, for example, bemoans the knowledge requirement as being based on a ‘mis-judgement’ by the BGB legislator.246 He argues that a person who has strong grounds to suspect that he has obtained something without a proper legal basis ought to make the necessary enquiries. This duty – Ossenbühl and others suggest – is not one confined to the public law sphere.247 In England, as elsewhere in the Common Law world, the questions, highlighted here by the discrepancies between private and public restitutionary claims in German law, are still in a state of flux. The disqualification threshold for the ordinary change of position defence seems to be settling somewhere above simple negligence and below actual knowledge.248 Similarly, non-disenriching changes of position appear to be on the cards. Elise Bant has made a forceful argument for their recognition.249 There is therefore less need for public law to ‘iron out’ perceived deficiencies of the private law defence. Indeed, if anything is to be learnt from the German experience, it is that – vis-à-vis private defendants – the rules governing the defence ought not to be lightly varied according to whether the wider context happens to be public or private law. On the other hand, however, there is a lot to be said for excluding the defence altogether where restitutionary claims against the state are at issue.250 If, contrary to Deutsche Morgan Grenfell, mistake-based claims were barred See the text to and following n 106. See the text to n 231. 245 But see M Weber (n 65) 406–7, who argues for an application of § 818 III BGB in the context of § 37 II AO. Note, however, that this suggestion is primarily motivated by the lack of any provision protecting a recipient’s legitimate expectations in the context of the statutory claim under § 37 II AO. The purpose of the argument is to enable a private defendant to plead change of position in the first place, not to restrict the circumstances under which he can do so. 246 Ossenbühl (n 55) 434: ‘Fehlwertung des Gesetzgebers’. 247 ibid. cf also H Weber (n 222) 35. 248 Lipkin Gorman (n 198) 580 (Lord Goff: ‘bad faith’); Niru Battery Manufacturing Co v Milestone Trading Ltd [2002] EWHC 1425 (QB Comm) [122]–[135] (Moore-Bick J: ‘dishonesty’); [2003] EWCA Civ 1446, [2004] QB 985 (CA) [148]–[162] (Clarke LJ: ‘unconscionability’). A comparative fault analysis like the one operated in New Zealand (cf National Bank of New Zealand Ltd v Waitaki International Processing (NI) Ltd [1999] 2 NZLR 211) has not found favour: Dextra Bank & Trust Co Ltd v Bank of Jamaica [2002] 1 All ER (Comm) 193 (PC) [40]–[46]. 249 E Bant, The Change of Position Defence (Oxford, Hart Publishing, 2009) 126–38, suggesting an ‘irreversibility’ test. 250 In the context of taxes levied in breach of European law, the Court of Justice’s recent jurisprudence suggests that the only defence ever available to the state is a direct version of the so-called ‘passing-on’ defence, which – taken at face value – would necessarily preclude any argument based on disenrichment or change of position. See Cases C-398/09 Lady & Kid A/S v Skatteministeriet, CJEU 6 September 2011, [2012] STC 854, and C-310/09 Ministre du Budget, des Comptes publics et de la Fonction publique v Accor SA , CJEU, 15 September 2011. For a 243 244

270 Birke Häcker whenever the Woolwich principle applied,251 this result could be reached in a relatively straightforward manner.252 Yet the Supreme Court’s decision and dicta in FII would seem to preclude such a development for the foreseeable future.253 Quite apart from such doctrinal subtleties, however, it looks distinctly odd for the state to invoke a defence which revolves essentially around the idea of reasonable reliance and legitimate expectations. Thus, even if it could be shown that – in a given case – there was a direct causal link between the enriching receipt and certain budgetary decisions,254 the economic burden of any extra expenditure ought to be borne by society as a whole (being the ultimate beneficiary of all state spending) rather than the individual reclaiming a performance either mistakenly rendered or having been unlawfully exacted.

F. CONCLUSIONS

It has been seen that there are a number of striking parallels between the history and development of so-called ‘public law restitutionary claims’ in Germany and the recognition of Woolwich-type claims in England. In both jurisdictions, the rule of law argument plays a crucial role in justifying not only the state’s liability to make restitution for unlawfully exacted taxes and similar enriching receipts (irrespective of any mistake), but also the reversal of any ultra vires transfers effected by the state. However, an important point of divergence between the two legal systems lies in the fact that English law does not operate any doctrine comparable to that of the ‘administrative act’ in German law. As a result, courts adjudicating on the equivalent of public law restitutionary claims in England are typically thrown back onto issues of substantive law, often many years after the events in question. In Germany, by contrast, an ‘administrative act’ is usually interposed between the substantive right to claim or the duty to render a specific performance and the performance as such. By furnishing the performance with a legal basis even where the decision to require or make the performance was flawed as a matter of substantive law, the administrative act entails a degree of legal certainty and finality that would otherwise be hard to achieve. Without a comparable doctrine, English law is well-advised to tighten up its timelimits on restitutionary claims in the public law context.255 As far as the scope of public law restitutionary claims and applicable defences are concerned, each system has something to learn from the other. English law appears to be ahead of German law on the issue of interest. While the House of Lords in Sempra established that full restitution may in an appropriate case require the state as defendant to pay compound interest to a private individual, the highest German courts still tend to adhere to the unsupportable basic proposition that the public purse does not benefit at all from having money at its disposal for a given length critical commentary cf R Williams, ‘Lady & Kid A/S v Skatteministeriet (C-398/09) and Ministre du Budget, des Comptes Publics et de la Fonction Publique v Accor SA (C-310/09): Unjust Enrichment and the Court of Justice of the European Communities, a Loss of National Competence and Principle?’ [2011] British Tax Review 631. 251 See the text to nn 133–35. 252 See the text to nn 204–11. Yet note that Rebecca Williams speaks out for allowing public bodies to rely on the change of position defence in principle: Williams, Unjust Enrichment and Public Law (n 7) 136–49. 253 See the text to nn 136 ff. 254 cf Henderson J’s observations in FII (Ch) (n 204) [343]–[346]. 255 On the question of time limits cf also Williams, Unjust Enrichment and Public Law (n 7) 128–35.

‘Public Law Restitutionary Claims’ 271

of time. On the other hand however, the German refusal to afford a disenrichment or change of position defence to the state seems preferable to the current English position which still leaves such a defence open in the context of mistake-based unjust enrichment claims.

13 Overpaid Taxes and Constitutional Redress in Ireland NIAMH CONNOLLY

A. INTRODUCTION

The seminal Irish case of Murphy v Attorney General condemned as unconstitutional an income tax regime which subjected married couples to a heavier burden of taxation than the unmarried.1 A precursor to Woolwich,2 Murphy is authority that, where an exaction was not lawful, then, in principle, the State is not entitled to it and ought not retain it at the expense of the citizen. Henchy J evoked ‘the prima facie right of the taxpayers to be recouped’.3 Murphy establishes three key propositions in Irish constitutional law. First, an unconstitutional statute is, in principle, void ab initio. Secondly, there is a prima facie right of redress for citizens who have suffered due to actions taken under the void unconstitutional measure. Thirdly, in certain exceptional cases, the State may be exempted from making full redress or returning all the benefits it has received, for reasons related to the risk of fiscal chaos. Despite the outcome on the facts – which provided a very limited restitutionary remedy, confined to the plaintiffs and after they challenged the law’s validity – the legal principles expounded in Murphy are expansive in orientation. Murphy conforms to a view that the right of recovery in overpaid taxation cases ought to be the general rule, and any curtailment of that right a strictly-limited exception.4 As with Woolwich, the simplicity of the general principle of recovery conceals difficult theoretical questions. It is debated whether it belongs to the law of unjust enrichment or to public law.5 Birks proposed an automatic right of recovery to take account of the Murphy v Attorney General [1982] 1 IR 241 (SC) (hereafter ‘Murphy’). Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] 1 AC 70 (HL) (hereafter ‘Woolwich’); P Birks, ‘Restitution from the Executive: a Tercentenary Footnote to the Bill of Rights’ in PD Finn (ed), Essays on Restitution (Sydney, Law Book Co, 1990) 185 ff. 3 Murphy (n 1) 320. 4 Kingstreet Investments Ltd v New Brunswick (Department of Finance) 2007 SCC 1, [2007] 1 SCR 3 (hereafter ‘Kingstreet’); cf Air Canada v British Columbia (Attorney General) [1989] 1 SCR 1161 (hereafter ‘Air Canada’); see F Bachand, ‘Restitution of Unlawfully Levied Taxes: Survey and Comparative Analysis of Developments in Canada, Australia and England’ (2001) 38 Alberta Law Review 960, 985; JD McCamus, ‘Restitutionary Recovery of Moneys Paid to a Public Authority under a Mistake of Law: Ignorantia Juris in the Supreme Court of Canada’ (1983) 17 University of British Columbia Law Review 233, 257. 5 Birks (n 2) 204; N Cleary, ‘Public and Private Law Principles: Murphy v Attorney General Reassessed’ (2011) 34 Dublin University Law Journal 155. 1 2

274 Niamh Connolly distinctive character of the relationship between public bodies and citizens, but considered it debateable whether it has ‘a purely public-law character’.6 Both the decision of the Supreme Court of Canada in Kingstreet and Williams’ recent book propose that the right of restitution from the executive of unlawfully exacted charges is a matter of public law.7 In contrast, Smith criticises Kingstreet as ‘deeply puzzling’, stating that previous case law did not suggest that this remedy was external to unjust enrichment law.8 He views the Woolwich principle as a private law rule modified to take account of special considerations arising from the distinctive character of public bodies. In light of the Murphy case, this chapter argues in favour of the characterisation of the Woolwich principle9 and its homologues in other common law jurisdictions as constitutional remedies, mandated by the principle of legality.10 This constitutional value is common to the disparate common law jurisdictions, despite the manifest diversity of constitutional regimes. The principle of legality supports a general public law principle that State actions which are not mandated by law should in principle be undone and a remedy granted to persons adversely affected. Although the Murphy decision also speaks of duress colore officii and change of position,11 it may be viewed as providing clear support for the constitutional approach.12 A key paragraph from the decision of Henchy J in the Irish Supreme Court sets out this rationale unambiguously: Once it has been judicially established that a statutory provision enacted by the Oireachtas13 is repugnant to the Constitution, and that it therefore incurred invalidity from the date of its enactment, the condemned provision will normally provide no legal justification for any acts done or left undone, or for transactions undertaken in pursuance of it; and the person damnified by the operation of the invalid provision will normally be accorded by the Courts all permitted and necessary redress.14

The Murphy principle extends beyond restitution to embrace other forms of redress for actions purportedly justified by invalid laws, such as the release from imprisonment of persons convicted under invalid statutes.15 Its extensive scope strongly supports the constitutional interpretation: the restitutionary application, as merely one dimension of a wider public law solution, cannot logically be regarded as falling solely within the ordinary private law of unjust enrichment. Interestingly, subsequent Irish authorities also support the existence of a free-standing compensatory remedy within constitutional law.16 This enhances the coherence of the proposed constitutional right to a remedy.

Birks (n 2) 168, 204. Kingstreet (n 4); R Williams, Unjust Enrichment and Public Law (Oxford, Hart Publishing, 2010); see also A Burrows, ‘Understanding the Law of Restitution: a Map through the Thicket’ in A Burrows (ed) Understanding the Law of Obligations: Essays on Contract, Tort and Restitution (Oxford, Hart Publishing, 1998) 56–57. 8 L Smith, ‘Public Justice and Private Justice: Restitution after Kingstreet’ (2008) 46 Canadian Business Law Journal 11, 16, 18, 20. 9 Woolwich (n 2). 10 Kingstreet (n 4); Murphy (n 1). 11 Cleary (n 5). 12 See G Jones, Restitution in Public and Private Law (London, Sweet & Maxwell, 1991) 20. 13 The Oireachtas is the Parliament of the Republic of Ireland – Bunreacht na hÉireann 1937, art 15.1.1. 14 Murphy (n 1) 313. 15 Director of Public Prosecutions v Cunningham [2012] IECCA 64 (hereafter ‘Cunningham’). 16 Blehein v Minister for Health and Children [2010] IEHC 329 (hereafter ‘Blehein (HC)’); cf Smith (n 8) 26. 6 7

Constitutional Redress in Ireland 275

Contrary to Williams’ view, these cases need not necessarily be conceived as a hybrid of public law and private law.17 This argument is partially an endorsement of the approach of the Supreme Court of Canada in Kingstreet,18 which happily repudiated Air Canada,19 while insisting that the answer lies in public law.20 Ordinary private law principles fail to take account of considerations which are specific to the case of exactions by the State. More importantly, the crucial reasons for restitution from the executive for demands made without valid authority are constitutional in nature. Equally, the reasons why the courts might wish to limit the right to restitution reflect values such as the common good, which have no place in the defences recognised in unjust enrichment law. This constitutional right should in principle be limited in application to cases where the public body is the defendant, and arising out of dealings in which it acts in exercise of public power. Public bodies may in some contexts deal with other legal persons on a relatively equal footing, and transactions of this latter kind should properly be governed by private law. Murphy also proposed a novel defence which may avail the State. Policy-based concerns arising from the role of the State as guardian of public funds on behalf of the nation may, in some strictly-limited circumstances, and whatever the ground of the prima facie right of recovery, supervene to deny the remedy warranted by justice. This defence does not conform to the logic of unjust enrichment law, such as by negativing enrichment or the unjust factor, but instead is permissibly influenced by policy considerations appropriate to public law. Although the concerns which inspire it are reminiscent of the fear of fiscal chaos evoked in Air Canada,21 the Murphy defence is strictly limited and exceptional.22 Fear of disruption cannot in normal circumstances outweigh those compelling imperatives which support the grant of a remedy: the principle of legality and the principle against unjust enrichment. The orientation of the Murphy solution is thus diametrically opposed to that of Air Canada.23 It follows the scheme advocated by McCamus: redress is the general rule, from which the possible denial of a remedy is a strictly exceptional derogation.24 Lastly, it is not necessary to exclude the orthodox causes of action as alternative grounds from cases concerning exactions by the State. Logically, the usual causes of action based on unjust factors including mistake or duress colore officii remain available to plaintiffs.25 Though the public law reason for restitution to uphold the rule of law is the more decisive rationale, frequently the plaintiff’s consent will also have been vitiated, so as to ground a claim under an unjust factor. However, it must be possible for the State to avail itself of Murphy’s exceptionally limited fiscal chaos-type defence, whichever the ground of the plaintiff’s action. This serves consistency and acknowledges that the same ‘transcendent considerations’26 may favour permitting the State to avoid full restitution in exceptional cases regardless of the cause of action. Williams (n 7) 36. Kingstreet (n 4) 350. 19 Air Canada (n 4). 20 Williams (n 7) 11. 21 Air Canada (n 4). 22 O’Rourke v Revenue Commissioners [1996] 2 IR 1 (HC) (hereafter ‘O’Rourke’). 23 M Pal, ‘The Supreme Court of Canada’s Approach to the Recovery of Ultra Vires Taxes: At the Border of Private and Public Law’ (2008) 66 University of Toronto Law Review 65, 75. 24 McCamus (n 4) 257. 25 Deutsche Morgan Grenfell Group plc v Inland Revenue Commissioners [2006] UKHL 49, [2007] 1 AC 558 (hereafter ‘DMG’). 26 Murphy (n 1) 313. 17 18

276 Niamh Connolly This chapter will first trace the development of claims for restitution from the State in Irish law and explore the competing private law and public law interpretations of the Murphy case. Then, it will examine the public law justifications for the right of recovery and seek to distill from the Irish approach principles relevant to other common law jurisdictions. The final part will outline the shape and scope of this constitutional cause of action, focusing in particular on the exceptional public-policy-motivated defence justified by the constitutional law solution.

B. IRISH APPROACHES TO RESTITUTION FROM THE STATE

(1) A Distinctive Irish Approach Prior to Murphy Before Murphy, the law governing the recuperation from the State of unlawfully exacted sums required reliance on consent-related unjust factors such as duress or mistake, subject to all the vagaries which the historically technical definition of these rules created.27 Dolan v Neligan28 and Rogers v Louth County Council 29 are the key cases in which the Irish courts offered expansive formulations of duress colore officii and mistake, emphasising the inequality which characterises relations between the State and private citizens.30 (a) Duress Colore Officii The Irish courts distanced the concept of duress colore officii from the requirements of the ordinary private law on duress, abrading the need for improper pressure resulting in compulsion.31 In Dolan v Neligan,32 Kenny J articulated an expansive conception of this ‘somewhat obscure’ cause of action.33 The plaintiff paid duties on the importation of alcoholic beverages before mounting a successful challenge to their imposition. Kenny J ruled that both duress colore officii and mistake would necessitate the return of the sums received, but that statute34 had excluded common law restitution.35 The Supreme Court implicitly approved his dicta on common law restitution36 and the judgment has been cited as authority in subsequent decisions.37 Kenny J formulated the doctrine of extortion colore officii as permitting recovery of, ‘moneys paid to a public official because of a demand made by him, when he can grant or withhold a privilege in the event of payment or non-payment’.38 Focusing on the power of Dolan v Neligan [1967] IR 247 (SC) (hereafter ‘Dolan’). ibid. 29 Rogers v Louth County Council [1981] IR 265 (SC) (hereafter ‘Rogers’). 30 See also McCamus (n 4) 247. 31 Woolwich (n 2); Williams (n 7) 22; see also Kingstreet (n 4) 364; RD Collins, ‘Restitution from Government Officials’ (1984) 29 McGill Law Journal 407, 428. 32 Dolan (n 27). 33 ibid 254. 34 Customs Consolidation Act 1876, ss 30 and 31. 35 Dolan (n 27) 264. 36 ibid 275. 37 Rogers (n 29) 271; O’Rourke (n 22) 10. 38 Dolan (n 27) 254; Kenny J cited Dew v Parsons (1819) 2 B & Ald 562, 106 ER 471; Morgan v Palmer (1824) 2 B & C 729, 107 ER 554; Steele v Williams (1853) 8 Ex 625, 155 ER 1502; Hooper v Mayor and Corporation of Exeter 27 28

Constitutional Redress in Ireland 277

the official authority to refuse to permit the importation of the plaintiff’s merchandise, Kenny J accepted that this constitutes exaction colore officii.39 The Supreme Court endorsed this view of duress colore officii in Rogers v Louth County Council.40 The case concerned a payment to the defendant local authority to redeem the annuity payable on a house. A calculation error resulted in overpayment.41 Griffin J adopted a broad, relational, conception of involuntariness, which ‘may also be deduced from the relationship of the parties’.42 Applying this expansive conception of involuntariness, the payment was not ‘voluntary’ because the parties ‘were not on equal terms’.43 The public authority had the power to deny the redemption of the annuity.44 On the Irish formulation, although duress colore officii is a species of duress, the criterion of involuntariness is readily met in many cases of payments to the State, as the inequality of the parties and the power of the official recipient to harm the interests of the payor negate voluntariness. The Supreme Court in Rogers did not demand that a plaintiff has protested the incorrect payment. This is undoubtedly appropriate since the plaintiff was completely unaware of the error.45 This indicates that protest or acquiescence are not factors relevant to duress colore officii in Irish law. However, O’Rourke v Revenue Commissioners, the 1996 case which directly approved Woolwich as good law in this jurisdiction, adopted a different view.46 The plaintiff in O’Rourke was a manager of a social welfare office, who had mistakenly been taxed as an employee instead of a self-employed person. Neither party was aware of the error until a court ruling seven years later. Keane J reasoned that, had the plaintiff protested, the money would have been taken colore officii. However, since the plaintiff acquiesced in the automatic deductions, this excluded duress. Protest is, of course, impossible if the error is unknown. On Keane J’s view, duress requires that the payor be aware that the payments are not due, but be forced to pay nonetheless. This view is inconsistent with both Rogers and Murphy, which considered such payments involuntary.47 Perhaps the decision reflects the court’s enthusiasm to align Irish law with Woolwich. If Keane J’s view on duress colore officii is correct, then it indicates either that the form of duress colore officii espoused in Dolan and Rogers is strictly limited to cases in which the payment is directly linked to the conferral or denial of a privilege by the State, excluding general taxation; or that, now that the Woolwich principle offers a straightforward solution, the Irish courts will resile from the expansive position favoured in those two cases. (b) Mistake Faced with the obstacle of the mistake of law bar, the Irish courts developed a liberal – though now old-fashioned – approach in cases such as Dolan v Neligan 48 and Rogers v (1887) 56 LJQB 457; T&J Brocklebank Ltd v The King [1925] 1 KB 52; and distinguished William Whiteley Ltd v The King (1909) 101 LT 741. 39 Dolan (n 27) 257. 40 Rogers (n 29). 41 Meade v Cork County Council (SC, 3 July 1974). 42 Rogers (n 29) 271; citing Morgan v Palmer (1824) 2 B&C 729, 107 ER 554. 43 Rogers (n 29) 271. 44 ibid 271; cf Jones (n 12) 14. 45 Rogers (n 29) 271. 46 O’Rourke (n 22). 47 ibid 271; Murphy (n 1) 317. 48 Dolan (n 27).

278 Niamh Connolly Louth County Council.49 In Dolan v Neligan,50 Kenny J strenuously disputed the existence of a bar on recovery for mistake of law as ‘an inaccurate simplification of a complex problem’,51 and, later, ‘grossly inaccurate’.52 Instead, he favoured recovery, provided that the defendant’s conduct rendered him primarily responsible for the mistake.53 Consequently, restitution could occur, ‘if the cause of the mistake were statements about the law made to him by the party receiving the money, or if the parties were not on equal terms at the time when the payment was made’.54 Rogers v Louth County Council adopted a similarly liberal approach.55 Griffin J stated that restitution could be granted where the plaintiff was not responsible for the mistake56 and the payment was made involuntarily, under legal compulsion.57 Restitution would not be allowed if the payment was deliberately made by a plaintiff who was aware that it might not be due or indifferent to that possibility.58 The first limb of the Dolan v Neligan mistake test justifies recovery where there is an inequality of information, coupled perhaps with a degree of culpability or at least responsibility on the part of the recipient, and the second, alternative, limb concentrates on inequality of status between the parties. While the first test is met in both Dolan and Rogers, the latter test, which seems to import into mistake elements of the duress colore officii ground, covers all cases of payments to the State. Only where the plaintiff knew to a high degree of certainty that the payment was unwarranted could he not avail himself of this route to recovery. This distinctive Irish approach to mistake of law will be redundant if the mistake of law bar no longer exists in Irish law, as logically implied by the development of a coherent framework of actions and defences in the modern law of unjust enrichment.59 Ireland has not yet expressly abolished the mistake of law rule,60 despite indications of its implicit abrogation.61 In Duff v Minister for Agriculture,62 the Supreme Court relied on Rogers,63 implying the retention of mistake of law as a particular class of mistake in restitution law. Similarly, Rogers (n 29). Dolan (n 27). 51 ibid 259. 52 Rogers (n 29) 273; see also McCamus (n 4) 235. 53 Kiriri Cotton Company Limited v Dewani [1960] AC 192 (PC) (hereafter ‘Kiriri Cotton’). 54 Dolan (n 27) 259. 55 Rogers (n 29). 56 Kiriri Cotton (n 53), approved by Kenny J in Dolan (n 27), cited by Griffin J in Rogers (n 29) 270; see also ibid 273; see also McCamus (n 4) 243. 57 Rogers (n 29) 271. 58 ibid 271. 59 Kleinwort Benson Ltd v Lincoln City Council [1999] 2 AC 349 (HL) (hereafter ‘Kleinwort Benson’) 373; Air Canada (n 4); David Securities Pty Ltd v Commonwealth Bank of Australia (1992) 175 CLR 353 (HC); Willis Faber Enthoven (Pty) Ltd v Receiver of Revenue (1992) (4) SA 202 (SC); G Virgo, The Principles of the Law of Restitution, 2nd edn (Oxford, OUP, 2006),142; J Convery, ‘Lord Goff ’s Swansong: Restitution, Mistake of Law, and the Retrospective Effect of Judicial Decisions’ (1999) 3 Edinburgh Law Review 202, 206; J Beatson and W Bishop, ‘Mistaken Payments in the Law of Restitution’ (1986) 36 University of Toronto Law Journal 149, 162, 175; M Bryan, ‘Mistaken Payments and the Law of Unjust Enrichment: David Securities Pty Ltd v Commonwealth Bank of Australia’ (1993) 15 Sydney Law Review 461. 60 Re Article 26 of the Constitution and the Health (Amendment) (No 2) Bill 2004 [2005] 1 IR 105 (SC) (hereafter ‘Health (Amendment) Bill Reference’); Duff v Minister for Agriculture (No 2) [1997] 2 IR 22 (SC) (hereafter ‘Duff’); O’Rourke (n 22) 10; E O’Dell, ‘Restitution’ in R Byrne and W Binchy (eds), Annual Review of Irish Law 1997 (Dublin, Round Hall, 1997) 607, 619. 61 Dublin Corporation v Building and Allied Trade Union [1996] 2 ILRM 547 (SC) (hereafter ‘Dublin Corporation’) 558; Harris v Quigley [2005] IEHC 81, [2005] IESC 79, [2006] 1 IR 165 (hereafter ‘Harris’) 189. 62 Duff (n 60). 63 Rogers (n 29). 49 50

Constitutional Redress in Ireland 279

in the Health (Amendment) Bill Reference, the Supreme Court flagged mistake of law among potentially relevant issues and cited Rogers.64 This indirect approval of the Rogers solution does not preclude the introduction of the internationally-accepted modern approach at the first opportunity. (c) The Intertwining of Duress Colore Officii and Mistake These key authorities do not differentiate clearly between the grounds of duress colore officii and mistake.65 In Dolan v Neligan, the test for recovery for mistake of law incorporated the inequality of the parties,66 and in Rogers, Griffin J’s treatment of mistake and duress was intertwined.67 Having identified the question as ‘whether a payment made, in circumstances such as the present, in mistake of law is recoverable’, Griffin J answered with reasoning related to colore officii.68 His ratio encompasses both grounds without acknow ledging that they are distinct alternatives.69 In both O’Rourke 70 and Harris v Quigley,71 the courts appear to present ‘colore officii’ as a means of rebutting the mistake of law bar, not necessarily an independent cause of action.72 This shows that the Irish courts, shortly before Murphy,73 were alive to the notion that cases involving overpaid taxes and other payments to the State constituted a particular category of unjust enrichment scenario, in which the inequality of the parties could not be ignored, even where the claim was formulated on the ground of mistake. The connexity of voluntariness and inequality of status in ‘the relationship of the parties’74 shifted the Irish conception of duress colore officii away from the necessity to establish specific compulsion overpowering the plaintiff’s will.75 Dolan and Rogers set the stage for the introduction of the policy ground in Murphy.

(2) Overpaid Taxes in Ireland in the Woolwich Era It is clear that current Irish law espouses a principle of repayment of exactions by the State (i) whether the payment takes the form of taxes or payments to a public authority for a service,76 and (ii) whether the payment is not due because the law is invalid or misapplied.77 In O’Rourke,78 Keane J rejected a proposed distinction between invalidity and misconstruction, finding that there was no meaningful distinction between the two causes of overpayment.79 Health (Amendment) Bill Reference (n 60) 177, 202. O’Rourke (n 22)10; see A Burrows, ‘Public Authorities, Ultra Vires and Restitution’ in A Burrows (ed), Essays on the Law of Restitution (Oxford, Clarendon Press, 1991) 49; McCamus (n 4) 246. 66 Dolan (n 27) 259. 67 Rogers (n 29) 270, 271. 68 ibid 271. 69 ibid 271. 70 O’Rourke (n 22). 71 Harris (n 61) 189. 72 O’Rourke (n 22) 10; Harris (n 61) 189. 73 Rogers was decided on 11 March 1981. 74 Rogers (n 29). 75 ibid 271. 76 Health (Amendment) Bill Reference (n 60). 77 O’Rourke (n 22); Harris (n 61). 78 O’Rourke (n 22). 79 ibid 9, 13; Harris (n 61) 179, 191; see also Woolwich (n 2). 64 65

280 Niamh Connolly There has never been any doubt that overpayments to State authorities in respect of services as opposed to taxation alone could be challenged on the same basis as taxation levies.80 Consistent with principle, this principle applies to a wide range of public bodies.81 (a) Reception of the Woolwich Principle and its Relationship with Home-grown Approaches O’Rourke, concerning small group of State officials mistakenly taxed as employees, was the first case to endorse Woolwich 82 as Irish law.83 Surprisingly, in incorporating Woolwich, O’Rourke does not acknowledge its similarity to Murphy. The judgment concentrates on rebutting the Murphy defence, rather than pointing to the primary right to redress which the earlier decision equally espouses. O’Rourke reflects a misinterpretation of Murphy, by which, rather than first recognising a cause of action and then, setting up a defence, it is seen as excluding the claim in the first place.84 In reality, Murphy is perfectly consonant with Woolwich. Both authorities are law in Ireland and both provide a general right of recovery of unlawfully demanded taxation payments. Murphy pre-dated Woolwich and emerged from the distinctive Irish legal and constitutional framework. It is possible either that the two authorities represent the same cause of action or that they are in fact discrete principles (as the express adoption of Woolwich85 into Irish law86 may imply). If Murphy stands for a public law approach to overpaid taxation, then either that proposition is relevant to other jurisdictions or Irish law has diverged from them as regards the nature of this claim. If the public law interpretation of Murphy is not correct, then, despite express approval of Woolwich, the Irish courts may in fact be applying a distinctive variant of duress colore officii rather than a policy-based reason for restitution. Secondly, while the main Irish cases prior to Murphy demonstrate an expansive, liberal approach to the doctrines of duress colore officii and mistake of law, there is some evidence post-Woolwich of a retreat from these generous stances. In O’Rourke, Keane J excluded the duress colore officii claim as formulated in Dolan v Neligan because there was ‘no specific element of compulsion or duress’. He also summarily excluded the possibility of a claim based on mistake of law, notwithstanding the endorsement of this cause of action in Rogers where the recipient was primarily responsible for the mistake and the inequality of the parties rendered the payment involuntary.87 In Harris v Quigley,88 Geoghegan J again approved the Woolwich principle but seems also to have reinstated the Dolan v Neligan approach, at least as representing the law prior to recent English developments, beginning with Woolwich.89

Rogers (n 29); Health (Amendment) Bill Reference (n 60) 192. ibid. 82 Woolwich (n 2). 83 See also Bank of Ireland Trust Services Ltd v Revenue Commissioners [2002] 4 IR 178 (HC); Harris (n 61). 84 O’Rourke (n 22) 13–14, 18. 85 Woolwich (n 2). 86 O’Rourke (n 22). 87 Rogers (n 29) 271. 88 Harris (n 61). 89 ibid 189. 80 81

Constitutional Redress in Ireland 281

(b) Defences Defences existing at different levels of generality might avail in a Murphy- or Woolwichtype case. First, there are the general legal or equitable doctrines which may serve as a defence to many diverse types of legal action.90 These include ‘prescription (negative or positive), waiver, estoppel, laches, a statute of limitation, res judicata’.91 Prescription is approved in the Health (Amendment) Bill Reference.92 Henchy J in Murphy invoked the doctrine of laches.93 His emphasis on notice, rather than the fact of expenditure, distinguishes his superficially similar reasoning from the ordinary defence of change of position and indicates that a wider equitable principle is at stake.94 As with the basis of the claim itself, there are two competing views as to the basis of the defence which Henchy J granted the State in Murphy v Attorney General.95 On one view, it is an adaptive application of the ordinary restitutionary defence of change of position, recognised in Irish law.96 Alternatively, it is a specific defence which is offered to the State for policy reasons so as to avoid extreme disruption resulting from a claim. In Murphy, Henchy J established the change of position defence by pointing to the good faith of the recipient, its reliance on the validity of the transaction, and the fact that it justifiably changed its position by spending the taxes collected.97 Furthermore, he indicated that permitting recovery would be inequitable, as this would require the imposition of special taxation on the current ranks of taxpayers. The divergence between this and the orthodox understanding of change of position may be explained by the fact that Murphy pre-dates Lipkin Gorman98 and by an inclination in the Irish courts to interpret change of position expansively.99 Equally, the view is advanced below that this is not true change of position but a novel policy-motivated defence. (c) The Limited Capacity of the Legislature to Extinguish Claims for Restitution The Irish courts have on a number of occasions disapproved of the possibility of the legislature limiting or excluding plaintiffs’ rights to a restitutionary remedy from the State. Walsh J in the Supreme Court in Dolan v Neligan100 evoked the possibility that it might be unconstitutional for the legislature to authorise a State body to refuse to repay monies levied in error.101 In Muckley v Attorney General,102 which arose shortly after Murphy, the Supreme Court ‘foiled’ Parliament’s efforts effectively to reinstate the original pre-Murphy

Burrows (n 65) 56. Murphy (n 1) 314, cited by Keane J in Dublin Corporation (n 61) 558; Health (Amendment) Bill Reference (n 60) 186. 92 Health (Amendment) Bill Reference (n 60). 93 Murphy (n 1) 318; Cleary (n 5) 157. 94 Murphy (n 1) 318. 95 ibid. 96 Dublin Corporation v Building and Allied Trade Union (High Court, 6 March 1996) at 75; E O’Dell, ‘Bricks and Stones and the Structure of the Law of Restitution’ (1998) 20 Dublin University Law Journal 101, 141–44. 97 Murphy (n 1) 318. 98 Lipkin Gorman (a firm) v Karpnale Ltd [1991] 2 AC 548 (HL). 99 O’Dell (n 96) 144–49. 100 Dolan (n 27). 101 ibid 275; cf O’Rourke (n 22) 18. 102 Muckley v Ireland [1985] IR 472 (SC) (hereafter ‘Muckley’). 90 91

282 Niamh Connolly tax demands against married couples who were in arrears.103 The Supreme Court engaged directly with the issue of the possibility of a statutory exclusion of claims for repayment in the Health (Amendment) Bill Reference.104 It ruled that a plaintiff’s right of recovery constitutes a chose in action, which is a constitutionally protected property right.105 A Bill purporting pre-emptively to extinguish all prospective claims arising from a mistakenly applied levy, without investigating whether valid and specific defences existed to meet them, was inconsistent with Murphy and unconstitutional.106

(3) The Murphy Debate: What is the Reason for Restitution? In Murphy v Attorney General,107 the Supreme Court ruled that there was a general right to recover unlawfully exacted taxes, subject to the restriction of this right when required by countervailing considerations.108 In that case, considering the massive scale of repayments required if all married couples in the State were reimbursed a decade’s worth of income tax,109 the Court restricted the remedy to recovery by the plaintiffs themselves of money paid in the two fiscal years since they first challenged the validity of the law.110 However, Henchy J made it quite clear that ‘the person damnified by the operation of the invalid provision will normally be accorded by the Courts all permitted and necessary redress’.111 The two principal competing views as to the basis of this right are that it is a manifestation of the doctrine of duress colore officii and that it results from constitutional principle. O’Dell and Cleary consider that Murphy represents an expansive application of duress colore officii.112 This view is consistent with prior authority and dicta in Murphy itself. This contrasts with the view adopted in this chapter that, while both duress colore officii and constitutional principle were presented as mandating the grant of a remedy, the policy argument, deriving from constitutional principle, was sufficient in itself to support the logic and solution of the Court. The terminology used betrays the case’s age.113 There was little argument as to the basis of the claim for restitution or the retrospective effect of a finding of invalidity.114 Contradictory dicta on the basis for recovery leave the decision open to alternative interpretations. Henchy J’s lead decision offers support for both a generous conception of duress colore officii and an independent policy rationale, which may be characterised as flowing from constitutional principle. He observed that normally, since an unconstitutional measure would provide no legal justification for any acts, persons affected by it could 103 Y Scannell, ‘The Taxation of Married Women – Murphy v Attorney General (1982)’ in E O’Dell (ed), Leading Cases of the Twentieth Century (Dublin, Round Hall Sweet & Maxwell, 2000) 327, 350. 104 Health (Amendment) Bill Reference (n 60). 105 ibid 180, 182, 196; Bunreacht na hÉireann 1937, arts 40.3.2 and 43.1. 106 Health (Amendment) Bill Reference (n 60) 192–93. 107 Murphy (n 1) 287. 108 Health (Amendment) Bill Reference (n 60) 192. 109 Murphy (n 1) 316. 110 ibid 323. 111 ibid 312. 112 E O’Dell, ‘Restitution – Debt and Taxes: Woolwich in Ireland’ (2002) 24 Dublin University Law Journal (ns) 295; Cleary (n 5) 160. 113 1982: Murphy (n 1) 320. 114 Murphy (n 1) 287–92, 315, 316, 334; Scannell (n 103) 349.

Constitutional Redress in Ireland 283

seek redress.115 Kenny J vigorously defended the view that an unconstitutional statute is void ab initio116 and the plaintiff’s right to recover the unlawfully exacted payments, as a matter of justice and as money demanded colore officii, subject only to the ordinary timelimit of six years.117 O’Higgins CJ, who dissented, disputing the retrospective effect of declarations of unconstitutionality, located most of his analysis within the field of constitutional law, but also evoked the idea of duress.118 (a) The Orthodox Reading of Murphy: Duress Colore Officii, or Something Like It Cleary views the Supreme Court judgment in Murphy 119 as the culmination of the Irish evolution of duress colore officii.120 In Murphy, Kenny J evoked the dual rationales of ‘elementary justice’ and an action for money received colore officii.121 Dicta from Henchy J also support the argument that the claim is based on duress colore officii. Both judges cited Mason v New South Wales.122 Henchy J considered it straightforward that the payments would ‘normally be recoverable as money exacted colore officii’, particularly since, having taken the form of compulsory deductions, they were in the strongest sense involuntary.123 Kenny J likewise reiterated an expansive formulation of the doctrine of duress colore officii,124 justified by presumed duress, which does not require protest or query before payment.125 If Murphy is to be interpreted as based on duress colore officii, it might be characterised as the offspring of duress colore officii: an irrebutable legal presumption of duress based on the inequality of status between State and taxpayer.126 This view is supported by the evolution of the terminology through Dolan, Rogers and Murphy. This incarnation of duress might be so far removed from the concept of compulsion of the plaintiff’s will as the logic justifying recovery that, despite the Court’s express recourse to the language of duress and case law on the point, the concept morphs in Murphy into an essentially new, policy reason for recovery.127 Birks considers that reference to duress adds nothing to our understanding of this right, which proceeds directly from the illegal governmental demand.128 This is consistent with Lord Browne-Wilkinson’s view that duress colore officii is merely an example of ‘a wider principle’ demanding repayment in such cases.129. (b) The Idea of Absence of Consideration Introduced via Woolwich If analysing Murphy itself reveals two possible justifications, the subsequent cases of O’Rourke and Harris v Quigley suggest a third possible basis for the policy ground. Keane J Murphy (n 1) 313. ibid 332. 117 ibid 335. 118 ibid 302. 119 ibid. 120 Cleary (n 5) 164. 121 Murphy (n 1) 334. 122 Mason v New South Wales (1959) 102 CLR 108 (HC) (hereafter ‘Mason v NSW’). 123 Murphy (n 1) 316. 124 Kenny J cites Sargood Brothers v The Commonwealth (1910) 11 CLR 258 (HC) and Mason v NSW (n 122). 125 Murphy (n 1) 334–35; cf ibid 302 (O’Higgins CJ dissenting). 126 Birks (n 2). 127 Murphy (n 1) 334. 128 Birks (n 2). 129 Woolwich (n 2). 115 116

284 Niamh Connolly in O’Rourke quoted at length from the speeches of the House of Lords in Woolwich130 and introduced to Irish law the idea that recovery of unlawful taxes might be akin to absence of consideration, as opposed to duress.131 Gilligan J took this a step further in Harris v Quigley.132 The plaintiff disputed his tax liability and succeeded before the Appeal Commissioners, but the defendant authority refused him the statutory repayment until all appeals were exhausted.133 Without engaging in detailed theoretical analysis, Gilligan J concluded that the plaintiff enjoyed a common law right of recovery because he ‘has made a payment for which there is no legal basis and therefore the defendants in their capacity as a public authority stand unjustly enriched by the payment’.134 Since there is no suggestion of Ireland adopting an absence of basis approach in private law, this is consistent with reading Murphy as either a policy-motivated unjust factor or a public law reason for restitution, rather than a species of duress. If the policy ground of recovery did indeed evolve from the colore officii doctrine, there is no longer any evidence of this pedigree in Gilligan J’s formulation. (c) The Public Law Reading of Murphy The alternative reading of Murphy advanced in this chapter is that the right to restitution emanates directly from fundamental public law principles. The decisions of the Supreme Court in Murphy itself support the view that a reason emanating from constitutional law is at least equally important as duress colore officii. For Henchy J, the ‘starting point’ in determining the extent to which things done under an unconstitutional statute may be undone is firmly constitutional; it ‘necessarily lies in the constitutional provisions’ governing judicial review of legislation.135 Then, having discussed the congenital invalidity of the unconstitutional enactment, Henchy J immediately and unequivocally asserted the principle of redress. He stated, concisely and clearly, that the condemned provision will normally provide no legal justification for any acts done or left undone, or for transactions undertaken in pursuance of it; and the person damnified by the operation of the invalid provision will normally be accorded by the Courts all permitted and necessary redress.136

It is precisely because the right to a remedy springs so easily from constitutional principle that Henchy J established it in so few words, and which may permit the constitutional reasoning to be overlooked. Similarly, when Henchy J conceded that ‘transcendent considerations’ may in some cases militate against the grant of a remedy, he framed the general principle as that, ‘what has been done in pursuance of a law which has been held to have been invalid for constitutional or other reasons will . . . give a good cause of action’.137 In this exposition of the principle, Henchy J did not invoke concepts such as duress and voluntariness. The material fact is the invalidity of the law, not the psychology of the plaintiff. O’Rourke (n 22) 11 ff. ibid 12, 13. 132 Harris (n 61). 133 ibid 171. 134 Harris (n 61) 179. 135 Murphy (n 1) 306. 136 ibid 313. 137 ibid 314. 130 131

Constitutional Redress in Ireland 285

Only subsequently did Henchy J adopt the terminology of private law, accepting that duress colore officii would ground a claim for the automatically deducted taxation.138 What is not fully clear from his judgment is the relationship between the analysis grounded in constitutional principle and that based on private law doctrines. On its face, the decision supports both the private law and the public law rationale, but the public law reasons predominate. This is compatible with Birks’ view that if duress is automatically presumed in State receipts cases, this renders reference to duress both redundant and a fiction.139 There are other indications that, despite reference to colore officii, the rationale is constitutional. Henchy J characterised the issue of when acts done pursuant to an unconstitutional law will not be disturbed as among ‘constitutional problems’.140 His formulation of the circumstances in which full redress may be denied was significantly wider than change of position in unjust enrichment law, embracing alien concepts such as the common good.141 He added that full restitution to all persons who had suffered loss would be ‘inequitable, if not impractical’ and ‘unjust and unreal’.142

C. IRISH LAW AS SUPPORT FOR A CONSTITUTIONAL PRINCIPLE OF REDRESS IN COMMON LAW JURISDICTIONS

(1) From National Solutions to Common Constitutional Ground Exploration of the public law ground for the remedy in Murphy will not necessarily translate to other countries’ Woolwich-type claims unless the public law rationale is compatible with their constitutional systems. Clearly a rule developed in one country’s constitutional system does not necessarily apply to others.143 The constitutional regime of the Republic of Ireland diverges markedly from that of the United Kingdom,144 as well as other jurisdictions in the common law world. The Irish perspective is heavily influenced by this jurisdiction’s constitutionalist tradition.145 The Irish superior courts are constitutionally empowered to engage in a strong form of judicial review of the constitutionality of legislation at the instigation of citizens.146 The legal system readily accommodates citizens taking plenary proceedings against the State and obtaining a full range of substantive remedies.147 Nonetheless, despite the obvious and significant differences between constitutional orders and approaches to constitutional justice, the commonality of shared and fundamental constitutional principles justifies the adoption of the public law reasoning modelled in Murphy in other common law jurisdictions. Our common constitutional inheritance includes the rule of law and the principle of legality, offering a shared constitutional ibid 317. Birks (n 2) 194. 140 Murphy (n 1) 315. 141 ibid 324. 142 ibid 320. 143 ibid 294; see Jones (n 12) 5. 144 See O Dixon, ‘The Law and the Constitution’ (1935) 51 LQR 590. 145 State (Burke) v Lennon [1940] IR 136 (SC) 144; Ryan v Attorney General [1965] IR 294 (SC) (hereafter ‘Ryan’) 310; Crotty v An Taoiseach [1987] IR 713 (SC) 742; see also G Hogan and G Whyte (eds), JM Kelly: The Irish Constitution, 4th edn (Dublin, Tottel, 2003) preface to 1st edn, 98–99, 104; see also Jones (n 12) 3. 146 Bunreacht na hÉireann 1937, art 34.3.2. 147 Byrne v Ireland [1972] IR 241 (SC) (hereafter ‘Byrne’) 280; Hogan and Whyte (n 145) 786. 138 139

286 Niamh Connolly rationale for the policy ground for recovery endorsed in Murphy and Woolwich.148 These principles are fundamental in jurisdictions which adhere to parliamentary supremacy as much as those which permit the judicial review of the constitutionality of legislation. The initial legal analysis which may determine the invalidity of a measure must respect the framework of each country’s domestic constitutional law. Woolwich is a reminder that invalid instruments exist in English law despite the absence of judicial review of Acts of Parliament.149 Once a rule has been deemed invalid for whatever reason is compatible with the jurisdiction’s legal order, the next stage of analysis may be susceptible to a shared solution and reasoning which proves compatible with our different legal systems and cultures. The grant of a remedy to persons adversely affected by an invalid measure does not depend on the reason for the finding of invalidity, deriving instead principally from the proposition that State agents should not be permitted to act beyond the scope of powers and duties conferred by law.

(2) The Specificity of Public Bodies The fundamentally distinctive character of the State is central to the argument that restitutionary remedies for overpayments by subjects are mandated by public law. The State exerts a real power of implicit compulsion against private parties, so that its relations with them are inadequately described by the assumptions of private law.150 The principles governing consent and duress between ordinary private law bodies, which are presumed in theory to be free and equal, are unsuited to relations between sovereign and subject. The evolution of duress colore officii as an expansive doctrine responds to the distinctive character of the State as defendant.151 In principle, duress requires illegitimate pressure,152 which, Smith argues, renders duress inappropriate to the case of payments to the State.153 Traditionally duress colore officii required that the State have demanded payment in return for a service. Duress could also be defeated by voluntariness, inferred from payment without protest. These restrictions do not adequately reflect the relations between State and subjects. In the Health (Amendment) Bill Reference,154 the Supreme Court cast doubt on the inferences which might be drawn from citizens’ failure to protest a levy, considering patients’ vulnerability and lack of familiarity with the relevant statutes.155 These observations support the view that citizens ought not to be unduly penalised for acquiescing in demands which are discovered to be incorrect or unlawful. In rejecting a requirement to prove actual pressure and assuming compulsion based on the unequal status of the State and its subjects, Irish courts demonstrated an appreciation of the power imbalance which differentiates the State from ordinary private litigants.156 However, the significance of the State’s coercive power is more appropriately reflected in Birks (n 2); see Williams (n 7) 35. Woolwich (n 2). 150 Jones (n 12) 20; Collins (n 31) 429. 151 Dolan (n 27). 152 See Burrows (n 65). 153 Smith (n 8) 17; see Jones (n 12) 14; E Bant, ‘Change of Position as a Defence to Restitution of Unlawfully Exacted Tax’ (2012) Lloyd’s Maritime & Commercial Law Quarterly 122, 125–26. 154 Health (Amendment) Bill Reference (n 60). 155 ibid 178. 156 Dolan (n 27); Rogers (n 29). 148 149

Constitutional Redress in Ireland 287

public law than by modifying the doctrine of duress. Public law is designed to identify and resolve the kind of concerns at stake. The policy reason for repayment responds to this specificity by shifting the focus entirely away from the quality of the plaintiff’s consent. Its logic concerns the State and its relationship with those who are subject to its laws. Collins argues that the problem with restitution flowing directly from the unlawful demand is that it ignores the voluntariness of payment.157 However, because the reason for restitution is the illegality of the State’s action, the motivation of the plaintiff in meeting the demand is immaterial.158 The citizen is obliged to meet lawful demands, and cannot refuse to obey laws which he deems inappropriate; as a corollary, the State ought to undo acts done in compliance with unlawful commands. The compliant citizen should not be penalised when it comes to the consequences of invalidity.159 Whether explained as part of unjust enrichment law or constitutional principle, the Woolwich principle is explained by the distinctive relationship between the State and its subjects. Smith’s criticism of Kingstreet is rooted in modern Canadian unjust enrichment law. He argues that Canadian juristic reasons or analysis160 could accommodate a solution in Kingstreet without looking outside that branch of law.161 This point does not apply to other jurisdictions which have not endorsed absence of basis reasoning. Nor should the easy availability of a private law claim obscure the proper, principled focus on the constitutional reason for recovery. There are good reasons why, even if we reject absence of basis analysis in private law, restitution should follow automatically when the State makes a demand of its subjects without proper legal foundation.162 Moreover, the grant of a remedy when public bodies act without legal justification does not at all imply an absence of basis approach to ordinary private law cases. This proposed rule is specifically designed to sanction excesses of State power, because the State enjoys a monopoly of power and must therefore be constrained to respect the boundaries of its power. Where the State acts without legal grounds this amounts to an abuse of power, whereas where an individual receives a benefit from another, it is appropriate for unjust enrichment law to inquire into the intentions behind the interaction, giving due respect to the free will of two free, equal parties and the principle of security of receipt.

(3) Constitutional Logic and the Right to a Remedy The most important constitutional reason for restitution of overpaid taxes is the imperative to constrain the exercise of power according to the rule of law. The orientation of constitutional law in a democracy is essentially restrictive towards those to whom we delegate the right to exercise power. Bastarache J articulates this view, describing the claim in Kingstreet as concerning, ‘a public law remedy in a constitutional principle stemming from democracy’s earliest attempts to circumscribe government’s power within the rule of Collins (n 31) 412. Kingstreet (n 4) [53]; T Hill, ‘Restitution from Public Authorities and the Treasury’s Position: Woolwich Equitable Building Society v IRC’ (1993) 56 MLR 856, 861–62. 159 Jones (n 12) 28; Collins (n 31) 421; McCamus (n 4) 247. 160 Garland v Consumers’ Gas Co 2004 SCC 25, [2004] 1 SCR 629. 161 Smith (n 8) 20. 162 McCamus (n 4) 256. 157 158

288 Niamh Connolly law’.163 A key difference between the private and public law explanations is that the constitutional solution does not just aim to reverse unjust enrichment but also reflects the need to impose a discipline on State agents, so as not to give effect or legal sanction to illicit uses of State power. If it is accepted that public law contains its own reasons for redress, this allows for two distinct explanations of the relationship between the Murphy claim and the structure of unjust enrichment. One possibility is the hybrid approach, according to which constitutional principle could supply the reason for restitution, but such cases would otherwise remain within the structures of the law of unjust enrichment. An alternative conceptual model sees a parallel principle against unjust enrichment operating in a public law remedies context. This implies the adaptive duplication of some of the rules concerning unjust enrichment within public law. Pal argues that ‘public law principles simply do not fit well within a private law framework’.164 However, the issue is not chiefly that unjust enrichment law is ‘ill-suited’ to resolve the issues in taxation cases. The constitutional argument for a remedy is not the only possible basis for restitution and conceptual difficulties with applying concepts such as enrichment to the public coffers need not in reality constitute an obstacle.165 Ordinary unjust enrichment law can still be applied to cases against the State, provided the Murphy defence is allowed. However, locating the solution within unjust enrichment law ignores the important public law reason for recovery and possible public-law reasons for limitation of that right. If Woolwich is regarded as a private law action inspired by policy considerations, the constitutional rationale remains implicit and anomalous. In Murphy, the determination that an unconstitutional taxation statute was void ab initio set the scene for a second stage of analysis as to the consequences for affected persons of this invalidity. The affirmation that the availability of a remedy in Murphy flows directly from constitutional principle does not conflate the issues of voidness and redress.166 There must be two separate stages of analysis. The first determines the validity of the measure and the temporal limits of any invalidity, the second considers the extent to which any acts purportedly authorised by that measure must be undone. The Murphy solution is that there should be redress for unlawful actions: that there is a general prima facie right of recovery does not negate the existence of this second, separate inquiry. There are, it is proposed, three reasons within constitutional law for this right of recovery. These reasons may be discerned in Murphy, Woolwich and Kingstreet. They have a broader application than overpaid taxation cases alone.167 They support a general principle of attempting to undo what has been done under an invalid, and particularly, an unconstitutional measure.168 The public law principles which support restitution are: the need to vindicate citizens’ rights through the grant of appropriate redress for infringements; the imperative to uphold the principle of legality by denying practical effect to illegal measures; and a principle analogous to the principle against unjust enrichment, which derives support from the guarantee of private property rights.

Kingstreet (n 4) [40]. Pal (n 23) 79. 165 See eg Bant (n 153). 166 cf Cleary (n 5) 162. 167 Murphy (n 1). 168 ibid 306. 163 164

Constitutional Redress in Ireland 289

(a) Giving Effect to a Specific Constitutional Rule at Issue First, if the State were permitted to retain unlawful taxes, the specific breach of rights which led to the finding of unconstitutionality or invalidity would be permitted to persist. It is, as Bastarache J states in Kingstreet, the central concern of the courts to ‘guarantee respect for constitutional principles’.169 The harm suffered through the infringement of a citizen’s rights must be undone in order for those rights to be vindicated. A right implies enforcement and an effective remedy.170 Thus, Dickson J proposed that permitting the retention of money unlawfully collected ‘would be tantamount to allowing the provincial Legislature to do indirectly what it could not do directly’.171 Within the domain of constitutional law, there is an imperative that a remedy be granted to realise the full practical voidness of an unlawful measure.172 This rationale is also expressed in European Union law, where the repayment of unlawful charges is described as ‘a consequence and complement of the rights conferred on individuals by Community provisions’.173 In a number of opinions, Advocate General Reischl expressly derived the right to restitution from the direct effect of European Union law, its conferral of individual rights on citizens and the obligation of national courts to vindicate those rights.174 The substantive rights at issue will vary between jurisdictions. Whatever the substance of the rights at issue, general right to redress is necessary to ensure their effective protection.175 If possible, that which was done under an unconstitutional law should be undone, to assure proper effect to the constitutional rule in question. Where restitution is not possible or appropriate, compensatory damages may be mandated to vindicate the violated right. (b) The Principle of Legality The second public law reason for the Woolwich principle is the principle of legality.176 This principle requires that every act of ‘governmental power . . . have a strictly legal pedigree’.177 Dixon regarded it as inherent in the supremacy of the law that ‘the Courts shall disregard as unauthorized and void the acts of any organ of Government, whether legislative or administrative, which exceed the limits of the power that organ derives from the law’.178 169 Kingstreet (n 4); see also Lambert JA’s dissent in Air Canada v British Columbia (Attorney General) (1986) 4 BCLR 356 (BCCA); Byrne (n 147) 281; Redmond v Minister for the Environment (No 2) [2004] IEHC 24, [2006] 3 IR 1 (hereafter ‘Redmond ’) 3; Meskell v Córas Iompair Éireann [1973] IR 121 (SC) (hereafter ‘Meskell ’); see also Bunreacht na hÉireann 1937, arts 34.5.1 and 40.3.1; Hogan and Whyte (n 145) 1296. 170 Meskell (n 169) 133; see also Joined Cases C-6/90 and C-9/90 Francovich v Italian Republic [1991] ECR I-5357 [33]. 171 Amax Potash Ltd v Government of Saskatchewan [1977] 2 SCR 576 (hereafter ‘Amax’) 590. 172 See also United Nations International Covenant on Civil and Political Rights 1966, art 2.3; P Blanchard, ‘The New Zealand Bill of Rights: Where Have We Got To After 16 Years?’ [2008] New Zealand Law Review 263, 271. 173 Joined cases C-397/98 and C-410/98 Metallgesellschaft Ltd and others v Inland Revenue Commissioners, Hoechst AG and another v Inland Revenue Commissioners [2001] ECR I-1727; see also Case 199/82 Amministrazione delle Finanze dello Stato v San Giorgio SpA [1983] ECR 3595. 174 Case 77/76 Ditta Fratelli Cucchi v Avez SpA [1977] ECR 987, Opinion of AG Reischl, 1020; Case C-68/79 Hans Just I/S v Danish Ministry for Fiscal Affairs [1980] ECR 501, Opinion of AG Reischl. 175 Cases C-46/93 and C-48/93 Brasserie du Pêcheur SA v Bundesrepublik Deutschland [1996] ECR I-1029 (hereafter ‘Brasserie du Pêcheur’) 1149. 176 Birks (n 2) 204. 177 HWR Wade and CF Forsyth, Administrative Law, 10th edn (Oxford, OUP, 2009) 17; see also Kennedy v Law Society of Ireland [2002] 2 IR 458 (SC) 486. 178 Dixon (n 144) 596; see R v Secretary of State for the Home Department, ex parte Pierson [1998] AC 539 (HL).

290 Niamh Connolly Ho characterises the Woolwich principle as ‘a distinct principle of restoring legality’ and argues credibly that the Court’s concern in Woolwich is not ‘reversal of unjust enrichment but with the control of public power, a fundamental objective of public law’.179 In a constitutionalist jurisdiction which subscribes to the hierarchy of norms, the principle of legality implies, first, that unconstitutional statutes are void ab initio,180 and, consequently, that unlawful laws should be deprived of practical effect. The latter implication of the principle applies whatever the reason for invalidity and should apply even in nonconstitutionalised legal systems. Logic dictates that a void law or other State measure, whatever the reason for its voidness, must be rendered fully ineffective – that is, deprived of practical effects – in addition to a declaration of invalidity. In Kingstreet, Bastarache J stated that the retention of taxes claimed under ultra vires legislation ‘undermines the rule of law’.181 The prohibition of the collection of taxes without legislative authority is a specific manifestation of the principle of legality. The simplicity of this constitutional reason for recovery is clear from the judgments of both Bastarache J in Kingstreet and Henchy J in Murphy. For Bastarache J, the constitutional approach permits the claims to be dealt with ‘more simply’ than unjust enrichment law.182 Henchy J’s logic is that an unconstitutional law ‘will normally provide no legal justification for any acts done’, so that the person who suffers by the implementation of an invalid measure ‘will normally be accorded by the Courts all permitted and necessary redress’.183 The straightforward constitutional basis for the remedy has been explained by Bastarache J as, ‘the Crown should not be able to retain taxes that lack legal authority’;184 by Wilson J as, ‘[w]here the payments were made pursuant to an unconstitutional statute there is no legitimate basis on which they can be retained’;185 and by Dickson J as, ‘[i]f a state cannot take by unconstitutional means it cannot retain by unconstitutional means’.186 The logic of this position is irresistible. As a matter of constitutional law, State organs cannot justify the retention of sums charged under a void provision, because to do so would effectively permit them to circumvent the legal delimitation of their powers. As a candidate for a common constitutional rationale for repayment, legality has an advantage over the vindication of the substantive right affected by the underlying cause of unconstitutionality. The principle of legality is shared by the diverse jurisdictions which recognise Woolwich or equivalent principles, whereas substantive compliance with constitutional provisions and the effect of non-compliance necessarily relate to matters of domestic law. (c) Unjust Enrichment as a Tenet of Public Law The third rationale for recovery, which may exist within public law as well as private, is the prevention of unjust enrichment.187 First, this serves the interests of justice. Lord Goff’s HL Ho, ‘Beyond Restitution and into Public Law’ [1993] Singapore Journal of Legal Studies 583, 584. Murphy (n 1); cf Air Canada (n 4); Hogan and Whyte (n 145) 900. 181 Kingstreet (n 4); see also Jones (n 12) 18; Birks (n 2). 182 Kingstreet (n 4) [13]. 183 Murphy (n 1) 313; see also Brasserie du Pêcheur (n 175) 1144. 184 Kingstreet (n 4) [53]; see also PW Hogg, ‘Recovery of Unauthorised Taxes: a New Constitutional Right’ (2008) 46 Canadian Business Law Journal 5, 5. 185 Air Canada (n 4); N Rafferty, ‘The Role of the Supreme Court in the Development of a Canadian Law of Restitution’ (1994) 32 Alberta Law Review 557, 571. 186 Amax (n 171). 187 See Hogan and Whyte (n 145) 187. 179 180

Constitutional Redress in Ireland 291

allusion to ‘common justice’ in Woolwich may be accommodated here. The values, such as corrective justice, which underpin the principle against unjust enrichment as the central concept in the private law of restitution must surely also operate within public law.188 In fact, it would be more surprising if justice did not mandate a concept analogous to the principle against unjust enrichment in cases where the beneficiary happens to be a public body. More particularly, the respect of citizens’ property rights is itself a widespread constitutional value which demands that de facto expropriation not be permitted.189 In Ireland, the private law principle against unjust enrichment already has as its corollary in public law the constitutional imperative to vindicate individuals’ property rights. In the Health (Amendment) Bill Reference,190 the Supreme Court found that the plaintiff’s right to claim restitution was a constitutionally protected property right under Articles 40.3.2 and 43.1 of the Irish Constitution.191 Consequently, the same logic that underpins unjust enrichment law supplies an ancillary reason for offering a restitutionary remedy in these public law cases. Dr Williams favours a hybrid approach in which restitution of overpaid taxes is fully a creature of neither private law nor public law. She reasons that it cannot be a matter of public law because sanctioning illegality ‘cannot satisfactorily explain why the remedy for the ultra vires taxation must take the form of return of the money to the claimant’.192 There are, however, two reasons why the appropriate sanction for the State’s excess of power should be restitutionary, as opposed to compensation or another form of sanction for excess of power. First, this is the clearest, most direct way to undo the ultra vires act193 and return the State to the proper scope of its power. Secondly, public law also encompasses a principle against unjust enrichment. Thus, the restitutionary solution simultaneously achieves two important aims. Restitution couples disgorgement with compensation.194 The principal focus is on the defendant public body and its duty to undo the act which was done without legal authority; that this may take the form of returning a benefit to the plaintiff reflects a secondary and compatible concern with the prevention of unjust enrichment. For Smith, the need for wrongful conduct to prompt disgorgement renders the concept unsuited to cases of the invalid exercise of public power. However, such acts may be viewed through a constitutional lens as ‘wrongful’ in a relevant sense – that is, objectively unlawful and in breach of constitutional limits − despite the absence of reprehensible conduct.

(4) Restitutionary Redress is One Example of a Wider Constitutional Principle This chapter argues that the restitutionary taxation cases are one incarnation of a wider principle. This is widely supported by subsequent Irish authority and offers a convincing argument in favour of the public law reading of Murphy. Henchy J clearly envisaged that L Smith, ‘Restitution: The Heart of Corrective Justice’ (2001) 79 Texas Law Review 2115, 2144. See Article 1 of Protocol 1 to the Convention for the Protection of Human Rights and Fundamental Freedoms, on the Enforcement of certain Rights and Freedoms not included in Section I of the Convention, Paris, 20 March 1952. 190 Health (Amendment) Bill Reference (n 60). 191 ibid 180, 182, 196. 192 Williams (n 7) 37. 193 Ho (n 179) 586. 194 McCamus (n 4) 255–56. 188 189

292 Niamh Connolly the question of the consequences of a finding of unconstitutionality extends beyond the taxation example, as it involves the further question whether, and to what extent, what has been done pursuant to, or what has happened on foot of, an unconstitutional enactment may be revoked, annulled, rectified, or made the subject of a claim for damages or for some other form of legal redress.195

It follows from the location of the rationale in the principle of legality that the undoing of acts done by the State without proper authority would equally require the release of persons convicted under invalid laws or the reparation of loss caused to citizens through the application of invalid measures. Undoing all that was mandated by a void provision may in many cases be difficult. O’Neill adverts to the ‘spectre’ of unravelling things which have been done in the past.196 Irish cases have raised the prospect that elections and jury trials were unlawfully carried out.197 The repayment of overpaid taxes must surely lie at one end of the spectrum as a case in which it is straightforward to quantify and undo the actions undertaken in application of the invalid law. Offering compensation to mitigate other forms of prejudice and hardship is more difficult, but not impossible. (a) The Compensatory Constitutional Remedy Smith argues that the absence of a compensatory remedy within constitutional law logically undermines the case for a public law restitutionary remedy.198 In most jurisdictions, there is ‘no free-standing public law compensation claim’.199 However, recent developments in Irish law suggest that in one jurisdiction at least, such a claim does exist. Irish law long hinted at the existence of a ‘constitutional tort’.200 In Meskell v Córas Iompair Éireann,201 Walsh J proposed that citizens may initiate proceedings to protect a constitutional right even if the case falls outside recognised common law causes of action.202 He reasoned that the existence of the constitutional right necessarily entails the right to enforce that substantive right and receive a remedy.203 Redmond v The Minister for the Environment (No 2)204 expressly countenances the award of compensatory damages, including for the breach of constitutional rights through the enactment of unconstitutional legislation.205 In Keating v Crowley,206 the High Court awarded damages for the infringement of a person’s constitutional rights pursuant to an unlawful interim barring order, though Murray CJ in the Supreme Court207 expressed doubts about awarding damages where State agents had implemented laws they believed to be valid. Murphy (n 1) 306. A O’Neill, ‘The Effect of a Finding that Legislation is Unconstitutional: The Approach of the Irish Supreme Court’ (2007) 36 Common Law World Review 220, 221. 197 McMahon v Attorney General [1972] IR 69 (SC) (hereafter ‘McMahon’); De Búrca v Attorney General [1976] IR 38 (SC) (hereafter ‘De Búrca’); Redmond (n 169); see Hogan and Whyte (n 145) 897. 198 Smith (n 8) 26. 199 ibid 27; see Burrows (n 65). 200 Meskell (n 169); see Hogan and Whyte (n 145) 1304. 201 ibid. 202 ibid 132–33. 203 ibid 133. 204 Redmond (n 169). 205 ibid 3. 206 Keating v Crowley [2010] IESC 29. 207 ibid. 195 196

Constitutional Redress in Ireland 293

In Blehein v Minister for Health,208 the plaintiff sought compensation for his involuntary detention in a residential facility under an unconstitutional statute.209 Laffoy J distinguished cases in which the circumstances were such that a declaration of invalidity would suffice to vindicate plaintiffs’ rights. Relying heavily on Murphy, she rejected an argument that the obligation to obey laws in force precludes the award of a remedy when a statute is later invalidated.210 This interesting line of authorities has not yet firmly established itself, but the Irish courts are at the least willing to consider awarding damages against the State to compensate for the violation of citizens’ rights by the enactment of unconstitutional legislation. It is likely that Ireland will in future cases recognise and apply the concept of the constitutional compensatory remedy without express provision,211 based, in large part, on the Murphy redress principle.212 This constitutional compensatory remedy is precisely analogous to the constitutional restitutionary remedy devised in Kingstreet. It falls outside the bounds of tort law, based on the wrong of acting outside the confines of legal authorisation. This is not necessarily a culpable fault, but a grave transgression from the perspective of the rule of law.213 The Irish example reverses Smith’s logic: if there can be both compensatory and restitutionary damages to remedy the consequences of unconstitutional laws, it is more likely that these are mandated by the same constitutional principle than that the restitutionary response belongs to unjust enrichment law.

D. THE SHAPE OF THE PUBLIC LAW RESTITUTIONARY CLAIM

(1) Relationship with Private Law Claims Deutsche Morgan Grenfell is strong authority for the concurrent availability of different causes of action.214 This is not contingent on viewing Woolwich as a private law claim. The constitutional remedy may indeed be ‘superior’, in that it derives from a fundamental constitutional principle and because it is simple and effective for litigants, but this does not require that it displace other rights of action from the territory which it occupies. As Smith indicates, it is not apparent why private law claims in unjust enrichment should disappear simply because public law offers a better reason for a remedy.215 Contrary to the views of Dr Williams216 and Bastarache J in Kingstreet, it does not seem necessary that the existence of a constitutional remedy should preclude the alternative formulation of ordinary private law claims against the State. Bastarache J’s view that ordinary unjust enrichment law is inappropriate to State receipts cases because it ignores important Blehein (HC) (n 16). Blehein v Minister for Health and Children [2008] IESC 40. 210 Blehein (HC) (n 16) [10.6]; see McDonnell v Ireland [1998] 1 IR 134 (SC) (hereafter ‘McDonnell’). 211 cf Canada Act 1982, s 24(1); Constitution of South Africa, art 38. 212 cf Watkins v Secretary of State for the Home Department [2006] AC 395 (HL); see also G McLay, ‘Damages for Breach of the New Zealand Bill of Rights – Why Aren’t They a Sufficient Remedy?’ [2008] New Zealand Law Review 333, 334. 213 Jones (n 12) 38–39, 45. 214 DMG (n 25). 215 Smith (n 8) 29 ff. 216 Williams (n 7). 208 209

294 Niamh Connolly constitutional principles217 is not entirely convincing.218 The Murphy remedy can be explained as a matter of public law, justified by the principle of legality, but there is no reason why claims formulated according to the ordinary law of unjust enrichment might not be made against the State in the alternative, provided that, for reasons of logic, pragmatism and consistency, the exceptionally limited policy-based defence permitted to the State in Murphy also be allowed in response to other grounds. The State is subject in principle to the ordinary law of obligations, besides public law actions.219 Williams argues that if the claim is brought under private law principles, ‘the court will be prevented from seeing the case as anything more than purely private’.220 However, the sole concession which is required to be made where a case is brought against the State under an ordinary unjust factor is to allow an exceptional policy-motivated defence. Therefore the Murphy defence should be allowed to reflect the special status of the State as defendant, regardless of the basis of the claim. Smith affirms the availability of private law claims against the State as a general principle, with private law claims adapted where necessary to respond to the distinctive character of public bodies.221 Any additional ‘layer of analysis’ occurs within the private law rules.222 This favours the availability of actions based on unjust factors while denying the need to treat these cases as public at all. However, there is a distinction between cases which call for the application of modified private law to the State and those in which the fundamental basis of the claim derives specifically from the public law nature of the body. Both can coexist as alternatives for a restitutionary claim against public authorities, as indicated by the Irish Supreme Court in the Health (Amendment) Bill Reference.223 There are two reasons to insist on the existence of the constitutional reason for recovery in parallel to adapted private law causes of action. First, the strongest, most persuasive reason for restitution in all these cases is the principle of legality. Secondly, it improves the coherence of the constitutional law of remedies to recognise that the restitutionary remedy in these cases is simply one application of a wider principle of redress for the breach of individuals’ rights through an excess of State power. Equally, the maintenance of consentrelated claims in unjust enrichment is desirable because to exclude them risks prejudicing plaintiffs224 and tends towards the reintroduction of technical wrinkles in the fabric of restitution law by suggesting that the availability of ordinary unjust factors depends on the identity of the defendant. Moreover, the continued availability of a claim in unjust enrichment in the alternative to a public law claim answers fears that characterising the Woolwichtype claim as public law risks diluting the right of plaintiffs to a remedy.225 Smith argues that defences may operate differently when public bodies are parties to a private law claim.226 Logically, the Murphy defence must be permitted whether the claim Kingstreet (n 4) 358. Williams (n 7) 33. 219 Byrne (n 147) 280; Smith (n 8) 29. 220 Williams (n 7) 70 ff, 73; R Williams, ‘The Beginnings of a Public Law of Unjust Enrichment’ (2005) 16 Kings College Law Journal 195, 201. 221 Smith (n 8) 29. 222 ibid 30. 223 Health (Amendment) Bill Reference (n 60) [72]. 224 See G Virgo, ‘Restitution from Public Authorities: Past, Present and Future’ [2006] 11 Judicial Review 370; Bant (n 153) 125–26. 225 See N Cleary, ‘Restitution and Public Law’ [2012] Restitution Law Review 38. 226 Smith (n 8) 30; see also J Beatson, ‘Restitution of Taxes, Levies and Other Imposts: Defining the Extent of the Woolwich Principle’ (1993) LQR 401, 413. 217 218

Constitutional Redress in Ireland 295

brought against a public body is a constitutional one or in unjust enrichment. This leaves the State defendant in a somewhat privileged position in so far as the exceptional Murphy defence is available to it and never available to private parties.227

(2) The Scope of Application of the Constitutional Remedy (a) The State as Defendant The courts accepted in Auckland Harbour Board v R that constitutional principles necessitate the recovery of unauthorised payments by the State.228 This claim is presented as constitutional; this being so, it is perfectly consistent that the right to recover money from the State is also constitutional. Williams views the coercive power of the State as a neutral characteristic which triggers the public law solution, whether the State is plaintiff or defendant.229 However, the constitutional logic which explains a right to restitution of overpayments from a State defendant does not apply equally to the State as plaintiff. Public law is oriented to constrain the exercise of public power in the interest of protecting the liberty of citizens. The recovery of public funds paid out without authority does not flow from the same logic.230 It is more analogous to restitution following a transfer which is ineffective due to corporate incapacity – the concern is not coercive power but the proper stewardship of assets. Were it not for the precedent which establishes that the right to automatic recovery applies when the public body is the plaintiff, it would seem that, logically, the public law right of recovery might be triggered only in those cases where the public body is the defendant as opposed to the plaintiff. At the least, the constitutional rationale for recovery is different in the two cases.231 (b) Public Law Dealings The Murphy claim should exist only where the enrichment occurs in the context of an exercise of public power, not to all dealings between public bodies and other legal persons. The hallmark of public power, which justifies recourse to public law, is the exercise of a power to command. In many cases, depending on the features of their dealings, public or semi-public bodies’ relations with their clients or contracting partners fall squarely within private law.232 Williams argues that ultra vires payments by public bodies should trigger the application of constitutional law, so that the Woolwich principle would apply also to the swaps cases.233 This conclusion contrasts with that of Lord Goff in Kleinwort Benson v Lincoln County Council, who preferred the view that the swaps cases were private law transactions by public bodies.234 cf Cleary (n 5) 178. Auckland Harbour Board v The King [1924] AC 318 (PC) (hereafter ‘Auckland Harbour Board’) 326–27. 229 Williams (n 220) 205. 230 cf Kingstreet (n 4) [24]. 231 Burrows (n 65). 232 See Smith (n 8) 21. 233 Williams (n 220) 199. 234 Kleinwort Benson (n 59). 227 228

296 Niamh Connolly (c) The Misapplication of Valid Measures Another question is whether the constitutional restitutionary remedy would be limited to cases where an empowering measure was invalid, or whether it would also apply where a valid measure was misapplied. There might appear to be a distinction because the Murphy public law analysis starts from the Kelsenian model of statutes which are inconsistent with a higher norm and therefore void. However, the same constitutional logic applies in both cases.235 Acts done outside the scope of a valid measure are equally as unlawful236 as acts purportedly justified by an invalid instrument, so that in consequence, the principle that the rule of law requires redress is equally applicable to both sorts of unlawful exaction.237 In a similar vein, Irish authority suggests that compensatory damages for prejudice caused to citizens by unauthorised State actions might arise whether the excess is attributable to the executive or legislative power.238

(3) The Murphy Defence Cases of restitution from the executive raise specific public law and policy concerns which pull in both directions.239 The principle of legality demands restitution. Yet, Air Canada was correct that fiscal chaos may be a valid concern, if emphatically wrong in the weight accorded to it.240 Murphy offers a model for the appropriate resolution of these considerations, establishing a prima facie right of recovery, subject to a defence in exceptional cases, where restitution would entrain a financial crisis for the State.241 This may arise where ‘it would be impossible, or unjust, or contrary to the common good, to attempt to reverse or undo’ the effects of an invalid measure.242 The recent judgment in DPP v Cunningham reiterates that the duty of the courts to provide redress may be limited when, ‘the consequences of the full or even partial retroactive application of a finding of unconstitutionality might be so catastrophic for organised society that they could not be accepted’.243 (a) A Policy-Based Defence Cleary discerns two defences in Murphy: change of position and laches, which she explains responds to the equitable claim.244 Henchy J’s judgment referred to both these concepts. However, the express policy motivations for the defence in Murphy itself, as well as its Hogg (n 184) 8. See Williams (n 7) 54. 237 See also Jones (n 12) 10; Canadian Pacific Airlines Ltd v British Columbia (Attorney General) [1989] 1 SCR 1133; Chiasson v Canada (Attorney-General) (2008) 295 DLR (4th) 744 (FCTD); M McInnes, ‘Canada’ [2010] Restitution Law Review 117, 121; cf Merchant Law Group v Canada (Revenue Agency) 2009 FC 755, (2009) 311 DLR (4th) 124 (FCTD). 238 Blehein (HC) (n 16). 239 Jones (n 12) 21; Birks (n 2); McCamus (n 4) 256. 240 See F Bachand, ‘Restitution of Unlawfully Levied Taxes: Survey and Comparative Analysis of Developments in Canada, Australia and England’ (2001) 38 Alberta Law Review 960, 983. 241 Murphy (n 1); Health (Amendment) Bill Reference (n 60); McCamus (n 4) 257. 242 Murphy (n 1) 323. 243 Cunningham (n 15). 244 Cleary (n 5) 168 ff. 235 236

Constitutional Redress in Ireland 297

treatment in later cases, show that it is policy-motivated and distinct from change of position. By focusing on the policy rationale for the defence in Murphy, as opposed to the generally-available defence of change of position, O’Rourke indicated that the policy motivation was indeed predominant in Murphy.245 Henchy J’s decision explicitly recognised the defence of change of position, but was at least as strongly influenced by policy factors or ‘transcendent considerations’.246 He evoked the difficulty that tens of thousands of other taxpayers also paid the unconstitutional tax for more than a decade.247 Consequently, he limited the plaintiffs’ right of recovery to the time from which they objected by instituting proceedings.248 Taxpayers who had not instituted proceedings would not be reimbursed. According to Henchy J, taxpayers could be presumed to know that their taxes would immediately be spent,249 and he invoked the doctrine of laches to justify the extinction of the majority of taxpayers’ prima facie right to restitution.250 Muckley 251 confirms a constitutional law interpretation of Murphy according to which the Murphy defence is ‘an exception to the general principle that voidness means having no legal effect’.252 The Murphy defence is of broader application than simply to overpaid taxation cases.253 Cases such as McMahon 254 (elections), De Búrca255 (jury trials), Murphy 256 (taxation) and the A Case 257 (criminal conviction) all demonstrate exceptional circumstances in which the extremely grave consequences of fully upholding the rule of law might understandably permit an exception, by which an invalid law is artificially given life, from the time of its attempted enactment to the declaration of its invalidity. This exceptional defence responds to policy factors unknown to the private law of unjust enrichment.258 Unlike change of position, fiscal chaos does not serve to negative enrichment. While the reasons for it broadly mirror the security of receipt rationale underlying change of position,259 this defence need not conform to the contours of the law governing the ordinary private law defences. The Murphy defence differs from change of position because there is no question of requiring the State to show how the funds in question were expended in reliance on the validity of their receipt.260 This obviates the need to resolve the theoretical difficulties inherent in equating the exchequer with the assets of a private individual. Likewise, the Murphy defence does not present the problem raised by Henderson J in FII that the wrongful character of the unlawful levying of a tax excludes a change of position defence.261 Acknowledging that this is a distinct defence avoids the risk of distorting O’Rourke (n 22) 18. Murphy (n 1) 314. 247 See Scannell (n 103) 348. 248 Griffin J agreed with the reasoning of Henchy in Murphy (n 1) 330. 249 Murphy (n 1) 317. 250 ibid 318. 251 Muckley (n 102). 252 R Byrne and W Binchy (eds), Annual Review of Irish Law 2006 (Dublin, Round Hall Sweet & Maxwell, 2006) 176. 253 Cunningham (n 15). 254 McMahon (n 197). 255 De Búrca (n 197). 256 Murphy (n 1). 257 A v Governor of Arbour Hill Prison [2006] IESC 45, [2006] 4 IR 88 (SC) (hereafter ‘The A Case’). 258 Smith (n 8) 24. 259 Jones (n 12) 24, 44. 260 See Bant (n 153). 261 Test Claimants in the Franked Investment Income Group Litigation v HM Revenue & Customs [2008] EWHC 2893 (Ch), [2009] STC 254. 245 246

298 Niamh Connolly the law on change of position as it applies in normal private law unjust enrichment cases by stretching it to apply to the State. Burrows queries why the State should enjoy this special favour.262 The State is not owner, but steward, of public assets.263 It must take account of the general good. The privilege is mandated by the public interest, and circumscribed always by the exceptionally limited character which the Murphy defence must retain if it is not to undermine the primary right to a remedy. In truth, the denial of a remedy to persons other than those who initiated successful legal action is not motivated by the interests of justice.264 It is purely pragmatic.265 It permits an escape from fiscal chaos in Murphy, or embarrassment and public revulsion in the criminal context of The A Case.266 Kelly argues in favour of the pragmatism justification, citing The State (Byrne) v Frawley 267 and Muckley.268 Laffoy J in Blehein asked whether redress is ‘undesirable, impractical or impossible’.269 Public law reasonably involves an element of policy-based discretion which is alien to the private law of obligations.270 None of the fundamental rights expressed in the Irish Constitution is absolute.271 This is justified and inevitable given the necessity for the State to reconcile conflicting rights and the requirements of the common good.272 Invoking the decision of Barrington J in Muckley, Kelly reasons that, if the right to redress is itself constitutionally founded in Article 40.3, which requires the State to vindicate the rights of citizens, it follows that this right, like other constitutional rights, is not absolute.273 Consequently, it may be legitimate for the State to curtail the right to redress when it conflicts with other rights or pressing demands of the public interest. The Murphy defence may be justified as an exceptional and proportionate curtailment of the property rights of the taxpayer or other citizen.274 While none of the solutions advanced to prevent a declaration of unconstitutionality from having unacceptably grave consequences is entirely compatible with constitutional principle, the Murphy approach of permitting a public interest defence in exceptional cases is less illogical than either (i) suspending the declaration of unconstitutionality or (ii) deeming the law to have been valid until the moment of its condemnation.275 It is better to accommodate the pressing policy concerns which influenced the court as a defence to a valid claim than to deny the voidness of the unconstitutional measure. Burrows’ suggestion that Parliament could legislate to prevent restitutionary claims would be unconstitutional in Irish law.276

Burrows (n 65). Auckland Harbour Board (n 228); McCamus (n 4) 257. 264 Cunningham (n 15). 265 Birks (n 2). 266 The A Case (n 257) 128. 267 The State (Byrne) v Frawley [1978] IR 326 (SC) (hereafter ‘Frawley’). 268 Muckley (n 102); Hogan and Whyte (n 145) 908. 269 Blehein (HC) (n 16) [12.7]. 270 Williams (n 7) 33. 271 Ryan (n 145); Hogan and Whyte (n 145) 1268. 272 Bunreacht na hÉireann 1937, art 43.2.2. 273 Hogan and Whyte (n 145) 908. 274 The A Case (n 257) 128. 275 Birks (n 2) 198; cf Kingstreet (n 4) 351. 276 Burrows (n 65). 262 263

Constitutional Redress in Ireland 299

(b) An Exceptional, Limited Defence The defence must be extremely limited.277 Ireland’s leading constitutional law text advocates setting the threshold at the point where ‘serious embarrassment’ would be caused to the State by giving full effect to a finding of unconstitutionality.278 In O’Rourke,279 the defendant argued the Murphy defence in opposition to a Woolwich claim.280 Keane J distinguished Murphy’s justification for the defence on the basis of the exceptional ramifications envisaged in the earlier case, concerning the taxes of tens of thousands of citizens, whereas O’Rourke could affect fewer than a hundred individuals.281 Consequently, the policy imperatives were ‘legitimately outweighed by the injustice to the plaintiff’.282 While this solution, which distinguishes between the scale of the liability in Murphy and O’Rourke, accords with common sense, it leaves open the question of how and where the boundary might be drawn between the case where restitution would have a severe impact on State finances and a negligible one. Further, might this dividing line shift according to the health of the State’s finances? Burrows reasonably considers a special fiscal chaos defence to be vague and arbitrary.283 In subsequent Irish authorities, Murphy has sometimes unfortunately been mischaracterised as authority against the grant of a remedy where a law is void for unconstitutionality. Far from ‘immunity’ for public authorities, the general principle espoused in Murphy is recovery. It is only because of the immense scale of the adverse effects on the exchequer of full restitution in the Murphy case that the negative dimension of the rule came into play.284 It is the reverse of the Air Canada position.285 Fiscal chaos is the exception rather than the rule. It should be demonstrated, not presumed. Duration is one factor, but so are the sums at issue. (c) A Challenge to Murphy: A v Governor of Arbour Hill Prison Murphy established the ‘orthodoxy’ in Irish law that unconstitutional statutes are void ab initio.286 It was applied in Muckley and expressly approved by Barrington J in McDonnell v Ireland.287 However, the notorious Supreme Court decision in A v Governor of Arbour Hill Prison challenged the Murphy recipe of voidness, a general right of recovery, and an extremely limited policy defence. More recently, in DPP v Cunningham,288 the Court of Criminal Appeal has distinguished The A Case and reiterated core aspects of Murphy. In McDonnell,289 O’Flaherty J suggested that the striking down of legislation ‘can only crystallise’ in respect of the litigation before the court.290 The A Case concerned a man con De Búrca (n 197) 72. Hogan and Whyte (n 145) 906. 279 O’Rourke (n 22). 280 ibid 13–14. 281 ibid 17. 282 ibid 18. 283 Burrows (n 65). 284 O’Rourke (n 22). 285 Air Canada (n 4). 286 O Doyle, Constitutional Law: Text, Cases and Materials (Dublin, Clarus, 2008) 450. 287 McDonnell (n 210). 288 Cunningham (n 15). 289 ibid. 290 See Byrne and Binchy (n 252) 178. 277 278

300 Niamh Connolly victed of unlawful carnal knowledge of a child, under a statute which was subsequently declared unconstitutional because it did not provide for a defence of mistake as to the victim’s age.291 The applicant in The A Case did not mistake the child’s age. In a decision which undermined Murphy while purporting to follow it, the Supreme Court refused to release him. In Doyle’s words, the Court ‘reversed the principle and the exception’.292 Murray CJ’s decision is a highly expansive reading of the Murphy defence, to the detriment of Murphy’s primary principle of redress. Denham J asserted that the general principle is that a finding of unconstitutionality will only have prospective effect, barring exceptional circumstances.293 The majority considered the interests of the common good superior to the logic of the hierarchy of norms.294 The Supreme Court decision in The A Case is not popular among legal scholars.295 The majority decisions purport to maintain the validity of a law which is incompatible with the constitution, contrary to the hierarchy of norms and constitutional principle. It entails the courts, which are charged with vindicating the constitution, knowingly tolerating the continued infringement of other citizens’ rights in exactly the same way as has led to a determination of unconstitutionality. Unlike his colleagues, Hardiman J acknowledged the true significance of Murphy.296 He identified its ratio as being that unconstitutional statutes are invalid from the beginning and that the finding of invalidity will normally entrain redress.297 Hardiman J properly qualified the refusal of a remedy as exceptional, but refused redress to the deeply unsympathetic applicant.298 His analysis proves that the concerns which dominate Murray CJ’s decision might instead have been met by adopting the Murphy defence.299 Less harm would have been done by forcing open the Murphy escape hatch on a one-off basis than throwing into question the conceptual foundations of the constitutional legal order. At least the defence approach better respects the rule of law by strictly circumscribing and labelling as exceptional the cases in which the void measure seems to be given effect. Fortunately, more recent authorities indicate that The A Case is an aberration, likely due to the exceptionally unsympathetic applicant. DPP v Cunningham300 quashed the conviction of a person convicted of money laundering because the search warrant had been issued under legislation which was subsequently ruled unconstitutional.301 This ruling reiterates that unconstitutional statutes are void ab initio. It demonstrates that Murphy both continues to be good law and applies beyond the restitutionary context, citing a line of constitutional authorities from Byrne 302 to Murphy 303 and A.304 Lastly, it reaffirms that the consequences of actions done under an unconstitutional law will not always be unravelled. CC v Ireland [2006] IESC 33, [2006] 4 IR 1 (SC). Doyle (n 286) 451. 293 The A Case (n 257) 128, 158. 294 ibid 113. 295 R Fanning, ‘Hard Case; Bad Law? The Supreme Court Decision in A v Governor of Arbour Hill Prison’ (2005) 40 Irish Jurist 188, 207– 8; Byrne and Binchy (n 252) 178. 296 The A Case (n 257) 170. 297 ibid 170. 298 ibid 188. 299 S Ring, ‘Collateral Challenges to a Declaration of Unconstitutionality: A v Governor of Arbour Hill Prison’ (2006) 14 Irish Student Law Review 245, 253. 300 Cunningham (n 15). 301 Damache v Director of Public Prosecutions [2012] IESC 11. 302 Frawley (n 267). 303 Murphy (n 1). 304 The A Case (n 257). 291 292

Constitutional Redress in Ireland 301

D. CONCLUSION

Before Murphy, the Irish Courts developed expansive formulations of duress colore officii and mistake of law. This conception of duress takes account of the relationship of the parties,305 and in particular the capacity of the State organ to grant or withhold a benefit,306 in order to demonstrate the involuntary character of a payment. They permitted recovery for mistakes of law where the recipient of the payment was responsible for the plaintiff’s mistake or the parties were not on equal terms,307 so that the parties were not in pari delicto.308 This principle would apply in the vast majority of cases against the State, particularly given the emphasis on the inequality of the parties. The intertwining of the considerations relating to duress and mistake foreshadows the move towards the treatment of claims against the State as a distinctive category of case requiring special rules, and the consequent identification of a policy ground, not based on defective consent, for such claims. Irish law in this area remains dominated by the landmark case of Murphy.309 Counterintuitively, Murphy is an expansive principle, recognising in the first place a prima facie right to recover, before allowing that this right may, in exceptional cases, be restricted for reasons of public policy. The positive interpretation of the Murphy principle was accepted by the Supreme Court in the Health (Amendment) Bill Reference.310 O’Rourke 311 emphasises the narrowness of the Murphy defence, excluding it where the scale of repayments was clearly manageable. Murphy explicitly identified two rationales for the primary right of redress against the State, namely duress colore officii, in an ever more expansive conception, and the inability of an invalid measure to provide justification for the exaction, from which logically flows the State’s duty to return it. This chapter argues that the latter reason is sufficient to explain the outcome in Murphy without recourse to the alternative ground of duress colore officii. This is a public law rationale, which may equally apply in other jurisdictions which subscribe to the principle of legality. Like Murphy, Kingstreet and Woolwich itself point to the concepts of legality and upholding constitutional principles as reasons for a right of restitution from the executive. Other public law reasons for granting restitution are the effective vindication of a substantive constitutional right and the principle against unjust enrichment, which ought properly be known to the public law as well as to private law. A key feature of the Irish Murphy analysis is its formulation of a wider principle, extending beyond the case of receipt of benefits, to favour redress, be it restitutionary or compensatory, in cases of unlawful State actions. If this principle were to become more firmly established in Ireland and extend to other jurisdictions, it would have a dramatic effect for this area of law. Although, it is suggested, this restitutionary claim is constitutional in nature, Kingstreet’s exclusion of private law claims as an alternative cause of action in cases where the public law claim exists is not warranted. In addition, the legality rationale for the constitutional Rogers (n 29). Dolan (n 27). 307 ibid. 308 Rogers (n 29); see Burrows (n 65) 50. 309 Murphy (n 1). 310 Health (Amendment) Bill Reference (n 60). 311 O’Rourke (n 22). 305 306

302 Niamh Connolly right to a remedy does not explain the extension of this claim to all cases involving public bodies as either plaintiffs or defendants. Like the cause of action itself, the Murphy defence, in which a remedy may be denied in the most exceptional of cases, so as to avoid catastrophic consequences which would result if the State offered full redress, is a public law defence rather than change of position. It is a necessary, if rarely used, tool to take account of the occasional need to protect the public exchequer from a catastrophic scale of claim. This defence has no connection to unjust enrichment and is a pure expression of public policy concerns. It should be available to the State to defend restitutionary claims based on any of the available unjust factors. Given that some recent Irish case law on the restitution of overpaid taxes seemingly privileges Woolwich over Murphy,312 Murphy might at first be thought outmoded; progressive for its time but subsequently replaced by an articulation of a principle fully-situated within the modern structures of the law against unjust enrichment. A further challenge was posed to the Murphy principle by the mis-characterisation of its constitutional reasoning in The A Case.313 However, developments outside the strictly-defined territory of restitutionary claims for overpayment of taxes reaffirm Murphy’s continued vitality in Irish law.314 The Murphy torch is carried by the innovative line of authority providing for compensatory, as opposed to restitutionary, damages against the State for acts done under invalid laws. These authorities considering possible constitutional remedies for unlawful State actions are doubly important. Not only do they demonstrate conclusively that the primary right to redress envisaged in Murphy survives the challenges of recent years, they also illustrate public law’s embrace of a remedy analogous to the private law of tort, thus supporting the view that public law also encompasses a restitutionary liability analogous to unjust enrichment.

ibid. The A Case (n 257). 314 Cunningham (n 15). 312 313

14 Restitution of Overpaid Tax in Canada ROBERT CHAMBERS

A. INTRODUCTION

There appears to be a big difference between the ways in which courts deal with overpaid taxes in England and Canada. In Deutsche Morgan Grenfell Group plc v IRC,1 the House of Lords rejected the argument that there are only two ways to recover overpaid tax: either by statute or under the Woolwich2 principle. Taxpayers are entitled to pursue their private law claims for restitution of money paid by mistake of law, which in England is clearly regarded as a form of unjust enrichment. Just three months later in Kingstreet Investment Ltd v New Brunswick,3 the Supreme Court of Canada came to the opposite conclusion, unfortunately without taking notice of Deutsche Morgan Grenfell. The court created a public law of restitution, relying in part on Woolwich. Bastarache J said: In Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] AC 70, the House of Lords has also recognized a right to restitution for payments made pursuant to ultra vires taxes. Without even referring to unjust enrichment, Lord Goff held, at p 172, that restitution was available as a matter of ‘common justice’. 4

He went on to say: Actions for recovery of taxes collected without legal authority and actions of unjust enrichment both address concerns of restitutionary justice, but these remedies developed in our legal system along separate paths for distinct purposes. The action for recovery of taxes is firmly grounded, as a public law remedy in a constitutional principle stemming from democracy’s earliest attempts to circumscribe government’s power within the rule of law.5 So, what can be learned from the Canadian experience? What is the significance of their rejection of unjust enrichment as a basis for restitution of overpaid tax? The short answer to both questions is not as much as one might hope. The main difficulty for the English Deutsche Morgan Grenfell Group plc v Inland Revenue Commissioners [2006] UKHL 49, [2007] 1 AC 558. Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] 1 AC 70 (HL) (hereafter ‘Woolwich’). 3 Kingstreet Investment Ltd v New Brunswick (Department of Finance) 2007 SCC 1, [2007] 1 SCR 3, 276 DLR (4th) 342 (hereafter ‘Kingstreet’, with references to pages and paragraphs in the DLR). 4 ibid 352, [17]. 5 ibid 360–61, [40]. 1 2

304 Robert Chambers observer is that the Canadian conception of unjust enrichment is very different from its English counterpart. An additional difficulty is caused by the way in which law develops in Canada, where judicial precedent does not carry as much weight in its highest court. The Supreme Court of Canada sometimes ignores its own judgments, which means that the law today might not be the law tomorrow. The Canadian law of restitution has not evolved slowly and steadily, but has moved forward (and back) in unexpected jumps. Therefore, one must take care not to place too much weight on a particular judgment, which might not have lasting significance. These two factors mean that, although the Canadian rejection of unjust enrichment and its replacement with a form of public law restitution of ultra vires taxes appear to open a vast gulf between the English and Canadian law on the subject, the gap is much less significant than it seems.

B. WHAT IS UNJUST ENRICHMENT IN CANADA?

In England, unjust enrichment is regarded as a cause of action or perhaps as a category of causes of action, although it is also referred to as a principle or concept. It is an interesting, but perhaps insignificant question whether the various ‘unjust factors’ are separate causes of action within the category of unjust enrichment or merely different ways of establishing an element of the cause of action for unjust enrichment. Either way, it is now well accepted that unjust enrichment occurs when a defendant is enriched at the claimant’s expense, there is some reason why that enrichment is unjust, and there is no defence to the claim for restitution.6 Unjust enrichment is a cause of action in Canada. In Garland v Consumers’ Gas Co, Iacobucci J said: As a general matter, the test for unjust enrichment is well established in Canada. The cause of action has three elements: (1) an enrichment of the defendant; (2) a corresponding deprivation of the plaintiff; and (3) an absence of juristic reason for the enrichment.7

If the plaintiff can prove these three elements, the burden then shifts to the defendant to establish a defence, such as change of position. This appears to be similar to the English version, except for the Canadian switch from unjust factors to absence of basis, which was confirmed in Garland. However, things are not as they appear. In England, ‘the mistaken payment of a non-existent debt’ is regarded as the ‘core case’ of unjust enrichment,8 while in Canada, it appears to stand outside the law of unjust enrichment. Five years after Garland, the Supreme Court of Canada dealt with a case of mistaken payment without citing Garland or even using the words ‘unjust enrichment’. In

6 See Rowe v Vale of White Horse District Council [2003] EWHC 388 (Admin), [2003] 1 Lloyd’s Rep 418, [11]; Greater Manchester Police v Wigan Athletic AFC Ltd [2008] EWCA Civ 1449, [2009] 1 WLR 1580, [38], [62]; Benedetti v Sawiris [2010] EWCA Civ 1427, [104]; Investment Trust Companies (in liq) v HM Revenue & Customs [2012] EWHC 458 (Ch) [38]. 7 Garland v Consumers’ Gas Co 2004 SCC 25, 237 DLR (4th) 385, 397, [30] (hereafter ‘Garland’, with references to pages and paragraphs in the DLR). 8 P Birks, Unjust Enrichment, 2nd edn (Oxford, OUP, 2005) 3.

Restitution of Overpaid Tax in Canada 305

BMP Global Distribution Inc v Bank of Nova Scotia,9 Deschamps J relied instead on Barclays Bank Ltd v WJ Simms Son & Cooke (Southern) Ltd,10 as authority for the plaintiff’s right to restitution of money paid under mistake of fact. It was surprising that the court offered no explanation why the ‘absence of juristic reason’ approach did not apply, but it was plain that it did not. Garland provided a detailed procedure for dealing with claims of unjust enrichment, which is very different from the test set out in Simms, and adopted in BMP, for dealing with mistaken payments. This requires us to re-evaluate the categories of causes of action listed by Bastarache J in Kingstreet: There are at least two distinct categories of restitution: (1) restitution for wrongdoing; and (2) restitution for unjust enrichment . . . Restitution for ultra vires taxes does not fit squarely within either of the established categories of restitution. The better view is that it comprises a third category distinct from unjust enrichment.11

There appear now to be at least four categories of reasons for restitution in Canada: wrongdoing, unjust enrichment, ultra vires taxes, and mistaken payments.12 In England, the first of these categories is very different from the other three, which all require an enrichment at the claimant’s expense. For wrongdoing, the breach of duty replaces the need for any expense to the claimant. The same is true in Canada.13 What distinguishes the other three categories from each other lies primarily in what the plaintiff must do to establish a reason for restitution. According to Garland, to prove unjust enrichment, the plaintiff must show that no juristic reason from an established category exists to deny recovery . . . The established categories that can constitute juristic reasons include a contract, a disposition of law, a donative intent, and other valid common law, equitable or statutory obligations. If there is no juristic reason from an established category, then the plaintiff has made out a prima facie case under the juristic reason component of the analysis.14

When it comes to taxes, ‘a citizen who has made a payment pursuant to ultra vires legislation has a right to restitution’.15 This is not much different from showing that there is no juristic reason for the payment, except that, as discussed below, it is not enough to show there was no statutory obligation to pay the taxes. The plaintiff must also show that they were collected under an ultra vires statute.16 For mistaken payments, the plaintiff must show that the mistake caused the payment, but the claim will yet fail if the defendant can show that it was ‘made for good consideration’.17 In other words, the defendant bears the onus of establishing a juristic reason for the payment notwithstanding the mistake. This is not all that different from the second stage in a claim for unjust enrichment, as set out in Garland. If the plaintiff can 9 BMP Global Distribution Inc v Bank of Nova Scotia 2009 SCC 15, [2009] 1 SCR 504, 304 DLR (4th) 292 (hereafter ‘BMP’, with references to pages and paragraphs in the DLR) [21]–[22]. 10 Barclays Bank Ltd v WJ Simms Son & Cooke (Southern) Ltd [1980] QB 677 (Comm Ct) (hereafter ‘Simms’). 11 Kingstreet (n 3) 358–59, [33], [40]. 12 Also see Bell Mobility Inc v Anderson, 2012 NWTCA 4, [9]–[10], where the Northwest Territories Court of Appeal said that ‘many of the common law causes of action encompassed by the category of “restitution” can be notionally considered together because they rest on the same general principle of “unjust enrichment” ’. 13 Soulos v Korkontzilas [1997] 2 SCR 217, 146 DLR (4th) 214, 229–30, [43]. 14 Garland (n 7) 402, [44] (Iacobucci J). 15 Kingstreet (n 3) 351, [15] (Bastarache J). 16 Alberta v Elder Advocates of Alberta Society 2011 SCC 24, [2011] 2 SCR 261, [91]. 17 Simms (n 10) 695 (Robert Goff J), quoted in BMP (n 9) [22].

306 Robert Chambers negate all the established categories of juristic reasons, ‘there is a de facto burden of proof placed on the defendant to show the reason why the enrichment should be retained’.18 From an English perspective, ultra vires taxes and mistaken payments are sub-sets of a larger category of unjust enrichment. It would, however, be wrong to conclude that the Canadian concept of unjust enrichment is merely smaller than the English one. It is not smaller, but different. The vast majority of cases of unjust enrichment in Canada concern the division of the family home at the end of a marriage or similar relationship. It was in this context that the test for unjust enrichment set out in Garland was first laid down by the Supreme Court of Canada in Pettkus v Becker.19 Unjust enrichment is used as the justification for declaring a constructive trust of the family home even where the enrichment consists of services that are not connected to the acquisition or improvement of the home.20 Similar outcomes are achieved in family property cases in other common law jurisdictions, including England, but not on the basis of unjust enrichment. Outside Canada, constructive trusts arise to fulfil common intentions or reasonable expectations regarding ownership of the family home.21 Reasonable expectations are also important in Canada, but working under the guise of unjust enrichment as the absence of juristic reason. Dickson J said in Pettkus v Becker: As for the third requirement, I hold that where one person in a relationship tantamount to spousal prejudices herself in the reasonable expectation of receiving an interest in property and the other person in the relationship freely accepts benefits conferred by the first person in circumstances where he knows or ought to have known of that reasonable expectation, it would be unjust to allow the recipient of the benefit to retain it.22

While unjust enrichment may well be present in these cases, the declaration of trust does not effect restitution of that enrichment, or at least is not restricted to that role. It fulfils expectations. In Sorochan v Sorochan, Dickson J said that in assessing whether a constructive trust remedy is appropriate, we must direct our minds to the specific question of whether the claimant reasonably expected to receive an actual interest in property and whether the respondent was or reasonably ought to have been cognizant of that expectation.23

As Dr Elias pointed out, this trust is ‘perfectionary’, not restitutionary.24 This does not mean that unjust enrichment in Canada is wholly different from its English counterpart. Garland, for example, concerned unlawfully collected penalties for late payments, which would fall well within the English concept of unjust enrichment.25 However, Garland (n 7) 402, [45] (Iacobucci J). Pettkus v Becker [1980] 2 SCR 834, 117 DLR (3d) 257, 273–74 (hereafter ‘Pettkus’, with references to pages in the DLR). For the Garland test see text accompanying n 7. 20 Peter v Beblow [1993] 1 SCR 980, 101 DLR (4th) 621. 21 See J Mee, The Property Rights of Cohabitees (Oxford, Hart Publishing, 1999); Law Commission, Cohabitation: The Financial Consequences of Relationship Breakdown – A Consultation Paper (LCCP no 179, 2006) 154–61, [A.23]–[A.45], cf also the ensuing Report on Cohabitation: The Financial Consequences of Relationship Breakdown (Law Com no 307, 2007); Jones v Kernott [2011] UKSC 53, [2011] 3 WLR 1121, [24]. 22 Pettkus (n 19) 274. 23 Sorochan v Sorochan [1986] 2 SCR 35, 50, 29 DLR (4th) 1, 12. 24 G Elias, Explaining Constructive Trusts (Oxford, Clarendon Press, 1990) 157; see R Chambers, ‘Constructive Trusts in Canada’ (1999) 37 Alberta Law Review 173, 201–7, reprinted in (2001) 15 Trust Law International 214, (2002) 16 Trust Law International 2. 25 The actual outcome of Garland bore little resemblance to restitution of unjust enrichment. The $22m settlement of the class action was distributed as follows: $2m for potential plaintiffs, $9m to charity to help the poor with heating bills, and the remaining $11m for legal fees and disbursements: Garland v Enbridge Gas Distribution Inc 2006 CanLII 41291 (Ont SCJ). 18 19

Restitution of Overpaid Tax in Canada 307

a search for ‘unjust enrichment’ in a database of Canadian cases shows that its natural habitat in that country is the family home. So, while it appears that ultra vires taxes and mistaken payments have been moved out of the realm of unjust enrichment in Canada, it would be more accurate to say that unjust enrichment has moved over the last 30 years, leaving the other two categories behind.

C. WHY WAS UNJUST ENRICHMENT REJECTED IN KINGSTREET?

One might think that an ‘absence of basis’ approach to unjust enrichment would be ideally suited to the recovery of overpaid taxes. It would obviate the need to rely on fundamental constitutional principles, deemed mistakes of law, or implied compulsion to achieve the result that seems obvious to the layperson asking the simple question: was the tax due? One might also think that Kingstreet provided the Supreme Court of Canada with the perfect opportunity to put the Garland test for unjust enrichment to good use. So, why did Bastarache J say repeatedly in Kingstreet that it is ‘inappropriate’ in this context? The answer, perhaps surprisingly, is that it is too complex. In cases involving ultra vires taxes, ‘the unjust enrichment framework adds an unnecessary layer of complexity to the real legal issues’.26 In Canada, the law of unjust enrichment is used to solve a wide variety of legal problems, from the division of family assets to class actions by consumers. In Garland, Iacobucci J described unjust enrichment as ‘an equitable remedy that will necessarily involve discretion and questions of fairness’.27 When considering whether the defendant can establish a juristic reason for retaining the enrichment, ‘courts should have regard to two factors: the reasonable expectations of the parties, and public policy considerations’.28 They must then ‘weigh the equities in the particular circumstances of each case’.29 This is inappropriate in cases where taxpayers are entitled to restitution of ultra vires taxes as of right. The Canadian rejection of unjust enrichment as a basis for restitution of ultra vires taxes is based primarily on problems with that body of law as it has developed in Canada. These are not problems in England where, as Lord Goff said in Lipkin Gorman v Karpnale Ltd: The recovery of money in restitution is not, as a general rule, a matter of discretion for the court. A claim to recover money at common law is made as a matter of right; and even though the underlying principle of recovery is the principle of unjust enrichment, nevertheless, where recovery is denied, it is denied on the basis of legal principle.30

While the arguments in Kingstreet against unjust enrichment have no application in England, perhaps something can be learned from the arguments made in favour of a public law of restitution. Bastarache J said that ‘a remedy as a matter of constitutional right . . . is in fact the only appropriate remedy because it raises important constitutional principles Kingstreet (n 3) 358, [35]. Garland (n 7) 401, [44]. 28 ibid 402, [46]. 29 Kingstreet (n 3) 359, [36] (Bastarache J). 30 Lipkin Gorman (a firm) v Karpnale Ltd [1988] UKHL 12, [1991] 2 AC 548, 578. But see Foskett v McKeown [2000] UKHL 29, [2001] 1 AC 102, [127], where Lord Millett said that the law of property is different from the law of unjust enrichment because: ‘Property rights are determined by fixed rules and settled principles. They are not discretionary. They do not depend upon ideas of what is “fair, just and reasonable”’. 26 27

308 Robert Chambers which would be ignored by treating the claim under another category of restitution’.31 There may be symbolic significance to a public vindication of the plaintiff’s constitutional rights (even if, as in Kingstreet, it is only a company’s right not to pay indirect sales taxes to a provincial government). However, other than the constitutional question that must be answered at the start of the proceedings, it is difficult to see what else might be special about the public law aspects of the claim.32 Bastarache J rejected the public law concerns of the majority of the Supreme Court of Canada in Air Canada v British Columbia33 that restitution of ultra vires taxes might lead to fiscal chaos and inefficiency. He also rejected the defence of passing on, although it had ‘developed almost exclusively in the context of recovery of taxes’.34 The normal six-year limitation period for all restitution claims applied. So even though the province had exacted taxes from the plaintiff by exceeding the constitutional limits of its taxing powers, it was entitled to rely on its own limitation statute to retain them. The constitutional right to restitution does obviate the need to consider whether a taxpayer was mistaken or under compulsion,35 but that should also be true of a claim based on unjust enrichment. Where money is paid under a contract or as a gift, the plaintiff may need to prove mistake, compulsion, or undue influence to invalidate the contract or gift as a juristic reason for the payment, but the right to restitution then follows from the absence of juristic reason. For ultra vires taxes, the invalidity of the taxing statute is unrelated to the taxpayer’s state of mind, which should therefore be wholly irrelevant even if recovery was based on unjust enrichment. Unfortunately, Garland makes these issues relevant to the second stage of the inquiry, regarding the reasonable expectations of the parties. In Garland, a utility company collected late payment penalties which were authorised by a provincial energy board pursuant to a provincial statute. However, the penalties could produce an effective rate of interest that exceeded 60 per cent per annum, which is a criminal rate of interest under paramount federal legislation. There was no juristic reason for the amounts collected in excess of the 60 per cent limit, but the company was entitled to retain the money collected while it was unaware of the invalidity of the provincial authority. This is because the company reasonably expected that the order of the energy board was valid and its customers reasonably expected to pay the penalty.36 In other words, the mistaken belief in the validity of the energy board order provided a juristic reason for the payments. The Garland test for unjust enrichment produces a strange inversion of the English approach, since the plaintiff’s mistake is no longer a ground of recovery but may be a reason for denying it. The defendant’s mistake is relevant in both countries for the defence of change of position, which is not normally available to a defendant who depletes the enrichment after becoming aware it is unjust. In Garland, the company could not ‘avail itself of the defence because it is not an “innocent” defendant given that the benefit was received as a result of a Criminal Code violation’.37 Nevertheless, it could retain the enrichment it expected to receive despite its wrongdoing and regardless of whether it remained enriched. As Professor McInnes said, ‘[c]rime evidently may pay as long as the perpetrator believes it Kingstreet (n 3) 358, [34]. See R Williams, Unjust Enrichment and Public Law (Oxford, Hart Publishing, 2010) 32–35. 33 Air Canada v British Columbia (Attorney General) [1989] 1 SCR 1161, 59 DLR (4th) 161. 34 Kingstreet (n 3) 362, [45]. 35 ibid 360, [39] and 365, [53]. 36 Garland (n 7) 405–7, [55]–[61]. 37 ibid 409, [66] (Iacobucci J).

31 32

Restitution of Overpaid Tax in Canada 309

is acting lawfully’.38 This convoluted approach and odd outcome cannot be reconciled with recovery as of right and was no doubt a major factor in the rejection of unjust enrichment in Kingstreet.

D. PUBLIC LAW OF RESTITUTION IN CANADA

Kingstreet Investment Ltd v New Brunswick appears to create a separate, public law of restitution of overpaid taxes in Canada. However, the change is not as significant as it seems, for two reasons. First, it applies only when taxes are collected pursuant to an ultra vires statute. Secondly, the relief granted in Kingstreet is not a special public law remedy, but appears (at least to an English private lawyer) to be a perfectly ordinary order for restitution of unjust enrichment. If taxes are overpaid by mistake or collected improperly under a valid taxing statute which provides a mechanism for obtaining a refund, that mechanism displaces any common law right to recovery, including the principle laid down in Kingstreet. For example, the Excise Tax Act states: Except as specifically provided in this Part, the Customs Act or the Financial Administration Act, no person has a right to recover any money paid to Her Majesty as or on account of, or that has been taken into account by Her Majesty as, tax, net tax, penalty, interest or any other amount under this Part.39

In Sorbara v Canada (Attorney General), the Ontario Court of Appeal held that the Act ‘provides a complete statutory framework with respect to a taxpayer’s claim for a rebate of GST’ (Goods and Services Tax) and that the ‘constitutional cause of action’ in Kingstreet does not apply when a taxpayer ‘claims a right to recover tax assessed under a misapplication or misinterpretation of a taxing statute’.40 The Federal Court of Appeal came to the same conclusion in Merchant Law Group v Canada Revenue Agency, where Stratas JA said: [I]n Kingstreet, the Supreme Court did not create a new, sweeping constitutional remedy to recover tax assessed under a misapplication or misinterpretation of a taxing statute. It certainly did not create a new, sweeping constitutional remedy that would allow aggrieved taxpayers to bypass all of the legislative schemes in force across the country that govern the recovery of tax assessed under a misapplication or misinterpretation of a taxation statute.41

Even in the absence of a statutory mechanism for the recovery of taxes, the Kingstreet principle will not apply unless the case concerns an ultra vires statute. The test for unjust enrichment set out in Garland is used in other contexts. In Alberta v Elder Advocates of Alberta Society,42 the Alberta government applied to strike out a class action for restitution of accommodation charges levied for elderly patients in 38 M McInnes, ‘Unjust Enrichment, Juristic Reasons and Palm Tree Justice: Garland v Consumers’ Gas Co’ (2004) 41 Canadian Business Law Journal 103, 114. 39 RSC 1985, c E-15, s 312. 40 Sorbara v Canada (Attorney General) 2009 ONCA 506, 309 DLR (4th) 162, [9] and [5]; leave to appeal refused: 2009 CanLII 61389 (SCC). 41 Merchant Law Group v Canada Revenue Agency 2010 FCA 184, [21]. 42 Alberta v Elder Advocates of Alberta Society 2011 SCC 24, [2011] 2 SCR 261.

310 Robert Chambers nursing homes. It was alleged that the charges were excessive and therefore unlawfully discriminated against the patients on the basis of their age or disability, contrary to section 15 of the Canadian Charter of Rights and Freedoms. If true, this would not affect the validity of any statute, but could invalidate the provincial regulation that set the maximum accommodation charges allowed. The Supreme Court of Canada held that Kingstreet did not stop the plaintiffs from proceeding with their claims for restitution of unjust enrichment. McLachlin CJ said: In my view, Kingstreet stands for the proposition that public law remedies, rather than unjust enrichment, are the proper route for claims relating restitution of taxes levied under an ultra vires statute, on the ground that the framework of unjust enrichment is ill-suited to dealing with issues raised by a claim that a measure is ultra vires. However, Kingstreet leaves open the possibility of suing for unjust enrichment in other circumstances. The claim pleaded in this case is not for taxes paid under an ultra vires statute. It is not therefore precluded by this Court’s decisions in Kingstreet. The pleading should be allowed to go to trial, at which point the propriety of the claim for unjust enrichment may be explored more fully in the context of the evidence adduced.43

This contrast between public law remedies in Kingstreet and unjust enrichment in Alberta v Elder Advocates of Alberta Society is difficult to understand. The central question in the Elder Advocates case was whether the Alberta government had violated the constitutional equality rights of the elderly patients in its care. This was surely a public law matter if ever there was one. Conversely, the remedy granted in Kingstreet was a simple order for the repayment of a sum of money received, plus interest and subject to the normal six-year limitation period applicable to all claims for restitution. There was nothing especially public about it. In contrast, the issues of fairness, reasonable expectations, and public policy, which were used to determine the measure of relief available in Garland, seem to belong naturally in a public law context. It is unsatisfactory to classify the recovery of taxes as either public law or unjust enrichment depending on the particular reason why the taxes are not due. Far better is the ‘hybrid approach’ advocated by Dr Rebecca Williams, in which public law explains why the tax is not due, and unjust enrichment provides the mechanism for recovery: [T]he reason for restitution, the ‘unjust factor’, is provided by the existence of a public law ultra vires event, leaving the definition of enrichment, the claimant’s expense and relevant defences to be worked out by the law of unjust enrichment in the usual manner.44

For Dr Williams, ‘a public law ultra vires event’ is not confined to cases involving ultra vires statutes, but is used in a broader sense ‘simply to indicate that there has been an event which could give rise to a successful action for judicial review, an instance of public law illegality of some kind’.45 Therefore, Garland, Kingstreet, and Alberta Elders would all be amenable to her analysis, and so would any claim for the recovery of taxes that is not subject to an exclusive statutory regime. Prior to Kingstreet, we can find cases approaching the problem in precisely this way.

43 ibid [91]. This was followed in Sivia v British Columbia (Superintendent of Motor Vehicles) 2012 BCSC 1030, [96]–[97], where the British Columbia Supreme Court held that Kingstreet principle applied only to taxes and not to regulatory charges. 44 Williams, Unjust Enrichment and Public Law (n 32) 38. See also Rebecca Williams’ contribution in chapter 2 of the present volume. 45 Williams, Unjust Enrichment and Public Law (n 32) 12.

Restitution of Overpaid Tax in Canada 311

For example, in Momi v Canada (Minister of Citizenship and Immigration),46 the plaintiffs sought certification of a class action for the recovery of fees collected from applicants for permanent residence in Canada. They alleged that the government had been unjustly enriched because the fees exceeded the limits set by the Canadian Financial Administration Act.47 The motion was dismissed because the fees had been levied in accordance with regulations in force under the Immigration and Refugee Protection Act,48 and the plaintiffs had failed to seek judicial review of those regulations in time. In other words, it was necessary to establish in public law that the fees were unauthorised before it could be said they were unjust for the purposes of the law of unjust enrichment. Dr Williams sees the role of public law in this context as providing the ‘unjust factor’ needed for the law of unjust enrichment. Alternatively, its contribution could be seen as the means to invalidate the legal justification for the tax and thus negate any juristic reason for the enrichment. The latter approach would be consistent with the development of the law of unjust enrichment in Canada and may one day be the direction taken in England. The law of unjust enrichment is not concerned with providing the various justifications for enrichment, but gets involved only when they are missing. It is for public law to determine whether a particular statute, regulation, or government act is valid, just as it is contract law which governs the validity of contracts, and the law of trusts, gifts, or wills which governs the validity of those transactions. Unjust enrichment can be seen as a relatively small body of law designed to help clean up after failures in other areas. As Professor Robert Stevens has said: In one sense, the law of unjust enrichment is different from, say, the law of contract, as what links the subject together is a lack, an absence or a negative, rather than a positive reason for recovery. It is for this reason that there was until very recently, hardly any attempt to theorise about the basis of the subject. There is not very much to theorise about.49

This simple, hybrid approach to the problem of restitution of overpaid tax would work well in Canada but for the problems created by Garland with its attempt to construct a modern law of unjust enrichment out of a body of law designed primarily for the equitable division of the family home. As a result, Canadian courts have struggled to find other solutions to deal with what should have been light work for a coherent law of unjust enrichment based on absence of juristic reasons.

E. CONCLUSION

We sometimes look to the law of another jurisdiction hoping to find a better solution to a common legal problem. Unfortunately, the Canadian experience with overpaid tax has little to offer the English lawyer except perhaps a valuable lesson in what not to do. The 46 Momi v Canada (Minister of Citizenship and Immigration) 2006 FC 738, 55 Imm LR (3d) 133, [2007] 2 FCR 291 (Fed Ct). 47 cf RSC 1985, c F-11, s 19(2): ‘Fees and charges for a service . . . provided by or on behalf of Her Majesty in right of Canada . . . may not exceed the cost to Her Majesty in right of Canada of providing the service . . . to the users or class of users’. 48 cf SC 2001, c 27. 49 R Stevens, ‘Is there a Law of Unjust Enrichment?’ in S Degeling and J Edelman (eds), Unjust Enrichment in Commercial Law (Sydney, Lawbook Co, 2008) 11, 33.

312 Robert Chambers separation of the cases into wholly separate bodies of law, depending on the reason why the tax is invalid, is difficult to understand. Why is the distribution of taxing powers in a federal system too important a constitutional principle to be left to the law of unjust enrichment, when the constitutional protection of the aged and infirm is not? Why is a straightforward claim for restitution not available in every case in which it turns out that the tax is invalid (unless barred or limited by statute)? The recovery of an invalidly collected tax should not be classified as either public law or unjust enrichment. In reality it is both. The validity of the tax is purely a public law question, and the right to recover it is (or at least ought to be) a simple matter of restitution of unjust enrichment. The law of unjust enrichment has nothing whatsoever to say about the public law question, but is wholly dependent on the answer. If the tax is not due, then unjust enrichment can provide an effective solution to the problem in the absence of a statutory regime of recovery. The real problem in Canada is the current state of its law of unjust enrichment. The added elements of equity, fairness, reasonable expectations, and public policy have proved useful in a family law context, but have produced a complex mixture that is unfit for what should be its primary purpose. It now has difficulty coping with a simple matter of money paid without legal basis. The decision in Kingstreet to abandon unjust enrichment in favour of a new form of public law restitution was perhaps a sign of despair, but at least it provided under a different name the solution that should have been obtained easily through a proper law of unjust enrichment. The decision in Alberta Elders to confine that solution to ultra vires statutes has not helped matters. The extent of the problem is revealed by two decisions of the British Columbia Supreme Court in which taxpayers disputing the validity of municipal taxing bylaws were allowed to pay the tax demanded into court because the law in this area was unclear.50 Since it was uncertain whether Kingstreet applied, there was a risk that the tax could not be recovered if paid to the municipality and therefore the taxpayers might suffer irreparable harm if they did. As a final note, the problems in Canada should not be blamed on the shift from unjust factors to absence of juristic reasons. As the English experience shows, unjust factors do not provide easy answers in this area. There would have been no need to invent the Woolwich principle or deal with the complexities of deemed mistakes of law if the right to restitution depended on the answer to the simple question whether the tax was due.

50 TimberWest Forest Corp v Campbell River (City) 2009 BCSC 1862 (Slade J); Port Alberni (City) v Catalyst Paper Corp 2010 BCSC 402 (Voith J).

15 Restitution of Unlawfully Exacted Tax in Australia: The Woolwich Principle SIMONE DEGELING*

A. INTRODUCTION

In Woolwich Lord Goff held, as is well known, that ‘money paid by a citizen to a public authority in the form of taxes or other levies paid pursuant to an ultra vires demand by the authority is prima facie recoverable by the citizen as of right’.1 Lord Goff supported this ‘simple call of justice’2 in part by reference to the Bill of Rights 1688 and the legal reality that when a public authority makes a demand for payment, ‘that demand is implicitly backed by the coercive powers of the state and may well entail . . . unpleasant economic and social consequences if the taxpayer does not pay’.3 Article 4 of the Bill of Rights 1689 states: [L]evying of money for or to the use of the Crown, by pretence of prerogative, without grant of Parliament, for longer time, or in other manner than the same is or shall be granted is illegal.4

In Lord Goff’s judgment, a right to restitution was required in order to give ‘full effect’ to that ‘fundamental’ principle.5 The common law thus developed to include a policymotivated unjust enrichment claim which was informed by the relevant constitutional arrangements.6 The limited purpose of this chapter is to consider the status of the so-called Woolwich principle in Australian law.7 Despite slim Victorian authority apparently applying the * I am grateful to Mark Aronson, Jeremy Kirk SC, Andrew Lynch and Keith Mason AC QC for their comments and for discussing some of the ideas in this chapter with me. All errors are of course my own. I am also indebted to Miriam Orski and Magdy Morcos from UNSW Law Library for their assistance. 1 Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] AC 70 (HL) 177 (hereafter ‘Woolwich’). 2 ibid 172. 3 ibid. Confirmed by Test Claimants in the FII Group Litigation v HMRC [2012] UKSC 19, [2012] 2 AC 337. As is discussed elsewhere in this volume, the formal requirement that there be an ultra vires ‘demand’ has now been dispensed with, in favour of an unlawful ‘exaction’. 4 Passage extracted from Glidewell LJ’s judgment in Woolwich Equitable Building Society v Inland Revenue Commissioners [1992] AC 70 (CA) 79. 5 Woolwich (n 1) 172. 6 In answer to the objection that recognising the right to restitution was tantamount to judicial legislation, Lord Goff replied: ‘[A]lthough I am well aware of the existence of the boundary, I am never quite sure where to find it. Its position seems to vary from case to case’: ibid 173. 7 See generally D Wong, ‘The High Court and the Woolwich principle: Adoption or Another Bullet that Cannot be Bitten?’ (2011) 85 Australian Law Journal 597; K Mason, J Carter and G Tolhurst, Mason and Carter’s Restitution Law in Australia, 2nd edn (Sydney, LexisNexis Butterworths, 2008) ch 20; J Edelman and E Bant, Unjust Enrichment

314 Simone Degeling House of Lords decision in Woolwich and allowing restitution of an increase in rent pursuant to a demand rendered void by statute (albeit in a case where it was not clear that the recipient, a housing association, was in any sense a public body),8 there is as yet no formal reception of the Woolwich principle in Australian law. There have been judicial statements which seem to assume the principle already operates in Australia,9 is likely to be adopted10 or that there is no barrier to adoption.11 However, as a matter of legal technicality, the principle has yet formally to be applied. At the outset, it should be stated that the normative question of whether Australian law should adopt such a rule is not directly addressed in this chapter. There has been some scholarly debate on the issue,12 and whilst the constitutional arrangements of the Commonwealth of Australia are obviously different to those of the United Kingdom, the author’s position is that Lord Goff’s observations about the coercive powers of the state and the attendant policy considerations have equal force in the Australian context. As recently emphasised by Crennan J in Williams v Commonwealth of Australia,13 the Commonwealth’s capacities are ‘capable of being utilised to regulate activity in the community in the course of implementing government policy’ including, relevantly, the power to impose coercive measures, attach penal consequences and the power to convict and impose punishment. Such considerations ‘highlight . . . the importance of . . . mechanisms for responsible government designed to protect the community from arbitrary government action’.14 A Woolwich-type claim would be an important bulwark against the improper use of state power in exacting tax and other dues. The discussion which follows therefore examines the possible routes for recovery of unlawfully exacted tax in the Australian context.

B. WOOLWICH AND POLICY-MOTIVATED CLAIMS

Woolwich and its sequelae are examples of policy-motivated claims: unjust enrichment claims where the unjust factor is not a function of the plaintiff’s intention to transfer wealth but rather down to a specific legal policy. Another example of a policy-motivated in Australia (Melbourne, OUP, 2006) 297–98; P Butler, ‘Restitution of Overpaid Taxes, Windfall Gains and Unjust Enrichment’ (1995) 18 University of Queensland Law Journal 318; T Voon, ‘Restitution from Government in Australia: Woolwich and its Necessary Boundaries’ (1998) 9 Public Law Review 15. 8 Jeffrey v Fitzroy Collingwood Rental Housing Association Ltd [1999] VSC 335, [44] (Harper J). 9 Lamesa Holding BV v Commissioner of Taxation [1999] FCA 612, [88]. In Roxborough v Rothmans of Pall Mall Australia Ltd [1999] FCA 1535, [30] and [59], Lehane and Hill JJ seemed to assume that Woolwich already formed part of the law of Australia, but on the facts of Roxborough it did not apply because the case concerned a claim between two non-state actors who were parties to a contract. 10 Chippendale Printing Co Pty Limited v The Commonwealth of Australia and the Commissioner of Taxation [1996] FCA 1259, [20] (Lehane J). Emmett J at first instance in Roxborough v Rothmans of Pall Mall Australia Ltd (1999) 161 ALR 253, [1999] FCA 107, [53], stated ‘[t]he principle in the Woolwich case may have been applicable if a claim was made against the State of New South Wales for recovery of licence fees paid. That, of course, is not this case’. Yet in the earlier decision of Hollis v Atherton Shire Council [2003] QSC 147, [45], Jones J had stated: ‘This so called “Woolwich principle” has not yet been adopted into Australian law’. 11 State Bank of New South Wales Ltd v Commissioner of Taxation of Commonwealth of Australia [1995] FCA 1652, [16] (Wilcox J); ACI Operations Pty Ltd v Commonwealth of Australia [1995] FCA 1770, [20] (Wilcox J). 12 See esp Wong (n 7); Mason and Carters’ Restitution Law in Australia (n 7) ch 20. 13 Williams v Commonwealth of Australia [2012] HCA 23 (hereafter ‘Williams’) [521]. 14 ibid [521]. See also [152] (Gummow and Bell JJ).

The Woolwich Principle in Australia 315

claim is that which arguably exists to recover the value of benefits conferred in a situation of necessity, in order to encourage interventions in aid of the property or health of another.15 The policy-motivated category is by definition a category of fragments. There is little to unite the members of the category other than that they are unjust factors which are not about the plaintiff’s intention. These specific legal policies are unconnected and beyond this very basic characteristic are unconnectable. The category is systematically unstable and this instability is visible at various levels of analysis.16 If we deal simply at the level of the unjust enrichment architecture, the liability model, we know that that there is no attempt to theorise at a general level about what specific legal policies might or might not be protected or vindicated via a claim in unjust enrichment. The category is simply a miscellaneous list.17 Even on the higher normative plane, the policy-motivated category is unstable. There is no coherent unifying explanation to account for its contents.18 Noticing the instability in the policy-motivated category is important. It will be argued that there are two possible routes in Australian law for the development of a Woolwich claim. The first is a common law unjust enrichment claim, parasitic on some constitutional invalidity. If this path is pursued, the indeterminacy of the policy-motivated category as a whole is once again in view. The effect would be to add an additional claim to the policymotivated category. However, the primacy of the Constitution in the Australian legal system to an extent mitigates against this instability. We should not be surprised to see the common law develop to protect and vindicate constitutional principles. The second possibility is a direct constitutional right of action. If this path is adopted, the problem of ultra vires demands is taken out of unjust enrichment proper. The work would be allocated to a department of constitutional law, and any restitutionary response would be triggered not by unjust enrichment but, in the language of Birksian events and responses,19 by some other (constitutional) event.20 In other words, we would have restitution, but not in response to unjust enrichment. In considering the existence of a Woolwich-style claim in Australian law, it is necessary to understand the relationship between the Constitution and the common law. This was articulated in Lange v Australian Broadcasting Corporation.21 The prime minister of New Zealand, David Lange, was the subject of an Australian Broadcasting Corporation (ABC) 15 See generally A Burrows, The Law of Restitution, 3rd edn (Oxford, OUP, 2011) ch 18; F Rose, ‘Restitution for the Rescuer’ (1989) 9 Oxford Journal of Legal Studies 167. 16 See generally S Degeling, ‘Understanding Policy Motivated Unjust Factors’ in C Rickett and R Grantham (eds), Structure and Justification in Private Law: Essays for Peter Birks (Oxford, Hart Publishing, 2008) ch 14. 17 P Birks, Unjust Enrichment, 2nd edn (Oxford, OUP, 2005) 107. 18 There has been much work done to demonstrate that a theory of corrective justice might be used to give a normative account of unjust enrichment. Professor Weinrib relies on Aristotle’s conception of corrective justice to argue that unjust enrichment restores the parties to their pre-transactional equality, drawing on Kantian right to elucidate the parties pre-transactional equality in the sense that they are both free-willed self-determining agents: see E Weinrib, The Idea of Private Law (Cambridge MA, Harvard University Press, 1995); E Weinrib, ‘Correctively Unjust Enrichment’ in R Chambers, C Mitchell and J Penner (eds), Philosophical Foundations of the Law of Unjust Enrichment (Oxford, OUP, 2009) ch 2; L Smith, ‘Restitution: The Heart of Corrective Justice’ (2001) 79 Texas Law Review 2115. However, as Professor Smith has shown, in the policy-motivated category it is not possible to use Kantian notions of right to inform the corrective justice analysis in the way suggested by Weinrib. Non-Kantian norms have a role to play. For example, Woolwich itself recognises the violation of a norm of public law, and other policy-motivated claims are similarly explicable as violations of non-Kantian norms: see Smith (above) 2145. See also C Mitchell, The Law of Contribution and Reimbursement (Oxford, OUP, 2003). 19 P Birks, ‘The Law of Restitution at the End of an Epoch’ (1999) 28 Western Australian Law Review 13. 20 cf Kingstreet Investments Ltd v New Brunswick (Department of Finance) 2007 SCC 1, [40]. 21 Lange v Australian Broadcasting Corporation (1997) 189 CLR 520 (hereafter ‘Lange’). See also Pape v Commissioner of Taxation of the Commonwealth of Australia (2009) 238 CLR 1.

316 Simone Degeling television programme which dealt with his performance, conduct and fitness for office. He commenced defamation proceedings against the ABC in New South Wales and in the High Court ‘argument . . . was conducted on the footing that the plaintiff’s action was to be determined solely by regard to the defamation laws of [NSW]’.22 The ABC pleaded various defences including an assertion that the defamatory matter was published pursuant to a freedom guaranteed by the Constitution of the Commonwealth. It therefore became necessary to consider the existence of an implied freedom of communication and its relationship to the common law, specifically defamation. Drawing on the text and structure of the Constitution,23 the High Court held the system of representative and responsible government provided for in the Constitution requires freedom of communication on matters concerning government or political matters.24 As is discussed below, the Constitution has not been interpreted to confer private rights and as such contained no ‘express right of freedom of communication or expression . . . [C]ommunications are free only to the extent that they are left unburdened by laws that comply with the Constitution’.25 The High Court therefore had to consider whether the law of New South Wales infringed the requirement of freedom of communication imposed by the Constitution. It was in this context that the Court considered the relationship between the common law and the Constitution. In the Court’s unanimous judgment: Of necessity, the common law must conform with the Constitution. The development of the common law in Australia cannot run counter to constitutional imperatives. The common law and the requirements of the Constitution cannot be at odds . . . [T]he common law rights of persons defamed may be diminished by statute but they cannot be enlarged so as to restrict the freedom required by the Constitution. Statutes which purport to define the law of defamation are construed, if possible, conformably with the Constitution. But, if their provisions are intractably inconsistent with the Constitution, they must yield to the constitutional norm . . . The common law may be developed to confer a head or heads of privilege in terms broader than those which conform to the constitutionally required freedom, but those terms cannot be any narrower.26

Against this background, as has been said, there are at least two possible paths for the development of a Woolwich-style claim. The first is the recognition of a common law claim which is parasitic on some constitutional invalidity. The second is a more difficult and controversial possibility, which is that the source of the right to restitution arises under the Constitution itself. Each will be considered in turn.

(1) Parasitic Common Law Claim The case which could have been the vehicle for a proper examination of the issue was British American Tobacco Australia Ltd v The State of Western Australia.27 This case was the Lange (n 21) 569. ibid 566–67. 24 ibid 567. Specific reference was made to s 7 (the Senate), s 24 (House of Representatives), s 64 (Governor General and Executive Government), s 128 (Alteration of the Constitution) and related sections of the Australian Constitution. 25 Lange (n 21) 567. 26 ibid 566. See also Kirby J in BAT (2003) 217 CLR 30, 80 and AID/WATCH Incorporated v Commissioner of Taxation of the Commonwealth Bank of Australia (2010) 241 CLR 539, [44] French CJ, Gummow, Hayne, Crennan and Bell JJ who stated: ‘the Constitution informs the development of the common law’. 27 British American Tobacco Australia Ltd v The State of Western Australia (hereafter ‘BAT ’) (2003) 217 CLR 30. 22 23

The Woolwich Principle in Australia 317

sequel to the Roxborough litigation28 in which a tobacco retailer and wholesaler entered into contractual arrangements to facilitate paying a New South Wales licence fee. This licence fee was imposed by the Business Franchise Licenses (Tobacco) Act 1987 (NSW) and was collected by passing on the ultimate burden of the tax to the consumer. In respect of each consignment of tobacco, the retailer paid a sum to the wholesaler on account of the licence fee (it being anticipated that the wholesaler would then pay these amounts to the New South Wales government) and then recouped this amount in the form of higher tobacco prices paid by tobacco consumers. The High Court in Ha v New South Wales29 declared that the licence fee was a duty of excise under section 90 of the Australian Constitution (dealing with the exclusive power of Commonwealth Parliament to impose duties of customs and excise), and was thus rendered invalid. Roxborough was a successful claim by the retailer in seeking restitution of amounts paid to the wholesaler, in other words a claim between two ‘private citizens’.30 In awarding the remedy, the High Court held that the basis of the transfer had failed. As explained by Gummow J, there was a ‘failure to sustain itself of the state of affairs contemplated as a basis for the payments the appellants seek to recover’.31 There were similar statutory licensing schemes in other Australian States, all of which were affected by the constitutional invalidity revealed by Ha.32 In Western Australia, the invalidity was discovered at a time when funds had already been paid by the tobacco wholesaler, British American Tobacco (BAT), to the Commissioner of Taxation in that State. The case thus concerned a ‘restitutionary claim against government’.33 BAT sought to recover these payments in unjust enrichment. The claim was brought in the Supreme Court of Western Australia against the State of Western Australia and the Commissioner of State Taxation. The defendants argued that the Crown Suits Act 1947 (WA) applied, such that – pursuant to section 6 then in force – ‘no right of action lies against the Crown’ unless the party proposing to take action gave, as soon as practicable or within three months after the cause of action accrued, written notice to the defendant. The notice period had not been fully complied with and the case went to the High Court on the issue of whether or not the Crown Suits Act 1947 was relevant to the proceedings. The High Court held that the right to proceed against Western Australia was conferred as a matter of federal law as it was a matter arising under the Constitution.34 Given that the West Australian court was exercising federal jurisdiction and the right to sue ‘[did] not depend upon, and [was] not subject to, the Crown Suits Act’,35 the notice period provided by that Act did not apply. BAT was free to sue the West Australian Commissioner of Taxation. Following this decision, it might have been thought that a claim against the Roxborough v Rothmans of Pall Mall Australia Ltd (2001) 208 CLR 516 (hereafter ‘Roxborough’). Ha v New South Wales (1997)189 CLR 465. 30 Roxborough (n 28) [29] (Gleeson CJ, Gaudron and Hayne JJ). 31 ibid [104]. 32 These covered – in addition to tobacco – petroleum products and liquor. The list of State Franchise Laws in section 4 of Franchise Fees Windfall Tax (Collection) Act 1997 (Cth) lists a number of State Acts caught by the constitutional invalidity. 33 Roxborough (n 28) [29] (Gleeson CJ, Gaudron and Hayne JJ). 34 BAT (n 27) (2003) 217 CLR 30. The legislative invalidity was based upon section 90 of the Constitution (Federal Parliament has exclusive power to impose duties of customs and excise). The dispute was one arising under the Constitution within the meaning of section 76 (i) (Parliament may make laws conferring original jurisdiction on the High Court in any matter arising under the Constitution or involving its interpretation), and under section 77 (iii) of the Constitution the Western Australian court was invested with federal jurisdiction. 35 BAT (n 27) 48 (Gleeson CJ). 28 29

318 Simone Degeling Commissioner would be relatively straightforward. However, no litigation ensued and the possibility of a Woolwich-style claim was not tested. This was perhaps because BAT had passed on the burden of the invalid tax so that in a cash-flow sense it was not out of pocket.36 The real risk, both legal and commercial, lay in the possibility that parties lower down the passing on chain (for example other intermediate legal entities or tobacco consumers) might after Roxborough bring proceedings to recover amounts they had paid on account of the invalid tax. In effect, BAT was simply protecting its right to recover against the Commissioner in case it should ever be required. Additionally, following Roxborough, specific legislation was passed by the Commonwealth Parliament which presumably provided a strong disincentive to sue. The Franchise Fees Windfall Tax (Collection) Act 1997 (Cth) provides in section 6 that a ‘taxable amount’ is an amount which a state is liable to repay to a person (taxpayer) because the state ‘law is wholly or partially invalid because of section 90 the Constitution’, and section 4 of the Franchise Fees Windfall Tax (Imposition) Act 1997 (Cth) imposes a windfall tax thereon at the rate of 100 per cent. Leaving aside the obvious question as to the constitutional validity of this legislation, there was presumably no point suing the Commissioner after Roxborough because any funds recovered would immediately be owed as tax. The possibility of a Woolwich claim was therefore never explored. However, BAT is useful as the facts illustrate the possibility of a constitutional invalidity which may found a common law claim. Establishing the constitutional invalidity is relatively straightforward. The difficulty is to make an argument on a principled basis for the recognition of an unjust enrichment claim which flows from this invalidity. Justice Gummow in Roxborough referred to Lord Mansfield’s statement of principle in Moses v Macferlan37 explaining that Lord Mansfield was giving examples of when an action for money had received might be available, but that this list was not exhaustive of the scope of the action. As stated by Justice Gummow in Roxborough: Lord Mansfield emphasised that he had stated what were but examples of ‘the gist of this kind of action [namely] that the defendant, upon the circumstances of the case, is obliged by the ties of natural justice and equity to refund the money’. The particular examples which Lord Mansfield gave, of which failure of consideration is one . . . each have developed their own body of authority in which their meaning and scope is expounded . . . However, the specific examples Lord Mansfield gave are not exhaustive of the scope of the action . . . Further, in deciding cases such as the present which question the boundaries of the established categories, recourse should be had to the general considerations referred to in Moses v Macferlan.38

To the extent that we may draw on Lord Mansfield’s ‘ties of natural justice and equity’ there is certainly scope to provide a foundation for such a development. Despite the High Court’s conservative approach when it comes to the law of unjust enrichment, it may be possible to give life to a Woolwich-style claim. The constitutional principle identified by Lord Goff in Woolwich could equally be identified in the Australian context. There are certainly judicial statements which indicate that the state cannot retain money without entitlement. For example, Mason CJ in Commissioner of State Revenue (Vict) v Royal Insurance

36 This was as a matter of commercial reality only. Passing on is not a defence to a claim in unjust enrichment in Australia. See Commissioner of State Revenue (Victoria) v Royal Insurance Australia (1994) 182 CLR 51. 37 Moses v Macferlan (1760) 2 Burr 1005, 97 ER 676. 38 Roxborough (n 28) 553, [95].

The Woolwich Principle in Australia 319

Australia Ltd identified an ‘important constitutional value’ that no tax can be levied by executive government without the authority of Parliament.39 In addition, in the same way that a constitutional freedom of political communication was identified in Lange, the text and structure of the Constitution must surely require that the citizen be entitled as of right to demand repayment of money levied otherwise than in conformity with the Constitution. The constitutional imperatives identified by Kirby J in BAT in support of a direct constitutional right of action (discussed below) arguably also apply to common law protection. We must be careful to observe, however, that any such right is sourced within the common law, while the policy which dictates the recognition of that right could be found (or perhaps even implied) into the Constitution. As emphasised by the High Court in the BAT case, although the claim in that case ‘arises under the Constitution’ in the sense that section 90 of the Constitution invalidated the Western Australian statute, the source of any right to restitution remained with the common law.40 The plaintiff’s action was pleaded in mistake of law and duress colore officii. Whilst the High Court has rejected attempts to invoke ‘constitutional reasons’ for treating ‘restitutionary claims against governments differently from claims against private citizens’ (emphasis added),41 it has not yet taken the next step and identified a unique policy-motivated unjust factor. An alternative common law claim would of course be one which does not involve the recognition of a novel policy-motivated unjust factor in the style of Woolwich. This is the possibility that an exaction of tax which is subsequently found to be unconstitutional may be recoverable on a failure of basis, or in the language of Roxborough, ‘a failure to sustain itself of the state of affairs contemplated as a basis for the payments’.42 The mirror position, contemplating a claim by the government, was referred to in passing in Williams v Commonwealth.43 That case concerned the executive power of the Commonwealth to enter contracts and expend public money (section 61 of the Constitution) on chaplaincy services in state high schools in circumstances where there had been no specific Act of the Commonwealth Parliament authorising it to enter funding agreements to spend public funds in this way. A majority of the Court struck down the funding agreement as being beyond the executive power of the Commonwealth.44 Citing Roxborough, Gummow and Bell JJ noted in passing the possibility that payments which have been made . . . [by the Commonwealth] under the Funding Agreement are, by reason of an absence of an appropriation, moneys had and received by . . . [the defendant] to the use of the Commonwealth and recoverable as such.45

Arguably, a claim for failure of basis should similarly be available in a claim against the government. It is mentioned here only for completeness, and because it flows naturally 39 Commissioner of State Revenue (Vict) v Royal Insurance Australia Ltd (1994) 182 CLR 51 (hereafter ‘Royal Insurance’) 69. See also Dawson J (at 99) who was of the view that ‘clear words are required to authorise the retention of moneys received without any entitlement’; Roxborough (n 28) 570 (Kirby J); Commissioner of Stamps (SA) v Telegraph Investment Co Pty Ltd (1995) 184 CLR 453 (hereafter ‘Telegraph Investment’) 467 (McHugh and Gummow JJ). 40 BAT (n 27) esp [39]–[41] (McHugh, Gummow and Hayne JJ), [6], [11], [15]–[16] (Gleeson CJ), [168] (Callinan J). 41 Roxborough (n 28) [29] (Gleeson CJ, Gaudron and Hayne JJ). 42 ibid [104] (Gummow J). 43 Williams (n 13). 44 ibid, esp 000. 45 ibid [116].

320 Simone Degeling from the discussion of Roxborough. Conceptually this claim belongs in the discussion of ‘Other Claims’ below.

(2) Direct Constitutional Claim A direct constitutional right to restitution may also be recognised. However, it must be conceded that the weight of authority is strongly against a private right of action being enlivened to allow an individual directly to enforce a constitutional right. There are many examples where individual litigants have invoked a provision of the Constitution to invalidate a law which is said to be in contravention of the Constitution, but little evidence of any cases where an individual has invoked the Constitution and sued the Commonwealth directly, for example for a money remedy. There are said to be no constitutional causes of action. Typical of the sentiment is the statement of Dixon J in James v The Commonwealth: [P]rima facie a Constitution is concerned with the powers and functions of government and the restraints upon their exercise. There is, in my opinion, no sufficient reason to regard section 92 [freedom of interstate trade and commerce] as including among its purposes the creation of private rights sounding in damages. It gives to all an immunity from the exercise of governmental power. But to find whether a governmental act be wrongful the general law must be applied. Sec. 92 will do no more than nullify an alleged justification. The plaintiff cannot, therefore, recover damages under sec. 92 independently of any tort by the Commonwealth.46

Against this is the strong view of Justice Kirby in the BAT case. The linchpin of his argument is the ‘important constitutional value’47 identified by Mason CJ in Commissioner of State Revenue (Vict) v Royal Insurance Australian Ltd : ‘There is [a] fundamental principle of public law that no tax can be levied by the executive government without Parliamentary authority, a principle which traces back to the Bill of Rights 1688 (Imp)’.48 In Justice Kirby’s judgment, allowing a direct right of recovery from the ‘polity that exacts an unconstitutional tax’49 is effective and appropriate in protecting and enforcing this constitutional value and discouraging any future breaches of it. As Mason CJ observed in Royal Insurance, [i]t would be subversive of an important constitutional value if [the High Court] were to endorse a principle of law which, in the absence of such circumstances, authorised the retention by the executive of payments which it lacked authority to receive.50

According to this view, the prohibition in section 90 the Constitution (about Federal Parliament having exclusive power to impose duties of customs and excise) necessarily carries the power of its own vindication and enforcement. The Constitution should not only be read as a document about government, but also as a source of ‘rights and obligations of the people, not necessarily dependent for their vindication upon legislation vulnerable to 46 James v The Commonwealth (1939) 62 CLR 339, 362. See also Werrin v The Commonwealth (1938) 59 CLR 150; Antill Ranger & Co Pty Ltd v Commissioner of Motor Transport (1955) 93 CLR 83, 98–99 and 101; Kruger v The Commonwealth (1997) 190 CLR 1. These cases all deal with a right to recover damages for loss. See also Park Oh Ho v Minister for Immigration and Ethnic Affairs (1988) 14 ALD 787, 789–90 (Davies J at first instance). Under the proposed claim, the plaintiff would not be suing on a (civil) wrong, but vindicating a constitutional right. 47 cf BAT (n 27) 80. 48 Royal Insurance (n 39) 69. 49 BAT (n 27) 69. 50 Royal Insurance (n 39) 69.

The Woolwich Principle in Australia 321

change or reliance upon governmental action susceptible to the power of interests that may not always coincide with those of the citizens’.51 Justice Kirby’s separate opinion in BAT is of course useful because the facts raise the precise paradigm problem under consideration. The more general issue of whether private rights of action derive from the Constitution itself remains unresolved. An illustration is ICM Agriculture Pty Ltd v Commonwealth of Australia52 which was a case under section 51(xxxi) of the Constitution. That case concerned bore water licences held by the plaintiff which were, by New South Wales legislation, replaced by a new system of licences which permitted the plaintiff to take less water than had been allowed under the bore licences. The NSW legislation was enacted pursuant to an agreement with the Commonwealth and provided for ‘structural adjustment payments’, the funds for which were to be provided equally by the NSW and Commonwealth governments. The plaintiff argued that the structural adjustment payments were not adequate and therefore commenced proceedings relying on section 51(xxxi) of the Constitution which is directed to laws made by the Parliament of the Commonwealth with respect to the acquisition of property on just terms. Section 51(xxxi) operates such that it forbids ‘laws with respect to acquisition [of property] on any terms that are not just’.53 The plaintiff did not simply argue the invalidity of the legislative framework under which the bore licences were cancelled, but rather argued that it had an implied right under the Constitution to recover from the Commonwealth such compensation for the loss of bore licences as would constitute ‘just terms’. What was contemplated was a direct private right of recovery. The High Court ruled that the replacement of the plaintiff’s bore licences did not constitute an ‘acquisition of property’ within the meaning of the Constitution, so this point was not decided, but the weight of authority is against a private right of action. The point arose in AID/WATCH Inc v Commissioner of Taxation in which a majority of the Court held that ‘personal rights of action are not . . . bestowed upon individuals in the manner of the Bivens action known in the United States’.54 Although State governments are arguably bound by the same constitutional imperative so that they should not levy tax without Parliamentary authority,55 there are few grounds on which to challenge the validity of a State legislative act, assuming that it does not fall foul of the Commonwealth Constitution. States have very broad plenary powers and most legislative acts will fall within ‘peace, order and good government’, making it difficult to challenge them. Against this State constitutional law background it is also necessary to take into consideration various pieces of human rights legislation which have been enacted in some states: the Charter of Human Rights and Responsibilities Act 2006 (Vict) and the Human Rights Act 2004 (ACT). Neither is particularly fertile ground for the recognition of an unjust enrichment claim of any kind, let alone in respect of tax demanded ultra vires. The Victorian Charter of Human Rights and Responsibilities Act 2006 states in section 20 that ‘a person must not be deprived of his or her property other than in accordance with law’. An ultra vires demand by a public authority could arguably fall within this 51 BAT (n 27) 81 (Kirby J), relying also on Bivens v Six Unknown Named Agents of Federal Bureau of Narcotics 403 US 388. 52 ICM Agriculture Pty Ltd v Commonwealth of Australia [2009] HCA 51. 53 Bank of New South Wales v The Commonwealth (1948) 76 CLR 1, 349–50 (Dixon J). Section 51(xxxi) speaks of ‘[t]he acquisition of property on just terms from any . . . person for any purpose in respect of which the Parliament has power to make laws’. 54 AID/WATCH Inc v Commissioner of Taxation (2010) 241 CLR 539, [44] (French CJ, Gummow, Hayne, Crennan and Bell JJ). On the so-called ‘Bivens action’, see n 51. 55 Stephenson v The Queen (1865) 2 WW & A’BL 143.

322 Simone Degeling prohibition. The Charter envisages that the main mechanism for enforcement lies in the jurisdiction of the Supreme Court of Victoria to declare that a particular statutory provision cannot be interpreted consistently with a human right56 and the possibility of subsequent amendment of that legislation by Parliament, or an explanation by Parliament as to why the legislation should stand despite the incompatibility.57 The remedial provisions in section 39 are quite restrictive. As explained by Maxwell P in Director of Housing v Sudi, in order to claim under the Charter, a plaintiff must show that independently of the Charter, ‘any relief or remedy in respect of an act or decision of a public authority on the ground that the act or decision was unlawful’. If – but only if – that condition is satisfied, then section 39(1) enables that person to seek ‘that relief or remedy’ on a supplementary ground of unlawfulness, that is, unlawfulness arising because of the Charter.58

The fact that Australian law does not currently admit the Woolwich principle therefore limits any protection which might be available under the Charter. The Charter itself creates no independent cause of action. Section 39(3) provides that a person is not entitled to be awarded damages because of a breach of the Charter, and it seems unlikely that any claim for restitution would be entertained.59 Of course arguments could be made that an award of restitution is not an award of damages, but it seems relatively clear that the legislative intent is to exclude money remedies, including arguably restitution of unjust enrichment. In the Australian Capital Territory, although section 40B of the Human Rights Act 2004 provides that Public Authorities must act consistently human rights, the rights listed in Part 3 do not deal directly with ultra vires demands. The trigger would have to be indirect, for example a right to recognition and equality before the law (section 8). Section 40C allows the victim of a public authority’s contravention to bring proceedings in the Supreme Court which may grant the relief it considers appropriate except damages, unless they are recoverable independently of the breach of the Human Rights Act.60 There is no mention of restitution and – similarly to the Victorian legislation – it seems likely that the intention is to exclude all money remedies.

C. OTHER CLAIMS

In the absence of a prima facie right to restitution based simply on the ultra vires nature of the demand, the Australian position is a convoluted web of indirect protections for the plaintiff. The administrative law regime and common law claims for duress colore officii, mistake and failure of basis all form part of the picture.

56 Charter of Human Rights and Responsibilities Act 2006 (Vict), s 36. See Momcilovic v R [2011] HCA 34, where the High Court clarified that a declaration under s 36 is limited to matters not in federal jurisdiction. 57 Charter of Human Rights and Responsibilities Act 2006 (Vict), s 31. 58 Director of Housing v Sudi [2011] VSCA 266, [96]. See also Sabet v Medical Practitioners Board of Victoria (2008) 20 VR 414, [104]–[105] (Hollingworth J). 59 Section 39(4) protects any right a person may have to damages apart from the operation of s 39. For a critique of the remedial provisions see J Gans, ‘The Charter’s Irremediable Remedies Provision’ (2009) 33 Melbourne University Law Review 105. 60 See Eastman v Chief Executive, Dept of Justice and Community Safety [2011] ACTSC 33, [23] (Mansfield J).

The Woolwich Principle in Australia 323

(1) Statutes and the Administrative Law Dimension61 Statutes may within their regime provide for the return of funds overpaid or otherwise demanded without foundation. For example, the Sales Tax Assessment Act 1992 (Cth) in Part 4 provides for a credit for sales tax overpaid.62 It was similarly on the basis of a statutory provision that Royal Insurance brought proceedings against the Victorian Commissioner of State Revenue.63 Section 111(1) of the Stamps Act 1958 (Vict) provided: [W]here the [Comptroller] finds in any case that duty has been overpaid, whether before or after the commencement of the Stamps Act 1978, he may refund to the company, person or firm of persons which or who paid the duty the amount of duty found to be overpaid.

It was common ground between the parties that Royal had overpaid stamp duty on policies of insurance against liability for workers compensation and that section 111(1) applied. The Comptroller decided not to make a refund and Royal commenced proceedings to recover these funds. A majority of the court held that section conferred a discretionary power on the Commissioner [the statutory successor to the Comptroller] to refund overpaid money, but created no duty to do so.64 Brennan, Toohey and McHugh JJ held that there was no residual discretion to deny a refund once a finding about the overpayment had been made. Chief Justice Mason opined that the discretion under section 111 was to be exercised in accordance with the principles of law of restitution . . . [and] that discretion was exercised erroneously. On the basis on which the case was fought in the courts below . . . Royal was entitled to recover the overpayments in conformity with the law of restitution.65

The remedy awarded was an order in the nature of mandamus, an order commanding the fulfilment of a duty of a public nature that remained unperformed and for which no other specific legal remedy is available.66 Strictly speaking, the source of the right was the Stamp duties legislation itself. As explained by Dawson J: Were the Comptroller to be governed by the common law rather than s. 111(1) with regard to her obligation, if any, to refund the overpaid stamp duty, the remedy available to the respondent would be of a quite different nature. The respondent would then have an action for money had and received based upon a right to restitution. On the other hand, were the Comptroller’s exercise of authority discretionary under s. 111(1), the remedy available to Royal would be confined in the first place to requiring the Comptroller to exercise their discretion and then to contesting the ability of its exercise by way of judicial review. The confined grounds upon which an administrative 61 The judicial review jurisdiction is conferred with respect to Federal Jurisdiction: under s 75(v) of the Constitution and s 39B of the Judiciary Act. State Supreme Courts inherited the inherent judicial review jurisdiction of all the superior courts at Westminster, although some now have specific legislation also conferring specific powers. See generally M Aronson et al, Judicial Review of Administrative Action (Sydney, Law Book Co, 2009) ch 2. There is also a statutory scheme of cross-vested civil jurisdiction. 62 Considered in Avon Products Pty Ltd v Commissioner of Taxation [2006] HCA 29. 63 Royal Insurance (n 39). 64 The majority were Mason CJ, Brennan, Toohey and McHugh JJ. Justice Dawson alone held that the permissive ‘may’ in section 111(1) merely conferred authority on the Commissioner to refund, but that once the condition necessary for the exercise of that authority was found to exist (ie the overpayment) the section created a duty to refund. There was no residual discretion. See Royal Insurance (n 39) 102. 65 ibid 79. 66 cf R Creyke and J McMillan, Control of Government Action (Sydney, Butterworths, 2009) 1103, citing R v War Pensions Entitlement Appeal Tribunal, ex parte Bott (1933) 50 CLR 228.

324 Simone Degeling decision may be reviewed by a court would preclude the substitution of a decision based upon restitutionary principles . . . But that is something different from an application of the common law relating to restitution. 67

In any case, Dawson J did not regard ‘s. 111(1) as conferring a discretion. Once the Comptroller found that duty had been overpaid, she was under an obligation to refund it’.68 The High Court in Pape v Commissioner of Taxation of the Commonwealth of Australia articulates an alternative claim which may have applied in Royal Insurance. Applying the reasoning of the Parke B in Shepherd v Hills,69 Gummow, Crennan and Bell JJ suggest that an action in debt, founded on the statutory entitlement, may also have been possible: ‘There is no doubt that whenever an Act of Parliament creates a duty or obligation to pay money, an action will lie for its recovery, unless the Act contains some provision to the contrary’. The result is that the payment of the tax bonus involves, contrary to the submission of the plaintiff, more than the receipt of a mere gratuity; the payment is the discharge of a legal obligation. There is a concomitant obligation to repay to the Commonwealth overpayments.70

It is a difficult question whether the Administrative Decisions (Judicial Review) Act 1977 (Cth) confers a power to order restitution independent of any other right to relief. Assuming that the Act applies (a decision of an administrative character under a relevant enactment by a Commonwealth authority or an officer of the Commonwealth), the remedial provisions in section 16 appear quite wide. Specifically, section 16(1)(d) empowers the court to make ‘an order directing any of the parties to do, or a refrain from doing, any act or thing the doing, or the refraining from doing, of which the court considers necessary to do justice between the parties’. The High Court in Park Oh Ho v Minister for Immigration and Ethnic Affairs 71 (a judicial review action of the decision under the Migration Act to detain Park) observed in passing that an award of damages for false imprisonment was not an appropriate remedy in judicial review proceedings. In the Federal Court phase of the proceedings, Morling J observed that remedies under section 16(1) ‘do not include the making of an award of damages’.72 This is apparently because of the view that [a] claim for damages from the Commonwealth must be based upon a civil wrong or a breach of contract. An applicant who merely establishes a ground of review under section 5 of the ADJR Act is not thereby entitled to an award of damages. The remedies of judicial review are those in the nature of certiorari, prohibition, mandamus, injunction and declaration as section 16 of the ADJR Act makes plain.73

It is obviously possible to place a clear line between such statements and claims for restitution, on the ground that what is sought is not founded upon a civil wrong, but rather restitution of unjust enrichment, and highlighting the appropriate remedy is simply the reversal of the unlawfully exaction. However, the cases apparently continue to insist that an independent basis of a right to recovery must be established. For example, in the Kawasaki Royal Insurance (n 39) 101. ibid 102. 69 Shepherd v Hills (1855) 11 Ex 55, 67; 156 ER 743, 747. See also Mallinson v Scottish Australian Investment Co Limited (1920) CLR 66, 70, relied on in The Commonwealth v SCI Operations Pty Ltd (1998) 192 CLR 285, [65] (McHugh and Gummow JJ). 70 Pape v Commissioner of Taxation of the Commonwealth of Australia (2009) 238 CLR 1, [140]. 71 Park Oh Ho v Minister for Immigration and Ethnic Affairs (1989) 167 CLR 637 72 Park Oh Ho v Minister for Immigration and Ethnic Affairs (1988) 20 FCR 104, 126. 73 Davies J at first instance (n 46) 789–90. 67 68

The Woolwich Principle in Australia 325

case74 Hill and Heerey JJ held in respect of a claim for the repayment of customs duty, that section 16(1)(d) did not in the circumstances of itself grant the power to make an order directing the decision maker to make a (re)payment. However, the section could ‘in an appropriate case . . . extend to directing the decision-maker to make a payment that had been refused contrary to law’.75 At the very least, this would extend to a statutory entitlement to money (Creyke and McMillan give as an example an entitlement to a social security benefit)76 and perhaps also encompass a statutory entitlement to be repaid. In any event, the possibility of a restitutionary remedy under the Administrative Decisions (Judicial Review) Act 1977 is at best a limited and indirect protection for a plaintiff who has paid pursuant to an ultra vires demand.

(2) Common Law Claims: Mistake, Duress and Colore Officii As shown by Federal Airports Corporation v Aerolineas Argentinas 77 in which the Federal Court reviewed a determination fixing a charge under section 56 of the Federal Airports Corporation Act 1986 (Cth), a parallel process to any review under the ADJR Act is the pursuit of a common law claim.78 In this respect, Australian law in basically resembles English law pre-Woolwich. Without a causative mistake79 or some kind of improper pressure, whether duress to goods80 or a demand colore officii,81 the plaintiff cannot succeed. The possibility of a Roxborough-style claim for failure of basis has already been noted.82

D. CONCLUSION

The doctrinal possibilities for a right to recover unlawfully exacted tax are relatively clear. Leaving to one side the labyrinth of indirect protections in the form of the old claims for compulsion and mistake, and the particular administrative law complications, it seems that there is scope either for the development of a common law claim or some type of direct constitutional claim. The High Court of Australia has been prominent in its conservative approach to unjust enrichment law. There has been little discussion of policymotivated unjust factors, although there does seem to be some acceptance of necessity83

74 Comptroller General of Customs v Kawasaki Motors Pty Ltd (No 2) (1991) 32 FCR 243 (hereafter ‘Kawasaki’) 266. 75 ibid 265–66. 76 Creyke and McMillan (n 66) 1139. 77 Federal Airports Corporation v Aerolineas Argentinas (1997) 147 ALR 649. 78 The common law claim in that case was on the basis of duress colore officii, mention having been made (ibid 663) of Bell Brothers Pty Ltd v Shire of Serpentine-Jarrahdale (1969) 121 CLR 137. Similarly in Kawasaki (n 74) 264. 79 See David Securities Pty Ltd v Commonwealth Bank of Australia (1992) 175 CLR 353 (hereafter ‘David Securities’). 80 See Mason v State of New South Wales (1959) 102 CLR 108 (hereafter ‘Mason’) 144 (Windeyer J). Technically, the ground of relief relied on by the plaintiff in Mason was colore officii, but the High Court acknowledged that it could have been duress of goods. See also Sargood Brothers v The Commonwealth (1910) 11 CLR 258. 81 See Bell Brothers Pty Ltd v Shire of Serpentine-Jarrahdale (1969) 121 CLR 137, 145 (Kitto J); Mason (n 80). 82 See the text accompanying nn 27–45. 83 Cook v Lumbers (2008) 232 CLR 635.

326 Simone Degeling and also illegality.84 Allowing a direct constitutional right of action is clearly more significantly controversial than allowing the development of the common law of unjust enrichment. With the Supreme Court of Canada and the Supreme Court of the United Kingdom having selected divergent paths,85 Australian law awaits an appropriate case to open the issue. The patchy protection offered by common law claims based on mistake and duress, and any possibility of an indirect remedy in a judicial review action, do little to provide the plaintiff who pays pursuant to an ultra vires demand with a meaningful remedy. As stated by Mason CJ in the Royal Insurance case, it is an ‘important constitutional value’ that no tax can be levied by executive government without the authority of Parliament.86 The corollary of this principle must surely be that such amounts are recoverable.87

David Securities (n 79) 379 (Mason CJ, Deane, Toohey, Gaudron and McHugh JJ). For an account of the Canadian position and how it differs from the English case law see esp Robert Chambers’ contribution in chapter 14 of the present volume. 86 Royal Insurance (n 39) (1994) 69. See also Dawson J (ibid 99) who was of the view that ‘clear words are required to authorise the retention of moneys received without any entitlement’; Roxborough (n 28) 570 (Kirby J); Telegraph Investment (n 39) 467 (McHugh and Gummow JJ). 87 In this vein also Wong (n 7) 603–6; Mason and Carter’s Restitution Law in Australia (n 7) 796–99. 84 85