Extending Working Life for Older Workers: Age Discrimination Law, Policy and Practice 9781509905768, 9781509905799, 9781509905782

Citation preview

EXTENDING WORKING LIFE FOR OLDER WORKERS The UK population is ageing rapidly. While age discrimination laws are seen as having broad potential to address the ‘ageing challenge’ and achieve instrumental and intrinsic objectives in the context of employment, it is unclear what impact they are having in practice. This monograph therefore addresses two overarching questions in the employment field: How are UK age discrimination laws operating in practice? How (if at all) could UK age discrimination laws be improved? A reflexive law theoretical standpoint is employed to investigate these issues, applying a mixed methods research design that engages qualitative, quantitative, doctrinal and comparative elements. This book demonstrates the substantial limitations of the Equality Act 2010 (UK) for achieving instrumental and intrinsic objectives. Drawing on qualitative expert interviews, statistical analysis and organisational case studies, it illustrates the failure of age discrimination laws to achieve attitudinal change in the UK, and reveals the limited prevalence of proactive measures to support older workers. Integrating doctrinal analysis, comparative analysis of Finnish law, and the Delphi method, it proposes targeted legal and policy changes to address demographic change, and offers an agenda for reform that may increase the impact of age discrimination laws, and enable them to respond effectively to demographic ageing.

ii

Extending Working Life for Older Workers Age Discrimination Law, Policy and Practice

Alysia Blackham

OXFORD AND PORTLAND, OREGON 2016

Hart Publishing An imprint of Bloomsbury Publishing Plc Hart Publishing Ltd Kemp House Chawley Park Cumnor Hill Oxford OX2 9PH UK

Bloomsbury Publishing Plc 50 Bedford Square London WC1B 3DP UK

www.hartpub.co.uk www.bloomsbury.com Published in North America (US and Canada) by Hart Publishing c/o International Specialized Book Services 920 NE 58th Avenue, Suite 300 Portland, OR 97213-3786 USA www.isbs.com HART PUBLISHING, the Hart/Stag logo, BLOOMSBURY and the Diana logo are trademarks of Bloomsbury Publishing Plc First published 2016 © Alysia Blackham Alysia Blackham has asserted her right under the Copyright, Designs and Patents Act 1988 to be identified as Author of this work. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage or retrieval system, without prior permission in writing from the publishers. While every care has been taken to ensure the accuracy of this work, no responsibility for loss or damage occasioned to any person acting or refraining from action as a result of any statement in it can be accepted by the author, editors or publishers. All UK Government legislation and other public sector information used in the work is Crown Copyright ©. All House of Lords and House of Commons information used in the work is Parliamentary Copyright ©. This information is reused under the terms of the Open Government Licence v3.0 (http://www.nationalarchives. gov.uk/doc/open-government-licence/version/3) excepted where otherwise stated. All Eur-lex materials used in the work is © European Union, http://eur-lex.europa.eu/, 1998–2015. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library. ISBN: HBK: 978-1-50990-576-8 ePDF: 978-1-50990-578-2 ePub: 978-1-50990-577-5 Library of Congress Cataloging-in-Publication Data Names: Blackham, Alysia, author. Title: Extending working life for older workers : age discrimination law, policy and practice / Alysia Blackham. Description: Oxford ; Portland, Oregon : Hart Publishing, 2016.  |  Includes bibliographical references and index. Identifiers: LCCN 2016028878 (print)  |  LCCN 2016029332 (ebook)  |  ISBN 9781509905768 (hardback : alk. paper)  |  ISBN 9781509905775 (Epub) Subjects: LCSH: Age discrimination in employment—Law and legislation—Great Britain. Classification: LCC KD3103.O43 B57 2016 (print)  |  LCC KD3103.O43 (ebook)  |  DDC 344.4101/398133—dc23 LC record available at https://lccn.loc.gov/2016028878 Typeset by Compuscript Ltd, Shannon

For Shelda: my mentor, role model and friend.

vi

PREFACE

2016 marks the tenth anniversary of the Employment Equality (Age) Regulations 2006 (SI 2006/1031) which, for the first time, prohibited age discrimination in employment in the UK. Since that time, not much appears to have changed in relation to the employment of older workers: demographic change is still posing serious issues for employers and policy makers; and many older workers continue to struggle to find and retain employment. It is, then, more than timely to think about the success (or otherwise) of age discrimination legislation in the UK—is it working to support the employment of older workers and extend working lives? If not, why not—and what can be done to improve the situation? This book seeks to answer these questions using an empirical mixed m ­ ethods research design. It is based on my PhD research, conducted at the University of Cambridge from 2011 to 2014 under the skilled supervision of Professor ­Catherine Barnard and Professor Simon Deakin; and with funding and support from the WM Tapp Studentship Fund at Gonville and Caius College, University of ­Cambridge; the Poynton Cambridge Australia Scholarship and Cambridge Australia Scholarships; the University of Cambridge; the UK Socio-Legal Studies Association; and the Cambridge Political Economy Society Trust. This book builds on preliminary work published during those studies.1 Fieldwork for the comparative elements of the study was conducted while I was a visiting scholar at the Faculty of Law at Lund University, and the Faculty of Law at the University of Helsinki. I am eternally grateful to all those who helped me over the course of this study, both formally and informally, and who provided feedback on the substantive content of this book. An empirical study of this nature draws heavily on the time and energy of research participants. I would like to thank all the experts and organisations that so generously agreed to participate in this research, both in the UK and in Finland. Without their contribution, this research would be impossible, and I am

1  A Blackham, ‘Unfinished Business in Compulsory Retirement Ages’ (2012) 71(3) Cambridge Law Journal 495; A Blackham and A Ludlow, ‘Introduction’ and A Blackham, ‘Using the Delphi Method to Advance Legal Reform: A New Method for Empirical Labour Law Research?’, both in A Ludlow and A Blackham (eds), New Frontiers in Empirical Labour Law Research (Oxford, Hart, 2015); A Blackham, ‘Tackling Age Discrimination Against Older Workers: A Comparative Analysis of Laws in the United Kingdom and Finland’ (2015) 4(1) Cambridge Journal of International and Comparative Law 108; A Blackham, ‘Emerging Options for Extending Working Lives: Results of a Delphi Study’ in S Manfredi and L Vickers (eds), Challenges of Active Ageing: Equality Law and the Workplace ­(Basingstoke/New York, Palgrave Macmillan, 2016).

viii 

Preface

e­ xceedingly grateful for their time, insight and endurance over the course of the research process. This book will be of particular interest to labour and discrimination lawyers, socio-legal and empirical legal scholars and those interested in law in context. While particularly pertinent for scholars and practitioners in the UK and Finland, it also offers useful comparative insights to those operating in other jurisdictions, particularly in the area of age discrimination and equality law. Beyond its substantive content, this book also offers a model for the use of mixed methods to answer legal questions. Thus, it will be particularly helpful to master’s and PhD students and more senior scholars in law and law-related fields embarking upon empirical projects. Finally, this monograph will be of interest to government, law reform bodies and non-governmental organisations, and all those with an interest in the practical impact of discrimination law and legal regulation more broadly. This research was conducted in a period of economic and social turbulence in the UK and EU. Deregulation and austerity are recurring issues throughout this work. Thus, the findings of this research are a product of their times. In this book, I put forward a programme of reform that is feasible despite deregulation, economic crisis and a retreat from government intervention under Conservative governments. However, I also raise ideas and options that are not currently seen as feasible, but are certainly of merit. With any luck, times will change. At some stage, and sooner rather than later, I hope we will see a return to prosperity and an increased willingness to consider government intervention in employment. There is much we can do to support older workers and increase the effectiveness of age discrimination legislation. While this work provides a framework for change in times of austerity, it also provides broader options that can be considered in years to come. I trust we can make more progress in increasing the effectiveness of age discrimination laws in the next 10 years. Alysia Blackham January 2016

CONTENTS

Preface���������������������������������������������������������������������������������������������������������������������� vii List of Defined Terms and Abbreviations������������������������������������������������������������������xv

1. Introduction��������������������������������������������������������������������������������������������������������1 I. Implications of Ageing for Labour Law����������������������������������������������������2 II. The Role of Discrimination Law for an Ageing Population��������������������6 A. Instrumental Ends�����������������������������������������������������������������������������7 B. Intrinsic Ends�������������������������������������������������������������������������������������9 C. Reconciling Conflicting Objectives�������������������������������������������������12 III. Research Questions����������������������������������������������������������������������������������13 IV. Chapter Overview������������������������������������������������������������������������������������14 2. Methodology�����������������������������������������������������������������������������������������������������17 I. Methodology��������������������������������������������������������������������������������������������17 A. Systems Theory��������������������������������������������������������������������������������18 B. Reflexive Law������������������������������������������������������������������������������������19 II. Research Framework�������������������������������������������������������������������������������24 A. Empirical Research��������������������������������������������������������������������������24 B. Mixed Methods Approach���������������������������������������������������������������26 (i) Advantages of Mixed Methods Designs��������������������������������26 (ii) Limitations of Mixed Methods Research������������������������������27 C. Doctrinal Research���������������������������������������������������������������������������29 D. Quantitative Research����������������������������������������������������������������������31 E. Qualitative Research������������������������������������������������������������������������32 F. Integrating Different Research Methods����������������������������������������33 III. Research Design���������������������������������������������������������������������������������������34 A. Qualitative Expert Interviews����������������������������������������������������������36 B. Quantitative Statistical Analysis������������������������������������������������������37 C. Comparative Analysis����������������������������������������������������������������������37 D. Organisational Case Studies������������������������������������������������������������38 E. The Delphi Method�������������������������������������������������������������������������39 IV. Conclusion�����������������������������������������������������������������������������������������������41 3. A Doctrinal Critique of UK Age Discrimination Laws�����������������������������������42 I. Legal Regulation of Age Discrimination������������������������������������������������42 A. EU Regulation����������������������������������������������������������������������������������42

x 

Contents (i) (ii)

Mandatory Retirement���������������������������������������������������������44 Legitimate Aims��������������������������������������������������������������������45 (a) ‘Fair Innings’�����������������������������������������������������������������46 (b) Maintaining ‘Dignity’ for Long-standing Employees���������������������������������������������������������������������48 (c) Workforce Planning�����������������������������������������������������49 (iii) Appropriate and Necessary���������������������������������������������������49 B. UK Regulation���������������������������������������������������������������������������������52 (i) Mandatory Retirement���������������������������������������������������������55 (ii) Legitimate Aims��������������������������������������������������������������������58 (a) Intergenerational Fairness�������������������������������������������59 (b) Dignity��������������������������������������������������������������������������60 (iii) Justification in a Specific Case����������������������������������������������60 (iv) Proportionality����������������������������������������������������������������������61 (v) Consequences of the Seldon Decision����������������������������������61 II. Limitations of Existing Laws�������������������������������������������������������������������68 III. Ways Forward�������������������������������������������������������������������������������������������72 A. Doctrinal Reform�����������������������������������������������������������������������������72 B. Systems Theory and Reflexive Law�������������������������������������������������74 (i) A Systems Theory Critique���������������������������������������������������74 (ii) A Reflexive Law Model����������������������������������������������������������75 (iii) Reforms to Strengthen Reflexive Regulation in the Equality Act 2010��������������������������������������������������������80 IV. Conclusion�����������������������������������������������������������������������������������������������83 4. Expert Perspectives on the Operation of Age Discrimination Laws��������������85 I. Expert Views on the Efficacy of Law�������������������������������������������������������86 A. Doctrinal Merits������������������������������������������������������������������������������86 B. Practical Implementation����������������������������������������������������������������89 (i) Implementation by Employers���������������������������������������������90 (a) Differences between Employers�����������������������������������90 (b) Prevalence of ‘Compliance’ Approaches���������������������90 (c) Role of Law�������������������������������������������������������������������92 (ii) Practical Impact of Legal Uncertainty���������������������������������93 (iii) Persistence of Negative Attitudes������������������������������������������94 (iv) Enforcement of Age Discrimination Legislation�����������������97 C. Reforms to Improve the Law and its Implementation�������������������98 (i) Reducing Legal Uncertainty�������������������������������������������������98 (ii) Operating Without a DRA��������������������������������������������������100 (iii) Need for a National Dialogue���������������������������������������������101 (iv) Improving Attitudes towards Older Workers��������������������102 (v) Promoting a Holistic Approach������������������������������������������104 D. Responsibility for Reform�������������������������������������������������������������104

Contents

 xi

II. Analysis���������������������������������������������������������������������������������������������������105 III. Conclusion���������������������������������������������������������������������������������������������109 5. Generalising the Qualitative Findings—Quantitative Analysis of WERS6������������������������������������������������������������������������������������������110 I. The WERS6 Data Set�����������������������������������������������������������������������������111 II. ‘Compliance’ Approaches in UK Workplaces���������������������������������������113 III. Age-Aware Workplaces��������������������������������������������������������������������������115 A. Theoretical Considerations and Past Research�����������������������������115 B. Bivariate Analysis���������������������������������������������������������������������������119 C. ‘Best Practice’ Organisations���������������������������������������������������������122 D. Multivariate Models of the Relationship between Organisational Characteristics and AAPS������������������������������������123 (i) Ordinal Models�������������������������������������������������������������������123 (ii) Count Models����������������������������������������������������������������������125 IV. Age-Aware Practices Over Time������������������������������������������������������������127 V. Workplaces that Adopt Age-Aware Practices Over Time���������������������130 A. Ordinal Models������������������������������������������������������������������������������130 B. Count Models��������������������������������������������������������������������������������132 VI. Conclusion���������������������������������������������������������������������������������������������134 6. A Comparative Perspective—The Case of Finland���������������������������������������136 I. Finnish Legal Provisions������������������������������������������������������������������������138 A. Non-Discrimination����������������������������������������������������������������������138 B. Mandatory Retirement������������������������������������������������������������������142 C. Positive Responsibilities�����������������������������������������������������������������143 D. Financial Incentives�����������������������������������������������������������������������145 II. Practical Implementation����������������������������������������������������������������������145 A. Enforcement of Age Discrimination Legislation��������������������������146 B. Involvement of the Social Partners�����������������������������������������������148 C. Implementation by Employers������������������������������������������������������150 D. Attitudinal Change������������������������������������������������������������������������154 III. Comparative Analysis of the UK and Finland��������������������������������������157 A. Differences between the Jurisdictions�������������������������������������������157 B. Similarities between the Jurisdictions�������������������������������������������160 C. Evaluating the Different National Approaches�����������������������������160 (i) Creating Opportunities for Older Workers to Work�����������������������������������������������������������������161 (ii) Improving the Quality of Employment for Older Workers���������������������������������������������������������������163 (iii) Achieving Equality for Older Workers�������������������������������166 IV. Conclusion���������������������������������������������������������������������������������������������169

xii 

Contents

7. Explaining the Quantitative Findings—Organisational Case Studies�����������������������������������������������������������������������������������������������������171 I. Why Organisations Adopt ‘Best Practice’���������������������������������������������172 II. Overview of the Case Studies����������������������������������������������������������������178 III. Overview of the Case Study Organisations������������������������������������������179 A. Case Study 1: UK Small Enterprise�����������������������������������������������179 B. Case Study 2: UK Large Enterprise�����������������������������������������������181 C. Case Study 3: UK Council�������������������������������������������������������������183 D. Case Study 4: Finnish Local Council��������������������������������������������185 E. Case Study 5: Finnish Manufacturing Company�������������������������188 IV. Case Study Findings������������������������������������������������������������������������������190 A. Drivers of Best Practice�����������������������������������������������������������������190 (i) Business Factors������������������������������������������������������������������190 (ii) Organisational Leadership��������������������������������������������������191 (iii) Organisational Culture�������������������������������������������������������192 (iv) Law and Legislation�������������������������������������������������������������192 (v) Unions����������������������������������������������������������������������������������195 V. Discussion����������������������������������������������������������������������������������������������197 A. Drivers of Best Practice�����������������������������������������������������������������197 B. Impact of Organisational Characteristics on Best Practice������������������������������������������������������������������������������198 C. Impact of National Context on Best Practice�������������������������������202 VI. Conclusion���������������������������������������������������������������������������������������������204 8. Constructing Future Scenarios—The Delphi Method���������������������������������206 I. The Delphi Survey���������������������������������������������������������������������������������206 II. Results����������������������������������������������������������������������������������������������������210 A. Scenario 2: Each Workplace Adopts an Age Management Approach to its Workforce�������������������������������������210 B. Scenario 4: Easily Accessible Information and Guidance about Managing and Working in an Ageing Workforce is Available to Employers and Employees.�����������������������������������212 C. Scenario 5: Effective Leadership Ensures the Ageing Workforce is an Issue of High Priority�����������������������������������������214 D. Scenario 6: A National Dialogue Regarding what Rights, Obligations and Privileges Should be Granted to Individuals at Different Ages is Undertaken����������������������������214 E. Scenario 7: Employers Have a Positive Duty to Achieve Employment Equality, Including for Older Workers��������������������������������������������������������������������������215 F. Scenario 8: Employers have a Responsibility to Consult with Employees and/or their Representatives in Relation to Equality Matters��������������������������216 G. Do Nothing������������������������������������������������������������������������������������217

Contents

 xiii

III. Analysis���������������������������������������������������������������������������������������������������217 IV. Conclusion���������������������������������������������������������������������������������������������218 9. Conclusion������������������������������������������������������������������������������������������������������220 I. How are the Laws Operating in Practice?���������������������������������������������220 II. How might the Laws be Improved?������������������������������������������������������222 III. Conclusion���������������������������������������������������������������������������������������������228 Appendix 1—Glossary of Statistical Terms and Tests������������������������������������������230

Index�����������������������������������������������������������������������������������������������������������������������233

xiv

LIST OF DEFINED TERMS AND ABBREVIATIONS

AAPS

Age-aware practice score

Acas

UK Advisory, Conciliation and Arbitration Service

BIS

UK Department for Business Innovation & Skills

CBI

Confederation of British Industry

CIPD

Chartered Institute of Personnel and Development

CJEU

Court of Justice of the European Union

DRA

Default Retirement Age

DTI

UK Department of Trade and Industry (now BIS)

DWP

UK Department for Work and Pensions

EAT

Employment Appeal Tribunal

Economic inactivity  Not in employment and either not seeking work or not ­available for work EHRC

UK Equality and Human Rights Commission

EJRA

Employer-Justified Retirement Age

EqA

Equality Act 2010

ET

Employment Tribunal

EU

European Union

FIOH

Finnish Institute of Occupational Health

Framework Directive Council Directive 2000/78/EC of 27 November 2000 establishing a general framework for equal treatment in ­employment and occupation GDP

Gross Domestic Product

HR

Human Resource

IQR

Inter-quartile range

LE

Large enterprise

xvi 

List of Defined Terms and Abbreviations

Member State

Member State of the European Union

NPAW

Finnish National Programme for Ageing Workers

NRA

Normal Retirement Age

OECD

Organisation for Economic Co-operation and Development

ONS

UK Office for National Statistics

PLC

Public Limited Company

PMGOL

Professional Match Game Officials Ltd

PSED

Public Sector Equality Duty

PWD

Public Works Department at the Finnish Council

SME

Small and medium enterprise

SPA

State Pension Age

TAEN

The Age and Employment Network

UK

United Kingdom

US/USA

United States of America

WERS

Workplace Employment Relations Study

WIER

Warwick Institute for Employment Research

1 Introduction The UK population is ‘ageing rapidly’.1 Government projections indicate that England will experience a 51 per cent increase in those aged over 65 and a 101 per cent increase in those aged over 85 from 2010 to 2030.2 As the ‘baby boom’ generation of the 1960s approaches the state pension age (SPA), the median age of the UK population is projected to rise from 39.7 years in 2010 to 42.2 years by 2035.3 Population ageing is likely to have consequences for industry and government services, including housing, pensions, health and social care, and employment. While legislative and policy measures have been introduced to respond to these challenges, government and society are still regarded as being ‘woefully underprepared’ for demographic change due to a ‘collective failure to address the implications’ of ageing.4 This study critically examines the efficacy of one area of government policy in addressing the ‘ageing challenge’. Using legal doctrinal and empirical mixed ­methods, I investigate whether age discrimination laws are effectively addressing ageing issues and challenges in the context of employment. Age discrimination laws are seen as having broad potential to achieve instrumental and intrinsic objectives. However, my research demonstrates the substantial limitations of age discrimination laws for fulfilling these aims. Drawing on qualitative expert interviews, statistical analysis and organisational case studies, I illustrate the failure of age discrimination laws to achieve attitudinal change in the UK, and reveal the limited prevalence of proactive measures to support older workers at an organisational level. Integrating doctrinal analysis, comparative analysis of Finnish law, and the Delphi method, I develop an agenda for reform and propose targeted legal and policy changes to address these limitations.

1  Select Committee on Public Service and Demographic Change, ‘Ready for Ageing?’, Report of ­Session 2012–13, HL Paper 140 (London, The Stationary Office, 14 March 2013) 7. 2 ibid. 3 Office of National Statistics, ‘UK Population Projected to Reach 70 Million by Mid-2027’ (26–26 2011): www.ons.gov.uk/ons/rel/npp/national-population-projections/2010-based-projections/ sum-2010-based-national-population-projections.html; see also European Commission, Active Ageing and Solidarity between Generations: A Statistical Portrait of the European Union 2012 (Publications Office of the European Union, 2011) 27. 4  Select Committee on Public Service and Demographic Change (n 1 above) 7.

2 

Introduction

In this chapter, I examine why ageing is likely to be an issue of increasing significance for labour law, and consider the role of age discrimination legislation in addressing the ageing ‘challenge’. I also provide an overview of the book’s structure.

I.  Implications of Ageing for Labour Law Older workers were traditionally a feature of the UK labour market. It was only in 1908 that a state pension was introduced for the ‘very old, the very poor, and the very respectable’ at a level (just) sufficient for survival.5 For most of the UK’s ­history, older workers have been expected to remain in employment until no longer capable of working, thereafter becoming dependent on family support or (in recent times) the welfare state. In the 1970s, the growth of early retirement schemes, private pensions and widespread redundancies targeting older workers caused participation rates for older workers to decline significantly.6 By the end of the twentieth century, a third of individuals in the UK aged between 50 and the SPA did not participate in paid employment.7 In the last decade, government and employer programmes have focused on counteracting this culture of ‘early exit’ and again increasing employment rates for older workers, both by introducing measures to retain older workers and changing pension entitlements to deter early exit.8 However, early exit remains a significant feature of the UK labour market.9 Tables 1 and 2 show the employment and inactivity rates for those aged over 50 in the UK. While levels of participation for older workers have increased dramatically since 1992,10 the tables still demonstrate that a substantial proportion of older workers are currently not participating in employment. Indeed, over 90 per cent of those over 65 years of age are not involved or seeking to be involved in the labour market. 5 L Hannah, Inventing Retirement: The Development of Occupational Pensions in Britain (Cambridge, Cambridge University Press, 1986) 15–16; P Thane, Old Age in English History: Past Experiences, Present Issues (Oxford, Oxford University Press, 2000) 225. 6  A-M Guillemard and M Rein, ‘Comparative Patterns of Retirement: Recent Trends in Developed Societies’ (1993) 19 Annual Review of Sociology 469, 474; F von Nordheim, ‘Responding Well to the Challenge of an Ageing and Shrinking Workforce: European Union Policies in Support of Member State Efforts to Retain, Reinforce and Re-Integrate Older Workers in Employment’ (2004) 3 Social Policy and Society 145, 145. 7  S Fredman, ‘The Age of Equality’ in S Fredman and S Spencer (eds), Age as an Equality Issue: Legal and Policy Perspectives (Oxford, Hart Publishing, 2003) 25. 8 C Phillipson, ‘Work and Retirement Transitions: Changing Sociological and Social Policy Contexts’ (2004) 3 Social Policy and Society 155, 156. 9  von Nordheim (n 6 above) 145; W Loretto et al, ‘Older Workers and Options for Flexible Work’, EOC Working Paper No 31 (Equal Opportunities Commission, 2005) 1. 10  This may be partly connected to the changing gender composition of the workforce, and increased participation rates for women.

Implications of Ageing for Labour Law

 3

Table 1:  UK Employment Rates for those over 50 (Source: ONS Labour Force Survey) Employment Rate (per cent) 50–64 65+

1992

2001

2013

56.6

62.0

67.4

5.6

4.9

9.5

Table 2:  UK Economic Inactivity Rates for those over 50 (Source: ONS Labour Force Survey) Economic Inactivity Rate (per cent) 1992

2001

2013

50–64

38.5

36.0

29.2

65+

94.1

95.0

90.2

Despite this low level of participation, it appears that older workers are again likely to become a feature of the UK labour market. First, advancements in medical care and improved living conditions mean that individuals are living longer and can reasonably expect substantially more productive, healthy years in their old age.11 Using 2008–10 mortality rates, a man aged 65 can expect to live for another 17.8 years and a woman for another 20.4 years.12 A substantial proportion of these years are likely to be enjoyed in reasonable health:13 in 2011, UK women had an average expectancy of 11.9 healthy life years at age 65 and men had 11.1 years (up from 9.5 and 8.2 respectively in 2001).14 Therefore, individuals are likely to be capable of working for a longer period in old age. Secondly, there is evidence that a substantial number of individuals will have inadequate income in retirement, meaning they will be financially compelled to continue in employment. In 2008–09, UK government pension benefits15 for a median earner represented only 37 per cent of average UK earnings, one of the lowest wage replacement rates in the Organisation for Economic Co-operation and Development (OECD).16 The UK is significantly more reliant on private 11 

European Commission (n 3 above) 18. ONS, ‘UK Interim Life Tables, 1980–82 to 2008–10’ (29 September 2011): www.ons.gov.uk/ons/ rel/lifetables/interim-life-tables/2008-2010/sum-ilt-2008-10.html. 13 MW Riley and JW Riley Jr, ‘Age Integration and the Lives of Older People’ (1994) 34 The ­Gerontologist 110, 110; European Commission (n 3 above) 28–29. 14  Eurostat, ‘Healthy Life Years (1995–2003)’ (Eurostat, 22 May 2013): open-data.europa.eu/en/data/ dataset/SkByEcjkoZu4yUcxSRNLw; Eurostat, ‘Healthy Life Years and Life Expectancy at Age 65, by Sex’ (Eurostat, 17 October 2013): epp.eurostat.ec.europa.eu/tgm/table.do?tab=table&init=1&plugin=1&la nguage=en&pcode=tsdph220. 15  Including the flat-rate basic pension, plus any earnings-related additional pension and pension credit. However, the figure does not include mandatory or voluntary private pensions. 16  OECD, ‘Pensions at a Glance 2011: Retirement-Income Systems in OECD and G20 Countries’ (OECD, 2011) 118–19. 12 

4 

Introduction

pension provision than other countries.17 However, the ‘truly spectacular flight away’ from defined benefit (so-called ‘final salary’) pension schemes18 in the private sector19 is likely to jeopardise this reliance on private pension provision. At 31 March 2011, only 16 per cent of defined benefit occupational schemes were still open to new members.20 Some employers are now offering their employees defined contribution pension schemes, where the risks of longevity and investment returns are shifted to the individual, rather than employers or the ­government.21 Other employers had (until recently) ceased to make any provision for their employees’ retirement: the 2009 Employers’ Pension Provision Survey of 2,519 UK organisations showed that employers making any pension provision for their employees had declined from 41 per cent in 2007 to 28 per cent in 2009, with occupational pension schemes being provided in only two per cent of private ­sector firms (compared with five per cent in 2007).22 The roll-out of pension ‘automatic enrolment’ under the Pensions Act 2008, section 3 from October 2012 may help to extend private pension coverage to many employees. Employers are progressively being required to enrol workers into a workplace pension scheme if they are aged between 22 and the SPA, earn more than a specified minimum wage and work in the UK. However, the level of saving required under the scheme (which is eventually increasing to eight per cent of an employee’s income) is unlikely to address chronic issues of ‘under-saving’ for retirement.23 According to 2012 figures, 10.7 million individuals in Great Britain are likely to have inadequate income in retirement.24 Participation rates for older workers are likely to increase as individuals face stark financial choices and defined benefit pensions become historical relics. Thirdly, and relatedly, while the low level of public pension provision in the UK has ensured that the state pension system is more sheltered from the effects of

17  S Diepeveen, ‘Impact of the Recession on Age Management Policies: United Kingdom’ (Dublin, Eurofound, 2012) 10. 18 Defined benefit or salary-related pension schemes pay pensions based on individuals’ pensionable earnings and years of service. The schemes pay a guaranteed pension upon retirement, irrespective of investment returns. In contrast, in defined contribution schemes individuals build up a pension ‘pot’ which may be used to purchase an annuity upon retirement. The level of pension is not guaranteed and is linked to investment returns. 19 C Kilpatrick, ‘The New UK Retirement Regime, Employment Law and Pensions’ (2008) 37 ­Industrial Law Journal 1, 21. 20  The Pensions Regulator, ‘Annual Report and Accounts 2010–2011’ (30 June 2011) 15. 21  D Hirsch, ‘Crossroads after 50: Improving Choices in Work and Retirement’, Research Report (York, Joseph Rowntree Foundation, 2003) 7; P Thornton, ‘A Note on the Investment Management of Defined Contribution Schemes’ in P Thornton and D Fleming (eds), Good Governance for Pension Schemes (Cambridge, Cambridge University Press, 2011) 265. 22  J Forth and L Stokes, ‘Employers’ Pension Provision Survey 2009’, Research Report No 687 (DWP, 2010) 15–16. 23  Select Committee on Public Service and Demographic Change (n 1 above) 10. Individuals may also ‘opt-out’ from the scheme, which may reduce its effectiveness. 24  ibid 7.

Implications of Ageing for Labour Law

 5

an ageing population than other EU countries,25 the UK is still facing a recurrent ‘pensions crisis’.26 As the ‘baby boom’ generation ages, the number of people over the SPA in the UK is expected to increase by 28 per cent, rising from 12.2 million people in 2011 to 15.6 million people by 2035.27 Further, people are living longer, which means they are likely to draw on state pension entitlements for a longer period of time. Indeed, some individuals will spend nearly a third of their life in retirement.28 National Insurance Fund expenditure is projected to increase from around five per cent of GDP in 2008–09 to eight per cent in 2070–71 as a result of the ageing population.29,30 To mediate this pension ‘crisis’, and promote the long-term sustainability of the UK public pension system, the coming years will see a number of increases to the SPA. For women, the SPA will increase from 60 to 65 by 2018.31 For both women and men, the SPA will increase to 66 by 2020 and again to 67 years of age between 2026 and 2028.32 It is anticipated that this latter increase will save the government around £60 billion in today’s prices between 2026–27 and 2035–36.33 The age at which a company or personal pension can be claimed was also increased in 2010 from 50 to 55 years old, encouraging workers with other pension entitlements to remain in employment for longer. Fourthly, some older workers may actually want to stay in employment into old age. Previous studies have found that many older workers remain in employment as they enjoy work, feel loyalty to the organisation, have a sense of contributing to society or a cause, or feel their work has intrinsic value.34 Therefore, older workers may wish to remain in work for its intrinsic or social value. 25  C Phillipson, ‘Transitions from Work to Retirement: Developing a New Social Contract’ (Bristol, Policy Press, December 2002) 11–12. 26  J Harris, ‘The Roots of Public Pensions Provision: Social Insurance and the Beveridge Plan’ in H Pemberton et al (eds), Britain’s Pensions Crisis: History and Policy, British Academy occasional paper 7 (Oxford, Oxford University Press, 2006) 34. 27  ONS (n 3 above). 28  Riley and Riley Jr (n 13 above) 110. 29 Government Actuary’s Department, ‘Government Actuary’s Quinquennial Review of the National Insurance Fund as at April 2005’ (London, The Stationary Office, March 2010) 16. 30  To put this in perspective, in 2009 public health care spending accounted for 8.2% of UK GDP (dropping slightly to 8.0% in 2010): U Qaiser, ‘Expenditure on Healthcare in the UK’ (May 2011) 4. This is projected to rise to 8.8% of GDP in 2062–63, with an additional 2.4% being spent on long-term care: Office for Budget Responsibility, ‘Fiscal Sustainability Report’ (London, The Stationery Office, July 2013) 80. 31  Pensions Act 2011. 32  HM Treasury and HM Parliament, Autumn Statement 2011 (London, The Stationery Office, 2011) 23. 33  ibid 6. 34  H Barnes et al, ‘Working after State Pension Age: Qualitative Research’, Research Report No 208 (DWP, 2004) ch 4; J Drew and M Drew, The Process of Participation and Phased Retirement: Evidence from Mature-Aged Workers in Australia (Brisbane, Post Pressed, 2005) 101–2; W Loretto and P White, ‘Work, More Work and Retirement: Older Workers’ Perspectives’ (2006) 5 Social Policy and Society 495, 499; S McNair, ‘How Different is the Older Labour Market? Attitudes to Work and Retirement among Older People in Britain’ (2006) 5 Social Policy and Society 485, 486; S Vickerstaff, ‘Entering the

6 

Introduction

Fifthly, the decline in the proportion of ‘prime age workers’ in their 30s and 40s as the population ages may force employers to draw on an older workforce. Labour supply of 20–64 year olds across the EU is predicted to decrease by 24.5 million people between 2010 and 2050.35 While this is unlikely to create a generalised labour shortage, particularly as there is already a large pool of unemployed and inactive workers who could be integrated into the labour market,36 employers will need to look beyond the traditional labour market to supplement their ­workforce,37 including by employing older workers. Thus, older workers are again likely to become a significant feature of the UK labour market.

II.  The Role of Discrimination Law for an Ageing Population While older workers may want or need to remain in work, and governments are seeking to extend working lives into old age, it is unclear whether there are adequate opportunities for older workers to be gainfully employed in the UK. There is limited evidence of whether employers are prepared for an ageing workforce or willing to accommodate the needs of older workers. Indeed, ageism and a preference for ‘prime age’ workers may significantly limit opportunities for older workers in the labour market.38 In this context, age discrimination laws are seen as a key mechanism for addressing negative attitudes towards older workers and encouraging employers to adopt a more rational (and less age-discriminatory) approach to their workforce. Age discrimination legislation was first introduced in the UK in 2006, implementing Council Directive 2000/78/EC of 27 November 2000 establishing a general framework for equal treatment in employment and occupation (‘the Framework ­Directive’)39 in UK law (see Chapter three). Previous government attempts to ‘persuade’ organisations not to discriminate on the basis of age prior to the introduction of age discrimination legislation, including via a Voluntary Code of Practice on ­ etirement Zone: How Much Choice Do Individuals Have?’ (2006) 5 Social Policy and Society 507, 509; R S Vickerstaff, ‘Older Workers: The “Unavoidable Obligation” of Extending Our Working Lives?’ (2010) 4 Sociology Compass 869, 873, 875; G Morrell and R Tennant, ‘Pathways to Retirement: The Influence of Employer Policy and Practice on Retirement Decisions’, Research Report No 673 (DWP, 2010) 26–27. 35 European Commission, ‘The 2012 Ageing Report: Underlying Assumptions and Projection Methodologies’, European Economy 4/2011 (European Commission, 2011) 95. 36  M Herrmann, ‘Population Aging and Economic Development: Anxieties and Policy Responses’ (2012) 5 Journal of Population Ageing 23, 28. 37  European Commission (n 3 above) 10; D Schiek, ‘Age Discrimination before the ECJ—­Conceptual and Theoretical Issues’ (2011) 48 Common Market Law Review 777, 779. 38 See Age UK, ‘Grey Matters—a Survey of Ageism across Europe: EU Briefing and Policy ­Recommendations’ (June 2011) 11–12. 39  Council Directive 2000/78/EC of 27 November 2000 establishing a general framework for equal treatment in employment and occupation [2000] OJ L303/16.

The Role of Discrimination Law for an Ageing Population

 7

Age ­Discrimination, made limited progress at addressing age discrimination in employment.40 Thus, the UK provides significant evidence that the market is unable to resolve issues of age discrimination without legislative intervention: a 2003 study of 1,804 adults aged 18–64 in work or looking for work found that three per cent of respondents reported being bullied or harassed and six per cent reported being discriminated against on the basis of age at work in the previous 12 months.41 Thus, while age discrimination legislation is merely one tool for improving the treatment of older workers, and can be used in conjunction with other initiatives, it is clear that soft law alone is not sufficient for addressing age discrimination in employment. Therefore, age discrimination legislation has a fundamental role to play in achieving concrete organisational change. Age discrimination laws may achieve two broad aims: first, they may secure instrumental economic ends, in being used as a means of relieving pressure on pension systems or the labour market from demographic ageing; and, secondly, they may achieve intrinsic ends, having value in and of themselves in acknowledging the dignity and inherent worth of older workers. These dual aims, and how they may interact, are explored below.

A.  Instrumental Ends Age discrimination laws are focused on increasing participation rates for older workers by preventing employers from irrationally discriminating against workers on the grounds of age. In theory, by preventing irrational age discrimination, and reducing discriminatory barriers that prevent older people from entering and remaining in employment, the laws should increase opportunities for older workers to gain and remain in work and promote more efficient employment practices. By increasing opportunities for older workers, age discrimination laws may help to reduce the social costs related to longer life spans, extended periods in retirement and inadequate pension provision.42 Further, by encouraging employers to retain older workers and look beyond the traditional ‘prime age’ workforce, the laws may promote ‘a skilled, trained and adaptable labour force’ with a broader range of experience and expertise.43 This reflects ‘crowding theory’, the idea that employers will benefit from a broader pool of employee talent if discriminatory barriers 40 D Grant, ‘Older Women, Work and the Impact of Discrimination’ in M Sargeant (ed), Age Discrimination and Diversity: Multiple Discrimination from an Age Perspective (Cambridge, Cambridge University Press, 2011) 42; see also CIPD, ‘Tackling Age Discrimination in the Workplace: Creating a New Age for All’ (October 2005). 41  Ipsos MORI, ‘Workplace Discrimination’ (13 October 2003): www.ipsos-mori.com/ researchpublications/researcharchive/948/Workplace-Discrimination.aspx. 42  S Fredman, Discrimination Law, Clarendon Law Series, 2nd edn (Oxford, Oxford University Press, 2011) 103. 43  B Hepple, ‘Age Discrimination in Employment: Implementing the Framework Directive 2000/78/ EC’ in S Fredman and S Spencer (eds), Age as an Equality Issue: Legal and Policy Perspectives (Oxford, Hart Publishing, 2003) 71–74.

8 

Introduction

are eliminated,44 and the belief that businesses will retain important skills and ­knowledge by retaining older workers.45 Therefore, age discrimination laws may have significant instrumental benefits in promoting efficiency at the ­organisational and national level. These aims are reflected in the rationale for UK age discrimination laws: the Regulatory Impact Assessment for the Employment Equality (Age) Regulations 2006, SI 2006/1031 says: The legislation should, both by providing protection to those subject to discriminatory practices and by stimulating a cultural change, increase the participation of older and younger workers in the economy, while at the same time helping employers draw on a wider pool of workers.46 Discrimination results in poorer quality matches in labour markets, which leads to lower national output … The risk that exists currently is that [older and younger workers] are not participating fully in the workforce and so human resources are wasted.47

Age discrimination laws are geared, therefore, to achieving organisational and economic efficiency, both by promoting workplace efficiency and by avoiding market failure due to irrational prejudice. Efficiency is viewed as a rationale, justification and outcome of discrimination law, via the ‘business case’ for age equality: Age discrimination prevents people of all ages from realising their full potential in the workplace. This in turn prevents employers from getting the best performance out of their business and delivering the best service to their customers.48

These statements indicate that instrumental economic ends drive UK age discrimination laws. However, it is clear that age discrimination laws can only go so far to increase employment rates for older workers: individual decisions to retire (and, in particular, to retire early) may turn on the availability of a pension, employer policies (such as flexible work practices), enjoyment of work and loyalty to the organisation, personal health, domestic caring responsibilities and relationships and whether work is viewed as having intrinsic value.49 At the 44 ACL Davies, Perspectives on Labour Law, Law in Context, 2nd edn (Cambridge, Cambridge ­University Press, 2009) 129. 45 Fredman, Discrimination Law (n 42 above) 105. 46 Regulatory Impact Assessment for the Employment Equality (Age) Regulations 2006, SI 2006/1031, [3]. 47  ibid [5]. 48  DTI, ‘Equality and Diversity Coming of Age: Consultation on the Draft Employment Equality (Age) Regulations 2006’ (July 2005) 6. 49  C Phillipson, ‘The Transition to Retirement’ in G Cohen (ed), Social Change and the Life Course (London, Tavistock Publications, 1987) 161; M Kohli and M Rein, ‘The Changing Balance of Work and Retirement’ in M Kohli et al (eds), Time for Retirement: Comparative Studies of Early Exit from the Labor Force (Cambridge, Cambridge University Press, 1991) 9–11; Fredman, ‘The Age of Equality’ (n 7 above) 26; Barnes et al (n 34 above) ch 4; Drew and Drew (n 34 above) 98–106; Loretto and White (n 34 above) 498–500, 503; McNair (n 34 above) 486; Vickerstaff, ‘Entering the Retirement Zone: How Much Choice Do Individuals Have?’ (n 34 above) 509; Vickerstaff, ‘Older Workers’ (n 34 above) 873, 875; Morrell and Tennant (n 34 above) 26–27, 30, 43, 45. While recognising the broad range of factors that may affect and influence older workers, this book focuses solely on age discrimination laws.

The Role of Discrimination Law for an Ageing Population

 9

same time, age d ­ iscrimination laws are a key means of ensuring that employers provide opportunities for older workers to stay at work, and can increase individual motivation to work. Conversely, experiencing age discrimination is associated with intention to retire.50 Therefore, and despite their limitations, age discrimination laws have a key role to play in achieving instrumental economic ends.

B.  Intrinsic Ends Through their pursuit of age ‘equality’, age discrimination laws also recognise the inherent dignity and worth of individuals of all ages and the injustice of age discrimination.51 While equality is seen as both an individual and social good, this has two fundamental limitations. First, age discrimination may be necessary or justified in some circumstances.52 Age discrimination is seen as ‘different’ to other forms of discrimination: unlike other grounds, age is not an unchanging or acquired attribute53 and changes over time, being a universal and temporal phenomenon with no fixed characteristics.54 Further, everyone has an age, meaning there is no ‘single, clearly defined, oppressed group’ who may benefit from ­protection.55 While this brings non-disabled, heterosexual white men ‘into the fold of discrimination law lobbyists’,56 it also undermines the age discrimination cause by accommodating those in a comparatively advantageous position who may be perceived as less ‘deserving’ of protection by equality law. Age may also be viewed with ambivalence due to an enduring social fear of the ageing process and the assumed association between ageing and declining capacity.57 Therefore, while age discrimination is undesirable, it is often seen as less undesirable than other

50  E Snape and T Redman, ‘Too Old or Too Young? The Impact of Perceived Age Discrimination’ (2003) 13 Human Resource Management Journal 78, 86. 51 Fredman, Discrimination Law (n 42 above) 104. 52  ibid 198. 53  M Sargeant, ‘Mandatory Retirement Age and Age Discrimination’ (2004) 26 Employee Relations 151, 154. 54  N Adnett and S Hardy, ‘The Peculiar Case of Age Discrimination: Americanising the European Social Model?’ (2007) 23 European Journal of Law & Economics 29, 38; M Sargeant, ‘The European Court of Justice and Age Discrimination’ [2011] Journal of Business Law 144, 146. See also Seldon v Clarkson Wright & Jakes (A partnership) [2012] UKSC 16 (25 April 2012), [2], [4]; Homer v Chief Constable of West Yorkshire Police [2012] UKSC 15, [2012] ICR 704. However, this may equally apply to other protected characteristics: gender is also a social construction, rather than a biological given, and (particularly with advances in medical technology) is not necessarily a fixed or unchanging attribute: see, eg, J Lorber, Paradoxes of Gender (New Haven CT, Yale University Press, 1994) 13–36. As a result, age may not be inherently different from other protected characteristics. 55  C Duncan, ‘Assessing Anti-Ageism Routes to Older Worker Re-Engagement’ (2003) 17 Work, Employment & Society 101, 108. 56  Schiek (n 37 above) 778. 57  B Bytheway, ‘Ageism’ in ML Johnson et al (eds), The Cambridge Handbook of Age and Ageing (Cambridge, Cambridge University Press, 2005) 339; C Duncan, ‘The Dangers and Limitations of Equality Agendas as Means for Tackling Old-Age Prejudice’ (2008) 28 Ageing & Society 1133, 1143.

10 

Introduction

forms of discrimination and, indeed, potentially beneficial and efficient in an array of circumstances.58 The issue, then, is how to distinguish between legitimate and illegitimate age discrimination.59 The law’s current approach to this issue is ­discussed in Chapter three. In summary, while age discrimination is prohibited in general, both direct and indirect discrimination can be justified. The balance that has been struck clearly reflects the fundamental tension between the instrumental and intrinsic ends of age discrimination laws.60 Secondly, there is limited agreement as to what ‘equality’ entails at both the UK and EU level.61 ‘Equality’ is not a unitary concept, and what it entails in practice is not straightforward.62 While the choice between different conceptions of equality is ultimately a matter for policy and value judgements, not logic,63 UK governments do not appear to have made a choice between competing interpretations. Indeed, Hepple identifies seven meanings of ‘equality’ evident in the Equality Act 2006 and government equality reviews: —— Respect for equal worth, dignity and identity as fundamental human rights; —— Eliminating status discrimination and disadvantage; —— Consistent treatment/formal equality; —— Substantive equality of opportunity; —— Equality of capabilities; —— Equalities of outcomes; and —— Fairness.64 Alternatively, equality could be defined as encompassing: —— Consistency (eg like individuals being treated alike, ‘formal equality’); —— Individual merit (eg treating individuals according to merit, free from ­stereotypical assumptions); —— Treating individuals differently according to their needs; —— Achieving a fair distribution of social resources (eg preventing certain groups from bearing particular burdens on the grounds of group membership); —— Equality of opportunities (eg giving individuals an equal set of alternatives from which to choose to pursue their idea of the ‘good life’); 58  Sargeant (n 53 above) 154; S McNair and M Flynn, ‘The Age Dimension of Employment Practices: Employer Case Studies’, Employment Relations Research Series No 42 (DTI, June 2005) 11. 59 This is explicitly acknowledged in the Preamble to the Framework Directive, which states: ­‘differences in treatment in connection with age may be justified under certain circumstances and therefore require specific provisions which may vary in accordance with the situation in Member States. It is therefore essential to distinguish between differences in treatment which are justified … and discrimination which must be prohibited’ Recital [25]. See Chapter 3. 60 Fredman, Discrimination Law (n 42 above) 106. 61  C O’Cinneide, ‘The Uncertain Foundations of Contemporary Anti-Discrimination Law’ (2011) 11 International Journal of Discrimination and the Law 7, 10–11. 62  See Fredman, ‘The Age of Equality’ (n 7 above) 37–46. 63 Fredman, Discrimination Law (n 42 above) 2. 64  BA Hepple, Equality: The New Legal Framework (Oxford, Hart Publishing, 2011) 13–24.

The Role of Discrimination Law for an Ageing Population

 11

—— Treating individuals with equal dignity and concern; and/or —— Full participation and inclusion in social institutions.65 Reconciling these forms of equality, and the substantive underpinnings of equality law, Fredman advocates a ‘multi-dimensional concept’ of substantive equality, p ­ ursuing redistributive, recognition, transformative and participative dimensions.66 These varied interpretations of equality continue to be evident in UK age ­discrimination laws.67 UK government policy is ‘built on two principles of ­equality: equal treatment and equal opportunity’.68 Equality is seen as ‘giving ­everyone an equal right to be treated fairly as an individual … and giving them an equal opportunity to progress’.69 However, government documents increasingly prioritise formal equality of opportunity over other conceptions of equality: indeed, the Regulatory Impact Assessment for the Employment Equality (Age) Regulations 2006, SI 2006/1031 says: The aim of the legislation is to maximise the participation and economic (and social) contribution of groups that are currently subject to discriminatory practices both inside and outside the labour market because of their age … the Government [aims] to improve opportunities and choice for individuals, and encourage labour market participation, whilst still allowing employers to manage their businesses effectively.70

This is consistent with other policy statements: [Our] goal … is prosperity for all. We want to promote the opportunity for all to contribute to and share in that prosperity. This is why we are committed to tackling age discrimination. It is only fair that those who experience age discrimination should have the same opportunities as others.71

However, there is also recognition that equality of opportunity does not mean uniformity of treatment. Rather, ‘it means giving everyone an equal right to be treated fairly as an individual, recognising both their needs and their talents and giving them an equal opportunity to progress’.72 In contrast to this minimalistic focus on equal treatment and formal equality of opportunity, Fredman argues for a more outcomes-focused approach, where 65  See Fredman, ‘The Age of Equality’ (n 7 above) 37–46; see also Fredman, Discrimination Law (n 42 above) 8–19. 66 Fredman, Discrimination Law (n 42 above) 25–33. See also B Hepple, ‘Equality at Work’ in B Hepple and B Veneziani (eds), The Transformation of Labour Law in Europe: A Comparative Study of 15 Countries 1945–2004 (Oxford, Hart Publishing, 2009) 154–60 on the emergence of transformative equality. 67  See Fredman, Discrimination Law (n 42 above). 68 HM Government, ‘The Equality Strategy—Building a Fairer Britain’ (Govt Equalities Office, December 2010) 4. 69  ibid 9. 70 Regulatory Impact Assessment for the Employment Equality (Age) Regulations 2006, SI 2006/1031, [2]. 71  DTI (n 48 above) 4. 72  HM Government (n 68 above) 9.

12 

Introduction

the aim of equality is to ‘facilitate equal participation of all in society, based on equal concern and respect for the dignity of each individual’.73 Age discrimination legislation should arguably recognise the equal worth of human beings of all ages74 and the need for dignity, autonomy and respect for all age groups.75 By focusing on equal treatment and equality of opportunity, age discrimination laws may neglect and minimise intrinsic ends. Indeed, it is arguable that UK age discrimination laws are failing to achieve intrinsic ends in practice (see Chapter three).

C.  Reconciling Conflicting Objectives Even a minimalist understanding of ‘equality’ may come into conflict with instrumental ends. Some scholars argue that age discrimination law is unnecessary, ‘inefficient’ and even harmful, and that the market should be allowed to resolve any issues of discrimination through the operation of competitive pressures.76 In this view, ‘equality is costly’ and likely to limit organisational efficiency.77 While this likely overstates the issue, there may be some net costs associated with age equality78 and commercial rationality can co-exist with discrimination.79 ­Therefore, intrinsic and instrumental ends may well come into conflict. The real issue, then, is how the costs of equality should be allocated,80 and how age discrimination legislation should reconcile competing objectives. There has been very limited consideration of how conflicting policy objectives (or conflicting interpretations of different objectives) should be reconciled or managed.81 This reflects the failure of governments to put forward a vision or coherent ­strategy for addressing demographic change.82 Indeed, it is debatable whether labour law can or should be used to reconcile these competing objectives. In reality, where the pursuit of equality or intrinsic objectives is perceived as coming into conflict with organisational efficiency, the UK government has ­chosen to adopt a ‘light touch’ approach to discrimination legislation to

73 

Fredman, ‘The Age of Equality’ (n 7 above) 21. Smith, Basic Equality and Discrimination: Reconciling Theory and Law, Applied Legal ­Philosophy (Farnham, Ashgate, 2011) 2. 75  See R Cotterrell, ‘Is Law Just a Means to an End’ (2008) 4 Socio-Legal Review 1, 7. 76  R Posner, Aging and Old Age (Chicago IL, University of Chicago Press, 1995) 319; RA Epstein, Equal Opportunity or More Opportunity? The Good Thing about Discrimination (London, Civitas, 2002) 19–20. 77  Posner (n 76 above) 324. 78  Fredman, ‘The Age of Equality’ (n 7 above) 49. 79 W Loretto et al, ‘Ageism and Employment: Controversies, Ambiguities and Younger People’s ­Perceptions’ (2000) 20 Ageing and Society 279, 284–85. 80  Fredman, ‘The Age of Equality’ (n 7 above) 49–50. 81 In the EU context, see AT Williams, ‘Promoting Justice after Lisbon: Groundwork for a New ­Philosophy of EU Law’ (2010) 30 Oxford Journal of Legal Studies 663. 82  Select Committee on Public Service and Demographic Change (n 1 above) 8. 74 N

Research Questions

 13

­ inimise its detrimental impacts on business.83 The pursuit of equality is m therefore ­ ­ subservient to business or efficiency concerns. This reflects the ­dominant view that market c­oncerns ‘trump’ equality.84 Indeed, a renewed focus on the ‘business case’ for equality may have led a shift away from equality, towards ‘diversity management’ and a lesser concern with social justice and intrinsic ends generally.85 Therefore, instrumental economic ends may displace equality, though there has been limited consideration of why this should be the case.86 Weller argues that the parameters of the equality discourse should be shifted to stress social and political concerns for ‘justice and universal rights to fair treatment’, rather than the economic business case for equality.87 If this were the case, ‘liberal notions of individual justice [could] challenge employers’ property rights’,88 creating the potential for a different weighting of societal objectives. In the chapters that follow, I explore the effectiveness of UK age discrimination laws in achieving both instrumental and intrinsic ends, and consider how the law reconciles these objectives where they conflict. However, the foregoing ­discussion demonstrates the tendency of UK age discrimination laws to prioritise ­instrumental ends over intrinsic objectives.

III.  Research Questions While age discrimination laws are seen as having broad potential to achieve instrumental and intrinsic objectives, it is unclear what impact they are having in practice. Serious questions remain as to the efficacy of these laws: are they promoting equality and reducing age discrimination in employment? Are they creating (quality) work opportunities for older workers? What can be done to improve the efficacy of the laws? In this book, I use empirical mixed methods to cast light on these complex legal issues. In doing so, I address two overarching questions, f­ ocusing on the employment context: —— How are UK age discrimination laws operating in practice? —— How (if at all) could UK age discrimination laws be improved?

83  See, eg, the Explanatory Memorandum to the Employment Equality (Age) Regulations 2006, SI 2006/1031, 2; Employment Relations Directorate, ‘Final Regulatory Impact Assessment: Age ­Discrimination—Summary’ (March 2006) 8. 84 Fredman, Discrimination Law (n 42 above) 35–37. 85  A Greene and G Kirton, ‘Trade Union Perspectives on Diversity Management’ in M Özbilgin (ed), Equality, Diversity and Inclusion at Work: A Research Companion (Cheltenham, Edward Elgar, 2009) 265–66. 86  S Fredman, Discrimination Law, Clarendon Law Series (Oxford, Oxford University Press, 2002) 25. 87  SA Weller, ‘Discrimination, Labour Markets and the Labour Market Prospects of Older Workers: What Can a Legal Case Teach Us?’ (2007) 21 Work, Employment & Society 417, 432. 88  ibid 432.

14 

Introduction

I approach these questions from an empirical perspective, using qualitative and quantitative research techniques and analysis to inform my normative views about how law should be (see Chapter two). In my research, I focus on demand-related issues facing older workers, with a particular emphasis on organisational practices and perspectives and how they impact upon an ageing workforce. While supply-related issues and individual perspectives are of fundamental importance to demographic ageing, they have been fairly comprehensively studied by other authors.89 In contrast, there has been a relative dearth of research focusing on employers and their response to the ageing workforce.90 This book therefore helps to correct this imbalance, through its focus on demand-side issues and employer practices and perspectives.

IV.  Chapter Overview In Chapter two I outline the reflexive law theoretical standpoint informing this book, and detail the ‘mixed methods’ research design employed in this study, which integrates qualitative, quantitative, doctrinal and comparative elements. I also explore how these various methods complement and enhance my overall research design. Drawing on the research framework in Chapter two, in Chapters three through to eight I present the results from each of the research methods. In Chapter three I detail doctrinal research on age discrimination legislation and case law to describe and identify potential limitations in the existing legal framework. I argue that a number of aspects of the law remain unclear and underdeveloped, particularly in relation to the use of employer-justified retirement ages (‘EJRAs’) and positive action in the UK. Further, broad exceptions to the prohibition of age discrimination may undermine the position of older workers. In Chapter four I present the results from 17 semi-structured qualitative expert interviews that explored expert views on UK age discrimination laws in late 2012. The interviews reveal that experts feel age discrimination laws in the UK are unclear and lack certainty. There was a strong feeling across many interviewees that employers are adopting a compliance-focused approach to

89 See, eg, Barnes et al (n 34 above) ch 4; Drew and Drew (n 34 above); Loretto and White (n 34 above) 499; McNair (n 34 above) 486; Vickerstaff, ‘Entering the Retirement Zone: How Much Choice Do Individuals Have?’ (n 34 above) 509; Vickerstaff, ‘Older Workers’ (n 34 above) 873, 875; S Manfredi, ‘Developing Good Practice in Managing Age Diversity in the Higher Education Sector: An Evidence-Based Approach’ (Centre for Diversity Policy Research and Practice, 2008); Morrell and Tennant (n 34 above) 26–27. 90  See W Loretto, ‘Work and Retirement in an Ageing World: The Case of Older Workers in the UK’ (2010) 5 Twenty-First Century Society 279, 280, 282.

Chapter Overview

 15

implementing age discrimination laws, meaning there is little proactive practice in workplaces to support older workers. Further, there was significant concern that attitudinal change is not keeping pace with legislative reforms, particularly among employers. While the experts generally did not feel that legal change was required, they noted that the implementation of age discrimination laws had ­significant limitations. Chapter five triangulates these qualitative and doctrinal findings with statistical analysis of the 2011 Workplace Employment Relations Study (WERS6), the flagship survey of employment relations in Britain. Corroborating the findings in Chapter four, I argue that employers are adopting a compliance-focused approach to implementing age discrimination laws, with limited evidence of proactive practice in workplaces to support older workers. While organisation size and type, union presence, and largest occupational group may be linked with the adoption of good practice, the ‘best’ WERS workplaces do not uniformly reflect these organisational characteristics. Panel data show that employers are adopting more ageaware policies and practices following the implementation of age discrimination legislation. However, past age-aware practices are strongly predictive of present practices, indicating that age discrimination legislation is having limited impact in some organisations. Chapter six considers a comparative perspective on age discrimination laws, with reference to the laws in Finland, to illuminate and critique the legal situation in the UK. Finland has the longest history of legal intervention in age discrimination in the EU91 and is notable for the relative success of its labour market interventions relating to older workers.92 Therefore, it is an interesting and worthwhile comparator for other EU countries. Drawing on comparative legal doctrinal analysis, 13 semi-structured qualitative interviews with Finnish experts and statistical data, I argue that Finnish laws demonstrate a more proactive, collective and employer-focused approach to preventing age discrimination. However, statistical data are ambivalent on relative outcomes in the two countries. Despite this, there are a number of lessons from the Finnish experience that may inform UK law and policy-making. Drawing on these results, Chapter seven details five organisational case studies of ‘best practice’ employers for older workers in the UK and Finland. It considers the role of law and age discrimination legislation in effecting organisational change, and draws conclusions on the impact of organisational structures and national context in promoting best practice. Law appears to be having limited impact on organisational practice, with business factors, organisational leadership and workplace culture as more significant drivers of good practice.

91 

Adnett and Hardy (n 54 above) 35. Sarfati, ‘Social Dialogue: A Potential “Highroad” to Policies Addressing Ageing in the EU Member States’ (2006) 59 International Social Security Review 49, 63. 92 H

16 

Introduction

Chapter eight draws together the results from previous chapters by presenting the findings from an online expert Delphi ‘roundtable’. Using this method, UK experts were asked to consider a range of scenarios derived from the literature, expert interviews and comparative analysis that might improve the operation of UK age discrimination laws. While most scenarios were regarded as being ­impracticable under a deregulatory governmental agenda, the respondents regarded some key changes as being important, desirable and practicable. I argue that these proposals should be seriously considered and vigorously pursued. Finally, Chapter nine concludes by evaluating the efficacy of UK age discrimination laws, and putting forward a programme for change.

2 Methodology In this book I employ empirical mixed methods to cast light on how UK age ­discrimination laws are operating in practice, and to develop ideas for their reform and improvement. This chapter sets out the methodological basis for this research, and provides an overview of the individual methods utilised as part of the research design. The research was conducted from a reflexive law theoretical standpoint, and employed a mixed methods research approach, incorporating qualitative, quantitative, doctrinal and comparative elements. Regulatory theory, including reflexive law, is broadly focused on how legal regulation may be made more effective.1 Therefore, it is well suited to addressing the research questions posed in this text.2

I. Methodology Legal scholars are often criticised for failing to adopt clear and transparent research methodologies.3 Few legal texts are self-reflective or self-aware regarding the assumptions that underlie the research process. This is of particular concern in legal empirical research, where clear research methodologies are essential to effectively frame, guide and justify appropriate research questions and methods and effectively interpret and comprehend empirical results. Research methodologies also offer additional insights and ideas that can enrich and consolidate the empirical research process. Therefore, in this book I explicitly acknowledge the theoretical and methodological assumptions underlying my research design and choices. By a ‘reflexive law’ standpoint, I mean a theory of law that focuses on the regulation of self-regulation, based on a view of law as an autopoietic system4 of ­communication.5 Reflexive law aims to cast light on the relationship between law 1 ACL Davies, Perspectives on Labour Law, Law in Context, 2nd edn (Cambridge, Cambridge ­University Press, 2009) 18. 2  See similarly C McCrudden, ‘Equality Legislation and Reflexive Regulation: A Response to the Discrimination Law Review’s Consultative Paper’ (2007) 36 Industrial Law Journal 255. 3  See L Epstein and G King, ‘The Rules of Inference’ (2002) 69 University of Chicago Law Review 1. 4  An autopoietic system is one capable of reproducing and maintaining itself via self-(re)production. 5  See R Rogowski and T Wilthagen, ‘Reflexive Labour Law: An Introduction’ in R Rogowski and T Wilthagen (eds), Reflexive Labour Law: Studies in Industrial Relations and Employment Regulation (Deventer, Kluwer Law and Taxation Publishers, 1994) 16.

18 

Methodology

and society by considering how law impacts upon other social systems. It also seeks to explain why law and legal regulation can have perverse and unexpected consequences. As a consequence, it is particularly relevant to questions of legal impact.

A.  Systems Theory Reflexive theory is grounded in systems theory, or a conception of law as a communicative system with self-referencing communications as its key element.6 Systems theory regards law and other social spheres (such as economic and political systems) as occupying functionally differentiated, discrete, closed, self-referencing and self-reinforcing autopoietic systems.7 Systems develop their own view of the world through ‘operationally closed autopoiesis’.8 While systems are operationally closed, and therefore lack a means of directly communicating, they are cognitively open9 and subject to ‘mutual interference’, as they ‘share the same world meaning’ and the elements of the systems are ‘essentially similar’.10 Cognitive openness means that systems are capable of evolving with reference to their external environment, and can respond to environmental inputs via a process of ‘structural c­ oupling’.11 While interference is mediated by the internal logic of each system, there can be some multivariate and non-linear causal relationship between systems,12 and systems can potentially co-evolve through a process of mutual ‘irritation’.13 Autopoietic systems theory posits that command and control regulation will have limited effectiveness in achieving change in other systems:14 it is difficult to predict the results of legal intervention, as it will be mediated by other social systems.15 Legal regulation therefore faces a ‘regulatory trilemma’ due to incongruences between the self-regulating systems being regulated and those doing the regulation.16 Regulatory failure may occur in three ways: a lack of response due to incongruence between law and society; ‘over-legalisation’ of society, endangering the self-reproduction of social systems; and ‘politicisation’ or ‘over-socialisation’ 6 

ibid 8. N Luhmann, Law as a Social System, Oxford Socio-Legal Studies, F Kastner et al (trans) (Oxford, Oxford University Press, 2004). 8  Rogowski and Wilthagen (n 5 above) 6. 9  N Luhmann, ‘The Self-Reproduction of Law and Its Limits’ in G Teubner (ed), Dilemmas of Law in the Welfare State (Berlin, Walter de Gruyter, 1986) 113. 10 G Teubner, Law as an Autopoietic System, The European University Institute Press Series, Z Bankowski (ed), A Bankowska and R Adler (trans) (Oxford, Blackwell, 1993) 88–89. 11  R Rogowski, Reflexive Labour Law in the World Society (Cheltenham, Edward Elgar, 2013) 33–34. 12  G Teubner, ‘After Legal Instrumentalism? Strategic Models of Post-Regulatory Law’ in G Teubner (ed), Dilemmas of Law in the Welfare State (Berlin, Walter de Gruyter, 1986) 309. 13  Rogowski (n 11 above) 34. 14  N Gunningham, ‘Introduction’ in N Gunningham and PN Grabosky (eds), Smart Regulation: Designing Environmental Policy, Oxford Socio-Legal Studies (Oxford, Clarendon Press, 1998) 12. 15 S Deakin et al, ‘Gender Inequality and Reflexive Law: The Potential of Different Regulatory Mechanisms for Making Employment Rights Effective’ (Working Paper, 2011) 4. 16  Teubner, ‘After Legal Instrumentalism?’ (n 12 above) 309. 7  See

Methodology

 19

of law, endangering the self-reproduction of the legal system.17 Systems theory may therefore be used as a critique of command and control regulation, and can be used as an explanatory framework for regulatory failure. This is developed further in Chapter three of this volume.

B.  Reflexive Law Drawing on systems theory, reflexive law acknowledges that legal impact is restricted and mediated by law’s autopoietic nature,18 and focuses instead on how legal regulation can promote and stimulate self-regulation within other systems19 via indirect control.20 Law, then, ‘becomes a system for the coordination of action within and between semi-autonomous social subsystems’21 and supports, rather than undermines, the self-referential structure of other systems.22 Therefore, reflexive law is concerned with legal process, rather than legal substance.23 By achieving regulatory aims through ‘enforced or stimulated self-regulation’, reflexive law encourages organisational change through legal stimuli.24 Like law, organisations are seen as ‘self-referential system[s] of interrelated ­communications’,25 which autonomously decide how open to be to their environment, and how to translate ‘environmental perturbations’ into their existing structure.26 Different organisations and organisational structures will be responsive to different types of regulation or ‘perturbations’. Regulation should therefore respond reflexively to differences in organisational structure and behaviour.27 Reflexive law may therefore be deployed as a model of regulation. Using reflexive law as a model involves mapping legal existing legal structures and provisions against archetypal reflexive law structures. However, reflexive law may be achieved in various ways. Thus, any mapping process must first clarify what we mean by a reflexive law ‘model’. First, according to Ayres and Braithwaite, reflexive regulation can be achieved through an ‘enforcement pyramid’, where escalating forms of government 17 

ibid 309, 311–12; Rogowski and Wilthagen (n 5 above) 16. Rogowski and Wilthagen (n 5 above) 6. ibid 7. 20  Teubner, ‘After Legal Instrumentalism?’ (n 12 above) 307. 21  G Teubner, ‘Substantive and Reflexive Elements in Modern Law’ (1983) 17 Law & Society Review 239, 242. 22  Teubner, ‘After Legal Instrumentalism?’ (n 12 above) 314. 23  Rogowski and Wilthagen (n 5 above) 7; see also Teubner, ‘Substantive and Reflexive Elements in Modern Law’ (n 21 above) 255. 24 C McLaughlin, ‘Equal Pay, Litigation and Reflexive Regulation: The Case of the UK Local ­Authority Sector’ (2014) 43 Industrial Law Journal 1, 4. 25  G Teubner, ‘Company Interest: The Public Interest of the Enterprise “in Itself ”’ in R Rogowski and T Wilthagen (eds), Reflexive Labour Law: Studies in Industrial Relations and Employment Regulation (Deventer/Boston MA, Kluwer Law and Taxation Publishers, 1994) 24. 26  ibid 25. 27  I Ayres and J Braithwaite, Responsive Regulation: Transcending the Deregulation Debate, Oxford Socio-Legal Studies (Oxford, Oxford University Press, 1992) 4. 18  19 

20 

Methodology

r­ egulation reinforce market self-regulation.28 Figure 129 provides an example of how an enforcement pyramid might operate in the contemporary equality context in the UK. The base of the enforcement pyramid involves ‘restorative dialogue’, focusing on information, persuasion and voluntary agreement.30 Where dialogue and self-regulation fail, regulation becomes increasingly interventionist, culminating in civil penalties. Reflexive theory suggests that voluntary measures are effective because of the prospect of more deterrent sanctions,31 and that government agencies ‘credibly asserting a willingness to regulate more intrusively’ will steer firms towards self-regulation.32 This interpretation of reflexive law therefore posits a dynamic model of regulation, which allows a responsive assessment of when to punish and when to persuade.33 It suggests that regulation should always start at the base of the pyramid, making dialogue and self-regulation the default form of intervention.34 This reduces the costs of government intervention, and can solve issues of limited state capacity.35 Further, it increases the legitimacy of government intervention (should it prove necessary) and is ultimately more likely to result in compliance.36 Therefore, the enforcement pyramid is an effective means of balancing the strengths and limitations of each method of enforcement37 and engages and ‘harness[es] the energy’ of those best able to effect change.38 Secondly, reflexive law may be conceived as a process of ‘directly deliberative polyarchy’ via an experimentalist architecture.39 Sabel and Simon have previously noted that, while experimentalism ‘bears a strong resemblance’ to responsive regulation, they do not themselves categorise experimentalism as a form of reflexive law.40 At the same time, the obvious synergies between experimentalism and reflexive law mean that an experimentalist architecture is a useful model for developing an alternative conception for how reflexive law might be achieved in practice, in contradistinction to the enforcement pyramid. In this conception, then, framework goals and measurement criteria are decided via joint action between central and lower units; lower-level units are then ‘given 28 

Ayres and Braithwaite (n 27 above). Based on B Hepple, ‘Enforcing Equality Law: Two Steps Forward and Two Steps Backwards for Reflexive Regulation’ (2011) 40 Industrial Law Journal 315. 30  ibid 321. 31  ibid 322. 32  Ayres and Braithwaite (n 27 above) 4. 33  J Braithwaite, Restorative Justice and Responsive Regulation (Oxford, Oxford University Press, 2002) 30. 34 ibid. 35  ibid 32–33. 36 ibid. 37  ibid 32. 38  S Fredman, ‘Breaking the Mold: Equality as a Proactive Duty’ (2012) 60 American Journal of Comparative Law 265, 272. 39  CF Sabel and J Zeitlin, ‘Learning from Difference: The New Architecture of Experimentalist Governance in the EU’ in CF Sabel and J Zeitlin (eds), Experimentalist Governance in the European Union: Towards a New Architecture (Oxford, Oxford University Press, 2012) 3. 40 CF Sabel and WH Simon, ‘Minimalism and Experimentalism in the Administrative State’ (2011–12) 100 Georgetown Law Journal 53, 55. 29 

Methodology

 21

Figure 1:  Enforcement Pyramid in UK Equality Law

the freedom to advance these ends as they see fit’.41 Lower-level units must report regularly on their performance against the agreed criteria, and the framework and criteria should be revised and updated regularly.42 In this model, dialogue is used to ‘disentrench settled practices’ and promote reconsideration of particular interests.43 At its core, deliberative polyarchy is concerned with learning from difference to solve similar problems,44 and is closely linked to organisational benchmarking.45 Experimentalist governance is complemented and advanced by ‘destabilization regimes’, which are collections of structures that make the current situation ‘untenable’ and encourage actors to adopt alternative and superior options.46 Destabilisation regimes may include a public justification requirement;47 the right

41 

Sabel and Zeitlin (n 39 above) 3; see also Sabel and Simon (n 40 above) 79. Sabel and Zeitlin (n 39 above) 3. ibid 5–6. 44  ibid 6. 45  See MC Dorf and CF Sabel, ‘A Constitution of Democratic Experimentalism’ (1998) 98 Columbia Law Review 267 and ch 7. 46  Sabel and Zeitlin (n 39 above) 13. 47  Such as the need to justify an EJRA as a proportionate means of achieving a legitimate aim to courts or ETs if challenged: see Chapter 3 of this volume. 42  43 

22 

Methodology

to challenge and a duty to explain;48 the use of penalty defaults, including the threat of engaging in traditional rule-making;49 or enticements to participate. The Athena SWAN Charter award scheme for gender equality is a strong example of how experimentalism and incentives might work in practice. The scheme recognises employment excellence for women in higher education. University institutions and departments can submit for awards at Gold, Silver and Bronze levels, based on a detailed self-assessment that is reviewed by a panel of UK higher education representatives.50 Thus, the award scheme is a form of voluntary peer review of gender equality measures. There is considerable qualitative evidence that the Athena SWAN scheme has had a positive impact on institutional practices within participating institutions.51 Further, the self-assessment process has encouraged institutions to collect equality data and allowed them to identify challenges to gender equality.52 Efforts are also being made to promote networking between institutions and the sharing of best practice.53 The scheme is now being linked to research funding, providing an additional positive incentive for participation. In sum, then, the experimentalist architecture reflects a more horizontal, nonhierarchical and peer-based model of reflexive law than the enforcement pyramid (see Figure 2). While more easily applied to EU or regional law- and policy-­ making,54 transnational institutions55 or tiered governance councils,56 the experimentalist conception of reflexive law could also be applied to discrimination law at a domestic level.57 In this context, the ‘lower-level units’ engaging in peer review

Figure 2:  Experimentalist Architecture

48  In the equality field, this is arguably reflected in Employment Tribunal (ET) claims and unlawful act notices. 49  This arguably occurred with the voluntary codes of conduct relating to age discrimination in the UK. 50 See further Equality Challenge Unit, ‘Athena SWAN Charter for Women in Science: Annual Report 2013’ (July 2014). 51 See F Munir et al, ‘Advancing Women’s Careers in Science, Technology, Engineering, Mathematics and Medicine: Evaluating the Effectiveness and Impact of the Athena SWAN Charter’ (Equality ­Challenge Unit/Loughborough University, 2013). 52  See ibid. 53  Equality Challenge Unit, ‘Evaluating the Athena SWAN Charter: ECU Response’ (May 2014) 3. 54  Sabel and Zeitlin (n 39 above) 3. 55 G de Búrca et al, ‘Global Experimentalist Governance’ (2014) 44 British Journal of Political Science 477. 56  Dorf and Sabel (n 45 above) 320. 57  Sabel and Zeitlin (n 39 above) 25–26.

Methodology

 23

and scrutiny would likely be networks of firms,58 akin to the process operating in the Athena SWAN scheme. According to Sabel and Zeitlin, experimentalist governance may emerge under two conditions: first, strategic uncertainty, where actors must work together to determine what their goals should be and how to achieve them; and, secondly, a ‘multipolar or polyarchic distribution of power’, where no actor may impose their own solution without considering the views of others.59 While EU equality law generally fulfills these requirements, it is more difficult to identify these conditions at the national level: employers can generally implement solutions to equality issues independently, and are rarely required to consult or consider the views of their peers, particularly given the limited role and membership of sectoral trade associations in the UK.60 While UK employers are increasingly networking and sharing best practice in relation to equality (see Chapters four, seven and eight), this does not mean there is a polyarchic distribution of power in the UK. Power still generally resides with the UK government, even where the government may be disinclined to exercise it. Thus, it is more likely that equality law in the UK reflects the use of state power in the shadow of hierarchy, where unilateral solutions are (almost) acceptable, than a destabilisation regime with peer review.61 At the same time, a reflexive perspective suggests that the established state-­ centred approach for addressing inequality in the UK may be ineffective in relation to structural and systemic inequality, raising the possibility of an experimentalist approach to address more deep-seated inequality.62 Further, an experimentalist approach may help to address issues of enforcement of equality law (see Chapters three and four), particularly as equality issues occur at workplaces which are ‘too numerous and dispersed’ for easy centralised monitoring of compliance.63 Even if directly deliberative polyarchy is not realised as a whole, it may still be evident in ‘partial variants’ in UK equality law.64 Thus, it is useful to consider what an experimentalist architecture can illuminate about UK equality law. This is pursued further in Chapter three. Reflexive law (in its various forms) thus provides an appropriate framework for modelling age discrimination laws and their efficacy, particularly when considering 58 

J Cohen and C Sabel, ‘Directly-Deliberative Polyarchy’ (1997) 3 European Law Journal 313, 327. Sabel and Zeitlin (n 39 above) 9. 60  See T May et al, ‘Business Representation in the UK since 1979: The Case of Trade Associations’ (1998) 46 Political Studies 260; RJ Bennett, ‘The Logic of Membership of Sectoral Business Associations’ (2000) 58 Review of Social Economy 17; J Clarke, ‘Trade Associations: An Appropriate Channel for Developing Sustainable Practice in SMEs?’ (2004) 12 Journal of Sustainable Tourism 194. 61 CF Sabel and J Zeitlin, ‘Experimentalism in the EU: Common Ground and Persistent Differences’ (2012) 6 Regulation & Governance 410, 413–14; Sabel and Zeitlin (n 39 above) 15. For a discussion of the limitations of experimentalist governance as a model of EU anti-discrimination law more generally, see G de Búrca, ‘Stumbling into Experimentalism: The EU Anti-Discrimination Regime’ in CF Sabel and J Zeitlin (eds), Experimentalist Governance in the European Union: Towards a New Architecture (Oxford, Oxford University Press, 2012). 62  de Búrca (n 61 above) 217–18. 63  Cohen and Sabel (n 58 above) 331. 64  See Dorf and Sabel (n 45 above) 323. 59 

24 

Methodology

the relationship between formal legal regulation and organisational self-regulation, and the potential of different modes of regulation to affect organisational behaviour. Reflexive law is particularly useful for considering how discrimination law is embedded at the level of the firm, which remains a key concern of both sociologists and legal practitioners.65 Therefore, in this book I draw on a reflexive law theoretical framework to inform and shape my approach.

II.  Research Framework The theoretical perspective guiding this study has influenced my methodological choices. More particularly, my research is posited in the field of empirical legal studies and employs a mixed methods approach, drawing on doctrinal, qualitative and quantitative research methods. Empirical mixed methods are particularly suited for considering complicated, evolving legal questions. Given that reflexive law posits that there may be some multivariate, non-linear causal relationship between law and other social systems, empirical mixed methods offer a means of obtaining a nuanced understanding of these complex and ­multifaceted ­dynamics.66 ­Further, reflexive law theory requires scholars to a­ nalyse both internal legal processes, through a study of doctrinal law, and law’s environment, through a sociological and economic study of law.67 Therefore, this theoretical perspective demands an interdisciplinary, mixed methods research design.

A.  Empirical Research Empirical legal research entails the study of law through direct methods68 to ­provide more developed and nuanced understandings of how law operates in practice. Empirical research methods are a key means of furthering socio-legal research into the ‘material realities of everyday life’.69 As noted by Genn, P ­ artington and

65  See, eg, E Kelly and F Dobbin, ‘How Affirmative Action Became Diversity Management: Employer Response to Antidiscrimination Law, 1961 to 1996’ (1998) 41 American Behavioral Scientist 960; G Kirton and A Greene, The Dynamics of Managing Diversity: A Critical Approach, 3rd edn (Oxford, Elsevier, 2010). 66  LB Nielsen, ‘The Need for Multi-Method Approaches in Empirical Legal Research’ in P Cane and HM Kritzer (eds), The Oxford Handbook of Empirical Legal Research (Oxford, Oxford University Press, 2010) 955; see also M McConville, ‘Development of Empirical Techniques and Theory’ in M McConville and WH Chui (eds), Research Methods for Law, Research Methods for the Arts and Humanities (Edinburgh, Edinburgh University Press, 2007) 219. 67  Rogowski (n 11 above) 37. 68  J Baldwin and G Davis, ‘Empirical Research in Law’ in P Cane and MV Tushnet (eds), The Oxford Handbook of Legal Studies (Oxford University Press 2003) 880–81. 69 P Hillyard, ‘Invoking Indignation: Reflections on Future Directions of Socio-Legal Studies’ (2002) 29 Journal of Law and Society 645, 652.

Research Framework

 25

Wheeler, ‘empirical research helps us to understand the law better and an empirical understanding of the law in action helps us to understand society better’.70 Therefore, empirical research recognises the importance of investigating beyond ‘law on the books’ to consider legal results.71 While empirical legal scholarship is growing in prevalence, and has arguably achieved a position ‘approaching stability’ in the ‘legal academy’s mainstream’,72 it is still unclear where the line should be drawn regarding what is ‘empirical’. For Heise, empirical scholarship is defined narrowly as that which ‘uses statistical techniques and analyses’.73 In contrast, Epstein and King see empirical research as ‘learning about the world using quantitative data or qualitative information’.74 Similarly, Cane and Kritzer define empirical research as ‘the systematic collection of information … and its analysis according to some generally accepted method’.75 Using such a broad definition, legal doctrinal research could also be seen as a form of empirical analysis, drawing on legislation and case law as ‘data’. However, Epstein and King argue that purely normative or theoretical legal research is not empirical,76 and many doctrinal scholars would not regard their research as empirical.77 Instead, legal doctrinal research is more analogous to a social science literature review78 that typically precedes empirical work. Empirical legal research is best viewed as a subset of socio-legal inquiry. Sociolegal research examines law in its social context, often by utilising perspectives and research techniques from the social sciences.79 Socio-legal research and theory bring social considerations into legal study as objects of rigorous inquiry80 and emphasise the importance of approaching legal research as an ‘empirical, ­systematic

70  HG Genn et al, Law in the Real World—Improving Our Understanding of How Law Works: Final Report and Recommendations (London, Nuffield Foundation, 2006) 1. See also E Mertz, ‘Conclusion: A New Social Constructionism for Sociolegal Studies’ (1994) 28 Law & Society Review 1243, 1261. 71  LE Teitelbaum, ‘An Overview of Law and Social Research’ (1985) 35 Journal of Legal Education 465, 466. 72  M Heise, ‘The Past, Present, and Future of Empirical Legal Scholarship: Judicial Decision Making and the New Empiricism’ [2002] University of Illinois Law Review 819, 820. 73  ibid 821. 74  Epstein and King (n 3 above) 1. 75  P Cane and HM Kritzer, ‘Introduction’ in P Cane and HM Kritzer (eds), The Oxford Handbook of Empirical Legal Research (Oxford, Oxford University Press, 2010) 4. 76 Epstein and King (n 3 above) 3. cf I Dobinson and F Johns, ‘Qualitative Legal Research’ in M McConville and WH Chui (eds), Research Methods for Law, Research Methods for the Arts and Humanities (Edinburgh, Edinburgh University Press, 2007) 19; L Webley, ‘Qualitative Approaches to Empirical Legal Research’ in P Cane and HM Kritzer (eds), The Oxford Handbook of Empirical Legal Research (Oxford, Oxford University Press, 2010) 927. 77  Cane and Kritzer (n 75 above) 5. 78  Dobinson and Johns (n 76 above) 22. 79 A Bradshaw, ‘Sense and Sensibility: Debates and Developments in Socio-Legal Research Methods’ in PA Thomas (ed), Socio-Legal Studies (Aldershot, Dartmouth, 1997) 99, 107, 109; P Thomas, ­‘Socio-Legal Studies: The Case of Disappearing Fleas and Bustards’ in PA Thomas (ed), Socio-Legal Studies (Aldershot, Dartmouth, 1997) 2–3. 80  R Cotterrell, ‘Subverting Orthodoxy, Making Law Central: A View of Sociolegal Studies’ (2002) 29 Journal of Law and Society 632, 636.

26 

Methodology

study of a field of social experience’.81 Similarly, law in context approaches critically examine law in its social, political and economic contexts, with the hope of bringing fresh perspectives to established legal issues and areas.82 While the fields of socio-legal research and law in context are easily able to accommodate empirical research methods, socio-legal inquiry and contextual legal research are not necessarily empirical. As a consequence, this book is located within the fields of socio-legal, law in context and empirical research, and addresses legal questions using social science empirical methods, both quantitative and qualitative.

B.  Mixed Methods Approach Mixed methods research combines qualitative and quantitative research ­methods to understand a research problem.83 In addition to collecting and analysing ­different forms of data, mixed methods research mixes the data sets in a ­meaningful way, for example, by comparing and contrasting the results achieved and/or presenting an overall interpretation of the findings.84 As a result, mixed methods research designs must integrate qualitative and quantitative approaches to be effective. That said, mixed methods designs generally give primacy to quantitative methods, using qualitative methods to merely ‘assist’ statistical analysis.85 Instead, this study focuses primarily on qualitative methods, which are more in keeping with traditional legal doctrinal scholarship and more a­ menable to a complex, multi-layered world view.86

(i)  Advantages of Mixed Methods Designs Mixed methods designs utilise and synthesise the different strengths and advantages of qualitative and quantitative research methods,87 significantly enhancing research validity88 and providing an enriched understanding of legal issues.89

81 

ibid 633. See further W Twining, Law in Context: Enlarging a Discipline (Oxford, Clarendon Press, 1997) 45. Plano Clark et al, ‘Mixing Quantitative and Qualitative Approaches: An Introduction to ­Emergent Mixed Methods Research’ in SN Hesse-Biber and P Leavy (eds), Handbook of Emergent Methods (New York, Guilford Press, 2008) 364. This may be compared with multimethods research, which uses two or more qualitative or quantitative methods (eg expert interviews and organisational case studies): SN Hesse-Biber, Mixed Methods Research: Merging Theory with Practice (New York, Guilford Press, 2010) 3. 84  Plano Clark et al (n 83 above) 364. 85  Hesse-Biber (n 83 above) 9, 14. 86  ibid 9. 87  MQ Patton, Qualitative Evaluation and Research Methods, 2nd edn (Thousand Oaks CA, Sage, 1990) 13. 88 J Kirk and ML Miller, Reliability and Validity in Qualitative Research, Qualitative Research Methods (Beverley Hills CA, Sage, 1986) 30. 89 JC Greene et al, ‘Toward a Conceptual Framework for Mixed-Method Evaluation Designs’ (1989) 11 Educational Evaluation and Policy Analysis 255, 258. 82 

83 VL

Research Framework

 27

While qualitative research is particularly suited to areas that are fluid and ­changing90 and allows issues to be examined in depth across a limited number of cases,91 quantitative research methods allow for the aggregation of many individual responses, enabling succinct comparisons and generalisation92 and the study of correlations. The combined use of quantitative and qualitative ­methods can achieve better results than using one method alone93 and can produce a more nuanced understanding of complex and multi-faceted issues.94 Therefore, a mixed methods design is the best option for overcoming the limitations and maximising the strengths of individual research methods. Similarly, mixed methods designs, incorporating both legal and social science research methods, recognise that law and the social sciences produce equally important constructions of the world. While neither law nor social science has privileged access to ‘reality’, the epistemic authority of both legal discourse and scientific discourse are noteworthy and of value.95 Therefore, mixed methods research can be used to reveal constructed ‘facts’ using different processes and procedures of factual inquiry, drawing on different disciplines or social s­ ystems96 and gaining value from multiple social spheres. As a result, mixed methods designs recognise the importance of both legal and social science ways of knowing, and can create new composite forms of knowledge by integrating ­multiple social ‘realities’.

(ii)  Limitations of Mixed Methods Research Despite the substantial benefits of empirical and mixed methods research designs, there is still a relative dearth of empirical legal scholarship in many areas.97 The use of mixed methods remains unusual in legal research, and few studies integrate mixed methods with doctrinal analysis. This is perhaps due to the complexity of mastering multiple research methods, particularly for legal scholars who are ­unfamiliar with social science research techniques and statistical analysis.98 In addition to these practical challenges, mixed methods research designs have a number of theoretical critics. In particular, the incompatibility thesis argues that qualitative and quantitative research methods cannot be used in the same study, as

90 JM Corbin and AL Strauss, Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory, 3rd edn (Thousand Oaks CA, Sage Publications, 2008) 13. 91  Patton (n 87 above) 14. 92 ibid. 93 JW Creswell and VL Plano Clark, Designing and Conducting Mixed Methods Research (Thousand Oaks CA, Sage, 2007) 5; AR Poteete et al, Working Together: Collective Action, the Commons, and Multiple Methods in Practice (Princeton NJ, Princeton University Press, 2010) 5. 94  Nielsen (n 66 above) 955; see also McConville (n 66 above) 219. 95 G Teubner, ‘How the Law Thinks: Toward a Constructivist Epistemology of Law’ (1989) 23 Law & Society Review 727, 743. 96  ibid 744. 97  Genn et al (n 70 above) 2, 6. 98  See Poteete et al (n 93 above) 15.

28 

Methodology

their foundational paradigms cannot be reconciled.99 This argument is explored further below: however, it suffices to say at this stage that the differences between qualitative and quantitative research methods may not be as fundamental as the incompatibility thesis claims. A similar disjunction may exist between legal ways of knowing and those in other social spheres: ‘world-views in different subsystems are thoroughly incompatible on the basis of their cognitive assumptions’.100 Indeed, a researcher may embody one disciplinary perspective to the exclusion of any other. Teubner describes the ‘homing instinct of lawyers, their natural inclination towards their own legal order’,101 which could potentially undermine a lawyer’s attempt to draw on multiple social systems and social science research methods in a mixed methods study. Similarly, Van Klink and Taekema argue that: As a researcher, you are educated and socialized in a particular practice, consciously and unconsciously internalizing its norms. Awareness of disciplinary differences does not automatically mean that you can also overcome those differences. Truly appreciating the work done in another discipline might require unlearning your own disciplinary perspective.102

While recognising these concerns, I would argue that a reflective awareness of the assumptions, conventions and norms of legal scholarship, and a critical scrutiny of method and methodology, go a long way to addressing and correcting the ‘homing instinct’ of lawyers. Further, it is at least arguable that the legal system is shaped by similar forces to other non-legal social phenomena.103 In this case, law can (and should) be ­examined using the same tools as other disciplines.104 Similarly, and consistently with a reflexive law theoretical standpoint, scholars’ declining focus on ‘state-­ centred’ law increases the scope for multi-disciplinary research.105 Therefore, a mixed methods study is entirely consistent with the theoretical underpinnings of this research. Indeed, engagement with a variety of methods (including legal ­doctrinal research) will ‘produce a stronger social science’.106 The ‘plurality of method’ made available by mixed methods approaches may pose difficult choices for researchers, and ‘challenges the sense of direction which

99 See NK Denzin and YS Lincoln, ‘Introduction: The Discipline and Practice of Qualitative Research’ in NK Denzin and YS Lincoln (eds), The Landscape of Qualitative Research, 4th edn (London, SAGE, 2013) 15. 100 Teubner, Law as an Autopoietic System (n 10 above) 62; see also Luhmann (n 7 above). 101 Teubner, Law as an Autopoietic System (n 10 above) 109. 102  B Van Klink and S Taekema, ‘A Dynamic Model of Interdisciplinarity: Limits and Possibilities of Interdisciplinary Research into Law’ (Tilburg University Legal Studies Working Paper, 9 June 2008) 8. 103 C McCrudden, ‘Legal Research and the Social Sciences’ (2006) 122 Law Quarterly Review 632, 641. 104 ibid. 105  ibid 644. 106  ibid 647.

Research Framework

 29

method claims to afford (where should one go if many roads can be followed?)’.107 Of course, research is ultimately about ‘finely judging the ability of a particular research tool to provide the data required’.108 While this task is made more complex by mixed methods approaches, it would not be eliminated by a purely qualitative or quantitative research design. All research designs involve complex choices about data collection and analysis: this challenge is not confined to mixed methods studies. Acknowledging these limitations and challenges, the sections that follow provide a broad overview of the key types of research deployed in this mixed methods design: doctrinal research, quantitative research and qualitative research. I also consider in more depth theoretical concerns regarding how to integrate these different forms of research in a meaningful way.

C.  Doctrinal Research Doctrinal research method is often an ‘implicit and … tacit’ process, as many legal scholars feel it is unnecessary to detail their research approach.109 However, it is imperative that legal scholars articulate their research methods, both to facilitate communication between disciplines and to reveal the assumptions that underlie the research process,110 thereby promoting rigour and accountability in legal scholarship. Legal research has traditionally been undertaken through the hermeneutic study of legal texts.111 Where discussed explicitly, doctrinal research is c­ onceptualised as a ‘two-part process’, involving (1) the identification and (2) the interpretation of legal texts.112 Once all relevant materials113 have been collected, doctrinal researchers create a hypothesis as to the validity and precise meaning of legal texts,114 ­combining ‘specific interpretations of legal principles, rules and concepts in a (newly) systematised whole’.115 Legal doctrinal method is therefore systematically interpretative, involving a search for coherence across legal texts and sources.116 107  S Glanert, ‘Method?’ in PG Monateri (ed), Methods of Comparative Law, Research Handbooks in Comparative Law (Cheltenham, Edward Elgar, 2012) 66. 108  JA Hughes, The Philosophy of Social Research, Aspects of Modern Sociology, 2nd edn (New York, Longman, 1990) 11. 109  T Hutchinson and N Duncan, ‘Defining and Describing What We Do: Doctrinal Legal Research’ (2012) 17 Deakin Law Review 83, 99. 110  P Chynoweth, ‘Legal Research’ in A Knight and L Ruddock (eds), Advanced Research Methods in the Built Environment (Oxford, Wiley-Blackwell, 2008) 37. 111  M van Hoecke, ‘Legal Doctrine: Which Method(s) for What Kind of Discipline?’ in M van ­Hoecke (ed), Methodologies of Legal Research: What Kind of Method for What Kind of Discipline?, European Academy of Legal Theory Series (Oxford, Hart Publishing, 2011) 17. 112  Hutchinson and Duncan (n 109 above) 110. 113  These might include normative cases, legislation and treaties and authoritative non-binding cases and scholarly writings: van Hoecke (n 111 above) 11. 114  ibid 14. 115  ibid 17. 116 ibid.

30 

Methodology

However, doctrinal method is rarely a straightforward process. ‘­Systematisation’ is grounded in legal positivism117 and ‘a view of the world where the law is objective, neutral and fixed’.118 Positivism continues to have ‘immense influence’ on legal research,119 which mostly reflects a modernist search for a singular, ‘true’ meaning of law.120 In practice, though, law is likely to be uncertain, highly contested and constantly evolving.121 Rather than being a fixed and static ‘truth’, law is the result of an ongoing and dynamic process of negotiation and compromise122 where meaning is culturally and historically dependent123 and in perpetual motion.124 While law is not necessarily ‘indeterminate’, it may be ‘underdeterminate’, as it is constantly reshaped and contested in practice.125 Further, the analysis of legal texts in isolation from their social and cultural context can be misleading and inadequate, emphasising the importance of using non-legal perspectives to build on an analysis of legal texts.126 Failing to consider the social aspects of law may produce an incomplete picture of legal change.127 Therefore, while hermeneutic legal doctrinal study is essential to clarify, identify and provide certainty regarding legal rules,128 it is also important to look beyond

117  Positivism is described by Morgenthau as a philosophy that ‘restricts the object of scientific knowledge to matters that can be verified by observation, and thus excludes from its domain all m ­ atters of an a priori, metaphysical nature’: HJ Morgenthau, ‘Positivism, Functionalism, and ­International Law’ (1940) 34 American Journal of International Law 260, 261; see also P Baert, Social Theory in the Twentieth Century (Cambridge, Polity Press, 1998) 174–82. Legal positivism, then, asserts the ‘separability of law and morality’ and focuses on law contained within legal texts: see ­further MH Kramer, In Defense of Legal Positivism: Law without Trimmings (Oxford, Oxford University Press, 2003). 118  Hutchinson and Duncan (n 109 above) 116. 119  G Samuel, ‘Does One Need an Understanding of Methodology in Law before One Can Understand Methodology in Comparative Law?’ in M van Hoecke (ed), Methodologies of Legal Research: What Kind of Method for What Kind of Discipline?, European Academy of Legal Theory (Oxford, Hart Publishing, 2011) 192. While there is some diversity of approach (eg, in theories based on feminism and critical legal studies), Samuel argues that this should not be overstated: ibid 206. 120  See P Dann, ‘Thoughts on a Methodology of European Constitutional Law’ (2005) 6 German Law Journal 1453, 1459. 121  Hutchinson and Duncan (n 109 above) 110, 116. 122  F Snyder, New Directions in European Community Law, Law in Context (London, Weidenfeld and Nicolson, 1990) 17; A Griffiths, ‘Legal Pluralism’ in R Banakar and M Travers (eds), An Introduction to Law and Social Theory (Oxford, Hart Publishing, 2002) 305; BG Carruthers and TC Halliday, ‘Negotiating Globalization: Global Scripts and Intermediation in the Construction of Asian Insolvency Regimes’ (2006) 31 Law & Social Inquiry 521. 123  J Shaw, ‘Postnational Constitutionalism in the European Union’ (1999) 6 Journal of European Public Policy 579, 593; TC Halliday and BG Carruthers, ‘The Recursivity of Law: Global Norm Making and National Lawmaking in the Globalization of Corporate Insolvency Regimes’ (2007) 112 American Journal of Sociology 1135, 1149–52. 124  See Carruthers and Halliday (n 122 above). 125  Mertz (n 70 above) 1246. 126  Hutchinson and Duncan (n 109 above) 115. 127  SLR Anleu, Law and Social Change, 2nd edn (London, SAGE, 2010) 247. 128  J Hage, ‘The Method of a Truly Normative Legal Science’ in M van Hoecke (ed), Methodologies of Legal Research: What Kind of Method for What Kind of Discipline?, European Academy of Legal Theory (Oxford, Hart Publishing, 2011) 42, 44.

Research Framework

 31

legal texts to consider the empirical context in which law operates.129 Even if only aiming to construct a coherent description of law, legal research and doctrine is more than just a hermeneutic discipline.130 Drawing on van Hoecke, I adopt an approach to legal research that consists of: —— a hermeneutic discipline, involving the study of legal texts and their interpretation;131 —— an empirical discipline, looking beyond legal rules to examine their operation within a social context;132 and —— a normative discipline, which requires choices to be made among values and interests to take a normative position.133 If viewed as a normative discipline, legal research must consider issues beyond legal texts and may necessitate empirical research to reach a valid and informed normative judgement.134 Therefore, meaningful analysis of legal provisions must involve consideration of outside sources, and may entail substantial empirical research in some cases. To explore research questions like those posed in this book, legal doctrinal research can only be the first step and, on its own, is far from sufficient for meaningful analysis. Therefore, in this study I also employ quantitative and qualitative social science methods.

D.  Quantitative Research Quantitative research methods allow for the aggregation of responses from many people to a limited set of questions, enabling succinct comparisons and generalisation across a specific set of issues.135 Statistical analysis may take two forms. First, descriptive analysis organises, summarises and describes data without inference or generalisation.136 This form of research is geared to describing ‘what is’, including via the numerical measurement of how variables are distributed across a population or phenomenon and/or the relationships between events and phenomena.137

129  PC Westerman, ‘Open or Autonomous? The Debate on Legal Methodology as a Reflection on the Debate on Law’ in M van Hoecke (ed), Methodologies of Legal Research: What Kind of Method for What Kind of Discipline?, European Academy of Legal Theory (Oxford, Hart Publishing, 2011) 108. 130  van Hoecke (n 111) 3, 6. 131  ibid 4. 132  ibid 6. 133  ibid 10. 134 ibid. 135  Patton (n 87 above) 14. 136  A Walsh, Statistics for the Social Sciences: With Computer Applications (New York, Harper & Row, 1990) 3–4. 137  WH Chui, ‘Quantitative Legal Research’ in M McConville and WH Chui (eds), Research ­Methods for Law, Research Methods for the Arts and Humanities (Edinburgh, Edinburgh University Press, 2007) 48; SB Merriam, Qualitative Research: A Guide to Design and Implementation, The Jossey-Bass Higher and Adult Education Series (San Francisco CA, Jossey-Bass, 2009) 5.

32 

Methodology

Secondly, explanatory analysis focuses on why phenomena occur,138 and allows inference or generalisation about a large group based on data taken from a subset of the population.139 Data analysis is focused on explaining variance, and using variables to explain observed variability.140 More particularly, quantitative research is concerned with deductively testing hypotheses derived from theory.141 Quantitative methods therefore attempt to control variables to measure causal relationships, including by ruling out rival or alternative explanations of variability.142 However, while correlations may be observed between variables, this does not imply causation.143 Causation requires that (1) the cause precedes the event, (2) there be an empirical association between the variables, and (3) there is no plausible alternative ­explanation for the relationship.144 Proving causation and temporal ordering is generally left to theory,145 not quantitative methods.

E.  Qualitative Research In contrast, qualitative research allows an issue to be examined in depth and detail across a limited number of cases.146 That said, what constitutes qualitative research—and what distinguishes it from quantitative research—is contentious.147 Qualitative research has evolved from, and is underpinned by, a wide range of ­disciplinary traditions, philosophies and methods.148 This pluralist framework and history is one of the strengths of qualitative research.149 Despite this disagreement, it is possible to identify five core characteristics of qualitative research methods: first, they are grounded in a broadly interpretivist philosophical framework,150 being concerned with how the social world is ‘interpreted, understood, experienced or produced’;151 secondly, they are based on flexible methods of data collection and generation, which are sensitive to

138 

Chui (n 137 above) 50. Walsh (n 136 above) 3–4. Rose and O Sullivan, Introducing Data Analysis for Social Scientists (Buckingham, Open University Press, 1993) 6–7. 141  ibid 10. 142  ibid 11. 143  ibid 12. 144  See Chui (n 137 above) 51. 145  Rose and Sullivan (n 140 above) 30. 146  Patton (n 87 above) 14. That said, qualitative research should still be generalisable, or have some ‘broader resonance’: J Mason, Qualitative Researching (London, SAGE, 1996) 6. 147  Mason (n 146 above) 3. 148 ibid. 149  ibid 4. 150  That is, the assumption that reality is socially constructed, and that there are multiple realities or interpretations of a given event: Merriam (n 137 above) 8. 151  Mason (n 146 above) 4; Merriam (n 137 above) 8–9. 139 

140 D

Research Framework

 33

their social context;152 thirdly, they use techniques of analysis that are oriented towards rich description153 and ‘complexity, detail and context’;154 fourthly, they use the researcher as the main ‘instrument’ for collecting and analysing data, and recognise the influence of a researcher’s subjectivities on the data produced;155 and, fifthly, they generally involve a process of inductive reasoning, gathering data to build concepts and theories, rather than deductively testing existing hypotheses.156 Ultimately, qualitative research is directed to ‘understanding the meaning people have constructed’ and how they understand and make sense of their world.157 Qualitative research may (simultaneously) include methods such as case ­studies, interpretative analysis, and interviewing.158 Indeed, qualitative research is ‘inherently multimethod in focus’, using multiple methods and sources of data to triangulate findings and add depth and richness to research.159

F.  Integrating Different Research Methods As noted above, it is at least arguable that the divergent foundational paradigms of qualitative and quantitative research methods cannot be reconciled.160 Each research ‘tool’ is embedded in particular epistemological foundations and ­justifications.161 Quantitative research methods are grounded in positivism162 and a focus on discovering ‘pure reality’ through objective research.163 Further, quantitative analysis views science as ‘value-free’ and having privileged access to objective knowledge of the world.164 In contrast, qualitative research is grounded in ‘complex interpretative practices’ and acknowledges the presence of ‘multiple, refracted realities’.165 Qualitative research situates the researcher in the world,166 acknowledges and integrates the researcher’s subjectivity,167 and embraces the ‘value-laden nature of inquiry’.168 However, these descriptions may overemphasise the epistemological differences between qualitative and quantitative methods. Quantitative data still requires 152 

Mason (n 146 above) 4. Merriam (n 137 above) 16. Mason (n 146 above) 4. 155  Merriam (n 137 above) 15. 156 ibid. 157  ibid 13. 158  Denzin and Lincoln (n 99 above) 5. 159  ibid 9–10. 160  See ibid 15. 161  Hughes (n 108 above) 11. 162  A Holliday, Doing and Writing Qualitative Research (London, SAGE, 2002) 2. 163  Chui (n 137 above) 49. 164 ibid. 165  Denzin and Lincoln (n 99 above) 10, 13. 166  ibid 6. 167  Merriam (n 137 above) 15. 168  Denzin and Lincoln (n 99 above) 17. 153  154 

34 

Methodology

human interpretation to be given meaning, and qualitative data may need to be quantified to explore the frequency of a theme or occurrence in the data.169 Further, qualitative research cuts across a range of methodological and theoretical traditions:170 it has no theoretical paradigm that is ‘distinctly its own’;171 and positivism is still influential in shaping qualitative research practices, lingering like a ‘long shadow’.172 It is therefore inappropriate to essentialise and juxtapose qualitative and quantitative research methods: both encompass varied and diverse traditions, and both may be used within a range of methodologies.173 As a consequence, in contrast to the incompatibility thesis described above, Teddlie and Tashakkori posit a compatibility thesis for mixed methods research.174 This reflects a form of paradigm pluralism, or a methodological eclecticism with a willingness to use and ‘synergistically integrate’ the most appropriate methods for investigating a particular research problem. Adopting this pragmatist paradigm allows mixed methods to be used.175 Sale, Lohfeld and Brazil suggest an alternative approach: while rejecting the pragmatist compatibility thesis, and regarding qualitative and quantitative approaches as ‘incommensurate’, the authors still believe mixed methods studies to be possible, though only for complementary purposes, as each method studies different phenomena.176 This is consistent with the research design adopted for this study, where mixed methods are employed to obtain complementary perspectives on different aspects of the research questions177 and to guide and develop different stages of the research design.178 This design is described in more detail in the next section.

III.  Research Design Within this empirical and mixed methods field, this study integrates qualitative, quantitative and doctrinal research methods. In adopting this research f­ ramework, 169 

Webley (n 76 above) 930. Denzin and Lincoln (n 99 above) 5. 171  ibid 11. 172  ibid 17–18. 173  Hesse-Biber (n 83 above) 11–13. 174  C Teddlie and A Tashakkori, ‘Overview of Contemporary Issues in Mixed Methods Research’ in A Tashakkori and C Teddlie (eds), Sage Handbook of Mixed Methods in Social & Behavioral Research, 2nd edn (Thousand Oaks CA, SAGE, 2010) 5, 8–9. 175 In contrast, Hesse-Biber describes this methods-driven research design as a ‘cart before the horse’ approach, and calls for a renewed focus on methodology in mixed methods research: Hesse-Biber (n 83 above) 10. 176  JEM Sale et al, ‘Revisiting the Quantitative-Qualitative Debate: Implications for Mixed-Methods Research’ (2002) 36 Quality and Quantity 43, 50. 177  J Brannen, ‘Combining Qualitative and Quantitative Approaches: An Overview’ in J Brannen (ed), Mixing Methods: Qualitative and Quantitative Research (Aldershot, Avebury, 1992) 12. 178  Greene et al (n 89 above) 260; Poteete et al (n 93 above) 12. 170 

Research Design

 35

this book makes an original addition to existing legal research, drawing on interdisciplinary approaches and qualitative and quantitative methods to complement robust doctrinal research. Further, this research design ensures that the various limitations and strengths of individual research methods are complemented by other sources and ways of ‘knowing’. Thus, this research framework allows a more robust and nuanced examination of the questions underpinning this book. Figure 3 illustrates the overall mixed methods design used in this study. The design includes triangulation (comparing and contrasting the results of different methods to validate data); exploratory elements (to generalise qualitative findings); and explanatory elements (to explain quantitative results).179 Figures 4 and 5 indicate how each method contributes to answering the ­questions that underpin this study.

Figure 3:  Overall Mixed Methods Design

Figure 4:  Mixed Methods to Analyse Question 1 179 

See Plano Clark et al (n 83 above) 381.

36 

Methodology

Figure 5:  Mixed Methods to Analyse Question 2

The subsections that follow provide a brief overview and justification of each research method.180

A.  Qualitative Expert Interviews Qualitative expert interviews were employed to explore the operation of UK age discrimination laws in their social context (Chapter four), and to triangulate ­comparative doctrinal analysis and examine the operation of Finnish age discrimination laws (Chapter six). Qualitative expert interviews were selected for this study as an effective and efficient means of canvassing expert opinions,181 developing new theoretical models182 and expanding theoretical thinking.183 ­ More particularly, semi-structured interviews were chosen for their flexibility, as they allow p ­ articipants to respond in their own terms and the interviewer to seek clarification and further exposition of answers, while still retaining a degree of structure to the interview.184 Semi-structured interviews optimise the time available for each interview185 but also allow unanticipated statements and ideas to

180  Detailed research procedures can be viewed elsewhere: see A Blackham, ‘Extending Working Life for Older Workers: An Empirical Legal Analysis of Age Discrimination Laws in the UK’ (University of Cambridge, 2014). 181  Webley (n 76 above) 937. 182 A Bogner et al, ‘Introduction: Expert Interviews—an Introduction to a New Methodological Debate’ in A Bogner et al (eds), Interviewing Experts, Research Methods (Basingstoke/New York, ­Palgrave Macmillan, 2009) 2. 183 P Dilworth-Anderson and MD Cohen, ‘Theorizing across Cultures’ in V Bengtson (ed), ­Handbook of Theories of Aging, 2nd edn (Berlin, Springer, 2009) 490. 184 T May, Social Research: Issues, Methods and Process (Buckingham, Open University Press, 1993) 93. 185  Patton (n 87 above) 285.

Research Design

 37

emerge.186 This style of interviewing encourages the active engagement of the interviewer and respondents, facilitating collaborative identification and analysis of issues.187

B.  Quantitative Statistical Analysis This study employs statistical analysis in two instances. First, quantitative statis­ tical analysis is used to describe and draw correlations between organisational age-aware practices and organisational characteristics, and to examine how ageaware practices have evolved over time with the passage of age discrimination legislation (Chapter five). Secondly, statistical analysis is used to explore the comparative impact of age discrimination laws in the UK and Finland (Chapter six). Secondary analysis of official government statistics provides rich, high-quality data to explore changing public opinions and chart trends over time.188 Secondary quantitative analysis is particularly suited to identifying differences in organisational practices,189 exploring changing practices and identifying correlations between organisational characteristics and practices. However, this form of analysis has a number of limitations. First, the topics selected and questions posed are those of interest, and acceptable, to government officials,190 limiting the diversity and objectivity of the data set and potentially omitting key variables.191 Secondly, as noted above, quantitative data pose a ‘fundamental problem of causal inference’192—while it is possible to determine if a relationship exists between variables, causality cannot be determined using statistical data. It is therefore necessary to supplement quantitative analysis with qualitative empirical research to posit causal linkages.

C.  Comparative Analysis In Chapter six I use explanatory comparative legal analysis of Finnish law to develop ideas regarding how UK law could accommodate an ageing workforce, and to cast light on how UK laws are operating in practice. This analysis is complemented by qualitative expert interviews and statistical analysis (discussed above) 186 K Charmaz, Constructing Grounded Theory: A Practical Guide through Qualitative Analysis ­(London, Sage Publications, 2006) 26. 187  H Simons, Case Study Research in Practice (London, SAGE, 2009) 43. 188  May (n 184 above) 52. 189  H Barnes et al, ‘An Ageing Workforce: The Employer’s Perspective’, Report 468 (Institute for Employment Studies, 2009) 7. 190  May (n 184 above) 52. 191  A Bryman, Social Research Methods, 4th edn (Oxford, Oxford University Press, 2012) 316. 192  L Epstein and AD Martin, ‘Quantitative Approaches to Empirical Legal Research’ in P Cane and HM Kritzer (eds), The Oxford Handbook of Empirical Legal Research (Oxford, Oxford University Press, 2010) 903.

38 

Methodology

to triangulate and verify the comparative doctrinal findings. Explanatory comparative legal analysis seeks to identify both similarities and differences between jurisdictions, and account for these variances.193 Through a doctrinal examination of Finnish legislation and case law, it is possible to create a ‘rational reconstruction’ of the law regarding older workers in Finland194 and compare it with that in the UK. My interest in Finnish law stems from what Örücü describes as a ‘problemsolving’ or sociological approach to comparative law, which examines how different legal systems have responded to similar problems in different ways: here, the challenges of an ageing workforce.195 The fact that both jurisdictions are facing a similar problem warrants the comparison196 and the significant differences in national context increase the potential for mutual learning.197 While comparative research is similar to traditional legal doctrinal research,198 it is important to be mindful of the specific cultural and social circumstances in which comparative law is embedded.199 As Adams notes, the problems and solutions addressed by law are ‘very much connected to the socio-cultural environment that gave rise to them. This environment should be actively and consciously engaged for meaningful comparison to become possible’.200 It is therefore necessary to learn about both law and the society in which it originates when conducting comparative legal analysis201 to respect ‘the complex and manifold relations [that] exist between laws and the local cultures from which laws emerge’ and, in turn, to avoid subverting these connections when borrowing ideas from other legal cultures.202

D.  Organisational Case Studies In Chapter seven I use explanatory organisational case studies to advance our understanding of why employers adopt ‘best practice’ measures in relation to

193  M Adams, ‘Doing What Doesn’t Come Naturally: On the Distinctiveness of Comparative Law’ in M van Hoecke (ed), Methodologies of Legal Research: What Kind of Method for What Kind of Discipline?, European Academy of Legal Theory (Oxford, Hart Publishing, 2011) 237. 194  J Bell, ‘Legal Research and the Distinctiveness of Comparative Law’ in M van Hoecke (ed), ­Methodologies of Legal Research: What Kind of Method for What Kind of Discipline?, European Academy of Legal Theory (Oxford, Hart Publishing, 2011) 167. 195  E Örücü, ‘Developing Comparative Law’ in E Örücü and D Nelken (eds), Comparative Law: A Handbook (Oxford, Hart Publishing, 2007) 52. 196 ibid. 197 D Schiek, ‘Enforcing (EU) Non-Discrimination Law: Mutual Learning between British and ­Italian Labour Law?’ (2012) 28 International Journal of Comparative Labour Law & Industrial Relations 489, 508. 198  Bell (n 194 above) 167. 199  ibid 168, 170; Adams (n 193 above) 230. 200  Adams (n 193 above) 231. 201  Bell (n 194 above) 168–69. 202  G Watt, ‘Comparison as Deep Appreciation’ in PG Monateri (ed), Methods of Comparative Law, Research Handbooks in Comparative Law (Cheltenham, Edward Elgar, 2012) 86.

Research Design

 39

age equality. These studies enrich and deepen the statistical findings presented in Chapter five and the comparative analysis in Chapter six. Case studies enable the complexity of the employment of older workers to be studied in depth and ­interpreted in the organisational context.203 This method is particularly effective for addressing contemporary ‘how’ and ‘why’ questions in real-life ­contexts,204 such as the questions pursued in this study. Further, case studies are well-suited for exploring the dynamics of change,205 and engage research participants in the research process,206 undertaking research with people not on people.207 In this study I utilised instrumental case studies, examining each organisation for insight into the broader issues surrounding the employment of older workers, rather than focusing on the case study organisations for their own sake.208

E.  The Delphi Method In Chapter eight I employ an online Delphi survey as a ‘capstone’ or final research tool to bring together the other research methods, and to evaluate and develop ideas for reform of UK age discrimination laws. In this sense, the Delphi method brings the results of my research ‘back’ to law, and considers how the preceding chapters may contribute to the law reform agenda. The Delphi method is a structured group communication process that allows individuals to deal with complex problems as a group.209 The RAND Corporation first used the Delphi method during the Cold War to estimate Soviet nuclear capabilities for the US Air Force.210 The Delphi method has since been used a handful of times in legal and socio-legal research, including for examining the impact of legal rules and how laws operate in practice,211 offering solutions to law-related problems,212

203  J Feagin et al, ‘Introduction: The Nature of the Case Study’ in J Feagin et al (eds), A Case for the Case Study (Durham NC, University of North Carolina Press, 1991) 7–13; Simons (n 187 above) 23. 204  RK Yin, Case Study Research: Design and Methods, 5th edn (Thousand Oaks CA, SAGE, 2013) 10. 205  Feagin et al (n 203 above) 12–13; Simons (n 187 above) 23. 206  Simons (n 187 above) 23. 207  McConville (n 66 above) 211. 208  R Stake, ‘Qualitative Case Studies’ in Y Lincoln and N Denzin (eds), Strategies of Qualitative Inquiry, 3rd edn (Thousand Oaks CA, SAGE, 2008) 123. 209  HA Linstone and M Turoff, ‘Introduction’ in HA Linstone and M Turoff (eds), The Delphi Method: Techniques and Applications (Reading MA, Addison-Wesley, 1977) 3. 210  N Dalkey and O Helmer, ‘An Experimental Application of the Delphi Method to the Use of Experts’ (1963) 9 Management Science 458. 211 JP van Gigch and R Hommes, ‘A Study of How Correctional Counsellors and Psychologists Agree upon Pre-Sentence Recommendations’ (1973) 15 Canadian Journal of Criminology and Corrections 93; B Dziurzynski, ‘FDA Regulatory Review and Approval Processes: A Delphi Inquiry’ (1996) 51 Food and Drug Law Journal 143. 212  HG McDonald and CP Kirsch, ‘Use of the Delphi Method as a Means of Assessing J­udicial ­Manpower Needs’ (1978) 3 Justice System Journal 314; SJ Young et al, ‘Best Case Scenario: The ­Development of a Teaching Tool for Sport Law’ (2004) 14 Journal of Legal Aspects of Sport 1.

40 

Methodology

developing taxonomies and definitions of legal phenomena213 and criteria to evaluate policy initiatives,214 and exploring the future of interactions between law and society.215 Where knowledge is uncertain or imperfect, the Delphi method can achieve reliable group consensus216 and identify divergence of opinion on hypothetical future scenarios.217 The Delphi method is effective for exploring solutions in policy areas with high levels of uncertainty218 and can eliminate distractions and distortions in group discussion.219 The method is therefore more effective at generating new ideas and exploring future scenarios in areas of uncertainty than traditional face-to-face communication.220 At a practical level, the Delphi method allows participants to think through responses, promoting careful and thoughtful contributions, while the anonymity of responses ensures open and honest discussion. While a range of Delphi structures can be used, the ‘policy Delphi’, which I adopted, is designed to identify, examine and estimate the impact, consequences and acceptability of particular policy options.221 Further, it allows respondents to react to and assess different viewpoints on policy issues.222 Therefore, it is particularly well suited to coordinating and structuring respondents’ thinking around how complex legal issues might develop and evolve in the future,223 and provides a constructive forum for discussion, the building of consensus224 and the clarification of different ideas and viewpoints.225 213  MF Hudson, ‘Elder Mistreatment: A Taxonomy with Definitions by Delphi’ (1991) 3 Journal of Elder Abuse & Neglect 1. 214  E Guglyuvatyy, ‘Identifying Criteria for Climate Change Policy Evaluation in Australia’ (2010) 7 Macquarie Journal of Business Law 98. 215  KN Wright, ‘A Delphi Assessment of the Effects of a Declining Economy on Crime and the Criminal Justice System’ (1982) 46 Federal Probation 36; LF Travis et al, ‘The Future of Sentencing and Parole: A Delphi Reassessment of Sentencing and Parole Reforms’ (1985) 10 Criminal Justice Review 45; SJ Young and LM Jamieson, ‘Perceived Liability and Risk Management Trends Impacting Recreational Sports into the 21st Century’ (1999) 9 Journal of Legal Aspects of Sport 151. 216  G Aichholzer, ‘The Delphi Method: Eliciting Experts’ Knowledge in Technology Foresight’ in A Bogner et al (eds), Interviewing Experts, Research Methods Series (Basingstoke/New York, Palgrave Macmillan, 2009) 252–53. 217  MK Rayens and EJ Hahn, ‘Building Consensus Using the Policy Delphi Method’ (2000) 1 Policy, Politics, & Nursing Practice 308, 308. 218  See A Rotondi and D Gustafson, ‘Theoretical, Methodological and Practical Issues Arising out of the Delphi Method’ in M Adler and E Ziglio (eds), Gazing into the Oracle: The Delphi Method and Its Application to Social Policy and Public Health (London, Jessica Kingsley, 1996) 42. 219  Aichholzer (n 216 above) 252–53. 220  E Ziglio, ‘The Delphi Method and Its Contribution to Decision-Making’ in M Adler and E Ziglio (eds), Gazing into the Oracle: The Delphi Method and Its Application to Social Policy and Public Health (London, Jessica Kingsley, 1996) 22. 221  M Turoff, ‘The Policy Delphi’ in HA Linstone and M Turoff (eds), The Delphi Method: Techniques and Applications (Reading MA, Addison-Wesley, 1977) 83. 222 ibid. 223  Aichholzer (n 216 above) 259. 224  Rayens and Hahn (n 217 above) 309. 225 W Rauch, ‘The Decision Delphi’ (1979) 15 Technological Forecasting and Social Change 159, 163.

Conclusion

 41

While the Delphi method has been used to answer a range of legal and quasilegal questions, it has rarely (if ever) been used to inform the process of law reform. This is an area where the Delphi method could add significant value. Therefore, this research uses the Delphi method in an innovative way to enhance and ­complement doctrinal and mixed methods research.

IV. Conclusion This chapter has articulated the reflexive law theoretical standpoint grounding this study and provided a detailed justification for the adoption of a mixed methods research design. In employing qualitative, quantitative, doctrinal and comparative elements, this research casts a nuanced and multi-faceted light on complex legal issues and provides new insights into the practical operation of age discrimination laws in the UK. Thus, this research design is well equipped to address the questions posed in this book. The chapters that follow present the results from each of the research methods described in this chapter.

3 A Doctrinal Critique of UK Age Discrimination Laws In this chapter I critically examine the doctrinal efficacy and limitations of age ­discrimination laws in the EU and UK in the context of employment, with a ­particular focus on compulsory retirement and default retirement ages. I argue that many areas of age discrimination law remain unclear and underdeveloped, particularly in relation to the use of EJRAs and positive action. The laws represent a partial and incomplete form of reflexive regulation, which is likely to impair their effectiveness. Further, the broad e­ xceptions to the prohibition of age discrimination undermine the laws’ instrumental and intrinsic objectives. Therefore, I argue that the law is in substantial need of reform if it is to be an effective means of addressing demographic ageing.

I.  Legal Regulation of Age Discrimination A.  EU Regulation Instruments to regulate age discrimination in employment were first introduced at the EU level in 2000. However, increasing the employment rate of older workers was an EU objective for a number of years prior to that,1 and measures to achieve that end featured in EU economic and employment strategies. Indeed, the Lisbon Strategy was designed, in part, as a response to the challenges of an ­ageing ­population.2 In 1997, the Treaty Establishing the European Community was amended by the Treaty of Amsterdam to include a new Article 6a,3 which empowered the Council of the European Union to take action to combat discrimination

1  See, eg, Commission of the European Communities, ‘Growth, Competitiveness, Employment: The Challenges and Ways Forward into the 21st Century’ (White Paper, 5 December 1993); European Council, ‘Presidency Conclusions—Meeting on 9 and 10 December 1994 in Essen’ (1994). See also Council Resolution 95/C 228/01 of 29 June 1995 on the Employment of Older Workers [1995] OJ C228/1. 2  A Coralia et al, ‘Lisbon Strategy Evaluation Document’ (SEC(2010) 114 final, 2 February 2010) 2. 3 Treaty establishing the European Community, art 13, now Treaty on the Functioning of the ­European Union, art 19.

Legal Regulation of Age Discrimination

 43

on a number of grounds, including age. Empowered by Article 19 of the Treaty on the Functioning of the European Union, in 2000 the Council of the European Union adopted the Framework Directive,4 which established a general framework for equal treatment in employment and occupation, including on the grounds of age. The Framework Directive gives specific expression to the general EU principle of non-discrimination on the grounds of age.5 While the Framework Directive prohibits direct and indirect age discrimination in employment and occupation,6 it also provides for a number of specific exceptions to the principle of equal treatment. Member States of the EU may provide that a difference of treatment does not constitute discrimination in three situations: where there are genuine occupational requirements (GORs); where the different treatment is objectively justified; and in relation to positive action. First, Member States may provide that occupational requirements are not discriminatory where ‘such a characteristic constitutes a genuine and determining occupational requirement, provided that the objective is legitimate and the requirement is proportionate’.7 Secondly, Article 6(1) of the Framework Directive provides that Member States may: provide that differences of treatment on grounds of age shall not constitute discrimination, if, within the context of national law, they are objectively and reasonably justified by a legitimate aim, including legitimate employment policy, labour market and vocational training objectives, and if the means of achieving that aim are appropriate and necessary.

The Framework Directive provides a non-exhaustive list of examples of differences of treatment that might fall within the provision, including: (a) the setting of special conditions on access to employment and vocational training, employment and occupation, including dismissal and remuneration conditions, for … older workers … to promote their vocational integration or ensure their protection; … (c) the fixing of a maximum age for recruitment which is based on the training requirements of the post in question or the need for a reasonable period of employment before retirement.

It is the responsibility of courts to determine the boundaries of what is justified under Article 6(1).8 This approach has two key limitations. First, the test is

4  Council Directive 2000/78/EC of 27 November 2000 establishing a general framework for equal treatment in employment and occupation [2000] OJ L303/16. 5  Case C-144/04 Mangold v Helm (22 November 2005), [2005] ECR I-9981, para 75; Case C-555/07 Kücükdeveci v Swedex GmbH & Co KG (19 January 2010), EU:C:2010:21, paras 21, 50. 6  arts 1, 2. 7  art 4(1) (the so-called ‘GOR’). 8 M Sargeant, ‘Distinguishing between Justifiable Treatment and Prohibited Discrimination in Respect of Age’ (2013) 4 Journal of Business Law 398, 401.

44 

UK Age Discrimination Laws: A Critique

unclear and provides insufficient certainty regarding when an exception will be acceptable.9 In particular, determining what will constitute a ‘legitimate aim’ and be ‘appropriate and necessary’ provides significant scope for judicial discretion and differing interpretations. By leaving these issues to be resolved by the courts, the Framework Directive makes frequent legal challenges inevitable, requiring significant time and expense to achieve a level of clarity. Secondly, the test affords considerable leeway for Member States to undermine the principle of equal treatment on the grounds of age,10 significantly limiting the protection available to older workers and effectively legitimising discrimination on the grounds of age.11 This undermines the Framework Directive’s capacity to support older workers or to uphold any notion of equality on the basis of age. It is therefore incompatible with the law’s instrumental and intrinsic objectives. The third exception to the equal treatment principle, Article 7(1) of the Framework Directive, makes provision for Member States to take positive action to achieve equality, by noting that ‘the principle of equal treatment shall not prevent any Member State from maintaining or adopting specific measures to prevent or compensate for disadvantages linked to [age] … with a view to ensuring full equality in practice’. While this raises the possibility of equality of outcomes, it does little to encourage Member States to address disadvantage: rather, such measures are merely not ‘prevented’ by the Framework Directive and need to be positively adopted by Member States. It is therefore unsurprising that there has been limited use of positive action measures in European countries.12

(i)  Mandatory Retirement The consequences of the broad exceptions in the Framework Directive to the principle of equal treatment have been explored most fully in relation to retirement ages. Mandatory retirement is coming under increased scrutiny as governments attempt to extend working lives13 and individuals resist retirement due to ­inadequate pension

9 See further E Muir, ‘Fine-Tuning Non-Discrimination Law: Exceptions and Justifications ­ llowing for Differential Treatment on the Ground of Age in EU Law’ (2015) 15 International Journal A of Discrimination and the Law 38. 10 Some might describe this as ‘flexibility’ to accommodate the ‘special’ nature of age discrimination. This is critiqued in Chapters 1 and 9 of this volume. 11  M Sargeant, ‘Age Discrimination’ in M Sargeant (ed), The Law on Age Discrimination in the EU (Deventer, Kluwer Law International, 2008) 3; M Sargeant, ‘The European Court of Justice and Age Discrimination’ [2011] Journal of Business Law 144, 148. 12  See B Hepple, ‘Equality at Work’ in B Hepple and B Veneziani (eds), The Transformation of Labour Law in Europe: A Comparative Study of 15 Countries 1945–2004 (Oxford, Hart Publishing, 2009) 151–54. 13 Performance and Innovation Unit, ‘Winning the Generation Game: Improving Opportunities for People Aged 50–65 in Work and Community Activity’ (April 2000) 47; European Commission, ‘Annexes to the Impact Assessment Accompanying the Document White Paper: An Agenda for Adequate, Safe and Sustainable Pensions’ (Commission Staff Working Document, 16 February 2012) 82–83.

Legal Regulation of Age Discrimination

 45

benefits.14 However, the Framework Directive is made without prejudice to national provisions laying down retirement ages.15 The Court of Justice of the European Union (CJEU) has explicitly endorsed the use of compulsory retirement ages if the provisions are objectively and reasonably justified by a legitimate aim and the means of achieving that aim are appropriate and necessary, in accordance with Article 6(1).16

(ii)  Legitimate Aims To be ‘legitimate’, the aims underlying retirement provisions must have a public interest nature beyond purely individual reasons particular to an employer’s ­situation.17 However, in pursuing legitimate aims, a national rule may allow a degree of flexibility for employers.18 These aims do not need to be explicitly ­specified in legislation so long as the general context of the provision allows for the aims to be identified.19 Further, the aims underlying a provision may change over time without affecting the validity of the law itself.20 In the context of retirement provisions, the CJEU has held that legitimate aims might include: —— the creation of a ‘favourable age structure’ to establish a balance between generations;21 —— distributing work and professional opportunities between generations;22 —— planning for staff departures and recruitment;23 —— encouraging recruitment and promotion of young people24 and other ­categories of workers;25 14 See, eg, E Whitehouse, ‘Pensions and the Crisis: How Should Retirement-Income Systems Respond to Financial and Economic Pressures?’ (OECD, 2009). This is discussed further in Chapter 1 of this volume. 15  See Recital 14 in the Preamble to the Framework Directive. However, this has not acted to exclude retirement ages from review under the Framework Directive: see further Muir (n 9 above) 43–44. 16  See, eg, Case C-411/05 Palacios de la Villa v Cortefiel Servicios SA (16 October 2007), [2007] ECR I-8531, para 73; Case C-388/07 R (Age Concern England) v Secretary of State for Business Enterprise and Regulatory Reform (5 March 2009), [2009] 3 CMLR 4; Case C-45/09 Rosenbladt v Oellerking ­Gebäudereinigungsges (12 October 2010), [2011] 1 CMLR 32; Case C-250/09 Georgiev v Tehnicheski universitet—Sofia, filial Plovdiv (18 November 2010), [2011] 2 CMLR 7; Cases C-159/10 and C-160/10 Fuchs v Land Hessen (21 July 2011), [2011] 3 CMLR 47. 17  Age Concern (n 16 above) para 46. 18 ibid. 19  Palacios (n 16 above) paras 54–57; Case C-141/11 Hörnfeldt v Posten Meddelande AB (5 July 2012), para 24; Case C-286/12 Commission v Hungary (6 November 2012), para 58. 20  Fuchs (n 16 above) paras 41–42. 21  Georgiev (n 16 above) para 46; Fuchs (n 16 above) paras 47, 49; Commission v Hungary (n 19 above) para 62. 22  Palacios (n 16 above) para 53; Georgiev (n 16 above) para 42; C-341/08 Petersen v ­Berufungsausschuss für Zahnärzte für den Bezirk Westfalen-Lippe (12 January 2010), [2010] 2 CMLR 31, para 65; Rosenbladt (n 16 above) para 43. 23  Rosenbladt (n 16 above) paras 60–62. 24  Georgiev (n 16 above) para 45; Rosenbladt (n 16 above) paras 43, 60–62; Fuchs (n 16 above) paras 47, 49; Hörnfeldt (n 19 above) para 29. 25  Palacios (n 16 above) para 65.

46 

UK Age Discrimination Laws: A Critique

—— avoiding legal disputes with older employees over their ability to perform their duties26 or the need to dismiss older employees on performance grounds,27 particularly in ‘situations which are humiliating for elderly workers’;28 and —— standardising the age-limit for compulsory retirement in a specific sector.29 However, aims beyond the scope of legitimate employment policy, labour market and vocational training objectives will not be accepted30 and budget savings on their own are not considered a legitimate aim.31 Member States may only take budgetary constraints into account if they are considered alongside other factors (such as social, political or demographic issues).32 Underpinning these justifications are three suppositions: the fair innings argument; the need to avoid humiliating a long-serving employee; and assumptions based on workforce planning. However, there are strong reasons for questioning the CJEU’s acceptance of these policy assumptions. (a)  ‘Fair Innings’ Justifying mandatory retirement because of the need to distribute work and professional opportunities between generations and encourage the recruitment and promotion of young people is grounded in the ‘fair innings’ argument, or the idea that policies should take into account all advantages an individual has experienced cumulatively over their life.33 This argument posits that ill treatment of or discrimination against older workers can be offset by perceived advantages they enjoyed earlier in life.34 By this reasoning, older workers, who have ‘had their chance’ in the labour market, need to retire to make way for the next generation. Mandatory retirement will therefore open up jobs for (younger) workers in a ‘fair’ and ‘bloodless’ way.35 The fair innings argument is ‘fundamentally unsound’.36 First, the argument assumes there are a fixed number of opportunities in the economy, and that employment of a new (younger) worker requires the removal of another (older)

26 

Fuchs (n 16 above) paras 47, 50. Rosenbladt (n 16 above) para 43. Hörnfeldt (n 19 above) para 34. 29  Commission v Hungary (n 19 above) para 61. See also Lady Hale’s helpful categorisation in Seldon v Clarkson Wright & Jakes (A partnership) [2012] UKSC 16 (25 April 2012) [50]. 30 C-447/09 Prigge v Deutsche Lufthansa AG (19 May 2011), [2011] IRLR 1052 para 82. 31  Fuchs (n 16 above) para 74. 32  ibid para 73. 33  S Fredman, ‘The Age of Equality’ in S Fredman and S Spencer (eds), Age as an Equality Issue: Legal and Policy Perspectives (Oxford, Hart Publishing, 2003) 47. 34  See, eg, Seldon (n 29 above) [4]. 35  R Posner, Aging and Old Age (Chicago IL, University of Chicago Press, 1995) 355; B Hepple, ‘Age Discrimination in Employment: Implementing the Framework Directive 2000/78/EC’ in S Fredman and S Spencer (eds), Age as an Equality Issue: Legal and Policy Perspectives (Oxford, Hart Publishing, 2003) 90–91. 36  Fredman, ‘The Age of Equality’ (n 33 above) 47; see also Performance and Innovation Unit (n 13 above) 39–40. 27  28 

Legal Regulation of Age Discrimination

 47

worker. This does not hold at the macro-economic level (though, significantly, it might at the level of an individual firm).37 Instead, jobs can create further jobs, meaning it is not necessary to introduce or retain compulsory retirement to promote the employment of younger workers.38 Indeed, countries with high levels of employment of older people also generally have high levels of employment of younger people.39 Secondly, applying retirement rules across the workforce is not ‘fair’, as they do not impact equally on all generations or individuals. Younger and older workers are not subject to the same provisions over the course of their working lives, as laws continually change and evolve.40 Indeed, even the same laws may affect generations differently due to historical and cultural differences between age groups.41 The same retirement rules may therefore have disparate practical impacts on individuals of different ages. The same laws may also affect individuals within generations differently, exacerbating social inequality and marginalisation.42 As a result, retirement rules may worsen other forms of inequality, both within and between generations, making them fundamentally unfair. Thirdly, removing older workers from employment does not guarantee that new (or younger) workers will replace them. Young workers may lack the skills or experience to fill a position vacated by an older worker. Further, particularly in times of austerity, vacant positions may not be filled at all. Given these concerns, Sargeant has criticised the CJEU’s ‘somewhat alarming readiness’ to accept fair innings aims as legitimate.43 By relying on a discredited ‘fair innings’ argument to justify retirement provisions, the CJEU has rested its judgments on precarious foundations.

37 E Dewhurst, ‘Intergenerational Balance, Mandatory Retirement and Age Discrimination in Europe: How Can the ECJ Better Support National Courts in Finding a Balance between the ­Generations’ (2013) 50 Common Market Law Review 1333, 1352–59. The degree to which the ‘fair innings’ argument holds at the level of the individual firm will depend on the internal labour market in each firm. There is wide variation in internal labour markets across and between firms and in different industries and occupations: PB Doeringer and MJ Piore, Internal Labor Markets and Manpower ­Analysis (Lexington MA, Heath Lexington Books, 1971) xi, 2. The fair innings argument is most likely to hold where the internal labour market of a firm has rigid ‘lines of progression’ and ‘internal career ladders’, limited ‘ports of entry’ and workers have limited mobility between firms. This is less likely to occur for professionals, where job markets tend to stretch across more than one establishment: ibid 3, 43. Where the ‘fair innings’ argument applies to an individual firm, mandatory retirement ages highlight a tension between individual employers’ interests and the interests of the economy as a whole. 38 Performance and Innovation Unit (n 13 above) 39–40; Fredman, ‘The Age of Equality’ (n 33 above) 47. 39  Fredman, ‘The Age of Equality’ (n 33 above) 46; see further Sargeant, ‘Distinguishing between Justifiable Treatment and Prohibited Discrimination in Respect of Age’ (n 8 above) 410–12. 40  Fredman, ‘The Age of Equality’ (n 33 above) 38. 41 ibid. 42  For example, mandatory retirement will disproportionately affect those with disrupted work histories and limited pension entitlements. This will mostly affect women, exacerbating existing gender inequalities. 43 Sargeant, ‘Distinguishing between Justifiable Treatment and Prohibited Discrimination in Respect of Age’ (n 8 above) 409.

48 

UK Age Discrimination Laws: A Critique

(b)  Maintaining ‘Dignity’ for Long-standing Employees Justifying retirement provisions due to the need to avoid ‘humiliating’ performance management for older workers is also controversial. While mandatory retirement is arguably a less intrusive mechanism for managing declining performance of senior workers than ‘degrading’ personal appraisals,44 this conflates age with capacity and wrongly assumes that ageing is necessarily a process of decline and deterioration.45 Age is not a good proxy for capability:46 if anything, older workers form a more heterogeneous group than younger workers, as people experience ageing at different rates and in different ways.47 Further, as Fredman argues, ‘it is also an affront to the dignity of the individual to assume that he or she automatically shares the characteristics of everyone else in his or her age group’.48 Rather than relying on stereotypical assumptions of capability based on age, people should be treated as individuals, and have their capabilities assessed on an individual basis. Performance appraisals will not be degrading or undermine older workers’ dignity unless declining capability with age is regarded as stigmatic.49 Indeed, if individual performance appraisals undermine the dignity of older workers, it is unclear why younger workers should be subjected to appraisals at the expense of their dignity. It is at least arguable that the dignity argument is concerned with respecting older workers’ dignity in recognition of their loyalty to the organisation after years of service. However, older workers will not always have longer periods of service or more loyalty to an organisation than younger workers. Further, it is possible to uphold employees’ dignity and respect employee loyalty while still addressing performance issues. While the ‘dignity’ argument is flawed, it may still influence employers’ practices: employers who view capability assessments as contrary to older workers’ dignity and loyalty may be reluctant to introduce performance appraisals for their entire workforce. Further, a failure to use performance appraisals in the past may mean that employers lack the capacity and skills to performance manage older workers effectively. Failed capability dismissals may well lead to an increase in unfair dismissal claims by older workers. Therefore, while flawed in theory, the ‘dignity’

44  See Posner (n 35 above) 351; RA Epstein, Equal Opportunity or More Opportunity? The Good Thing about Discrimination (London, Civitas, 2002) 29; R (Age UK) v Secretary of State for Business, Innovation & Skills [2009] EWHC 2336 (Admin), [2009] IRLR 1017, [72]–[76]. 45  See further KW Schaie, ‘Theories of Ageing’ in NJ Smelser and PB Baltes (eds), International Encyclopedia of the Social and Behavioral Sciences (Oxford, Elsevier Science Ltd, 2001) 317, 319–20; CL Estes et al, Social Theory, Social Policy and Ageing: A Critical Introduction (Buckingham, Open University Press, 2003) 18, 29. 46  S Harper and S Marcus, ‘Age-Related Capacity Decline: A Review of Some Workplace Implications’ [2006] Ageing Horizons 20 (Oxford Institute of Ageing). 47  JG Evans, ‘Age Discrimination: Implications of the Ageing Process’ in S Fredman and S Spencer (eds), Age as an Equality Issue: Legal and Policy Perspectives (Oxford, Hart Publishing, 2003) 16. For further discussion, see E Dewhurst, ‘Are Older Workers Past Their Sell-by-Date? A View from UK Age Discrimination Law’ (2015) 78 Modern Law Review 189, 194–97. 48  Fredman, ‘The Age of Equality’ (n 33 above) 45. 49  ibid 45.

Legal Regulation of Age Discrimination

 49

argument may continue to shape employer responses to the ageing ­workforce. Abandoning the argument may have significant practical consequences for individual employers.50 This is taken up further in Chapter four. (c)  Workforce Planning It is also questionable whether, as the CJEU asserts, mandatory retirement can (or should) assist with workforce planning, either by allowing employers to plan for staff departures and recruitment or by facilitating the creation of a ‘favourable [workforce] age structure’. The CJEU’s position is consistent with comments in the literature: mandatory retirement is seen as facilitating planning for individual careers, work allocation, and training and recruitment, as the workforce has a known attrition profile.51 Employee attrition is only marginally related to retirement: employees can leave an employer for a variety of reasons. On its own, mandatory retirement is unlikely to make workforce planning substantially easier, as essential roles and skills may equally be held by early- and mid-career workers who are ‘susceptible to recruitment by other employers’.52 Planning on the basis of a ‘known attrition profile’ ignores the broader transience of modern employment.53 Further, at a broader level, the idea that mandatory retirement facilitates planning in such a wide range of areas and workplace decisions (about training, work allocation, and career profiles) endorses the use of chronological age as a decision-making mechanism across all facets of work. This is clearly inconsistent with notions of equality and the desire to reduce discrimination on the grounds of age, and fundamentally undermines the intrinsic objectives of age discrimination laws. In sum, while there is legal clarity regarding what ‘legitimate aims’ may justify national retirement provisions, there are substantial grounds to question whether these aims are really ‘legitimate’. Indeed, the above discussion shows that they are likely to be flawed in many cases.

(iii)  Appropriate and Necessary Assuming Member States can produce an objective justification, they then have a broad discretion in defining measures to achieve a legitimate aim.54 ­However, 50  See S Manfredi, ‘Retirement, Collective Agreement, and Age Discrimination: Implications for the Higher Education Sector in the UK’ (2011) 11 International Journal of Discrimination and the Law 65, 76–77. 51  Posner (n 35 above) 324; Epstein (n 44 above) 28; Hepple, ‘Age Discrimination in Employment’ (n 35 above) 91. 52  See further A Rappaport et al, ‘The Aging Workforce Raises New Talent Management Issues for Employers’ (2003) 23 Journal of Organizational Excellence 55, 60. 53  M Kohli, ‘The World We Forgot: A Historical Review of the Life Course’ in VW Marshall (ed), Later Life: The Social Psychology of Aging (Beverley Hills CA, Sage, 1986) 295; M Kohli, ‘The Institutionalization of the Life Course: Looking back to Look Ahead’ (2007) 4 Research in Human Development 253, 261. 54  Palacios (n 16 above) para 68.

50 

UK Age Discrimination Laws: A Critique

in exercising this discretion, they must not frustrate the prohibition of age ­discrimination in the Framework Directive.55 Member States are required to ­balance the desirability of older workers remaining in employment with other (possibly divergent) interests, such as an individual’s desire to retire, and the need to promote young people’s entry into the labour market.56 For a measure to be appropriate and necessary it must not appear unreasonable in the light of the aim pursued and must be supported by evidence.57 A key consideration in applying the proportionality test is whether workers are entitled to a ‘not unreasonable’ pension following retirement:58 if so, retirement rules will not ‘unduly prejudic[e] the legitimate claims of workers’.59 Even if retirement income is deemed to be inadequate, retirement rules are likely to be proportionate if an employee is able to continue working after being retired, either with their current employer or with a different company60 or under a different type of employment arrangement, such as a fixed-term contract.61 While this reflects a concern with securing adequate income for the elderly—if not through a pension, then through the possibility of paid employment—older workers experience substantial difficulties in recruitment, and are unlikely to return to the labour market once they have lost their job.62 It is unrealistic to assume that retired workers will have the chance to return to work to secure an adequate income. Further, this argument implicitly acknowledges that older workers have no right or entitlement to remain in employment, as a retirement rule cannot ‘unduly prejudic[e]’ their ‘legitimate claims’. This reinforces the idea of a ‘fair innings’ in employment. A retirement rule is also more likely to be proportionate, according to the CJEU, if it has been the subject of collective bargaining and/or is tailored to the circumstances of the case.63 This reflects the argument that retirement ages should not be regarded as blanket age discrimination, but rather as part of a ­mutually agreed company personnel policy, or collective agreement, generally negotiated by ­individuals with reasonable bargaining power. [They] should only be banned if there are explicit reasons for governments to override such private contractual arrangements.64 55  Age Concern (n 16 above) para 51. That said, it is arguable that having a compulsory retirement age, in and of itself, frustrates this prohibition: M Sargeant, ‘United Kingdom’ in M Sargeant (ed), The Law on Age Discrimination in the EU (The Hague, Kluwer Law International, 2008) 224; cf M Gunderson, Banning Mandatory Retirement: Throwing out the Baby with the Bathwater, Backgrounder No 79 (Toronto ONT, CD Howe Institute, 2004) 6. 56  Palacios (n 16 above) paras 69, 71. This again raises consideration of the ‘fair innings’ argument. 57  Palacios (n 16 above) para 72; Fuchs (n 16 above) para 83. 58  Palacios (n 16 above) para 73; Georgiev (n 16 above) para 54; Rosenbladt (n 16 above) paras 43, 48; Fuchs (n 16 above) paras 66–67; Hörnfeldt (n 19 above) para 42. 59  Palacios (n 16 above) para 73; Georgiev (n 16 above) para 54; Fuchs (n 16 above) para 66. 60  Rosenbladt (n 16 above) paras 73–76; Fuchs (n 16 above) para 66. 61  Hörnfeldt (n 19 above) para 40–41. 62  See, eg, Sargeant, ‘United Kingdom’ (n 55 above) 224. 63  Palacios (n 16 above) para 74; Rosenbladt (n 16 above) paras 49–50, 67–69; Hörnfeldt (n 19 above) para 32. 64  Gunderson (n 55 above) 6.

Legal Regulation of Age Discrimination

 51

The CJEU’s position implies that age equality can be ‘trumped’ by freedom of ­contract, majority rule or bargaining,65 so long as the retirement rule is ­appropriately negotiated between the parties. However, in fact, retirement provisions are rarely imposed following ­meaningful individual consent. Many employees have no choice but to accept the terms on which employment is offered, and will have little or no opportunity to bargain over or amend a retirement age specified in an employment c­ ontract.66 Further, most union members will have only ‘diluted influence’ over the terms of a collective agreement, limiting the meaningfulness of their ­consent.67 A retirement age is also unlikely to be a primary concern or consideration of many (particularly younger) employees at the time when a contract is signed. In sum, consent to a retirement age is ‘largely illusory’ in most cases,68 undermining the argument that retirement ages should be seen as a negotiated private contractual arrangement. Finally, retirement rules are more likely to be proportionate where they apply to professions with a limited number of posts, where individuals cannot be promoted without a vacancy.69 Again, this reflects a questionable belief in the ‘fair innings’ argument and the idea that retirement rules help to facilitate the distribution of work and professional opportunities between generations. Overall, the application of the proportionality test does not subject retirement provisions to rigorous scrutiny or require substantial evidence of necessity from Member States.70 It appears likely that a retirement provision that reflects any of the above considerations will be deemed valid by the CJEU, providing little protection for workers who do not wish to retire. That said, Commission v Hungary71 may mark a shift towards more rigorous scrutiny of the proportionality of mandatory retirement ages. In that case, which involved the lowering of compulsory retirement ages for judges, prosecutors and notaries in Hungary from 70 to 62 years of age, the CJEU ruled that Hungary had ‘failed to provide any evidence’ that more lenient provisions could not have achieved the same aims,72 particularly given that other changes to increase public service retirement ages were being introduced there via a ‘gradual staggering’.73 Therefore, the provisions were not necessary to achieve the standardisation of retirement ages across the public service. Further, the provisions were not appropriate to achieve a more balanced 65 M Connolly, ‘The Coalition Government and Age Discrimination’ [2012] Journal of Business Law 144, 158. 66 ibid. 67 ibid. 68 ibid. 69 See Georgiev (n 16 above) para 52. 70  See E Dewhurst, ‘The Development of EU Case-Law on Age Discrimination in Employment: “Will You Still Need Me? Will You Still Feed Me? When I’m Sixty-Four”’ (2013) 19 European Law Journal 517, 526, 534–36. 71  Commission v Hungary (n 19 above). 72  ibid para 71. 73  ibid paras 73–74.

52 

UK Age Discrimination Laws: A Critique

age structure among judges, prosecutors and notaries:74 the sudden retirement of those aged between 62 and 70 would create a ‘very significant acceleration’ of turnover in positions in 2012, and a ‘radical slowing down’ thereafter, particularly as the retirement age would then be progressively increased (with the rest of the public service) to 65.75 Therefore, the change would not create a balanced age structure in the medium and long term.76 Commission v Hungary demonstrates that the CJEU will carefully examine the proportionality of retirement ages in some circumstances. However, that case is exceptional for three reasons. First, it was not a preliminary ruling, meaning the CJEU was not limited to providing guidance to national courts.77 Secondly, the case involved the abrupt lowering of compulsory retirement ages, rather than the maintenance of established retirement provisions. Unlike in other cases, the change therefore failed ‘to protect the legitimate expectations of the persons ­concerned’.78 Thirdly, the broader impact of the case may be limited by its particular political circumstances: prior to the CJEU’s decision, the Venice Commission had condemned Hungary’s changing of judicial retirement rules as potentially ‘open[ing] the way for undue influence on the composition of the judiciary’79 and raised concerns that the rules may be used ‘as a means to put an end to the term of office of persons elected or appointed under the previous Constitution’.80 The new retirement rules potentially undermined the independence of the judiciary and the rule of law. It is therefore unsurprising that the CJEU adopted a more rigorous approach in this case, as age discrimination law was being used to secure broader political ends. However, this more rigorous approach is unlikely to be extended to subsequent cases, particularly where a retirement age has already been in place for some time.81

B.  UK Regulation The Framework Directive was implemented in the UK by the Employment Equality (Age) Regulations 2006, SI 2006/1031 (‘the Regulations’), and later consolidated into the Equality Act 2010 (‘the EqA’). As noted in Chapter one, previous government attempts to ‘persuade’ organisations not to discriminate on

74 

ibid para 79. ibid para 78. 76  ibid para 77. 77  See further Dewhurst, ‘Intergenerational Balance, Mandatory Retirement and Age Discrimination in Europe’ (n 37 above) 1345. 78  Commission v Hungary (n 19 above) para 68. 79 European Commission for Democracy Through Law, ‘Opinion on the New Constitution of Hungary’ (Opinion, 20 June 2011) para 108. 80  ibid para 140. 81 Indeed, Commission v Hungary (n 19 above) was regarded as being of limited assistance in determining the proportionality issue in Seldon v Clarkson Wright & Jakes [2014] UKEAT/0434/13/RN (13 May 2014) [33]. 75 

Legal Regulation of Age Discrimination

 53

the basis of age, via soft law measures and a Voluntary Code of Practice on Age ­Discrimination, made limited progress at addressing age discrimination.82 Thus, the Regulations and EqA marked a significant turning point in UK law. The EqA prohibits direct and indirect discrimination, harassment and victimisation in the workplace on the grounds of age during recruitment, in setting the terms of employment, deciding to award promotions and provide training and in dismissal.83 However, protection from age discrimination is limited in three key ways. First, direct discrimination because of age can be objectively justified if the treatment is shown to be ‘a proportionate means of achieving a legitimate aim’.84 This does not apply to any other protected characteristic. Secondly, indirect discrimination may also be justified as a proportionate means of achieving a legitimate aim.85 While these tests sound similar, they are different in practice.86 Thirdly, Schedule 9 of the EqA includes a number of specific exceptions to the prohibition of age discrimination in employment, including for: —— occupational requirements which are a proportionate means of achieving a legitimate aim;87 —— service in the armed forces;88 —— benefits based on length of service that: —— relate to a period of service of up to five years duration; or —— relate to a period of service exceeding five years duration and which the employer reasonably believes fulfil a business need;89 —— the national minimum wage;90 —— enhanced redundancy payments;91 and —— contributions to personal pension schemes (by order of a Minister).92 As with Article 7(1) of the Framework Directive, the EqA allows (but does not require) employers to take positive action that is a proportionate means of: —— enabling or encouraging persons who share a protected characteristic to overcome or minimise disadvantage connected to that characteristic; —— meeting the needs of persons who share a protected characteristic which are different from the needs of persons who do not share the characteristic; or 82 D Grant, ‘Older Women, Work and the Impact of Discrimination’ in M Sargeant (ed), Age ­ Discrimination and Diversity: Multiple Discrimination from an Age Perspective (Cambridge, Cambridge University Press, 2011) 42; see also Charted Institute of Personnel and Development, ‘Tackling Age Discrimination in the Workplace: Creating a New Age for All’ (October 2005). 83  EqA s 39. 84  ibid s 13(2). 85  ibid s 19(2). 86 See Age Concern (n 16 above), paras 53–67 (especially at para 65); see also Seldon (n 29 above) [50]. 87  EqA sch 9, s 1(1). 88  ibid sch 9, s 4(3). 89  ibid sch 9, s 10. 90  ibid sch 9, ss 11–12. 91  ibid sch 9, s 13. 92  ibid sch 9, s 16.

54 

UK Age Discrimination Laws: A Critique

—— enabling or encouraging persons who share a protected characteristic to ­participate in an activity in which their participation is disproportionately low.93 Similarly, positive action may be taken in recruitment and promotion to address a disadvantage or disproportionately low participation.94 However, a person with a protected characteristic may only be treated more favourably if: they are as ­qualified as the other person (the so-called tie-break); the employer or company does not have a policy of treating persons who share the protected characteristic more favourably in recruitment or promotion; and the action is a proportionate means of overcoming or minimising the disadvantage, or promoting participation in the activity.95 While positive action is allowed under the EqA, most employers are unlikely to take advantage of the provisions: positive action is seen as too risky and resourceintensive to be beneficial, and may lead to a ‘potential minefield’ of legal action if employers ‘get it wrong’.96 Rather than being helpful to employers, the sections are a ‘trap for the well intentioned’.97 The drafting of the sections makes them ‘too dangerous [for employers] to use safely’,98 limiting any possibility of positive action in the UK.99 The EqA also establishes a public sector equality duty (PSED), requiring public authorities or people exercising public functions to, in exercise of their functions, have due regard to the need to: —— eliminate discrimination, harassment, victimisation and conduct prohibited by the EqA; —— advance equality of opportunity between persons who share and do not share a protected characteristic, including by: —— removing or minimising disadvantages suffered by persons who share a protected characteristic that are connected to that characteristic; —— taking steps to meet the particular needs of persons who share a protected characteristic; and 93 

ibid s 158(1)–(2). ibid s 159. 95  ibid s 159(3)–(4). 96 A Hoggarth and L Taft, ‘Positive Action: Tie Breaks in the Recruitment Process’ (Prolegal, Thomson Reuters, 2 May 2012): www.prolegal.co.uk/employment-lawyer/tie-breaks-in-therecruitment-process.htm. 97  A Hoggarth and L Taft, ‘Legal Insight: Positive Discrimination—a Trap for the Well Intentioned?’ (HRZone, 10 May 2012): www.hrzone.com/topic/recruitment/legal-insight-positive-discriminationtrap-well-intentioned/120274. 98  Kingsley Napley, ‘“Positive Action”—Will It Make Any Difference?’ (Employment Law Blog, 12 May 2011): www.kingsleynapley.co.uk/news-and-events/blogs/employment-law-blog/positive-actionwill-it-make-any-difference. 99  See further L Barmes, ‘Navigating Multi-Layered Uncertainty: EU Member State and Organizational Perspectives on Positive Action’ in G Healy et al (eds), Equality, Inequalities and Diversity: Contemporary Challenges and Strategies, Management, Work and Organisations (Basingstoke/ ­ New York, Palgrave Macmillan, 2010). 94 

Legal Regulation of Age Discrimination

 55

—— encouraging persons who share a protected characteristic to participate in public life or activities in which their participation is disproportionately low; and —— foster good relations between persons who share a relevant protected characteristic and persons who do not share it, including by tackling ­ ­prejudice and promoting understanding.100 A review of the PSED was conducted in 2013 ‘to establish whether the Duty is operating as intended’.101 The Steering Group conducting the review concluded that it was too early to make a final judgement about the impact of the duty, as it was introduced only in April 2011, and the available evidence was as yet inconclusive, particularly in relation to the associated costs and benefits of implementing the duty.102 The Group thus recommended that the government consider conducting a formal evaluation of the duty in 2016.103 In sum, the EqA provides less protection against age discrimination in employment than against other forms of discrimination, particularly as both direct and indirect age discrimination may be justified. This provides significant scope for employers to undermine the principle of equal treatment on the grounds of age. In the hierarchy of discriminatory grounds, age therefore ‘occup[ies] the lowest rung’.104 This is particularly evident in the case of mandatory retirement.

(i)  Mandatory Retirement Compulsory retirement is a key area in which employers might seek to justify direct age discrimination. Prior to the introduction of the Regulations, employers choosing to implement a normal retirement age (NRA) for their workforce were protected by legislation, with employees dismissed on the ground of retirement after reaching the NRA or age 65 being unable to claim unfair dismissal or redundancy payments.105 In drafting the Regulations, the UK government conducted extensive consultation on whether to introduce a national default retirement age (DRA).106 It was ultimately decided to include a DRA of 65 in the Regulations, but to allow employers to retain a lower NRA if it could be

100 

EqA s 149. Government, ‘Review of Public Sector Equality Duty’ (GOV.UK, no date): www.gov.uk/ government/policy-advisory-groups/review-of-public-sector-equality-duty-steering-group. 102 Government Equalities Office, ‘Review of the Public Sector Equality Duty: Report of the Independent Steering Group’ (6 September 2013) 11. 103 For a critique of the review, see M-A Stephenson, ‘Misrepresentation and Omission—an Analysis of the Review of the Public Sector Equality Duty’ (2014) 85 The Political Quarterly 75. 104  C McGlynn, ‘EC Legislation Prohibiting Age Discrimination: “Towards a Europe for All Ages”?’ in A Dashwood et al (eds), Cambridge Yearbook of European Legal Studies 2000, vol 3 (Oxford, Hart Publishing, 2000) 294. See further Age UK (n 44 above) [20]. 105  Employment Rights Act 1996, s 109. 106  DTI, ‘Equality and Diversity: Age Matters’ (July 2003); DTI, ‘Towards Equality and Diversity: Report of Responses on Age’ (June 2003); DTI, ‘Equality and Diversity Coming of Age: Report on the Consultation on the Draft Employment Equality (Age) Regulations 2006’ (March 2006). 101  HM

56 

UK Age Discrimination Laws: A Critique

objectively justified. Kilpatrick describes this decision as a ‘pragmatic concession to employer lobbying’ in which the government ‘buckl[ed] before employer pressure’ to introduce a DRA.107 The introduction of a DRA was extensively criticised for placing age equality second to business performance.108 Under the Regulations, employers were required to consider an employee’s request to work beyond retirement age and could only retire an employee in accordance with complicated procedural provisions.109 Employers’ failure to comply with these provisions resulted in a number of legal challenges to dismissals.110 An employer could also refuse to offer employment to an applicant who was over the employer’s NRA or, if the employer did not have a NRA, over the age of 65, or an applicant who would turn that age within six months.111 Thus, the Regulations effectively endorsed age discrimination in employment, in the form of mandatory retirement ages. The DRA and Regulations were subject to a legal challenge brought by advocacy group Age Concern England (now Age UK). The claimant argued that the degree to which the Regulations allowed derogation from the principle of nondiscrimination was ‘over-broad’ and, as a result, that the Regulations failed to give effect to the terms of the Framework Directive. The claimant took particular issue with regulation 3, which permitted employers to justify direct discrimination, and regulation 30, which provided that it was not unlawful discrimination to dismiss an employee over age 65 on the grounds of retirement (the DRA). In R (Age Concern England) v Secretary of State for Business Enterprise and ­Regulatory Reform,112 the CJEU held that the Regulations fell within the scope of the Framework Directive.113 The CJEU noted that Article 6(1) of the Framework Directive gives Member States the option to provide that some differences of treatment on the grounds of age do not constitute discrimination under national law if the differences of treatment are ‘objectively and reasonably’ justified.114 It was for the national court to determine whether the aims underlying the Regulations were legitimate social policy objectives and whether the national legislature or regulatory authority could legitimately consider the regulations to be

107 C Kilpatrick, ‘The New UK Retirement Regime, Employment Law and Pensions’ (2008) 37 Industrial Law Journal 1, 23. 108  M Flynn, ‘The United Kingdom Government’s “Business Case” Approach to the Regulation of Retirement’ (2010) 30 Ageing and Society 421, 423; M Harcourt et al, ‘The Effects of ­Anti-Age Discrimination Legislation: A Comparative Analysis’ (2010) 26 The International Journal of ­ Comparative Labour Law and Industrial Relations 447, 451. 109 Employment Equality (Age) Regulations 2006, SI 2006/1031, reg 47, sch 6. For further description and critique of these provisions, see Kilpatrick (n 107 above). 110 See, eg, Compass Group Plc v Ayodele [2011] IRLR 802; Howard v Campbell’s Caravans Ltd [2011] UKEAT 0609_10_1205 (12 May 2011); Bailey v R & R Plant (Peterborough) Ltd [2011] UKEAT 0370_10_1805 (18 May 2011). 111  Employment Equality (Age) Regulations 2006, SI 2006/1031, reg 7(4). 112  Age Concern (n 16 above). 113  ibid para 30. 114  ibid para 65.

Legal Regulation of Age Discrimination

 57

‘­appropriate and necessary’ to achieve those aims.115 Despite the broad discretion given to Member States in matters of social policy, the legitimacy of the aims pursued must be established by the Member State to a high standard of proof.116 Further, mere generalisations regarding how a measure might contribute to employment policy, labour market or vocational training objectives would be insufficient evidence of whether the measure was ‘appropriate and necessary’ to achieve the aim.117 Following this decision, in R (Age UK) v Secretary of State for Business, Innovation & Skills,118 the High Court of England and Wales held that the Regulations, and regulations 3 and 30 in particular, were supported by legitimate social policy concerns, namely protecting the integrity of the labour market and ensuring that employers had confidence in the labour market.119 The Court noted that the government had a ‘wide margin of appreciation’ in pursuing these social aims,120 even in the face of evidence to the contrary.121 While upholding DRAs generally as a legitimate ‘social choice’ made by governments,122 the Court had ‘concerns’ regarding the adoption of a DRA of 65 and whether that particular age was proportionate in light of ‘contemporary conditions’.123 While the Court declined to grant relief in the case, this was partly due to the government’s announcement of a review of the DRA.124 The Court stated: ‘I cannot presently see how 65 could remain as a DRA after the review’.125 Over 250 cases challenging retirement under the DRA were stayed pending the outcome in the Age Concern case.126 Prior to even passing the Regulations, the UK government had declared its intention of revisiting the DRA within five years with the ‘long-term aim [of consigning] fixed retirement ages to the past’.127 In July 2010, the government commenced a period of consultation regarding how the DRA should be phased out. In January 2011, the government announced its intention to phase out the DRA from April 2011, and passed the Employment Equality (Repeal of Retirement Age Provisions) Regulations 2011, SI 2011/1069 (‘the 2011 Regulations’) to facilitate the change. From 1 October 2011, it was no longer possible to retire an employee using the DRA. In addition to removing the DRA, the 2011 Regulations 115 

ibid paras 49–50, 52. ibid para 65. 117  ibid para 51. 118  Age UK (n 44 above). 119  ibid [90], [103]. 120  ibid [97]. 121  ibid [106]. 122  ibid [95]. 123  ibid [129]. 124  ibid [130]. 125  ibid [130]. 126  L Keldusild, ‘Retirement under the Age Equality Legislation: The Heyday Challenge’ (2009) 30 Business Law Review 2, 4. See, eg, Johns v Solent SD Ltd [2008] IRLR 88. 127  HM Government, ‘Opportunity Age: Meeting the Challenges of Ageing in the 21st Century’ (March 2005) 20. 116 

58

UK Age Discrimination Laws: A Critique

also removed retirement as a fair reason for dismissal in the Employment Rights Act 1996. While the national DRA was abolished by the 2011 Regulations, employers may still adopt an EJRA so long as it can be objectively justified as a proportionate means of achieving a legitimate aim. That said, it remains unclear when an EJRA will be ‘justified’, although the case of Seldon v Clarkson Wright & Jakes (A partnership)128 provides some limited clarification of the law in this area. In that case, the Supreme Court considered the appeal of a solicitor who claimed he was subject to direct age discrimination when compulsorily retired from the partnership at age 65 in accordance with the partnership deed. While Mr Seldon was retired prior to the abolition of the DRA, the DRA did not apply to members of a partnership129—therefore, retirement under the partnership deed had to be justified as necessary and proportionate in the interests of the partnership (the same test that now applies to all retirement ages). The issues to be considered by the Supreme Court included: (1) whether the aims identified by the Employment Tribunal (ET) were capable of being legitimate aims; (2) whether the firm had to justify the application of the retirement clause in this particular case; and (3) whether relying on the retirement clause in this case was a proportionate means of achieving those aims.

(ii) Legitimate Aims In relation to point (1), the Court concluded that the UK had decided to give employers and partnerships the flexibility to choose which objectives to pursue, so long as these objectives: (a) could count as ‘legitimate objectives of a public interest nature’; (b) were consistent with the state’s social policy aims; and (c) the means used to achieve the objectives were proportionate. Therefore, while it is for states to identify broad social policy aims, employers may articulate and apply those aims as they relate to their particular circumstances. This interpretation of ‘legitimate aims’ appears inconsistent with Article 6(1) of the Framework Directive, which provides that Member States, as opposed to employers, ‘may provide that differences of treatment on the grounds of age shall not constitute discrimination’.130 In considering the CJEU case law, the UK Supreme Court categorised legitimate aims as falling within two broad classes: first, intergenerational fairness; and, second, dignity.131 In relation to the actual aims identified by the ET in this case—ensuring associates were given the opportunity of partnership after a reasonable period; facilitating workforce planning; and limiting the need to use 128 

See n 29 above. 30, which established the DRA, provided that employees at or over the age of 65 could be dismissed on the basis of retirement. In the Seldon case, the firm’s partnership deed applied to ­partners (not employees). Therefore, the DRA and reg 30 did not apply. 130  See further C Barnard, ‘Retiring Gracefully’ (2011) 70 Cambridge Law Journal 304, 306. 131  Seldon (n 29 above) [56]–[57]. 129  Regulation

Legal Regulation of Age Discrimination

59

performance management to remove partners, thereby contributing to the firm’s ‘congenial and supportive culture’—the Court noted that each had been recognised by the CJEU as legitimate social policy aims. The aims could also be related to the circumstances of the firm, making them legitimate in this particular case.132 Lord Hope noted that while the aims were directed to the firm’s own best interests, this did not prevent them from being legitimate social policy aims. This decision means employers will be able to identify legitimate aims to support a retirement policy fairly readily. However, there are substantial grounds for concern in relation to both the intergenerational fairness and dignity aims recognised by the Supreme Court. (a) Intergenerational Fairness The Court regarded intergenerational fairness as a ‘comparatively uncontroversial’ legitimate aim.133 However, what is ‘fair’ between generations is not self-evidently ‘uncontroversial’. Indeed, the Court explicitly recognised multiple (potentially conflicting) interpretations of ‘intergenerational fairness’.134 Further, the ‘fair innings’ argument that underlies most notions of ‘intergenerational fairness’ is fundamentally unsound (see further above). In firms and industries with flexible internal labour markets, there is no justification for using intergenerational fairness as a rationale. Connolly further argues that government documents indicate a shift away from ‘dead man’s shoes’135 and intergenerational fairness as policy aims since the Supreme Court’s decision in Seldon, meaning this aspect of the judgment is ‘no longer good law’.136 Indeed, ‘job-blocking’ has been resoundingly rejected by government documents since at least 2000,137 meaning intergenerational fairness was not a legitimate social policy aim even prior to the Seldon judgment.138 Therefore, despite the CJEU’s acceptance of ‘intergenerational fairness’ as a legitimate aim, it should not be regarded as legitimate in the UK context for justifying mandatory retirement ages.139 132  133 

ibid [67]. ibid [56].

134 ibid.

135  That is, that compulsory retirement of older workers provides a clear and defined career path for more junior workers, facilitating recruitment and retention of younger employees: M Harman and M Hopkins, ‘Will You Still Need Me, Will You Still Feed Me?’ (Birkett Long, 12 August 2011): www.birkettlong.co.uk/site/library/legalnews/will_you_still_need_me_will_you_still_feed_me.html. 136  Connolly (n 65 above) 148. 137 Performance and Innovation Unit (n 13) 6; BIS, ‘Phasing out the Default Retirement Age: Government Response to Consultation’ (13 January 2011) 2; DWP, ‘Employing Older Workers: An Employer’s Guide to Today’s Multi-Generational Workforce’ (February 2013) 10. 138  See also M Connolly, ‘Ruling on Retirement: “A Major Disappointment for Older Workers and Seriously Flawed”’, HR Magazine, 14 May 2012. While government documents have rejected intergenerational fairness, job-blocking and ‘dead man’s shoes’ arguments as policy aims, the ‘fair innings’ argument may still hold at the level of the individual firm: see n 37 above. 139  Vickers and Manfredi argue that ‘intergenerational fairness’ should be replaced as a rationale by ‘intergenerational solidarity’: L Vickers and S Manfredi, ‘Age Equality and Retirement: Squaring the

60

UK Age Discrimination Laws: A Critique

(b) Dignity The second category of legitimate aims identified by the Supreme Court—dignity—was described by Lady Hale as ‘much more controversial’ than intergenerational fairness, particularly given that ‘the assumptions underlying these objectives look suspiciously like stereotyping’ as they conflate age with capacity.140 However, rather than question the CJEU’s decision that ‘the avoidance of unseemly debates about capacity is capable of being a legitimate aim’,141 the Supreme Court endorsed dignity as a legitimate aim, albeit reluctantly. ‘Dignity’ is now clearly regarded as a legitimate aim in the UK context. That said, the dignity argument is unsound (see above) and has been rejected in government policy documents.142 Connolly therefore argues that ‘dignity’ or ‘collegiality’ arguments are not social policy aims: ‘On the contrary, the Government’s view is that dignity is preserved by performance management’.143 In sum, both ‘intergenerational fairness’ and ‘dignity’ rest on precarious logical foundations, and are inconsistent with government social policy objectives. While the UK may have decided to give employers and partnerships the flexibility to choose which legitimate objectives to pursue, these objectives must still be consistent with the state’s social policy aims. Therefore, the Supreme Court in Seldon should not have accepted the dignity and intergenerational fairness arguments put forward by the partnership. By accepting these arguments, the Supreme Court adopted uncritically the social policy aims identified by the CJEU, without considering whether they were appropriate in the UK context.

(iii) Justification in a Specific Case In relation to the second issue to be considered by the Supreme Court in the Seldon case (whether the firm had to justify the application of the retirement

Circle’ (2013) 42 Industrial Law Journal 61, 67–69. However, intergenerational solidarity is an evolving and contested idea, which is unlikely to lend clarity to this area of law: see, eg, A Supiot, Homo Juridicus: On the Anthropological Function of the Law, S Brown (trans) (London, Verso, 2007) 209; VL Bengtson and PS Oyama, ‘Intergenerational Solidarity and Conflict’ in MA Cruz-Saco and S Zelenev (eds), Intergenerational Solidarity: Strengthening Economic and Social Ties (New York, Palgrave Macmillan, 2010) 38–39, 46–47; MA Cruz-Saco, ‘Intergenerational Solidarity’ in MA Cruz-Saco and S Zelenev (eds), Intergenerational Solidarity: Strengthening Economic and Social Ties (Basingstoke, Palgrave Macmillan, 2010) 11. 140  Seldon (n 29 above) [57]. This argument was also considered by the EAT, which found that there was no evidence of under-performance in this case to support a ‘dignity’ argument: Seldon v Clarkson Wright & Jakes [2008] UKEAT 0063_08_1912, [2009] IRLR 267, [70]. However, it considered that such an argument could be successful if supported by appropriate evidence: ibid [71]. According to the EAT: ‘It would not necessarily be surprising if there were to be general recognition that the contribution made by partners became less significant after a particular age. It may also be that the desire not to subject partners to a performance regime, possibly leading ultimately to expulsion, would justify imposing a general retirement age (which could in an appropriate case conceivably be below 65)’: ibid [71]. 141  Seldon (n 29 above) [58]. 142  BIS (n 137 above) 2; DWP (n 137 above) 16. 143  Connolly (n 65 above) 149–50.

Legal Regulation of Age Discrimination

 61

clause in that particular case), the Court held that where a general rule is justified, the application of the rule to a specific case would also generally be justified. Therefore, a rule will only need to be justified in the particular circumstances of the business, not upon each application of the rule. As a result, the firm did not need to justify the application of the retirement rule to Mr Seldon personally.

(iv) Proportionality Finally, in relation to point (3) (whether relying on the retirement clause was a proportionate means of achieving those aims), the Court held that the ET had not considered whether a retirement age of 65 was proportionate, as opposed to a retirement age more broadly: ‘there is a difference between justifying a retirement age and justifying this retirement age’.144 The case was referred back to the ET to consider this question. In May 2013, the ET held that the partnership’s retirement age of 65 was appropriate and reasonably necessary for achieving the aims of staff retention and planning for the future of the firm.145 Associates ‘should see that upon the retirement of partners opportunities were created for succession to partnership’146 and that there was a ‘realistic long-term expectation as to when and where vacancies will arise’.147 In deciding whether the age of 65 was proportionate, the ET considered the importance of consent,148 the existence of the DRA, the SPA, and the fact that the CJEU had considered 65 to be a proportionate age in the past. However, the ET also noted that the position ‘might have been different’ if Mr Seldon had been retired after abolition of the DRA and planned changes to the SPA.149 The ET’s decision on proportionality was upheld by the Employment Appeal Tribunal (EAT) in May 2014.150

(v)  Consequences of the Seldon Decision The Seldon decision has consequences for all employers wishing to retain a retirement age for their workforce, as it will guide the treatment of EJRAs generally going forward (a point expressly recognised by Lady Hale in Seldon).151 Using Seldon, employers will easily be able to identify legitimate aims to justify a retirement age, so long as those aims are relevant to their particular circumstances. This will be particularly straightforward where the organisation has a hierarchy with ­limited senior positions, as in a law firm or university. 144 

Seldon (n 29 above) [68]. Seldon v Clarkson Wright & Jakes [2013] UKET 1100275/2007 (14 May 2013). The collegiality or dignity aim was not raised due to a lack of evidence: ibid [8], [36]. 146  ibid [76]. 147  ibid [77]. 148  See further text surrounding n 197 to 208 below. 149  Seldon (n 145 above) [92]. 150  Seldon (n 81 above). 151  Seldon (n 29 above) [26]. 145 

62 

UK Age Discrimination Laws: A Critique

However, it will be more challenging to prove that the actual retirement age adopted is a proportionate means of achieving these aims. Following the abolition of the DRA, employers may find it difficult to justify 65 as a proportionate retirement age, as the government has deemed a DRA of 65 not to be appropriate for the general workforce. The ET’s further consideration of the Seldon case has provided limited guidance on this issue, as the DRA was still in place at the time of Mr Seldon’s retirement. The ET’s decision on proportionality is limited to a particular historical moment, before the introduction of the 2011 Regulations and changes to the SPA. Indeed, the ET explicitly acknowledged the relevance of the DRA and SPA to the issue of proportionality: ‘The determination of the issue is relevant only to the difference in treatment as applied to the Claimant at the relevant time’.152 Given that most employers have only needed to justify an EJRA since the passing of the 2011 Regulations,153 Seldon provides limited meaningful guidance on the proportionality issue. In argument before the Supreme Court, members of the Court themselves raised concerns that the proportionality test was very fact sensitive, requiring an ET ruling before individuals could ‘know where they stand’. While some general guidance on how the test might operate was obviously desirable, the Court noted that it was unsure how to give general guidance and, indeed, failed to provide any guidance in its final judgment. By narrowing the issues to be considered in this way, the Court missed an opportunity to provide clarity to employers and workers, making future challenges to retirement ages practically inevitable. Only seven other reported UK cases have considered whether a retirement age is proportionate.154 These cases may be classed into two groups: those relating to judicial and quasi-judicial roles, where claimants have included lay members of Tribunals,155 recorders,156 judges,157 and a parking adjudicator;158 and

152 

Seldon (n 145 above) [64], [92]. Though retirement ages under 65 also needed to be justified prior to the 2011 Regulations. 154  As at 8 January 2016. While Bonett v Prospects Services Ltd [2013] EqLR 432 considered whether an employer’s retirement age was objectively justified, the Tribunal in that case was not required to consider whether the retirement age was proportionate, as the employer had merely asserted that the retirement age was justified, rather than actually demonstrating its need for a retirement age. A slightly different issue was raised in West Midlands Police v Harrod [2015] IRLR 790, which involved reg A19 of the Police Pensions Regulations 1987, SI 1987/257. While reg A18 establishes a compulsory retirement age for the police force, reg A19 makes additional provision for compulsory retirement where retention is not ‘in the general interests of efficiency’ and the officer is entitled to a pension of not less than two-thirds of their average pensionable pay (which accrues after 30 years of service). The case was treated as one of indirect discrimination on the basis of age, rather than direct discrimination. 155  Lindsay v Department for Employment and Learning [2013] NIIT 356/12 (19 December 2013), [2014] EqLR 180; John v Ministry of Justice [2014] UKET 3300891/2012 and 1700191/2013 (17 December 2014). 156  Hampton v Lord Chancellor [2008] IRLR 258. 157  White v Ministry of Justice [2014] UKET 2201298/2013 (25 November 2014). 158  Engel v Transport and Environment Committee of London Councils [2013] UKET 2200472/2012 (26 April 2013). 153 

Legal Regulation of Age Discrimination

 63

those relating to sports referees, where the claimants have been football match officials159 and cricket umpires.160 In these cases, the retirement ages in question have ranged from 48 in Martin, to 65 in Hampton and Willey, to 70 in Engel, Lindsay, White and John. The claimant’s challenges were successful in Hampton, Martin and Engel. Intergenerational fairness was accepted as a legitimate aim in all seven cases. However, the retirement age was not shown to be a proportionate means of achieving this aim in Hampton, as there was no evidence that recorders would remain in their posts until the age of 70,161 steps could have been taken to remove recorders who were failing to satisfy the minimum sitting requirement,162 and other means were available to facilitate workforce planning.163 Similarly, in Martin it was held that the retirement age was not proportionate, as there was no evidence justifying 48 as a reasonable age for retirement, as opposed to any other age;164 Professional Match Game Officials Ltd (PMGOL) had failed to explain adequately its ‘thought processes’ for adopting and retaining the retirement policy;165 there was evidence of an alternative, merit-based ‘non-discriminatory system to encourage churn’ used by the Football Association;166 and PMGOL’s equivalent in the Netherlands had removed its upper age limit for referees in 2000.167 In Engel, the retirement age was not regarded as proportionate as there was no evidence that other, less discriminatory methods of achieving the aims would be ‘ineffective or unavailable or had some other flaw’;168 and, while there was no evidence of the age profile of the parking adjudicators169 or why the age of 70 was appropriate,170 there was evidence that most adjudicators left their role for reasons other than retirement.171 It appears even more difficult to justify a retirement age on the basis of dignity. On the facts in Willey, dignity was regarded as a legitimate aim, given the difficulty of assessing umpires’ performance, the collaborative approach adopted to the management of umpires,172 and the lack of uniform performance assessments.173 However, there was no evidence of age-related decline among umpires at the age of 65, meaning a higher retirement age could still uphold their dignity.174 159 

Martin v Professional Match Game Officials Ltd [2010] UKET 2802438/2009 (13 April 2010). Willey v England and Wales Cricket Board Ltd [2015] UKET 2201406/2014 (10 March 2015). 161  Hampton (n 156 above) [51]. 162  ibid [53]. 163  ibid [54], [56]–[57]. 164  Martin (n 159 above) [7.44]. 165  ibid [7.34]. 166  ibid [7.35]. 167  ibid [7.41]. 168  Engel (n 158 above) [47]. 169  ibid [55]. 170  ibid [69]. 171  ibid [53]. 172  Willey (n 160 above) [8.15], [8.16]. 173  ibid [44]. 174  ibid [45]–[48]. 160 

64 

UK Age Discrimination Laws: A Critique

It was therefore not appropriate or reasonably necessary to adopt a retirement age of 65.175 Proportionality arguments are more likely to succeed where performance ­management is difficult or impossible. In Lindsay, capability assessments for older panel members were not seen as a viable alternative to a retirement age: according to the ET, ‘there is no proportionate and suitable means of assessing, on a regular basis, the continuing fitness of panel members’ given the time and cost of capability processes, the possibility of litigation, the impact on panel members’ dignity, and risks to panel members’ independence.176 In White, performance management for judges was seen as ‘not only distasteful and undignified but potentially damaging to the rule of law’.177 One clear conclusion emerges from this case law: if a less discriminatory alternative to a retirement age is shown to be practicable, whether in the same organisation (as in Hampton), a similar organisation (like in Martin), or even hypothetically,178 it is highly unlikely that the retirement age will be proportionate. Conversely, if available alternatives would be ineffectual (as in Lindsay) or create ‘chaos’ (as in Willey), this will tend towards a finding of proportionality. Beyond this, the cases leave many issues unresolved. It is entirely unclear what evidence will be required to prove a retirement age is proportionate and to discharge the respondent’s burden of proof. In Seldon the EAT rejected the idea that ‘a tribunal must always have concrete evidence, neatly weighed, to support each assertion made by the employer’ linking declining performance and age in support of a dignity argument.179 Instead, the EAT could apply its knowledge of what was ‘plain’ and ‘self-evident’ common sense in relation to ‘how humans behave’ and ‘human nature’, while being ‘astute to differentiate between the exercise of their knowledge of and stereotyped assumptions about behaviour’.180 The EAT subsequently also endorsed the ET ‘implicitly taking into account its own view of the labour market and the world of employment’.181 Therefore, minimal evidence

175  Though the retirement age was upheld as a proportionate means of achieving intergenerational fairness. The strict scrutiny of the evidence in Willey may be compared with the less stringent approach in White: in that case, the ET held that the need to ensure that judges retired while ‘still at the top of their game’, particularly in the context of a non-fanciful ‘worry’ that judges might start ‘going downhill after 70’, justified a retirement age of 70: White (n 157 above) [47]. This decision relies heavily on the decline theory of ageing and problematic dignity arguments. Minimal evidence was produced to support the government’s (and ET’s) ‘mere generalisations’ about age and capacity: see further the critique in John (n 155 above). 176  Lindsay (n 155 above) 16. 177  White (n 157 above) [45]. 178 See Engel (n 158 above) [47]. 179  Seldon (n 140 above) [73]. See also Seldon (n 81 above), where the EAT stated: ‘[it] has often and rightly been said that a Tribunal does not have to dot every “i” nor cross every “t”’ using evidence: [32]. 180  Seldon (n 140 above) [73]. This is not unique to cases of age discrimination: the use of personal experience or ‘common knowledge’ can also be seen in other discrimination cases: see, eg, London Underground Ltd v Edwards (No 2) [1997] IRLR 157 (EAT), [23]; [1998] IRLR 364 (CA), [25] (Potter LJ). 181  Seldon (n 81 above) [32].

Legal Regulation of Age Discrimination

 65

is required to prove proportionality: instead, respondents can rely on assertions of what is ‘self-evident’ and consistent with ‘common sense’. This undermines the use of social science evidence to ground judicial decision-making, perpetuating the disconnect between judicial conceptions of ‘self-evident’ behaviour and those in non-legal spheres.182 In contrast, the ET in Engel pointedly rejected the argument that it could ­‘conclude that [70] was an appropriate age for retirement based on [its] own experience’.183 Assuming that 70 was a ‘reasonable retirement age’ without evidence ‘might be falling into the trap of stereotyping and/or discrimination’.184 This is reassuringly at odds with the EAT’s judgment in Seldon. However, the ET in Engel also expressed doubts about ‘the amount of evidence which a Tribunal should reasonably expect’ to discharge the respondent’s burden of proof, and noted that an appeal would allow more senior courts to give guidance on this point.185 The level of evidence (if any) required to discharge the respondent’s ­burden of proof remains an unclear and contested area. This caused serious concerns in John, where the ET’s decision in White was criticised for relying on opinion and ‘mere generalisations’ in its findings regarding judicial capacity.186 The ET defended its previous decision in John on the basis that, where there is an absence of empirical evidence, opinion evidence could still be ‘highly persuasive’, particularly where it accorded with (judicial) ‘common sense’.187 The proportionality cases also diverge in their degree of scrutiny of retirement ages. The ET in Seldon188 conducted a far less rigorous inspection of the partnership’s retirement age than that in Hampton, Engel, Martin or Willey.189 If the approach of the ETs in these cases were applied to Seldon, it would be ­necessary to consider whether partners at Clarkson Wright & Jakes were likely to work beyond the age of 65 and whether partnership positions could be ‘freed up’ by some means other than retirement. There is no concrete evidence of

182 See, eg, J Monahan and L Walker, Social Science in Law: Cases and Materials, University Casebook Series, 3rd edn (Mineola NY, Foundation Press, 1994); DK Flaks, ‘Gay and Lesbian Families: ­Judicial Assumptions, Scientific Realities’ (1994) 3 William and Mary Bill of Rights Journal 345. However, requiring social science evidence to prove ‘how humans behave’ would add an additional (costly) hurdle to the ET’s processes, which are intended to provide a quick, efficient and cost effective mechanism to resolve disputes: see BIS, ‘Resolving Workplace Disputes: A Consultation’ (January 2011) 2, 5. It would also impose a significant burden on small business. 183  Engel (n 158 above) [43]. 184  ibid [69]. 185  ibid [71]. 186  John (n 155 above) [19]–[20]. 187  ibid [31]. 188  Seldon (n 145 above). 189  That said, the approach in Lindsay (n 155 above) appears similar to that in Seldon: the ET noted (but did not scrutinise) general concerns regarding older workers’ declining health and less recent work experience: 17. No evidence of the actual situation among panel members was required. The approach in John and White was also less rigorous. This may be due to the fact that the cases relate to statutory provisions, which are subjected to a lower level of scrutiny: see White (n 157 above) [42]. See similarly West Midlands Police v Harrod (n 154 above).

66 

UK Age Discrimination Laws: A Critique

the ­situation at Clarkson Wright & Jakes regarding older partners. However, ­statistics relating to UK law firm partners generally suggest retirement rules are difficult to justify. The average age of partners in UK law firms in 2010 was 46,190 indicating that most are aged well below 65 and unlikely to work beyond that age.191 In 2012, only 10.8 per cent of the UK partner population were aged over 60, and 3.4 per cent were aged over 65.192 It is unlikely that partners are remaining in work into old age and limiting associates’ career progression: evidence of the specific situation at Clarkson Wright & Jakes would be necessary to rebut this conclusion. Further, as noted by the ET in Seldon, the number of positions in a partnership is not fixed, meaning it is not necessary to retire a partner for an associate to be promoted:193 in 2012–13, 2,795 new partnership posts were created across UK law firms.194 In addition, while some UK partnerships still adopt a retirement age,195 many firms have already removed retirement ages from their partnership deeds.196 Therefore, Clarkson Wright & Jakes could have been required to show why the non-discriminatory alternative adopted by other firms was not appropriate to its circumstances. Had a more rigorous approach been adopted in Seldon, it is unlikely the retirement age would have been found to be proportionate. It is arguable that the key difference between the cases, and the prime reason for the different approaches, relates to consent: in Hampton, Engel and Martin, the individual was an employee or in a position comparable to an employee, whereas Seldon related to a partnership deed.197 It is arguable that a partner in a law firm has greater capacity and scope to consent to and negotiate a retirement age than an employee. The fact that Mr Seldon had consented to the partnership deed was a significant factor for the ET: ‘The consent of all the partners and the signing of

190 K Wright, ‘Revealed: Top UK Firms Make Poor Showing in Female Partner Stakes’ (The Lawyer, 26 September 2010): www.thelawyer.com/revealed-top-uk-firms-make-poor-showing-in-femalepartner-stakes. 191  See also Engel (n 158 above) [53]. 192  Solicitors Regulation Authority, ‘Diversity Monitoring Statistics 2012’ (no date): www.sra.org. uk/sra/equality-diversity/diversity-monitoring/diversity-monitoring-2012.page. 193  Seldon (n 145 above) [39]. 194 Wilkins Kennedy, ‘Number of Partners at UK Law Firms up as “Firing and Retiring” Falls’ (Wilkins Kennedy, 16 January 2014): www.wilkinskennedy.com/news-press/wk-press-releases/ number-of-partners-at-uk-law-firms-up-as-firing-and-retiring-falls/. 195  See A Reynolds, ‘Employment Tribunal Rules in Favour of Law Firms in Key Age Discrimination Case’, Legal Week, 31 May 2013. 196 F Heine, ‘A&O Abolishes Mandatory Retirement Age as Top Law Firms Review Policies’, Legal Week, 10 February 2012. 197  Though compare Lindsay, where the retirement age was subjected to a lower level of scrutiny and the individual was in a similar position to the claimant in Engel. Hampton and Engel also involved government respondents perhaps raising the level of scrutiny of the retirement age: see, eg, Walsh v Tewkesbury Borough Council [2011] EqLR 1107; cf White (n 157 above) [42], which related to statutory provisions. However, this distinction was not raised in the decisions themselves. Again, compare Lindsay, which also involved a government respondent.

Legal Regulation of Age Discrimination

 67

the partnership deed in 2005 are not determinative but amount to strong factors to be taken into account in the determination of proportionality’.198 As noted above, the quality of ‘consent’ in relation to mandatory retirement provisions is questionable. In Seldon ‘there was no discussion’ regarding the retention of a retirement age in the partnership deed199 and the clause had remained in the deed for ‘as long as could be remembered’.200 The ET interpreted this as meaning that the retirement age was ‘not controversial’.201 It could equally mean that the partners did not turn their minds to the provision, undermining the quality of Mr Seldon’s consent. Indeed, the EAT noted: ‘it was not clear when the compulsory retirement age had first been agreed; it always appears to have been part of the partnership deed’.202 It is therefore difficult to argue that Mr Seldon meaningfully consented to the retirement age, though he did not object to it. Still, the original ET regarded the deed as a negotiated instrument: the partners had agreed a compulsory retirement age between themselves … Here were parties of equal bargaining power agreeing between themselves what appeared to be a sensible and acceptable set of provisions. It is not akin to the position of an employee subject to a contract of employment who is perforce in an unequal relationship with his employer.203

This finding was endorsed by the EAT: We think it is a legitimate consideration that a rule of this kind was agreed by parties with equal bargaining power … [W]e think that the principle of justification which, after all, seeks to look at all the circumstances, should in principle take into account the fact that the partners must have perceived the rule to be in their collective interests.204

It is possible that Mr Seldon’s superior bargaining position (as both a partner and former managing partner of the firm)205 justifies the difference of approach between Seldon and Hampton, Engel and Martin. This is supported by Willey, where the collaborative development of policies relating to the umpires’ employment,206 and the involvement of the claimants in discussions of the retirement age,207 was strongly influential in the ET’s reasoning. While the claimants in Willey ‘did not have the same leverage as Mr Seldon’ in negotiating their terms and conditions, they were ‘far from being pawns’.208

198 

Seldon (n 145 above) [82]. ibid [83]. 200  ibid [79]. 201  ibid [83]. 202  Seldon (n 140 above) [16]. 203  ibid [27]. 204  ibid [52]–[53]. 205  Seldon (n 81 above) [1]. 206  Willey (n 160 above) [12]. 207  ibid [16]–[18]. 208  ibid [13]. 199 

68 

UK Age Discrimination Laws: A Critique

II.  Limitations of Existing Laws While the discussion above has proffered consent as one basis on which we can potentially reconcile and offer a coherent account of the EJRA case law, this means of analysis can only take us so far. The idea of ‘consent’ in relation to retirement ages is largely illusory, and does not provide a solid foundation for legal development in this area. Further, while the cases can potentially be reconciled on this basis, the apparent inconsistency and enduring uncertainty in the case law is likely to cause significant problems for employers. Anecdotal evidence shows that some employers are abandoning fixed retirement ages, instead dealing with employees on a case-by-case basis.209 A 2013 survey of 307 employers found that only 2.9 per cent had retained a policy of mandatory retirement (compared with 69 per cent in a similar survey in 2011) and 72 per cent attributed their change of policy to the repeal of the DRA.210 Many did not feel positively about the change: 34 per cent of respondents felt the abolition of the DRA had negatively or very negatively impacted their organisation (compared with 14 per cent who felt it had had a positive impact) and 48 per cent thought the DRA should be reinstated (compared with 43 per cent who thought the law should remain as it is).211 A separate 2013 survey found that 19 per cent of respondents had retained a retirement age for their workforce, but most failed to provide a justification for their retirement age when asked.212 While these surveys indicate that employers are choosing to abandon retirement rules, significant sampling issues mean they cannot be regarded as indicative of UK employers as a whole.213 Therefore, it still remains unclear how far these trends extend and, indeed, whether they are trends at all. Regardless, legal uncertainty appears to be having a deleterious effect on a number of employers.

209  DWF, ‘De-Regulation of Retirement—One Year on’ (9 May 2012): web-archive-uk.com/uk/d/ dwf.co.uk/2012-07-19_157064_88/DWF_gt_De_regulation_of_Retirement_One_Year_On/. 210  Eversheds, ‘Eversheds UK HR E-Briefing: How Are Employers Managing without the Default Retirement Age?’ (April 2013) 2. 211  ibid 6–7. 212  R Thomas et al, ‘Recovery in Sight? The State of HR’ (2013) 9. This is consistent with a 2012 survey of 412 employers, which found that 18% of respondents had retained their retirement age. Of the 81.6% of employers who had chosen to remove their retirement age, 82.6% cited the change of the law in April 2011 as the reason for the change: Irwin Mitchell, ‘Retirement without the Default Retirement Age’ (2012) 7–8. 213  The first survey was sent to recipients of a law firm’s e-briefings newsletter (who are mostly likely to be larger employers): Eversheds (n 210 above) 1. The second was advertised online and emailed to a purchased list of senior HR manager contacts, and clients and contacts of Speechly Bircham and King’s College London: Email from M Clinton, ‘5th Annual State of HR Survey’ (21 May 2013). Neither survey had a randomly selected nor specifically stratified sample, and response rates are not available or applicable.

Limitations of Existing Laws

 69

In addition to legal uncertainty in relation to EJRAs, we can distil four key limitations of the current legal framework that deserve further discussion. First, the framing of age discrimination laws in the EU and UK offers limited protection to older workers in practice. At present, exceptions in the legislation allow governments (and employers) ‘to tolerate most forms of age discrimination indefinitely’.214 By allowing governments and employers to justify both direct and indirect age discrimination, the laws provide significant scope to undermine the principle of equal treatment on the grounds of age, particularly in relation to retirement ages. Further, by providing less rigorous protection from age discrimination than other forms of discrimination, the laws effectively legitimise discrimination on the basis of age.215 This is seen most clearly in relation to EJRAs: the retention of mandatory retirement affirms and legitimises age discrimination, and undermines, devalues,216 and stigmatises older workers on the basis of age.217 Thus, allowing employers to adopt an EJRA directly undercuts the intrinsic objectives of age discrimination legislation, and is incompatible with the broader governmental push to extend working lives and promote individual choice.218 If law is used for expressive purposes, to declare and affirm social values,219 then existing age discrimination legislation has likely failed miserably to promote new values in relation to age equality. Secondly, leaving the detailed determination of what is ‘justified’ discrimination to the courts has created an uncertain and contradictory legal landscape, as demonstrated most keenly by the EJRA case law.220 This is likely to have significant implications at the organisational level, as neither employers nor employees will be clear as to their legal rights and responsibilities. While this situation persists, age discrimination laws are unlikely to create opportunities for older workers to remain in or re-join the workforce. Indeed, as the discussion of retirement ages has shown, existing laws may seriously undermine the position of older workers. Thirdly, laws regarding positive action are fundamentally unclear, and unlikely to be utilised by employers. The current framing of the provisions is overly

214 House of Commons Education and Employment Committee, ‘Age Discrimination in Employment’ Seventh Report of Session 2000–01, HC 259 (27 March 2001) Appendix 17. 215 M Sargeant, ‘The Employment Equality (Age) Regulations 2006: A  Legitimisation of   Age  Discrimination in  Employment’ (2006) 35 Industrial Law Journal 209, 219. 216 A Walker, ‘Combating Age Barriers in Employment: European Research Report’ (1997) 23; A Walker, ‘Public Policy and the Construction of Old Age in Europe’ (2000) 40 The Gerontologist 304, 305. 217  See LM Friedman, ‘Age Discrimination Law: Some Remarks on the American Experience’ in S Fredman and S Spencer (eds), Age as an Equality Issue: Legal and Policy Perspectives (Oxford, Hart Publishing, 2003) 192; Sargeant, ‘United Kingdom’ (n 55 above) 224. 218 OECD, Live Longer, Work Longer, Ageing and Employment Policies (OECD, 2006) 12; see also European Commission (n 13 above) 82–83. 219  JR Gusfield, Symbolic Crusade: Status Politics and the American Temperance Movement, 2nd edn (Urbana-Champaign IL, University of Illinois Press, 1986) 4–5. 220  See further C O’Cinneide, ‘The Uncertain Foundations of Contemporary Anti-Discrimination Law’ (2011) 11 International Journal of Discrimination and the Law 7, 14–15.

70 

UK Age Discrimination Laws: A Critique

complex and unlikely to encourage proactive measures by risk-averse ­ ­employers.221 This may reflect the enduring ambivalence in the UK towards positive action, particularly given that positive action has traditionally not been a feature of UK equality law.222 Instead, equality law in the UK has generally been symmetrical, protecting disadvantaged and advantaged groups equally.223 Positive action is sometimes seen as ‘unfair’ and counter-productive, as it prevents disadvantaged groups from achieving success based on their own efforts and ‘merits’. That said, positive action is only unfair if advantaged and disadvantaged groups commence on an equal footing,224 which is unlikely to be the case in practice. Thus, to achieve substantive equality of opportunity, some tipping-point provision is likely to be both desirable and necessary. Clarifying and simplifying the existing provisions may promote a more proactive approach to demographic ageing on the part of employers. Fourthly, the EqA’s reliance on individual enforcement likely undermines its effectiveness in practice. Under the EqA, claims of age discrimination may be enforced: by individuals in ETs, with potential remedies including a declaration of rights, award of compensation or a recommendation;225 by individuals in the county court, to remove or modify discriminatory contractual provisions;226 or by the Equality and Human Rights Commission (EHRC), which is empowered to take enforcement action spanning investigations, unlawful act notices, compliance notices relating to the PSED, action plans, agreements, applications to restrain, conciliation, and legal proceedings.227 In practice, enforcement of the EqA is primarily dependent on individual litigation in ETs. Unlike the situation in other Member States, collective enforcement of equality matters is ‘very weak’ in the UK.228 While the EHRC is empowered to enforce age discrimination legislation, these powers are exercised infrequently: between October 2007 and March 2010, the EHRC considered enforcement action in only 23 age discrimination matters.229 While agency enforcement, such as that undertaken by the EHRC, is an important means of addressing structural discrimination and demonstrating the state’s commitment to addressing discrimination,230 significant budget and

221  See further L Barmes, ‘Equality Law and Experimentation: The Positive Action Challenge’ (2009) 68 Cambridge Law Journal 623. 222 ACL Davies, Perspectives on Labour Law, Law in Context, 2nd edn (Cambridge, Cambridge ­University Press, 2009) 134. 223 ibid. 224  ibid 127. 225 EqA s 124(2). The Deregulation Act 2015, s 2 removed the power of ETs to make wider recommendations. 226  EqA ss 142, 143. 227  Equality Act 2006, ss 16, 20–22, 32. 228 L Dickens, ‘The Road is Long: Thirty Years of Equality Legislation in Britain’ (2007) 45 British Journal of Industrial Relations 463, 481. 229 EHRC, ‘Legal Enforcement—Update from the Equality and Human Rights Commission’ (April 2010) 3. 230  Dickens, ‘The Road is Long’ (n 228 above) 475.

Limitations of Existing Laws

 71

staffing cuts have limited the efficacy of the EHRC’s operations.231 Instead, the ‘weight of enforcement’ has fallen on individual complainants and enforcement in the courts.232 Individual enforcement of age discrimination claims has a number of limitations, including the inherent reliance upon individuals’ awareness of their rights and their willingness or capacity to enforce them.233 Previous studies in the UK have found that a large minority of older workers have limited knowledge of their employment rights, and only a few have actually exercised their rights.234 Further, as Malmberg notes, relying on individual claims is unlikely to facilitate effective macro- or societal-level enforcement of legislation.235 Therefore, the UK approach to enforcement may result in discrimination not being addressed,236 and may severely impair the enforcement of age discrimination legislation in the UK.237 These challenges are compounded by government austerity measures in the UK, which have introduced a number of barriers to individual enforcement of employment rights. First, legal aid funding is now generally excluded for employment matters.238 While discrimination claims can still obtain legal aid funding, many discrimination claims are initially raised as employment matters, meaning the reforms are likely to have an indirect ‘chilling’ effect on individual age discrimination claims.239 Secondly, since 29 July 2013, ETs have charged fees to hear claims or appeals. While there is a fee remission system in place, and ETs may order employers to reimburse a successful claimant’s fees, it appears that the fee regime is deterring applicants in practice: from October to December 2013, ETs received 79 per cent fewer claims compared with the same period in 2012, and 75 per cent

231 ‘Budget Cuts May Leave to All White Management at EHRC’ HRreview, 10 September 2012: www.hrreview.co.uk/hr-news/diversity-equality/budget-cuts-may-leave-to-all-white-managementat-ehrc/38788. 232  Dickens, ‘The Road is Long’ (n 228 above) 475. 233  S Fredman, Discrimination Law, Clarendon Law Series (Oxford, Oxford University Press, 2002) 165; Dickens, ‘The Road is Long’ (n 228 above) 479. 234  D Smeaton et al, ‘Older Workers: Employment Preferences, Barriers and Solutions’, Research Report 43 (EHRC, December 2009) 89–94. 235  J Malmberg, ‘Effective Enforcement of EC Labour Law: A Comparative Analysis of Community Law Requirements’ (2004) 10 European Journal of Industrial Relations 219, 223. 236 See European Commission, ‘Joint Report on the Application of Council Directive 2000/43/ EC of 29 June 2000 Implementing the Principle of Equal Treatment between Persons Irrespective of Racial or Ethnic Origin (“Racial Equality Directive”) and of Council Directive 2000/78/EC of 27 November 2000 Establishing a General Framework for Equal Treatment in Employment and ­Occupation (“Employment Equality Directive”)’ 2014(COM) 2 final (17 January 2014) 4–5, 6. 237  See further Fredman, Discrimination Law (n 233 above) 161–74; L Dickens, ‘Introduction—Making Employment Rights Effective: Issues of Enforcement and Compliance’ in L Dickens (ed), Making Employment Rights Effective: Issues of Enforcement and Compliance (Oxford, Hart Publishing, 2012); GS Morris, ‘The Development of Statutory Enforcement Rights in Britain and Enforcement Mechanisms’ in L Dickens (ed), Making Employment Rights Effective: Issues of Enforcement and Compliance (Hart 2012). 238  See Legal Aid, Sentencing and Punishment of Offenders Act 2012, sch 1, pt 1, s 43. 239 EHRC, ‘Response of the Equality and Human Rights Commission to the Consultation on Reform of Legal Aid in England and Wales’ (15 November 2010) 11.

72 

UK Age Discrimination Laws: A Critique

fewer claims than in the previous quarter.240 More specifically, age discrimination claims dropped by 37 per cent between 2013 and 2015.241 These data indicate that fees may reduce the viability of individual enforcement of employment rights. Thirdly, from April 2014, employees must contact the Advisory, Conciliation and Arbitration Service (Acas) before making a claim to an ET,242 though conciliation is not compulsory. This may significantly lower the number of claims reaching ETs, further reducing the already limited amount of case law on age equality issues in the UK, and making it more difficult to predict individual case outcomes prior to a hearing. This, then, may increase the risks and costs of litigation and individual enforcement. These changes and limitations will make it significantly harder for individuals to bring a claim of age discrimination, reducing the microor individual-level effectiveness of the UK legislation.243

III.  Ways Forward A.  Doctrinal Reform If age discrimination laws are to achieve the instrumental and intrinsic objectives posited in Chapter one, legal reform in three areas is a necessary first step. First, we should limit or amend the existing exceptions to the legislation. Limiting exceptions for age discrimination would endorse an individual rights perspective and the intrinsic motivations of equality law, promoting the ending of age ­discrimination as a primary priority, rather than just a means of achieving ­workforce diversity and instrumental ends.244 Legal exceptions could be limited in four (potentially overlapping) ways: first, by introducing a specific and limited list of acceptable exceptions to age discrimination laws;245 secondly, by amending the legislation so that direct age discrimination can only be justified if it falls within another exception, bringing age into line with other protected characteristics; thirdly, by ­reducing the number of specific exceptions on the grounds of age in Schedule 9 of the EqA;246 and/or, fourthly, by adopting a two-stage

240 

Ministry of Justice, ‘Tribunals Statistics Quarterly: October to December 2013’ (13 March 2014) 8. H Anthony and C Crilly, ‘Equality, Human Rights and Access to Civil Law Justice: A Literature Review’, Research Report 99 (EHRC, October 2015) 93. 242  Enterprise and Regulatory Reform Act 2013, s 7; The Employment Tribunals (Early Conciliation: Exemptions and Rules of Procedure) Regulations 2014, SI 2014/254. 243  Malmberg (n 235) 223. For further on these reforms, see Anthony and Crilly (n 241 above). 244  Sargeant, ‘The Employment Equality (Age) Regulations 2006’ (n 215 above) 218; M Sargeant, ‘The Default Retirement Age: Legitimate Aims and Disproportionate Means’ (2010) 39 Industrial Law Journal 244, 247. 245  See, eg, House of Commons Education and Employment Committee (n 214 above) appendix 17; Sargeant, ‘The Employment Equality (Age) Regulations 2006’ (n 215 above) 220. 246  Sargeant, ‘The Employment Equality (Age) Regulations 2006’ (n 215 above) 227. 241 

Ways Forward

 73

­ roportionality test, considering (1) whether the basic retirement rule is prop portionate and (2) whether the process of applying the rule was a fair one.247 These reforms would require governments to consider explicitly what exceptions are acceptable and consistent with governmental objectives. They would also increase legal clarity and reduce reliance on judicial determinations. Finally, these reforms are consistent with Article 6(1) of the Framework Directive, which says that Member States, not employers, may provide that differences of treatment on the grounds of age shall not constitute discrimination in certain circumstances.248 As part of this process of exception rationalisation, I would also argue that the law should be amended to prevent employers from adopting EJRAs. In addition to promoting the intrinsic and instrumental aims of the legislation, prohibiting EJRAs would promote legal certainty and reduce employer confusion. Indeed, even where an EJRA is likely to be legally justified, the practical implementation of the policy can lead to legal challenges. In writing about the experience of adopting an EJRA at the University of Oxford, Emery notes: the EJRA Policy has caused conflict within the University—with concerns raised about why such a Policy is required, whether it can be justified in principle for academics, and complaints about how it is being implemented. There is clearly risk of a legal challenge.249

The implementation of an EJRA may require older workers to ‘over-perform’ compared with junior staff to justify their retention beyond retirement age.250 EJRAs therefore create a series of complex and ambiguous legal questions in their adoption and implementation, and can perpetuate and exacerbate unequal treatment. Thus, removing the option to adopt an EJRA would likely have benefits for employers and employees in the longer-term. Secondly, the positive action provisions in the EqA should be clarified and simplified, to encourage employers and legal advisors to regard positive action as a practicable initiative rather than an unacceptable risk. However, even if positive action provisions were clarified and simplified, it would still be unlikely that employers would make use of the provisions. A simplified legal structure will be effective only if employers have a pre-established commitment to equal ­opportunities.251 Thus, social ambivalence regarding positive action may mean

247  Vickers and Manfredi (n 139 above) 72–73. Vickers and Manfredi suggest that this fourth option alone may be sufficient to achieve the objectives of the EqA. However, with increasing recognition of the arbitrary discrimination and injustice inherent in mandatory retirement, and the extent to which it undermines efforts to increase workforce participation (see European Commission (n 13 above) 82–83), relying on a stronger proportionality test is unlikely to be a adequate solution. Instead, I argue that the ability to justify mandatory retirement should be removed entirely: see below. 248  See further Barnard (n 130 above) 306. 249 M Emery, ‘Academia, Oxford University and Retirement’ (Withy King Blog, 2 May 2014): www.withyking.co.uk/blog/academia-retirement-policy/. 250 ibid. 251  Davies (n 222 above) 134.

74 

UK Age Discrimination Laws: A Critique

law reform has limited impact in practice.252 Rather than relying on positive action to achieve change, reflexive law and positive duties may offer a better ­solution (see below). Thirdly, it is necessary to rethink the EqA’s reliance on individual enforcement as the primary means of addressing discrimination. Alternatives and complementary strategies to individual enforcement include: stronger enforcement by agencies such as the EHRC (though this would require appropriate resourcing of the EHRC, and for age to be seen as a strategic priority by agency officials); collective negotiation and bargaining around equality matters; and enforcement of equality claims by union officials. Given the limited unionisation of the UK workforce, particularly in the private sector, and current trends to reduce the EHRC’s budget and powers, these strategies are likely optimistic in practice.

B.  Systems Theory and Reflexive Law Age discrimination laws in the UK therefore stand at a difficult juncture: while their limitations are clear (and are having significant implications in practice), reform is likely to be difficult in a deregulatory climate. In this challenging context, systems theory and ‘reflexive’ regulation provide one route for critiquing, ­modelling and strengthening the EqA.

(i)  A Systems Theory Critique As noted in Chapter two, systems theory critiques the ability of command and control regulation to achieve change in other systems (such as organisations). Command and control regulation is based on deterrence theory, and emphasises precise and narrowly drawn rules, penalties for non-compliance, and a one-sizefits-all approach.253 In a command and control approach, government commands organisations to meet certain standards (directly via legislation or indirectly through delegated authority), and controls their behaviour using the threat of negative sanctions.254 This form of regulation can be seen in the EqA at least to some extent. The core provisions of the EqA are ‘negative duties prohibiting discrimination’.255 Thus, there are fixed rules and standards regarding what organisations can and cannot do, though the definition of ‘discrimination’ is not particularly precise or

252 

ibid 135. J Job et al, ‘Culture Change in Three Taxation Administrations: From Command-and-Control to Responsive Regulation’ (2007) 29 Law & Policy 84, 87. 254  D Sinclair, ‘Self-Regulation versus Command and Control? Beyond False Dichotomies’ (1997) 19 Law & Policy 529, 534. 255  B Hepple, ‘Enforcing Equality Law: Two Steps Forward and Two Steps Backwards for Reflexive Regulation’ (2011) 40 Industrial Law Journal 315, 318. 253 

Ways Forward

 75

narrowly drawn. Further, organisational behaviour is controlled via individual ­complaints in ETs and the scrutiny of the EHRC, and discriminatory behaviour can be ­penalised by an award of compensation and other remedies. Thus, as it stands, it is certainly arguable that age discrimination laws in the UK resemble command and control regulation and risk engaging the regulatory trilemma. As discrimination in the workplace is attributed increasingly to structural, relational and situational conditions, specific and fixed rules are likely to have even less success at effecting change.256 Command and control regulation is instead likely to ‘induce gestures of compliance’ in relation to equality.257 Systems theory provides a telling critique of the limits of ‘command and control’ equality law and may explain the limited effectiveness of the EqA in achieving concrete organisational change (see Chapters four and five).

(ii)  A Reflexive Law Model That said, it is also arguable that the EqA embodies partly a ‘reflexive law’ approach to addressing discrimination, attempting to tailor regulation to particular contexts and integrate formal legal devices with self-regulation. However, if reflexive law is used as a model of regulation, it appears that equality law in the UK cannot be mapped effectively against the enforcement pyramid or an experimentalist architecture. While there are some aspects of equality law that could be reflexive, these are generally unsupported by the institutional conditions necessary for their effective operation. First, modelling the EqA on the enforcement pyramid, the apex of the pyramid for equality law in the UK consists of civil penalties. In other regulatory contexts, and in other jurisdictions, the apex of the pyramid includes criminal penalties.258 According to Ayres and Braithwaite, ‘the greater the heights of tough enforcement to which the agency can escalate … the more effective the agency will be at securing compliance’.259 Therefore, compliance appears more likely where criminal sanctions are available. Given criminal penalties are not available in the UK, and the limitations of individual enforcement, reflexive law may be inappropriate as a model of UK equality law. Secondly, and relatedly, the focus of UK governments in relation to equality law has primarily been on ‘restorative dialogue’ and the use of information and ­persuasion.260 Table 3 provides examples of non-regulatory action undertaken by the UK government in relation to age equality in recent years. While ­dialogue

256  S Sturm, ‘Second Generation Employment Discrimination: A Structural Approach’ (2001) 101 Columbia Law Review 458, 460. 257  ibid 461. 258  See the case of Finland, discussed in Chapter 6 of this volume. 259 I Ayres and J Braithwaite, Responsive Regulation: Transcending the Deregulation Debate, Oxford Socio-Legal Studies (Oxford, Oxford University Press 1992) 6. 260  J Macnicol, ‘The Age Discrimination Debate in Britain: From the 1930s to the Present’ (2005) 4 Social Policy and Society 295, 298.

76 

UK Age Discrimination Laws: A Critique

and self-regulation are the default form of intervention in the enforcement pyramid,261 this must be paired with increasingly interventionist regulation where dialogue and self-regulation fail. The UK government has appeared particularly unwilling to intervene in issues of age discrimination. Indeed, as Sargeant notes, ‘it is … impossible to know whether the UK Government would have progressed to [legislative intervention] without the need to transpose the Framework Directive’.262 While the EqA has provided the opportunity for increasing levels of intervention, this remains problematic given the issues with enforcement of UK equality law. Without credible intervention beyond dialogue and self-regulation, UK equality law fails to comply with a model of reflexive law premised on the enforcement pyramid, and instead reflects a reliance on ‘soft’ law or ‘non-law’. Table 3:  Examples of Non-Regulatory Action in the UK Type of intervention

Examples

Consultation documents, strategies and good practice guides

Action on Age (1998) Consultation on a Code of Practice (1998) Building a Better Britain for Older People (1998) Code of Practice: Age Diversity in Employment (1999) Winning the Generation Game (2000) Equality and Diversity: Age Matters (2003) Equality and Diversity: Coming of Age (2005) Opportunity Age (2005–09) A Sure Start to Later Life: Ending Inequalities for Older People (2006) Preparing for Our Ageing Society (2008) Building a Society for all Ages (2009) The Equality Strategy—Building a Fairer Britain (2010) Phasing out the Default Retirement Age (2010) Flexible, Effective, Fair: Promoting Economic Growth through a Strong and Efficient Labour Market (2011) Fuller Working Lives: A Framework for Action (2014) UK Advisory Forum on Ageing (2011–current)

Public information campaigns to tackle workplace ageism

Age Positive (2000–current: see Chapter 4) (including employer case studies, conferences, websites, information, and appointment of Age Positive Employer Champions)1 ‘Employers for Carers’ (2009–current) (promotes the business benefits of supporting carers and provides advice and support for employers on carer-friendly policy and practice)2 (continued)

261  J Braithwaite, Restorative Justice and Responsive Regulation (Oxford, Oxford University Press 2002) 30. 262  Sargeant, ‘The Employment Equality (Age) Regulations 2006’ (n 215 above) 214.

Ways Forward

 77

Table 3:  (Continued) Type of intervention

Examples

Information and Online tools and information for managing an older guidance for employers workforce (websites include DWP, Gov.UK, Acas, and the EHRC) ‘Be Ready’ (2004–06) (promotion of information and guidance on good practice and preparing for the coming age legislation) Age Positive guidance and toolkits (see Chapter 4) Education and training Train to Gain (2006–10) (supported workers to improve skills for older workers and employment prospects)3 New Deal 50 Plus (targeted back-to-work help for people over 50: replaced by the Flexible New Deal from October 2009) Notes: 1 See DWP, ‘Building a Society for All Ages’ (2009) 31; HM Government, ‘Opportunity Age: Meeting the Challenges of Ageing in the 21st Century’ (March 2005) 20–21. 2 See DWP, ‘Report on Implementation of Opportunity Age Commitments’ (August 2009). 3 DWP, ‘Building a Society for All Ages’ (2009) 32.

Thirdly, and more generally, while reflexive regulation may increase equality law’s ‘regulatory potential’,263 it will be successful only in certain circumstances. To be effective, reflexive regulation must provide legal incentives for internal organisational action and scrutiny,264 mechanisms for effective deliberation and participatory decision-making (such as collective bargaining),265 and an external body to back-up and enforce regulations where voluntary methods fail.266 Barnard, Deakin and Hobbs have argued that the UK lacks the institutional structures for securing equality rights through reflexive law, particularly given the low levels of union representation in UK workplaces.267 McLaughlin also raises doubts about whether local actors have the skills and experience to negotiate effectively about equality matters.268 It is therefore unsurprising that the UK equality framework 263 G Teubner, Law as an Autopoietic System, The European University Institute Press Series, Z Bankowski (ed), A Bankowska and R Adler (trans) (Oxford, Blackwell, 1993) 97. 264  B Hepple, ‘Agency Enforcement of Workplace Equality’ in L Dickens (ed), Making Employment Rights Effective: Issues of Enforcement and Compliance (Oxford, Hart Publishing, 2012) 64–65. 265  C McCrudden, ‘Equality Legislation and Reflexive Regulation: A Response to the Discrimination Law Review’s Consultative Paper’ (2007) 36 Industrial Law Journal 255, 260; S Deakin et al, ‘Gender Inequality and Reflexive Law: The Potential for Different Regulatory Mechanisms’ in L Dickens (ed), Making Employment Rights Effective: Issues of Enforcement and Compliance (Oxford, Hart Publishing, 2012) 120. 266  Hepple, ‘Agency Enforcement of Workplace Equality’ (n 264 above) 55. 267  C Barnard et al, ‘Reflexive Law, Corporate Social Responsibility and the Evolution of Labour Standards: The Case of Working Time’, Working Paper No 294 (ESRC Centre for Business Research, University of Cambridge, December 2004) 23. See also C McLaughlin, ‘Equal Pay, Litigation and ­Reflexive Regulation: The Case of the UK Local Authority Sector’ (2014) 43 Industrial Law Journal 1, 6–7, 15. 268  McLaughlin (n 267 above) 27.

78 

UK Age Discrimination Laws: A Critique

has failed to model effective reflexive regulation in practice, with few legal incentives for internal scrutiny and action in the private sector, limited information ‘flows’ regarding equality matters,269 a significantly circumscribed PSED,270 no duty to consult interest groups in making equality decisions,271 and limited local expertise and capacity to deliberate effectively on equality matters.272 Therefore, while many commentators see ‘signs of a model of reflexive regulation’ in the EqA, this is seen ‘through a mirror, dimly’,273 and ‘reflexive’ measures may fail to deliver the outcomes intended by Parliament.274 Fourthly, the PSED may be too weak and inadequately framed to operate as a form of reflexive law.275 The PSED has the potential to require public sector organisations to develop their own solutions to equality issues, therefore avoiding the regulatory trilemma.276 However, the ‘due regard’ standard may be ‘too flimsy’ to encourage organisations to operate in this way.277 The duty is arguably insufficiently prescriptive and too open-ended in its pursuit of ‘equality’ to allow a determination of when it has been complied with or breached in practice, impairing its efficacy.278 Further, the use of judicial review and attempts to develop settled principles to monitor and enforce the PSED may have exacerbated the regulatory trilemma, either by encouraging ‘mere procedural compliance’ and ‘box ticking’ by public sector organisations,279 or by forcing public bodies to seek judicial determinations of whether they have complied with the PSED.280 Thus, the PSED does not represent an effective form of reflexive law.281 269  See S Deakin and R Hobbs, ‘False Dawn for CSR? Shifts in Regulatory Policy and the Response of the Corporate and Financial Sectors in Britain’ (2007) 15 Corporate Governance: An International Review 68, 70. 270 S Fredman and S Spencer, ‘Delivering Equality: Towards an Outcome-Focused Positive Duty—Submission to the Cabinet Office Equality Review and to the Discrimination Law Review’ (June 2006) 9; Dickens, ‘The Road is Long’ (n 228 above) 473; S Fredman, ‘The Public Sector Equality Duty’ (2011) 40 Industrial Law Journal 405; S Fredman, ‘Breaking the Mold: Equality as a Proactive Duty’ (2012) 60 American Journal of Comparative Law 265, 266. 271  See further B Hepple et al, Equality: A New Framework—Report of the Independent Review of the Enforcement of UK Anti-Discrimination Legislation (Oxford, Hart Publishing, 2000) 57–58; Hepple, ‘Enforcing Equality Law’ (n 255 above) 330–32; Hepple, ‘Agency Enforcement of Workplace Equality’ (n 264 above) 64–65. 272  McLaughlin (n 267 above) 24–25. 273  Hepple, ‘Enforcing Equality Law’ (n 255 above) 334. 274  McLaughlin (n 267 above) 28. 275  Fredman, ‘The Public Sector Equality Duty’ (n 270 above) 419. 276 ibid. 277 ibid. 278  Fredman, ‘Breaking the Mold’ (n 270 above) 271, 274; see also H Conley, ‘Using Equality to Challenge Austerity: New Actors, Old Problems’ (2012) 26 Work, Employment & Society 349, 357–58; S Fredman, ‘Addressing Disparate Impact: Indirect Discrimination and the Public Sector Equality Duty’ (2014) 43 Industrial Law Journal 349, 354–55. 279  Fredman, ‘The Public Sector Equality Duty’ (n 270 above) 420; Fredman, ‘Breaking the Mold’ (n 270 above) 276. 280  Fredman, ‘Breaking the Mold’ (n 270 above) 281. 281  Despite these limitations, the PSED may have had positive impacts on the behaviour of public bodies in practice: see the examples presented to the government review of the PSED, as described in Stephenson (n 103 above) 77.

Ways Forward

 79

If experimentalism is applied to UK equality law as a model, a number of experimentalist features can be discerned. In particular, the EqA provides limited guidance regarding how employers should comply with its terms, leaving significant discretion and scope for experimentation by non-state actors. However, beyond this, it is clear that UK age discrimination law does not adhere to this conception of reflexive law. First, it is difficult to identify framework goals or measurement criteria in the EqA. While employers are prohibited from discriminating against their employees in some circumstances, and this could constitute a broad ‘framework goal’ of equality legislation,282 it is unclear how this goal is to be measured. Indeed, the only measurement that might be utilised is the number of ET claims against a given employer. This is highly unlikely to satisfy the experimentalist notion of ‘measurement criteria’, particularly given that its measurement is dependent on individual enforcement of claims. Secondly, UK employers generally have no obligation to report on their age equality performance. While the PSED requires public sector employers to publish information to demonstrate their compliance with the general equality duty,283 there is no requirement for private employers to do the same. Perhaps as a result, there is a noticeable lack of a ‘benchmarking logic’ between employers in the UK (see Chapter seven) and an absence of discussion among employers generally about best practice in relation to age equality.284 Thus, key elements of an experimentalist framework are not evident in the UK. However, the passage of the Small Business, Enterprise and Employment Act 2015, section 147 may mark a move towards equality reporting in the UK, and the growth of an experimentalist ideology in UK governance. Moving away from the government’s previous preference for ‘a voluntary, business-led initiative’ where ‘companies can choose what [information] they make public, and where’,285 the 2015 Act requires regulations to be made under the EqA section 78 to provide for gender pay gap reporting by employers with 250 or more employees. In time, this could well come to constitute regular reporting on performance against agreed criteria, and could facilitate peer review of equality performance, at least in relation to equal pay. If equality reporting was extended to other grounds, such as age, a shift towards experimentalist structures might be possible in the future.

282 See G de Búrca, ‘Stumbling into Experimentalism: The EU Anti-Discrimination Regime’ in CF Sabel and J Zeitlin (eds), Experimentalist Governance in the European Union: Towards a New Architecture (Oxford, Oxford University Press, 2012) 216–17. 283  Equality Act 2010 (Specific Duties) Regulations 2011, SI 2011/2260, reg 2; see further Chapter 7. 284  J Cohen and C Sabel, ‘Directly-Deliberative Polyarchy’ (1997) 3 European Law Journal 313, 326. This is discussed further in Chapters 4, 7 and 8 of this volume. 285  GOV.UK, ‘Think, Act, Report—Creating a Fairer and More Equal Society’ (18 August 2014): www.gov.uk/government/policies/creating-a-fairer-and-more-equal-society/supporting-pages/ think-act-report.

80 

UK Age Discrimination Laws: A Critique

Thirdly, there is currently no body ‘whose explicit purpose is to expose and clarify difference’ between employers.286 Directly-deliberate polyarchy must occur within a frame of institutional arrangements, as these are ‘essential to the legitimate and successful operation of the new problem-solving arrangements’.287 Experimentalism combines decentralised operative control with centralised coordination of evaluation of results.288 While a variety of institutional arrangements can fulfil these functions,289 it is questionable whether the UK has sufficient centralised institutional arrangements in the field of equality to satisfy an experimentalist model. That said, the EHRC could perform this function with appropriate resourcing and powers. Current trends towards reducing the EHRC’s role and resourcing run directly counter to an experimentalist architecture. Thus, equality law in the UK is difficult to model as a form of reflexive law, whether we use the enforcement pyramid or an experimentalist architecture as a starting point. However, a modelling process using the enforcement pyramid draws attention to: the range of remedies and techniques that may be used to secure compliance with UK equality law, and the ability of government institutions to make use of these techniques; the role of unions in UK equality law, and the position of collective action in UK employment law more generally; and the role of dialogue and consultation in achieving equality. Similarly, a modelling process based on an experimentalist architecture encourages a focus on: setting clear and evolving goals and performance measures for employers; regular reporting and peer scrutiny of employer actions and outcomes; the pooling of information between employers; the use of incentives and disincentives to encourage participation; centralised coordination of the review of employer action (for example, by the EHRC); and promotion of employer benchmarking.

(iii)  Reforms to Strengthen Reflexive Regulation in the Equality Act 2010 Given these findings, three key reforms are required if the EqA is to be modelled as a form of reflexive regulation: the imposition of positive duties on private sector employers; the strengthening of the existing PSED; and the introduction of consultative duties. First, to achieve effective reflexive regulation in the UK, and strengthen the operation of the EqA, Hepple, Coussey and Choudhury recommend the imposition of positive duties on private sector employers to achieve employment equality or fair participation, in addition to a PSED.290 Imposing

286  CF Sabel and J Zeitlin, ‘Learning from Difference: The New Architecture of Experimentalist Governance in the EU’ in CF Sabel and J Zeitlin (eds), Experimentalist Governance in the European Union: Towards a New Architecture (Oxford, Oxford University Press, 2012) 4. 287  Cohen and Sabel (n 284 above) 327. 288 CF Sabel and WH Simon, ‘Minimalism and Experimentalism in the Administrative State’ (2011–12) 100 Georgetown Law Journal 53, 79. 289  CF Sabel and J Zeitlin, ‘Learning from Difference: The New Architecture of Experimentalist Governance in the EU’ (2008) 14 European Law Journal 271, 274. 290  Hepple et al (n 271 above) 59–65, 69–72.

Ways Forward

 81

a positive duty on employers, including on the grounds of age, would encourage a more proactive and preventative, and less compliance-focused, organisational response to the equality framework,291 and would overcome some of the difficulties inherent in the positive action provisions. Further, it would shift the responsibility to organisations to identify and address unlawful discrimination, irrespective of whether an individual complaint has been received,292 and encourage a renewed focus on changing organisations to accommodate individual needs, rather than just adapting individuals to fit existing structures.293 Thus, it would also help to address some of the limitations of individual enforcement. By not imposing such duties on the private sector, the EqA does not go far enough to achieve equality:294 the duties would change ‘the whole landscape of discrimination law’295 through a ‘radical strategy for tackling deep-seated discrimination’.296 Positive duties therefore have the potential to address a raft of doctrinal limitations inherent in the current legal framework. Imposing a positive duty on the private sector would be likely to generate significant consternation among employers and substantial political resistance, as it would be seen as increasing the ‘regulatory burden’ on employers.297 The EqA currently limits positive duties to public bodies, on the assumption it would be ‘unfairly onerous’ to apply them to the private sector.298 However, the private sector has ‘potential strategic importance’ in addressing inequality, given that it employs the vast majority of the UK workforce.299 Private and public sector equality duties ‘harness the energy’ of those in the best position to promote equality and achieve structural change.300 It is also potentially unfair and practically problematic to create a division between the public and private sectors.301 Therefore, the imposition of positive duties to achieve employment equality on private sector employers should be seriously considered.

291  L Dickens, ‘Equality and Work-Life Balance: What’s Happening at the Workplace’ (2006) 35 Industrial Law Journal 445, 447; Fredman and Spencer (n 270 above) 6–7; Dickens, ‘The Road is Long’ (n 228 above) 474; Fredman, ‘Breaking the Mold’ (n 270 above) 271. In their original report, Hepple, Coussey and Choudhury did not extend the proposed positive duty to the ground of age, limiting it only to race, sex and disability: Hepple et al (n 271 above) 59–65, 69–72. Hepple later justified this distinction on the basis that age may be a relevant consideration in some circumstances (the ‘age is different’ argument) and that it would be ‘difficult and arbitrary’ to treat people in specific age bands as ‘groups’ that should be fairly represented: Hepple, ‘Age Discrimination in Employment’ (n 35 above) 84–85. However, this does not address the broader issue of how age equality should be secured, if not through a positive duty of this nature. 292  Fredman and Spencer (n 270 above) 6–7; Fredman, ‘Breaking the Mold’ (n 270 above) 266. 293  Dickens, ‘The Road is Long’ (n 228 above) 472. 294  Hepple, ‘Enforcing Equality Law’ (n 255 above) 319. 295  Fredman, ‘Breaking the Mold’ (n 270 above) 271. 296  Davies (n 222 above) 135. 297  Hepple, ‘Age Discrimination in Employment’ (n 35 above) 85. 298 Fredman, Discrimination Law (n 233 above) 178. 299  Fredman and Spencer (n 270 above) 8; see also Dickens, ‘The Road is Long’ (n 228 above) 474. 300 Fredman, Discrimination Law (n 233 above) 176; Fredman, ‘Breaking the Mold’ (n 270 above) 271. 301 Fredman, Discrimination Law (n 233 above) 178.

82 

UK Age Discrimination Laws: A Critique

It is also necessary to consider the practical content of any positive duty that might be placed on employers, and how it would be enforced. Hepple, Coussey and Choudhury propose that a positive duty include requirements to:302 —— Conduct periodical reviews (once every three years) of employment ­practices to determine whether certain groups are enjoying fair participation in employment; —— If certain groups are not enjoying fair participation in employment, draw up and implement an ‘employment equity plan’ in consultation with interest groups, to address barriers to participation and make reasonable adjustments to secure fair participation; and —— Disclose the results of periodical reviews and equity plans in company reports, to employees and to employee representatives.303 This may facilitate peer review and scrutiny, consistent with an experimentalist architecture. A failure to comply with these requirements could be used as evidence in proceedings for unlawful discrimination. Further, the EHRC could secure written undertakings or deliver notices directing employers to comply with the requirements and, if not complied with, apply to the ET for enforcement. Like the PSED, this positive duty has the potential to require organisations to develop their own solutions to equality issues, thereby avoiding the regulatory trilemma.304 However, the practical content of the duty also risks exacerbating the regulatory trilemma, by encouraging ‘mere procedural compliance’ by ­organisations.305 Thus, while a stronger positive duty may help to model UK equality law as a form of reflexive regulation, it may also exacerbate the regulatory trilemma and devolve into proceduralism.306 Secondly, as noted above, the PSED is significantly circumscribed.307 By only requiring authorities to have ‘due regard’ to equality matters, the duty ­establishes a standard that is ‘too deferent and too narrowly cast’ to further equality.308 In contrast, a duty for the public sector ‘to achieve employment equality or fair participation’ (like that proposed above for the private sector) would significantly strengthen the obligations imposed on public employers. By imposing a duty on employers to take action, this formulation would avoid the ‘risk of proceduralism’ inherent in the PSED.309 However, it would also require a far clearer understanding and specification of what ‘equality’ means, which is far from obvious.310 302 

Hepple et al (n 271 above) 71. See also F McAndrew, ‘Workplace Equality: Turning Policy into Practice’ (Equality and Diversity Forum, November 2010) 19–20. 304  Fredman, ‘The Public Sector Equality Duty’ (n 270 above) 419. 305  ibid 420. 306  See Fredman, ‘Breaking the Mold’ (n 270 above). 307  Fredman and Spencer (n 270 above) 9; Dickens, ‘The Road is Long’ (n 228 above) 473; Fredman, ‘The Public Sector Equality Duty’ (n 270 above); Fredman, ‘Breaking the Mold’ (n 270 above) 266. 308  Fredman, ‘Breaking the Mold’ (n 270 above) 266. 309  Fredman, ‘The Public Sector Equality Duty’ (n 270 above) 427. 310  See ch 1; Fredman, ‘Breaking the Mold’ (n 270 above) 271. 303 

Conclusion

 83

Finally, to enhance the effectiveness of reflexive regulation, Hepple, Coussey and Choudhury propose a duty on employers to engage with other groups in ­decision-making.311 This could operate in a similar way to the duty to ­consult employees in relation to health and safety matters in the UK.312 Again, like the positive duty, this runs the risk of falling into the regulatory trilemma, and encouraging employers to adopt a procedural compliance approach. Further, it relies on local actors having the skills and experience (or developing the skills and experience) to negotiate effectively about equality matters, which is doubtful at present.313 However, the relative effectiveness of these provisions in the context of health and safety offers a potential starting point for the duty in relation to equality matters.314 The viability of these reforms is explored in more detail in Chapter eight.

IV. Conclusion UK age discrimination laws have fundamental limitations in achieving the ­instrumental and intrinsic objectives laid out in Chapter one. To address age discrimination effectively, and to secure quality employment opportunities for workers of all ages, the law must be amended to restrict the ability to justify direct age discrimination and remove the possibility of employers adopting an EJRA. These changes would send a strong message to employers about the importance of age equality and the government’s commitment to extending working lives. Further, they would bring the law into line with government objectives, which are presently at odds with employer justifications and ‘legitimate aims’ upheld in court decisions. However, law reform alone is not sufficient to achieve meaningful change in this area. Effecting change via legislation has a number of limitations, including the reliance on individuals to enforce their rights through litigation.315 ­Further, the tendency of most legislation to be ‘broad brush’ means that employers have

311 Hepple et al (n 271 above) 71; Hepple, ‘Agency Enforcement of Workplace Equality’ (n 264 above) 60. 312 See, eg, Health and Safety at Work etc Act 1974, s 2(6); Safety Representatives and Safety ­Committees Regulations 1977, SI 1977/500; The Health and Safety (Consultation with Employees) Regulations 1996, SI 1996/1513. 313  McLaughlin (n 267 above) 27. 314  For a critique, see P James and D Walters, ‘Non-Union Rights of Involvement: The Case of Health and Safety at Work’ (1997) 26 Industrial Law Journal 35; P James and D Walters, ‘Worker Representation in Health and Safety: Options for Regulatory Reform’ (2002) 33 Industrial Relations Journal 141; see also D Walters and T Nichols, ‘Representation and Consultation on Health and Safety in Chemicals: An Exploration of Limits to the Preferred Model’ (2006) 28 Employee Relations 230. 315  Harcourt et al (n 108 above) 454; see further Dickens, ‘Introduction’ (n 237 above); Morris (n 237 above).

84 

UK Age Discrimination Laws: A Critique

considerable latitude in determining how the legislation is applied.316 A shift towards reflexive regulation and positive duties may help to address some of these issues. At the same time, however, any process of reform must involve and include employers and organisations: significant changes in behaviour are more likely to come from modification of employers’ policies than legislative reform.317 Legal doctrinal research is only capable of presenting part of this complex picture. Drawing on this doctrinal analysis as a foundation, the following chapters consider how employers are actually implementing age discrimination laws, and examine the strengths and limitations of the laws in practice.

316 

Harcourt et al (n 108 above) 450; see further Chapter 7 of this volume. ‘Combating Age Barriers in Employment’ (n 216 above) 138; S Vickerstaff et al, ‘Employers and the Management of Retirement’ (2003) 37 Social Policy & Administration 271, 271. 317 Walker,

4 Expert Perspectives on the Operation of Age Discrimination Laws Legal doctrinal research has limited utility for evaluating the practical efficacy of age discrimination laws. While doctrinal research offers insights into what the practical limitations of law might be, we must look beyond traditional hermeneutic methods to actually study law in its social context. To examine and critique the practical operation of age discrimination laws in the UK, I conducted semistructured qualitative interviews with 17 UK experts in employment law and the employment of older workers between September and November 2012. ‘Experts’ were defined as those who were active participants in ageing issues and likely to have special knowledge regarding older workers.1 The experts were asked for their views and evaluation of the efficacy of age discrimination laws, and potential reforms that may help to enhance the laws’ operation. The experts were drawn from government, trade unions, employer groups, lobby groups, the judiciary and academia.2 Analysis of the expert interview data, using content analysis to identify, inductively code and categorise key patterns in the data,3 shows that many experts feel

1 M Meuser and U Nagel, ‘The Expert Interview and Changes in Knowledge Production’ in A Bogner et al (eds), Interviewing Experts, Research Methods Series (Basingstoke/New York, Palgrave Macmillan, 2009) 17, 24. 2  The experts represented the following organisations: Acas (Advisory, Conciliation and Arbitration Service), Age UK, CIPD (Chartered Institute of Personnel and Development), CBI (Confederation of British Industry), Department for Work and Pensions (DWP), Department of Business, Innovation and Skills (BIS), enei (employers network for equality and inclusion), Equality and Human Rights Commission (EHRC), Federation of Small Businesses, General Federation of Trade Unions, Grandparents Plus, the judiciary, Middlesex University, Oxford Brookes University, TAEN (The Age and Employment Network), and Trades Union Congress. Interviewees have been broadly classified as a representative of government or a non-departmental government agency (Government), academia (Academic) or a non-governmental organisation (NGO); and allocated a number. So the first government representative is identified with Government 1; the fourth non-governmental organisation representative with NGO 4. For full discussion of the research procedures, see A Blackham, ‘Extending Working Life for Older Workers: An Empirical Legal Analysis of Age Discrimination Laws in the UK’ (PhD Thesis, University of Cambridge, 2014). 3  MQ Patton, Qualitative Evaluation and Research Methods, 2nd edn (Newbury Park CA, Sage, 1990) 381.

86 

Expert Views on Age Discrimination Laws

there is significant legal uncertainty around age discrimination laws, with many employers choosing not to adopt proactive practices for fear they might be discriminatory. Further, there is recognition that employers are adopting a conservative or compliance-focused approach to implementing the laws. Experts also voiced significant concern that attitudinal change is not keeping pace with legislative reform, particularly among employers. At the same time, few ideas emerged from the interviews for promoting attitudinal change, with many experts expressing feelings of futility or ineffectiveness at the need to change underlying stereotypes of older workers. While the experts generally did not endorse legislative change, the interviews reveal significant limitations in the practical implementation of age discrimination laws.

I.  Expert Views on the Efficacy of Law A.  Doctrinal Merits Nine experts believed existing laws were broadly adequate for addressing age discrimination. While the law in the UK was seen as less proactive than in other countries (cf Government 1), there was recognition that ‘there are really no [legal] barriers for older workers now’, enabling (at least in theory) older workers’ full participation in the labour market (NGO 2).4 Respondents felt the law to be in ‘pretty good shape’ (NGO 2), with the EqA being ‘broadly positive’ and ‘a really good basis for progressing these issues’ (NGO 7; similarly Academic 2). According to one respondent, the EqA is ‘about as good as you are going to get’ (NGO 2), and respondents had few ideas for improving the legislative provisions.5 Indeed, some respondents regarded the EqA as a given which could not or would not be changed (NGO 5): ‘I don’t think you could do it differently’ (Government 4); ‘I haven’t really thought about how it would work differently’ (NGO 5). While most experts regarded the EqA in a positive way, five respondents held strongly contrary views, with one describing the law as ‘rubbish’ and Article 6 of the Framework Directive as ‘an appalling piece of legislation’ (Academic 1). Concerns were raised about the specific terms of the EqA and its application of the Framework Directive. First, the ability to justify direct discrimination under the EqA was regarded with unease: Now that’s there in the European basis on which we founded our law … I don’t think it’s entirely convincing, but it’s certainly pretty arguable that you need to do that for age 4  This closely echoes notions of formal equality of opportunity (discussed in Chapter 1 of this ­volume), though the respondents did not openly discuss their ideas of ‘equality’. 5  This may reflect a broader change fatigue among the respondents: JB Bernerth et al, ‘Change Fatigue: Development and Initial Validation of a New Measure’ (2011) 25 Work & Stress 321, 322.

Expert Views on the Efficacy of Law

 87

because it is a bit different, and there are going to be situations where it’s more palatable to do it on grounds of age. Whether it’s therefore appropriate to enshrine it in law, I’m not sure, but it’s more appropriate to allow for that possibility. (NGO 2)

Another respondent criticised the UK’s implementation of the Framework ­Directive in allowing the same justifications and legitimate aims to support direct and indirect discrimination: the EqA ‘in apparently putting the two [justifications] together is a bit of problem and it’s even more of a problem that the Court of Justice said it was alright!’ (Government 5) The expert argued that justifications for direct discrimination should be ‘of a different character’: ‘much more precise’ and ‘more narrow’ than those for indirect discrimination, in accordance with the Framework Directive. While noting the expert’s concern, it is arguable that the CJEU does not actually appear to be ‘putting the two’ tests together.6 Indeed, this was expressly denied by Lady Hale in Seldon v Clarkson Wright & Jakes (A partnership):7 It now seems clear that the approach to justifying direct age discrimination cannot be identical to the approach to justifying indirect discrimination and that regulation 3 (and its equivalent in section 13(2) of the Equality Act 2010) must be read accordingly.8

The expert’s error may reflect broader legal misunderstandings and uncertainty in the law. Secondly, one expert criticised the number of exceptions for age discrimination under the EqA, arguing that they undermined the EqA’s efficacy: If you are going to have exceptions to a rule then you should actually specify what those exceptions are … [rather than] just leaving it to the courts to decide and not setting the parameters for the courts … Clearly there is justifiable discrimination … I’m not disputing that but I think they ought to be prepared to specify the areas in which there are exceptions that are able to be made. (Academic 1)

The respondent also criticised the retention of EJRAs, arguing that ‘we have to just absolutely … abolish any possibility of retiring people … because they reach a chronological age, which just has to be unacceptable’ (Academic 1). Excluding older workers from employment on the basis of retirement is ‘treating [them] as second class citizens … because they have reached a certain age and I think that’s not acceptable’ (Academic 1). Thirdly, respondents expressed concern at the classification of ‘dignity’ as a legitimate aim. One questioned the ‘readiness with which the Luxembourg Court accepted the dignity arguments for justifying a fixed retirement age’, noting that it was difficult for UK courts ‘to say that this is not a legitimate aim if Luxembourg has said it is a legitimate aim’ (Government 5). Another thought the ‘dignity’ 6  See C-388/07 R (Age Concern England) v Secretary of State for Business Enterprise and Regulatory Reform (5 March 2009), [2009] 3 CMLR 4, paras 58–64. 7  Seldon v Clarkson Wright & Jakes (A partnership) [2012] UKSC 16 (25 April 2012). 8  ibid [51].

88 

Expert Views on Age Discrimination Laws

aim would perpetuate ageist stereotypes: ‘this has got perception stamped on it’ ­(Government 4). This may undermine the position of older and younger workers: ‘so if you’re 30 then I can crucify your dignity and boot you on the street and that’s okay socially. Wake up! Smell the coffee!’ (Government 4)9 Fourthly, even experts who had positive views about the EqA saw legal uncertainty as a significant issue. They argued that the law on EJRAs has ‘always been complicated’ (Government 3) and the Seldon decision had failed to clarify the ­proportionality issue: Is 65 proportionate? God knows. God knows is the short answer to that because nobody has said one way or another and the courts will take a long, long time to decide, if ever … And [proportionality] will change over time. (Government 4)

It was suggested that this confusion may reflect the law’s need to address ‘very, very complex sets of circumstances’: [It is] trying to indicate the way an employer should respond to these potential circumstances so that they don’t break the law. So it’s almost like trying to second guess what might happen and it becomes extraordinarily difficult to think your way through this stuff, you know, and again if that is the case then you need to resort to a lawyer and the lawyer tries to explain and you go: ‘Oh my god, I can’t cope with any more of this’. That’s a time consuming thing and you would kind of switch off. (NGO 4)

The respondents recognised that frequent changes to equality law and the DRA have compounded this confusion (NGO 4), leading ‘to perceptions of unfairness’ and giving ‘legislation a bad name’ (Government 4). For one respondent, this uncertainty will resolve itself over time as case law develops (NGO 7). However, relatively few age discrimination cases are emerging, limiting the courts’ capacity to provide effective guidance (NGO 2, Government 1). The experts’ doctrinal concerns echo the issues discussed in Chapter three, particularly in relation to the lack of legal clarity regarding EJRAs (and concerns regarding the retention of mandatory retirement more broadly), the acceptance of ‘dignity’ as a legitimate aim, and broad exceptions to the principle of age equality. These expert interviews indicate that doctrinal limitations are constraining the law’s ability to achieve its instrumental and intrinsic objectives. Two experts offered a different perspective, criticising the law for failing to protect employers’ interests and excessively regulating the employment of older workers. They noted that the abolition of the DRA means the ‘pendulum has swung towards giving the choice to the older workers as opposed to the employers so I think the relationship is slightly out of kilter’ (NGO 6). The substantial administrative costs involved in reading, understanding, seeking help and implementing government changes were seen as onerous, particularly for small business: ‘just

9  In contrast, one expert noted the need for ‘performance managing with dignity when it gets to say an older age group’ (NGO 6), implicitly endorsing ‘dignity’ as a legitimate objective of employment policy affecting older workers.

Expert Views on the Efficacy of Law

 89

how much of your time is supposed to be taken up with all of the changes and all of the new legislative requirements that are constantly coming down the path at us?’ (NGO 8) The experts recommended that government should instead adopt a less interventionist role: We would advocate that legislation should be a last resort rather than a first resort … the government feels that if something is just not working the way it wants then they have to legislate, legislate, legislate. And I think that we really have to be moving away from that and getting people to be sensible about what employment is actually all about. (NGO 8) We should strip out employment law, in an ideal world, so it’s a lot simpler than it is now and so the culture that goes with employing people is not one of risk and of the fear of tribunal, particularly for smaller businesses, it’s one where employers are free to think: Well, actually, we can do cool things with our workforce. (NGO 6)

This implies that law and government intervention may stifle organisational innovation and creativity, via the ‘regulatory burden’ of employment regulation on business.10 While previous research has found limited evidence of any impact of legislation on innovation or creativity,11 this may be because it has rarely been an object of study.12

B.  Practical Implementation While few experts desired legal change, most respondents regarded the EqA’s implementation as an ongoing challenge, noting it is now ‘really just a matter of embedding those policy changes and getting people to realise what freedoms they have’ (NGO 2). However, the law was seen as making ‘quite a lot of strides … in starting the conversation [about age discrimination] and setting it up as a business case approach’ (Academic 3). The real impact of the EqA was unclear at this stage: the introduction of age discrimination legislation clearly is important and I would have thought capable of being an improvement but it’s a bit too early to say how much of an improvement it will eventually turn out to be. (Government 5)

10 See R Blackburn and M Hart, ‘Employment Rights in Small Firms: Some New Evidence, ­ xplanations and Implications’ (2003) 32 Industrial Law Journal 60, 64–65; L Dickens and M Hall, E ‘Fairness—up to a Point. Assessing the Impact of New Labour’s Employment Legislation’ (2006) 16 Human Resource Management Journal 338, 346–47. 11  S Wood, ‘Overview of the Impact of Employment Relations Law: Post-1997’ in L Dickens et al (eds), Review of Research into the Impact of Employment Relations Legislation, Employment Relations Research Series No 45 (Department of Trade and Industry, October 2005) 90–91. 12  ibid 94. cf the perception of the impact of regulation: E Jordan et al, ‘Employment Regulation— Part A: Employment Perceptions and the Impact of Employment Regulation’, Employment R ­ elations Research Series No 213 (BIS, March 2013); L Dickens, ‘The Coalition Government’s Reforms to Employment Tribunals and Statutory Employment Rights—Echoes of the Past’ (2014) 45 Industrial Relations Journal 234, 240–41.

90 

Expert Views on Age Discrimination Laws

(i)  Implementation by Employers (a)  Differences between Employers A number of respondents (11) noted that the implementation of age discrimination legislation varied considerably between employers: I think it’s mixed; I think the evidence is showing that [there] is really, really good employer practice out there; I think there’s a lot of good will and there’s a lot of desire and a lot of employers who see the benefits and the need for adapting for an ageing workforce, but I also think there are a lot that don’t and don’t get it or don’t have time to get it, or get it but aren’t quite sure what to do about it. (Government 2)

Factors that were identified as influencing organisational responses included organisational size, sector, and employee job or occupational group. Three interviewees believed that larger organisations with a specialist human resource (HR) function were more likely to support older workers and be well informed of their own diversity obligations. However, two other respondents felt small organisations were more likely to manage their HR obligations informally, creating fewer diversity issues and encouraging a focus on whether a person can ‘do the job’ (NGO 8). Small organisations: know their workforce intimately, almost treating them as part of the family … Doesn’t matter what age, what their equality issues may or may not be, all [small businesses want] are people who we can train to do the job that we need to do to service our customers and clients. It’s as simple as that. (NGO 8)

The impact of organisational characteristics on responses to ageing is considered further in Chapters five and seven. (b)  Prevalence of ‘Compliance’ Approaches While recognising the diversity between UK workplaces, three experts noted the tendency of UK workplaces to adopt a compliance or ‘defensive’ approach to age discrimination laws, with limited innovation occurring beyond the legal minima. While the experts viewed compliance with the law as ‘obviously appropriate’, a compliance approach was not perceived to be sufficient to deal with the ‘bigger challenge’ of ageing (NGO 3).13 This was compared with the approach in other European countries, which was seen as more proactive: We find that employers in Europe are increasingly talking about ‘age management’ … And it really rather begs the question of what should employers be doing, going beyond the narrow compliance, legal sort of defensive postures that they adopt here in the UK. (NGO 3)

13  See also E Parry, ‘Managing an Ageing Workforce: The Role of Total Reward’, Research Insight (CIPD, January 2008) 9.

Expert Views on the Efficacy of Law

 91

There was a general consensus that many employers are not fully engaging with the challenges of ageing, instead focusing on how they can retire older workers: I don’t think any employers actually have done very much thinking about what accommodation of the needs of older workers might even involve so I think that they seem to me to be probably working on an ad hoc basis of the people they have got in front of them and not necessarily thinking about age as an issue that needs any adaptation actually … I would be concerned that employers are not very aware of age as an issue that needs accommodating and I think they think about it as about retirement and that’s it. (Academic 2)

The experts felt that many employers remain preoccupied with the DRA and retirement ages, reducing consideration of other issues that affect older workers, and encouraging a ‘compliance’ focus: In general when you speak to employers about older workers they throw back at you the DRA; that’s the thing they think of and they don’t seem to have moved on from the fact that this has now been repealed. And they think in terms of compliance with the law— which of course is about non-discrimination—so this creates a somewhat simplistic and spurious mind-set of ‘Oh, we can’t do anything that is different for older people because that would be age discrimination!’ (NGO 3).

According to one respondent, ‘employers latched onto [the DRA] because … they felt they were retaining the control in the situation’ (NGO 2). Following removal of the DRA, there is a perception ‘of the law around older workers that it is creating a problem for the employer that didn’t exist before when … everybody got sacked at 65’ (Government 4; similarly, NGO 8), making it more difficult for employers to manage older workers. However, experts felt most employers (and employers’ associations) now ‘quietly accept’ the need to extend working lives (Government 4): employers ‘are not squealing as much as one might think they were going to’ following the abolition of the DRA (Government 1). While one respondent predicted that most employers will retain an ‘indicative’ retirement age of 7014 ‘and that will be okay for the time being probably’ (Government 5), another felt it was unlikely that employers would risk imposing an EJRA given the degree of legal uncertainty (Government 3). Instead, most employers ‘are sitting there unsure of what to do’ (NGO 5).15 Despite this uncertainty, some UK workplaces are implementing good practices to support older workers: There are some workplaces that do accommodate older workers very positively, some actually. I mean the classic example is B&Q where they sort of go out of their way to recruit older workers and see them as a positive asset to the workforce … I know there are some employers who positively recognise that older workers actually have 14 

This presumably equates to an EJRA of 70. This is consistent with early anecdotal and survey evidence, which indicates that some employers are abandoning fixed retirement ages: see Chapter 3 of this volume. 15 

92 

Expert Views on Age Discrimination Laws

the memory of the organisation; they are often extremely experienced, they have a kind of institutional memory which can’t be offered by younger employees. And they recognise and respect and, you know, want to hang on to that, but it is very variable. (Government 1)

That said, even this example of good practice was contentious. Another respondent stated: I think we need to stop with the 87 year old man pushing trolleys in B&Q as being an example of extending working life and actually do proper examples where … we can demonstrate the value of working longer to people’s lives. (Government 2)

Respondents noted good employer practices such as actively recruiting older workers, offering bridge jobs in the lead-up to retirement and options to reduce working hours, adjusting pension schemes to reduce financial disincentives to work, not imposing mandatory retirement on employees and encouraging flexible working arrangements.16 While these practices may benefit older workers, experts noted the difficulty in evaluating their impact using employer testimonials: If you look at these case studies [of good practice] they are generally all compiled by the employer so it is kind of difficult to test them—you know, how objective they are and how relevant and valid the claims are which they make concerning their impact on employees. So this is the sort of problem that we have got here: we have to rely on what employers say without having any testimony from the workers’ side. (NGO 3)

(c)  Role of Law In sum, while there is some evidence of good practice, the interviews indicated that most employers are adopting a defensive or compliance-focused approach to age discrimination legislation, and there is limited evidence of proactive practice among UK employers. However, according to the experts, it is not the role of law to require innovative or good practice: law is a ‘framework’ (Government 5), ‘underpinning structure’ (Government 2) or ‘bedrock’ (NGO 2, Government 4) which ‘creates a floor for decent standards for the way people treat each other’ (Academic 1). In the experts’ view, upon this minimum ‘safety net’ (Government 2) employers can make further provision for their employees: legislation is ‘not restricting people from doing more than [is prescribed]’ (Government 2). However, for the experts, this additional provision was not the role of government. Further, the experts argued that law was only one piece of the puzzle for addressing age discrimination: ‘it can only set the framework within which one hopes employers can be good employers and employees can be good employees’ (Government 5). However, they felt that while ‘the law doesn’t solve the issue …

16  For further discussion of flexible work in this context, see A Blackham, ‘Rethinking Working Time to Support Older Workers’ (2015) 31 International Journal of Comparative Labour Law and ­Industrial Relations 119.

Expert Views on the Efficacy of Law

 93

it helps’ (Academic 1; similarly, NGO 7). This conception of law is consistent with the traditional laissez-faire, ‘hands-off ’ approach to government in the UK and a minimalist understanding of legal regulation, as noted by one expert: ‘the UK system tends to be one of general permissiveness … we don’t say what you can do we only say what you can’t do generally’. (Government 3) The experts’ views are not incompatible with a reflexive law approach: in addition to setting a minimum floor of rights, law could also incentivise employers to ‘do more’ for their employees. Indeed, it is arguable that the EJRA has had this effect: while employers are able to retain a retirement age, the process of justifying mandatory retirement may encourage employers to review and reconsider their retirement processes. At the same time, while reflexive law is not inconsistent with the experts’ views, the interviewees did not explicitly acknowledge that law may play this sort of role in encouraging good practice.

(ii)  Practical Impact of Legal Uncertainty The interviews indicate that legal uncertainty may be limiting employers’ responses to the ageing workforce and promoting a ‘compliance’ approach among UK workplaces. As one respondent noted, ‘even those workplaces where there’s a willingness to sit down and redress [sic] the issues and do something positive, there’s often still uncertainty about what they can lawfully do or not do’ (NGO 7). This is particularly the case regarding retirement ages and positive action: ‘Sometimes I hear of organisations that want to do something positive for older workers but they are afraid that that may mean they will get age discrimination challenges from younger workers’ (NGO 7). While many positive measures could probably be justified under the legislation, ‘there is just that fear around doing it’ (NGO 7; similarly, NGO 3). This is consistent with the doctrinal concerns raised in Chapter three: while positive action is allowed under the EqA, legal uncertainty and a risk of legal action mean employers are unlikely to make use of the provisions. Employers’ fear or anxiety regarding age discrimination laws was a recurring theme in the interviews.17 Experts believed employers to be anxious about: whether older workers could still be productive, or whether their ‘high performance organisation is going to be … dragged down by older workers’ (NGO 3); whether they ‘are not going to be able to get rid of people’ like they could with a DRA (NGO 3); being accused of age discrimination (NGO 5, 7); and being subject to a legal claim of age discrimination (Academic 1, NGO 4, 6). Many of these fears reflect negative stereotypes about older workers,18 indicating that the law is not addressing negative attitudes towards older workers. At the same time, a legal system that cultivates fear may ‘have the opposite effect to what you’ve intended’

17 

See similarly Jordan et al (n 12 above). VJ Roscigno et al, ‘Age Discrimination, Social Closure and Employment’ (2007) 86 Social Forces 313, 314–15, 323–24; TWH Ng and DC Feldman, ‘Evaluating Six Common Stereotypes about Older Workers with Meta-Analytical Data’ (2012) 65 Personnel Psychology 821. 18  See

94 

Expert Views on Age Discrimination Laws

(NGO 4), as employers fail to take any action to address ageing for fear of legal repercussions. This is consistent with previous studies of the implementation of diversity measures. McAndrew argues: many well-intentioned employers are held back on their journey to diversity excellence by fear; fear of getting things wrong, fear of appearing unskilled, fear of somehow precipitating litigation, and fear of transparency. It could be argued that the field of equality and diversity management has become overly complex. What is primarily a matter of common sense has been weighed down by a confusing legislative landscape and anomalies in rights; and by the jargon and politically charged arguments that accompany it. This escalates the sense of fear and apprehension that ultimately deters some employers from engaging with equality issues at all.19

(iii)  Persistence of Negative Attitudes The persistence of negative attitudes towards older workers, despite age discrimination legislation, was the most frequently raised concern in the interviews (noted by 16 experts). Age discrimination is still a significant issue in the UK (Academic 2, Government 1, NGO 1, 2, 7) and fighting it remains ‘an ongoing battle’ (NGO 1). Experts noted the need for attitudinal change among employers (Academic 1, Government 2, NGO 3, 5, 7, 9), line managers (NGO 5), trade unions, organisational management, the human resource profession, business leaders (NGO 3), employees and older workers themselves (Academic 3, Government 1, 2). While respondents recognised that significant attitudinal change has already been effected through government and third sector awareness-­ raising programmes, it is ‘still going to be a major, major issue for years to come’ (NGO 4).20 To some, attitudinal change is an insurmountable problem that is difficult (if not impossible) to address through government policy: ‘you can’t legislate for attitudinal change’ (Government 1; similarly, Government 2). Many experts were despondent at the lack of attitudinal change: ‘I don’t think there’s much that can be done. I don’t know. I don’t know how you change attitudes in society’ (Academic 1; similarly, NGO 1, 7, 9). This is also a live issue in the literature: there is a very real question about whether government interventions can promote attitudinal change. D’Addio and others found that countries with a strong public stance against age discrimination did not necessarily experience a decline in reports of age discrimination.21 Similarly, in their analysis of data 19 F McAndrew, ‘Workplace Equality: Turning Policy into Practice’ (Equality and Diversity Forum, November 2010) 8. See also E Kelly and F Dobbin, ‘How Affirmative Action Became Diversity ­Management: Employer Response to Antidiscrimination Law, 1961 to 1996’ (1998) 41 American Behavioral Scientist 960, 963–64. 20  See similarly E Dewhurst, ‘Are Older Workers Past Their Sell-by-Date? A View from UK Age ­Discrimination Law’ (2015) 78 Modern Law Review 189, 191–94. 21  AC D’Addio et al, ‘Population Ageing and Labour Markets’ (2010) 26 Oxford Review of Economic Policy 613, 627.

Expert Views on the Efficacy of Law

 95

from the European Social Survey 2008–09, Abrams, Vauclair and Swift identified country and individual indicators associated with more favourable attitudes towards older people:22 these indicators were mostly unrelated to government interventions. Difficulties achieving attitudinal change through legal intervention may be explained by systems theory: if law and other social areas occupy discrete, closed, self-referencing and self-reinforcing autopoietic systems,23 operational closure between systems will mean that law and society lack a means of directly communicating. Social attitudes and norms cannot be ‘regulated directly from the outside’ by law24 and it is necessary to revisit the idea of ‘linear causality’ between law and social change.25 However, cognitive openness between systems means it is possible for law to encourage attitudinal change indirectly.26 Two academic experts suggested that attitudes towards older workers might become more positive over time with removal of the DRA, demographic change, and labour and skills shortages (Academic 1, 3). As noted by another expert: Yeah I mean, peoples’ attitudes and the culture, how do you change that? I mean where do you start? I don’t know how to say it but every one of us is going to get older and more people are going to be older and you are hoping that eventually there’s going to be a shift but it doesn’t seem to be the case at the moment. (NGO 1)

This is consistent with current literature: as the workforce ages, economic realities may force employers to confront out-dated stereotypes of older workers.27 Other experts were more optimistic about the potential for government policy to shift people’s attitudes (Academic 1, Government 3). One respondent cited the example of laws requiring the use of seatbelts in the UK: while some individuals originally resisted wearing seatbelts, they have now become part of people’s daily routine, to the point where individuals ‘feel naked if [they] haven’t got a seatbelt on’ (Government 4):

22  D Abrams et al, ‘Predictors of Attitudes to Age across Europe’, Research Report No 735 (DWP, 2011). At the individual level, more favourable attitudes to older people are correlated with old age, being female, a higher level of education, subjective affluence, membership of an ethnic minority, and living in a rural area. Societal indicators associated with more favourable attitudes to older people include higher levels of GDP, higher levels of income inequality, lower unemployment rates, later state pension ages, a larger proportion of the population being aged over 65, and country values that emphasise personal autonomy. 23  See N Luhmann, Law as a Social System, F Kastner et al (trans), Oxford Socio-Legal Studies (Oxford, Oxford University Press, 2004) and Chapter 2 of this volume. 24 G Teubner, Law as an Autopoietic System, The European University Institute Press Series, Z Bankowski (ed), A Bankowska and R Adler (trans) (Oxford, Blackwell, 1993) 71. 25  ibid 74. 26  G Teubner, ‘After Legal Instrumentalism? Strategic Models of Post-Regulatory Law’ in G Teubner (ed), Dilemmas of Law in the Welfare State (Berlin, Walter de Gruyter, 1986) 307. 27 See, eg, BL Hassell and PL Perrewe, ‘An Examination of Beliefs about Older Workers: Do Stereotypes Still Exist?’ (1995) 16 Journal of Organizational Behavior 457, 457; RW Johnson, ‘Managerial Attitudes toward Older Workers: A Review of the Evidence’, Discussion Paper 07-05, The Retirement Project (Washington DC, The Urban Institute, 24 September 2007) 27.

96 

Expert Views on Age Discrimination Laws

For me that’s a clear demonstration of a law that in the first instance demanded ­behavioural constraint ie you will do this, like it or lump it and we will prosecute you if you don’t. [But now] it’s actually changed attitudes. Now there are some people out there who say you can’t change attitudes, you can only change behaviour and I would say well look at that. That changed behaviour. And then it changed attitudes. So with law, if you can change the behaviour of how we treat older workers, over time that will change ­attitudes, and once you change attitudes then you can, I think, start pulling back, if you like, relinquishing laws … But at the moment I don’t think that we’re anywhere near there for all sorts of … time-related reasons. (Government 4)

Respondents also noted the success of the right to request flexible working as a means of changing employers’ attitudes: while the right does not require employers to adopt flexible work patterns, it does require them to consider the possibility of flexible working (Academic 2, Government 1). Although the right to request was regarded as being ‘rather weak legislation’ (Academic 3) and ‘really feeble’ (Academic 2), ‘it’s actually done a lot of good because once you get employers thinking about [it] … once those ideas are considered for one group of workers then it’s easier to introduce for other groups of workers’ (Academic 3; similarly, Academic 2). From the experts’ perspective, the legislation puts in employers’ minds an expectation that they will receive the requests [for flexible working] and will hopefully then prompt them to think ahead about how they will respond to them and to configure workplaces … in a way that allows that to happen. (Government 1)

The extension of the right to request to all workers was therefore regarded as having the potential to engender ‘a wider cultural change where anyone is able to ask for flexible working’ (Government 2). This is consistent with academic studies of the right to request: in a January 2005 survey of 3,222 UK employees, 14 per cent of respondents had requested a change to their working arrangements in the past two years. Of these requests, 81 per cent had been fully or partly accepted, down from the number accepted in 2004 (86 per cent), but still above the number accepted before the law was introduced (77 per cent).28 The number of declined requests decreased from 20 per cent to 11 per cent after the law was passed.29 It appears that the obligation to consider requests for flexible working has engendered new attitudes towards flexible working, and even provided benefits to employees without a right to request under legislation.30 Therefore, even weak legislation (like the EqA) may encourage attitudinal change. However, given the EqA’s ambivalence towards age discrimination, and the symbolic endorsement of discrimination via

28  H Grainger and H Holt, ‘Results of the Second Flexible Working Employee Survey’ in Office for National Statistics: Labour Market Trends (July 2005) 297, 300. 29  ibid 301. 30  S Himmelweit, ‘The Right to Request Flexible Working: A “Very British” Approach to Gender (In) Equality?’ (2007) 33 Australian Bulletin of Labour 246, 254.

Expert Views on the Efficacy of Law

 97

mandatory retirement, it is questionable whether the EqA provides a sufficient symbolic impetus to change attitudes towards older workers. Even if legal change can effect attitudinal change, these changes will take time (and, often, a significant period of time): ‘I think eventually it’ll be fine but I doubt whether it’ll be in my lifetime and I’m not sure of [whether it will happen] in yours either!’ (Academic 1) Further, current government programmes were seen as inadequate for realising attitudinal change. One expert suggested that the government is ‘not really very interested’ in attitudinal change, and has produced limited material and only ‘a few case studies of dubious quality’ on the issue (NGO 5). Others concurred: while the removal of the DRA was ‘meant to be indicative of a cultural shift’ (Government 3) towards ‘social acceptance that older people are active members of society for much, much longer now’ (Government 1), the experts were unsure how effective it was at securing attitudinal change. One suggestion to achieve real attitudinal change was that there ‘be something more; more of a push from perhaps government and campaigning and everything to get up to the next level’ (NGO 7). However, one expert felt it was not the role of law to create this cultural shift: ‘it’s a cultural change that I would want to see shifting … I’m not saying the law is right but I don’t think it’s the main problem anymore’ (Government 1).

(iv)  Enforcement of Age Discrimination Legislation Concerns were also raised about the enforcement of age discrimination legislation in practice, with few claims successfully pursued in the courts: ‘It’s very difficult to prove it; it’s very difficult to enforce it … the framework is there in theory but I think in practice it maybe doesn’t work terribly well’ (NGO 5). This echoes concerns about the reliance on individual claimants to enforce legislation (NGO 7).31 Table 4 demonstrates that these concerns may be justified: in 2012–13, only four per cent of age discrimination claims were successful at hearing.32 While this is much higher than in previous years, it is still well below the 11 per cent success rate for all types of claims.33 That said, the percentage of age discrimination claims settled through Acas conciliation (43 per cent in 2012–13) is significantly higher than the figure for all types of claims (33 per cent in 2012–13). Therefore, the lower success rate at hearing may be offset by significantly higher numbers of conciliated settlements.

31  See also Chapter 3 of this volume; L Dickens (ed), Making Employment Rights Effective: Issues of Enforcement and Compliance (Oxford, Hart Publishing, 2012). 32  Further, where claims are successful, reductions to compensation may particularly disadvantage older claimants: see Dewhurst (n 20 above) 202–08. 33  Indeed, in 2011–12, only 1% of age discrimination claims were successful at hearing, compared with 12% for all claims.

98 

Expert Views on Age Discrimination Laws

Table 4:  UK Age Discrimination Claims and Outcomes by Year, 2007–13 (Source: Ministry of Justice) Year

Claims Accepted by ETs

Per cent Successful at Hearing

Per cent Acas Conciliated Settlements

2007–08

2,900

3

45

2008–09

3,801

2

40

2009–10

5,184

2

39

2010–11

6,821

2

35

2011–12

3,715

1

33

2012–13

2,818

4

43

Compounding concerns about individual enforcement through litigation, the experts also felt that many employers lacked awareness of their legal obligations (NGO 7), particularly in small and medium-sized organisations (Government 1), and that many employees were not aware of their rights (Government 1, NGO 5). If individuals are unaware of their rights they are less likely to enforce them in the courts, making reliance on individual enforcement futile. According to one expert, the compound effect of both employers and employees failing to understand the law is particularly risky: If those businesses are employing people who are very unaware of their own legal rights, the combination means that people just don’t pursue their rights at all, the employer is taking outrageous decisions, the employee, although wronged, has no idea they have been wronged and doesn’t take any action so the net result is that they [the employer] get away with discrimination. (Government 1)

It is therefore essential that employers and employees are made aware of their legal rights and obligations, and that legal uncertainty be addressed to facilitate broader understanding of the law. More generally, however, this reinforces arguments for a move away from relying on individual enforcement, including through a focus on reflexive law and positive duties.

C.  Reforms to Improve the Law and its Implementation (i)  Reducing Legal Uncertainty To reduce legal uncertainty and employer anxiety, the experts suggested creating additional guidance material on the operation of age discrimination laws. Two respondents identified existing Acas guidance around work conversations as being particularly helpful in clarifying employers’ obligations and encouraging discussions with employees throughout their working career (and not just at retirement) (Academic 3, Government 3). Respondents also proposed the introduction of a

Expert Views on the Efficacy of Law

 99

code of practice to sit alongside existing guidance and legislation ‘to give more guidance and examples about what is and isn’t possible’ (NGO 7). On its own, the production of more information and guidance will not improve the accessibility and availability of information or ‘equate to more and more competent employers or employees’.34 Once guidance material is developed, it needs to be made readily accessible and available (including for those who do not have the Internet), come from a reliable source, and be ‘translated into very easy-peasy stuff for [employers and practitioners] to understand’ (NGO 4; similarly, NGO 8). To achieve this: ‘you need a facilitator. The government could be a facilitator in helping and fostering that understanding and signposting people to relevant bits of information and by creating the right infrastructure for the appropriate responses’ (NGO 4). With recent cutbacks the government has retreated from a ‘signposting’ role, instead relying on the third sector to link employers to information. Information provision has become ‘disparate and not connected’ (NGO 4). To address this, respondents proposed the creation of an (online) ‘one-stop-shop’ to signpost relevant information about managing an ageing workforce (Government 2, NGO 4) and improve the accessibility and coherence of information provision. This challenge has been taken up by the Age Action Alliance (discussed further below), which launched a website on 23 May 2013 with thematic resource pages, including on ‘healthy workplaces’ and ‘attitudes to ageing’.35 The website is intended to create a central, highly visible and easily searchable repository of information. However, it is limited to specific topics, and resources have been compiled based on submissions from organisations, limiting its comprehensiveness. Even if the right information is available, the experts recognised that ‘you have to make sure that people actually get it and read it, which is quite challenging’ (NGO 5). Written guidance may therefore be insufficient for promoting improved practice. Instead, one expert felt employers should be given opportunities to work through issues collaboratively: ‘it’s the issue of people talking to each other, networking and so on; it helps to create their confidence in responding to these issues which are different’ (NGO 4). Dialogue between workplaces provides employers with clarification and support in working with an ageing workforce. The expert believed that employers should be given facilitated opportunities to network, discuss issues around ageing and share best practice (NGO 4). According to another expert, this is already evolving organically as employers and businesses form networks to facilitate discussion (Government 3). The government is also supporting the Age Action Alliance, an independent partnership of over 85636 public, private and voluntary organisations ‘working together [to] combine our thinking, experience and diverse skills to improve the 34 

McAndrew (n 19 above) 17. See: ageactionalliance.org/. 36  As at 11 January 2016. 35 

100 

Expert Views on Age Discrimination Laws

lives of older people’.37 The Alliance has the potential to provide facilitated networking ­opportunities for employers. In addition to the experts’ suggestions, businesses, especially small and medium enterprises (SMEs), could be given access to ‘trusted advisers able to deliver intelligent, relevant, simple and objective interpretation of the law specifically for their organisation and their issue … [pitched] somewhere between the overly complex legal documentation and the overly simplistic generic guides’.38 This would help marshal existing information to transfer ‘information and guidance … into useful knowledge’.39 A starting point for this service is the Acas Helpline, which offers free telephone advice and guidance to employers.

(ii)  Operating Without a DRA Abolishing the DRA has revealed broader limitations in employers’ HR capacities. It was evident from the interviews that many employers had come to rely on the DRA ‘in lieu of performance management’ (Government 2; similarly, Academic 1, 3, Government 3, NGO 5): ‘it’s easier to tolerate poor performance when you know there’s an end date’ (NGO 2). One expert saw some UK employers as struggling to implement effective performance management systems for their whole workforce (NGO 2). Further, while meaningful conversations between management and employees are essential for good HR practice (Academic 3, Government 2, 4, NGO 6), removal of the DRA may, according to some respondents, ‘lead to employers and employees just not talking about retirement plans [as] there would be too much risk involved in having such a conversation’ (Government 3). While there is ‘nothing against an employer discussing an employee’s future plans with them’ (Government 3), employers are not necessarily ‘clear and free to have those discussions in the workplace’ under the existing legal framework (NGO 6), and a fear of litigation is encouraging HR processes to become increasingly formalised and risk averse (Government 4). Conversely, it was also argued that while the removal of the DRA may prove ‘exquisitely challenging’ for employers (Government 4), it might also compel organisations to improve their HR practices more broadly: It’s all rooted in the good management of people generally, so, if somebody isn’t good at managing people generally they could maybe learn how to get in there, thin end of the wedge, by looking at some of these things through the … older age lens; then we can spotlight weaknesses and then begin to change their general responses. (NGO 4)

One expert believed that employers now have a (short) window to improve their HR practices, as they are unlikely to experience a surge of employees who are reluctant to retire immediately: employees coming up to retirement are likely to 37 Age Action Alliance, ‘Age Action Alliance—Improving Older People’s Lives … Together’ (no date): ageactionalliance.org/. 38  McAndrew (n 19 above) 17. 39 ibid.

Expert Views on the Efficacy of Law

 101

have had their retirement plans in place (or at least to have anticipated retiring) for a significant period of time (Government 4).40 Thus, the removal of the DRA may provide the opportunity and the impetus for employers to improve their HR practices. According to the respondents, conversations should be occurring between employees and line managers to discuss future career plans, work preferences, development and health needs and other concerns (Government 2, NGO 4, 6). They also noted that conversations should occur consistently across the workforce, ensuring equity across different age groups and addressing the needs of each employee at their individual career stage (Government 4). While formal conversations may occur once a year, good practice recommendations encourage employers to ‘tune in’ to employees far more frequently (NGO 4). Rather than employees adapting to employers’ needs, a more mature approach to HR management regards the relationship as one of ‘mutuality’, where employers also cater to employees’ needs (NGO 4).41 The mature employer asks: ‘how can I deal with people in the right kind of way so they perform better for me, they enjoy their job more than they otherwise would and we are all better off?’ (NGO 4; similarly, Government 3). To achieve this, the respondents recognised the need to build capacity within workplaces (Government 3, NGO 6), particularly for line managers. It is unclear who should take the lead in improving organisational capacity: respondents variously suggested that Acas (Government 3) and the employee relations community provide good practice guidance; that employers take part in leadership and management training; and that government signpost SMEs to quality leadership and management training (NGO 6). However, a government expert felt it was unlikely that there would be ‘a concerted governmental push’ to improve employer capacity (Government 3). Alternatively, Hepple argues that the EHRC should take a primary role in developing the capabilities of those it regulates (that is, ­employers).42 However, the capacity of the EHRC to fulfil this role is ‘seriously threatened’ by changes to the body’s legislative powers and drastic budget cuts.43 Therefore, it is unsurprising that respondents did not identify the EHRC as a key player in building employer capacity.

(iii)  Need for a National Dialogue Experts also identified the need for a social dialogue regarding what is acceptable age discrimination, what rights and obligations should be provided at different 40  Employees resisting retirement may not be an issue for employers at all: ‘people do know when it’s time to go’ (Government 1). As one expert noted: ‘there’s this sort of assumption that [employers] are going to be facing huge issues with capacity of older workers and having to force them out and it will be very embarrassing and undignified and I think experience shows that this is actually not borne out’ (Government 1). 41  This reciprocity is reflected in the notion of ‘work ability’, discussed in Chapter 6 of this volume. 42  B Hepple, ‘Enforcing Equality Law: Two Steps Forward and Two Steps Backwards for Reflexive Regulation’ (2011) 40 Industrial Law Journal 315, 334. 43  ibid 334.

102 

Expert Views on Age Discrimination Laws

ages and what exceptions should be allowed to age discrimination legislation. As one expert noted: there are differences in the way we treat people [on the grounds of age] … that we justify as being reasonable, proportionate and indeed what society wants us to do. Because I think social drivers like that would be the only way it would happen but we need to have that sort of, almost public debate to identify them and decide where the line lies. (Government 4)

Without this dialogue, ‘you get muddled policy’ as assumptions and values around age remain ‘hidden and unanswered’, creating further uncertainty in the minds of all concerned (Government 4). The experts therefore called for a broader social debate on what relevance (if any) age should have for social and employment rights and what benefits should accrue with age: Let’s start deciding, let’s start getting some social discourse on what it is likely to be in the future. (Government 4; similarly, Government 1) Okay, if you think it’s necessary to [discriminate], spell it out then we can debate it. (Academic 1)

This is consistent with prevailing arguments in the literature. In the context of pension reform, Walker identifies the need for ‘open public debate about the proposed direction of change’ which is ‘completely absent’ in the UK.44 This arguably extends to other areas of ageing policy. Sarfati has also called for: a broad and informed public debate to clarify the policy options … to achieve agreement among the different stakeholders about the kind of sustainable and socially acceptable reforms that correspond to the national context of the different countries [in relation to demographic ageing].45

According to Sarfati, countries that have debated legal and social reform publicly are more likely to adopt and carry out change successfully.46 Dialogue can help negotiate and accommodate the needs of different stakeholders, and promotes intergenerational and intra-generational solidarity.47 Thus, the experts’ proposed dialogue around ageing has significant merit.

(iv)  Improving Attitudes towards Older Workers As noted above, the experts saw attitudinal change as a key priority for addressing age discrimination. For some, employers’ attitudes towards older workers could be improved through a ‘heart and minds approach’ (NGO 2) where employers are persuaded to employ older workers through promotion of the business case for 44  A Walker, ‘Ageing in Europe: Policies in Harmony or Discord?’ (2002) 31 International Journal of Epidemiology 758, 760. 45 H Sarfati, ‘Social Dialogue: A Potential “Highroad” to Policies Addressing Ageing in the EU ­Member States’ (2006) 59 International Social Security Review 49, 50. 46 ibid. 47  ibid 53.

Expert Views on the Efficacy of Law

 103

diversity (Academic 3, Government 2, NGO 5, 7). This is consistent with regulatory theory, which suggests that ‘harnessing the enlightened self-interest of the private sector’ is often more effective at achieving change than command and control regulation.48 However, it was noted that ‘selling the message [of age equality] is quite difficult’ given the need to tailor messages to many different organisations and individuals (NGO 2, 5). Further, attitudes towards older workers were perceived as being ‘quite impressionistic’: ‘everybody will have a story’ about their experience of older workers, ‘and that’s the difficult thing to overcome’ (NGO 2). Changing attitudes, it was suggested, will also require addressing the existing retirement culture that ‘is still ingrained in people’ (Academic 1; similarly, Government 2). To improve attitudes towards older workers, respondents suggested the adoption of more positive and active depictions of the elderly in the media (Government 2), and the introduction of employer education programmes to dispel negative stereotypes about older workers (NGO 9). The respondents’ suggestions are partly encompassed by UK government initiatives: the Age Positive campaign launched in 2000 was intended to ‘tackle ageism in the workplace’ by raising awareness of the ‘business benefits of recruiting and retaining workers aged over 50’ and promoting good employer practice.49 However, Age Positive has now been scaled back, despite the government recognising that ‘[there] is still more to be done: employer attitudes to older people still need work’.50 Two other programmes continue to try to address societal attitudes towards older people: first, the Everyday Ageism Project aims to raise awareness of ageism by encouraging people to share their experiences of ageism online;51 and, secondly, the Age Action Alliance Working Group on Attitudes to Ageing is working to improve media attitudes to and depictions of older people. Despite these measures, the UK Business Champion for Older Workers has identified negative attitudes towards older workers as a major barrier to extending working lives.52

48  N Gunningham, ‘Introduction’ in N Gunningham and PN Grabosky (eds), Smart Regulation: Designing Environmental Policy, Oxford Socio-Legal Studies (Oxford, Clarendon Press, 1998) 12. ­However, self-regulation via the business case is unlikely to be sufficient for promoting structural change: SM Hart, ‘Self-Regulation, Corporate Social Responsibility, and the Business Case: Do They Work in Achieving Workplace Equality and Safety?’ (2010) 92 Journal of Business Ethics 585. See further Chapter 7 of this volume. 49  DWP, ‘Building a Society for All Ages’ (Cm 7655, 2009) July 2009, 31. 50  HC Deb 28 June 2012, vol 547, col 553. That said, the campaign ‘re-emerged as though it was here all along’ in a reduced form in June 2014: TAEN, ‘New “Fuller Working Lives Strategy” Revealed as Promising in Parts’ (TAEN Blog, 16 June 2014): taen.org.uk/blog/view/139. TAEN describes this as ‘an example of the “flavour of the month” approaches that have so often been followed’ by the UK ­government: ibid. See: www.gov.uk/government/collections/age-positive/. 51  See: www.everydayageism.blogspot.co.uk/. 52  R Altmann, ‘A New Vision for Older Workers: Retain, Retrain, Recruit: Report to Government by Dr Ros Altmann CBE Business Champion for Older Workers’ (March 2015) 23–25.

104 

Expert Views on Age Discrimination Laws

(v)  Promoting a Holistic Approach A number of respondents identified a need for government and employers to adopt a more holistic, ‘joined-up’ approach to their support of older workers. While the experts noted the introduction of health and wellbeing programmes and career reviews by some employers, ‘very few [employers] think about these things in the joined up holistic way that the best employers that we’ve seen in Europe are increasingly doing’ (NGO 3). They felt that the initiatives ‘lack the coherence’ demonstrated by the best companies globally (NGO 3). According to the experts, this lack of strategy is attributable to a failure to articulate and communicate the importance of age diversity and older workers to organisational leaders and ‘top teams’ (NGO 4). They therefore identified a need for dialogue within workplaces to identify what (if any) opportunities, standards and benefits should be age dependent: ‘being open about it at least starts moving the debate on’ and allows questions of fairness to be debated publicly (Government 4). This, then, would be complementary to the national dialogue proposed by the experts.

D.  Responsibility for Reform As noted above, there was a general consensus among the experts that law should play a minimal, standard-setting role. However, the role of government might be broader than just setting standards through legislation. Respondents believed government should play a leading role in: training employers to overcome ageist attitudes (NGO 1); encouraging good practice among employers (Academic 3); encouraging broader attitudinal change, including by producing materials and developing information campaigns (NGO 5; similarly, NGO 7); investing in midand late-career worker development (Academic 3); and providing support to employers to improve occupational health (NGO 8). While the experts felt some of these functions could be performed by third sector organisations, they thought the third sector would struggle to reach ‘the organisations that probably need the most support’ (Academic 3). Further, the experts noted that relying on businesses to address their own training needs would entail significant costs and fewer economies of scale (NGO 1). Therefore, the interviews indicate that government action is instrumental for achieving meaningful change for older workers. Indeed, one respondent noted the need for strong leadership from government to change employer attitudes and encourage organisations to adopt good practice (NGO 3): The government has an important role to play: government can be a very powerful catalyst of change, government can do things that, you know, can have an impact which in itself can then be magnified. Government can provide incentives, government can hold up companies to be exemplars, they can support employee awards or employer awards, you know, they can provide pots of funding that organisations can tap into sometimes and that can leverage other funds to become available and so forth and could be a whole range of things that government could do but … it has to be government that takes the lead on this really. (NGO 3)

Analysis

 105

However, government respondents were unwilling to attribute such a broad role to government. Rather, they emphasised the role of business in leading activities to spread best practice, with some facilitation from government (Government 3). While they thought the government might be responsible for ‘setting the themes’ and direction of public debate and reform ‘by removing or putting in legislation to facilitate it’ (Government 2), they did not feel it should be responsible for effecting or implementing change by itself: The role of the government [is] to just … set … where they think we should be going and why we need to do it but really we are working with others to make that change because on its own, it’s not going to do it on its own. (Government 2)

Thus, according to government respondents, the role of government is ‘facilitating, enabling, setting out what the end issues are, but actually working really closely with organisations who can influence employers and employers themselves, to generate that sort of culture change for themselves’ (Government 2). Other experts also noted that a broader role for government, however desirable, is unlikely in an era of government austerity and budget cuts (NGO 5; similarly, Government 3).53 Rather than relying on government, experts felt that employers could play a key role in addressing the ageing workforce, including by ‘maintaining their older workers’ (Government 2); thinking strategically to ‘use their ageing workforce in a way that helps their business’, including by ‘up-skill[ing] younger workers’ and introducing mentoring programmes (Government 2); ‘thinking about the biggest pool of talent that they can draw’ on in recruitment (Government 2); forming networks to share best practice (Government 3); training middle managers to have the skills to implement age diversity policies effectively (Government 2; similarly, NGO 5); training all employees on age diversity issues (NGO 9); investing in midand late-career development for workers (Academic 3); and providing and cultivating strong organisational leadership around age diversity (NGO 4). The experts viewed employers and government as occupying complementary and mutually important roles in addressing ageing.

II. Analysis The results of the interviews provide strong support for the findings of the doctrinal analysis presented in Chapter three. Where respondents supported legal change, this was entirely consistent with the arguments and findings from that study. However, the interviews also demonstrate that the majority of experts did 53  However, this is part of a broader historical deregulatory trend under conservative governments: ACL Davies, Perspectives on Labour Law, Law in Context, 2nd edn (Cambridge, Cambridge University Press, 2009) 13. Therefore, a broader role for government may be constrained by broader trends and changing norms in the political landscape, in addition to austerity measures.

106 

Expert Views on Age Discrimination Laws

not desire legal change, and appeared to be broadly satisfied with the legislative provisions in place under the EqA. While this may indicate that the law is fit for purpose, it may also reflect a broader ‘change fatigue’ among the respondents.54 Exhaustion can occur when individuals feel change is occurring too frequently or the pace of change is too fast.55 The legal landscape for older workers has changed significantly in the last decade, with major regulatory reforms being introduced in 2006, 2010 and 2011. It is possible that this change has been too rapid for many respondents, reducing any appetite for further legal change, even if it is theoretically desirable. This is supported by the results of the interviews. Most experts felt that the implementation of UK age discrimination laws had substantial limitations. According to the experts, law is having limited success at improving employer attitudes towards older workers, and is not encouraging proactive employer responses to an ageing workforce. Therefore, as noted by the experts, the practical efficacy of age discrimination laws remains doubtful, particularly in effecting attitudinal change. Change fatigue may explain why, in the face of these significant limitations, the expert respondents did not feel legal change was desirable or necessary. Alternatively, these findings may reflect the respondents’ limited conception of the role of law and legal regulation, seeing it as merely a tool for setting legal minima. If law plays a minimal part in addressing age discrimination, then legal reform will have limited significance in practice, and need not be regarded as a high priority. These findings challenge lawyers’ centralised and narrow focus on legal regulation as a tool for achieving social change, and reemphasise the importance of ‘soft’ law and ‘non-law’ for achieving broader social ends. The interviews also raise complex questions about ‘legal compliance’ and the nature of the ‘compliance approach’ that is being adopted by employers in relation to age discrimination. From the experts’ responses, it appears they conceive of a ‘compliance approach’ as being the bare minimum required to satisfy legal requirements. In the case of age discrimination laws, this may entail not openly discriminating on the basis of age, but would not require any further action to support, recruit or retain older workers. This is an approach that is consistent with the letter (but not the spirit) of the law (see further Chapter seven). However, in contrast to the views implicit in the interviews, legal ‘compliance’ is not a singular or simple concept or event.56 Compliance involves a ‘complicated process of adaptation, flexibility, reflection, and, above all, interpretation’,57 and an ‘extended and endless negotiation’.58 Ultimately, what compliance entails is ‘what 54 

Bernerth et al (n 5 above) 322. Huy, ‘Time, Temporal Capability, and Planned Change’ (2001) 26 The Academy of ­Management Review 601; Bernerth et al (n 5 above) 323, 332. 56 BM Hutter, Compliance: Regulation and Environment, Oxford Socio-Legal Studies (Oxford, Clarendon Press, 1997) 18. 57  ibid 3. 58  PK Manning, ‘Book Review: To Punish or Persuade’ (1988) 28 British Journal of Criminology 559, 561. 55 QN

Analysis

 107

agencies involved define it to be’59 via a process of negotiation and interaction with those they regulate.60 This raises two questions. First, what is the definition or process of ‘compliance’ in the context of age discrimination laws (and, relatedly, who defines what compliance entails)? Secondly, given this definition, what does it mean if employers are adopting a compliance-focused approach to law? While UK legislation has outlawed age discrimination in most circumstances, the law does little to define compliance.61 In the face of this legal uncertainty, the interpretation of ‘compliance’ has been left largely to organisations themselves and, in larger organisations, personnel managers.62 Therefore, there is scope for variation in how organisations achieve compliance.63 More particularly, employers may achieve compliance proactively (by making structural adjustments for groups of workers or to organisational structures) or reactively (by making individual adjustments as required).64 That said, Dickens and Hall argue that employers often respond reactively to law, rather than endeavouring to make structural changes to their processes and procedures.65 ‘Compliance’ with UK age discrimination laws is therefore characterised by a process of negotiation and interaction between employers, the EHRC, unions and the courts. However, as noted in Chapter three, the scope for the EHRC to monitor and interact with individual employers is limited, particularly in the face of dramatic budget cuts. Further, the minimal union presence in UK workplaces (and private workplaces in particular) has reduced the capacity for unions to exert influence on employers in this area.66 In 2013, only 25.6 per cent of UK employees were members of a trade union, down 6.8 per cent since 1995.67 Therefore, the definition of ‘compliance’ in relation to age discrimination laws is now left largely to employers, with some limited oversight by ETs and courts where claims actually progress to a hearing. In leaving the definition of ‘compliance’ to those being regulated, there is a risk that measures taken by employers to demonstrate ‘compliance’ and satisfy legal scrutiny will become divorced from the actual problem of age discrimination. In relation to sexual harassment law in the USA, Grossman identifies a ‘misguided culture of compliance’, where liability is measured by whether employers ‘comply 59 ibid. 60 

Hutter (n 56 above) 13. See similarly F Dobbin, Inventing Equal Opportunity (Princeton NJ, Princeton University Press, 2011) 3; see also Chapter 2. 62  ibid 19; see further Chapter 7. In contrast, it has also been argued that a ‘compliance’ approach is often the result of excessive regulation, which generates an employer ‘backlash’: A Klarsfeld et al, ‘Social Regulation and Diversity Management: A Comparative Study of France, Canada and the UK’ (2012) 18 European Journal of Industrial Relations 309, 310. 63  Hutter (n 56 above) 18. 64  Dickens and Hall (n 10 above) 348. 65 ibid. See also L Dickens and M Hall, ‘Legal Regulation and the Changing Workplace’ in W Brown et al (eds), The Evolution of the Modern Workplace (Cambridge, Cambridge University Press, 2009) 349–50. 66  See further Dickens and Hall (n 10 above) 350–51. 67  BIS, ‘Trade Union Membership 2013’ (Statistical Bulletin, May 2014) 5. 61 

108 

Expert Views on Age Discrimination Laws

with judicially created prophylactic rules’, rather than whether they successfully prevent harassment.68 In this context, ‘employers could conceivably insulate themselves from liability entirely without making a dent in the underlying problem’.69 This is consistent with the UK experience: ETs are more likely to look at whether employers have followed procedural rules that embody ‘equal treatment’ than whether they have addressed the disadvantages that older workers face or actually promoted age equality. Therefore, allowing employers to define ‘compliance’ may undermine the law’s instrumental and intrinsic objectives in practice. However, this may also reflect weaknesses in the law itself, which focuses on preventing discrimination rather than addressing inequality.70 It is possible to encourage employers to move beyond a compliance approach. According to Kelman, changes in attitudes and actions due to social influences may occur at three ‘levels’: compliance, where individuals accept social influence because they ‘[expect] to gain specific rewards or approval and avoid specific punishments or disapproval by conforming’, though they may not believe in the behaviour’s content; identification, where individuals accept social influence because they want ‘to establish or maintain a satisfying self-defining relationship to another person or a group’; and internalisation, where individuals accept social influence because the ‘induced behavior … is intrinsically rewarding’ and ‘congruent with [their] value system’.71 A compliance approach is likely to lead to behaviour ‘only under conditions of surveillance by the influencing agent’, an identification approach ‘only under conditions of salience of his relationship to the agent’, and an internalisation approach ‘under conditions of relevance of the issue, regardless of surveillance or salience’.72 Similarly to Kelman’s internalisation level, Barnes and Burke have examined the role of commitment, or the degree to which organisations embrace law’s ‘underlying social goals’, in encouraging ‘“beyond-compliance” behaviours’.73 More recent research has complicated these categorisations. Even a compliance approach may be driven by both affirmative motivations (arising from good intentions) and negative motivations (arising from a fear of consequences).74 Negative motivations are influenced by surveillance and inspection practices, whereas positive motivations are influenced by attitudes and beliefs and knowledge of legal rules.75 Therefore, if UK employers are adopting a c­ ompliance approach to e­ quality 68  JL Grossman, ‘The Culture of Compliance: The Final Triumph of Form over Substance in Sexual Harassment Law’ (2003) 26 Harvard Women’s Law Journal 3, 3. 69 ibid. 70  See Dickens and Hall (n 65 above) 349 and Chapter 3 of this volume. 71 HC Kelman, ‘Compliance, Identification, and Internalization: Three Processes of Attitude Change’ (1958) 2 Journal of Conflict Resolution 51, 52–53. 72  ibid 54. 73 J Barnes and TF Burke, ‘Making Way: Legal Mobilization, Organizational Response, and Wheelchair Access’ (2012) 46 Law & Society Review 167, 172. 74 PJ May, ‘Compliance Motivations: Affirmative and Negative Bases’ (2004) 38 Law & Society Review 41, 61. 75 ibid.

Conclusion

 109

legislation, change is likely to require both stronger surveillance and supervision to secure proper observance of legislative ­requirements,76 and a broader emphasis on attitudinal change and promoting knowledge of the law. That said, if employers can be encouraged to integrate age equality as an important factor in their relationships with others, or to adjust their value systems to make age equality intrinsically rewarding, this may encourage a response beyond ‘compliance’. How this might be achieved is examined in Chapter seven.

III. Conclusion While the majority of experts held positive views of UK age discrimination laws, implementation of the law is significantly constrained. Echoing the discussion in Chapter three, some experts noted significant legal uncertainty around positive action and EJRAs, and voiced concern that exceptions to age equality rules are undermining the EqA. Legal uncertainty, and a consequent fear of legal action, are encouraging a defensive, compliance-focused implementation of age discrimination laws in UK workplaces. However, it remains unclear how widespread this compliance-focused approach is among employers, or, alternatively, which employers are adopting proactive measures to address ageing. This is taken up in Chapter five, which deploys statistical analysis to evaluate prevailing organisational practices in relation to age. While there is limited evidence of proactive practice by UK employers, the experts did not feel it was the role of law to secure good practice. Further, government and non-government respondents were divided on whether the government could and should occupy a broader, more proactive role in addressing the ageing workforce, beyond standard-setting in legislation. The experts’ views of the role of law are consistent with a laissez-faire, ‘hands-off ’ approach to government and a minimalist view of legal regulation. Given that employers are not independently engaging with the needs of an ageing workforce, a laissez-faire approach is no longer adequate for achieving meaningful change. This chapter has identified a series of reforms that may help to improve the effectiveness of UK age discrimination laws, which will be investigated further using the Delphi method in Chapter eight. However, beyond these reforms, the potential for a broader and more ‘proactive’ role for government and legislation is worth considering. A comparative study of a proactive governmental approach to the ageing workforce is presented in Chapter six, where the case of Finland illustrates how financial incentives, collective negotiation and positive duties may be used to address age discrimination in employment. A proactive governmental approach, and broader view of ‘equality’, may remedy a number of the deficiencies of UK legislation. 76  Including, potentially, through peer review processes: see the discussion of experimentalism in Chapter 2 of this volume.

5 Generalising the Qualitative Findings—Quantitative Analysis of WERS6 The qualitative expert interviews reported in Chapter four suggest that UK age discrimination laws are failing to encourage proactive employer responses to an ageing workforce. Further, the implementation of age discrimination legislation appears to vary considerably between employers, though it is unclear whether large or small employers are more likely to adopt good practices. This chapter therefore focuses on these differences, with the goal of exploring the following questions: —— Are employers adopting a compliance-focused approach to the implementation of age discrimination laws? —— Does this approach vary across different types of employers? —— Is it relevant whether an employer is large or small; in the public or ­private sector; or listed on a stock exchange? —— Have organisational approaches changed since the introduction of age ­discrimination legislation? Drawing on cross-sectional data from the sixth Workplace Employment Relations Study (WERS6),1 the flagship national survey of employment relations in Great Britain at the workplace level, I use quantitative analysis to triangulate and corroborate the doctrinal and qualitative findings reported in earlier chapters. The results illustrate the extent to which UK organisations are aware of and are ­accommodating workers of different ages (‘age-aware’). Corroborating the

1  Department for Business, Innovation and Skills, Advisory, Conciliation and Arbitration Service and National Institute of Economic and Social Research, Workplace Employee Relations Study, 2011 [computer file], 3rd edn (Colchester, UK Data Archive [distributor], January 2014). SN: 7226. [Crown copyright held jointly with the Economic and Social Research Council, the UK Commission for Employment and Skills, the Advisory, Conciliation and Arbitration Service and the National Institute of Economic and Social Research. Crown copyright material is reproduced with the permission of the Controller of HMSO and the Queen’s Printer for Scotland. The original data creators, depositors or copyright holders, the funders of the Data Collections (if different) and the UK Data Archive bear no responsibility for the data’s further analysis or interpretation].

The WERS6 Data Set

 111

findings in Chapter four, the data show that employers are adopting a largely ­compliance-focused approach to implementing age discrimination laws, with limited evidence of proactive practice in workplaces to support older workers. Going beyond the findings in Chapter four, statistical analysis demonstrates that age-awareness is associated with large and public organisations and those with a union presence. Using two-wave panel data2 from WERS6, I also examine changes in organisational approaches to age over time, comparing organisational practices in 2004 (prior to the introduction of age discrimination legislation) with those in 2011. The data suggest that more employers are adopting age-aware practices over time (and following the implementation of age discrimination legislation). However, the adoption of age-aware practices is still dependent on organisations’ previous levels of age-aware practices. Further, large and public sector organisations and those with a union presence are still more likely to adopt age-aware practices. This indicates that age discrimination legislation is having limited impact on organisational practices in some types of organisation.

I.  The WERS6 Data Set Fieldwork for WERS6 was undertaken by NatCen Social Research between March 2011 and June 2012 across 2,680 workplaces in Great Britain with five or more employees, drawn from an overall population of around 750,000 workplaces.3 The panel data set was created through repeat interviews with 989 workplaces previously surveyed in WERS5 in 2004. While WERS6 covers a wide range of issues, this study focuses on the relationship between organisational characteristics and age-aware practices, and the evolution of age-aware practices over time, as evidenced by the WERS6 management questionnaire. Data in the management questionnaire were collected from each workplace via face-to-face interviews with the most senior manager responsible for employment relations, human resources or personnel. The management questionnaire contained seven questions relating to age-aware practices. In particular, respondents were asked to indicate whether they: 1. Had a formal written equal opportunity policy that explicitly mentioned age (‘formal policy’); 2  That is, measurements taken at two time points. Two-wave panel data cannot show the process or trajectory of change, and risks confounding true change with measurement error. It is also cannot accommodate elaborate statistical models or be used to model change over time. However, despite these limitations, two-wave panel data analysis is still more informative than using cross-sectional data alone. 3  B van Wanrooy et al, Employment Relations in the Shadow of Recession: Findings from the 2011 Workplace Employment Relations Study (Basingstoke, Palgrave Macmillan, 2013) 5–6. The sample excluded workplaces in agriculture, forestry, fishing, and mining and quarrying: see further ibid 5.

112 

Quantitative Analysis of WERS6

2. Had special procedures to encourage applications from older workers (‘encourage applications’); 3. Monitored recruitment and selection by age (‘monitor recruitment’); 4. Reviewed recruitment and selection procedures to identify indirect ­discrimination by age (‘review recruitment procedures’); 5. Monitored promotions by age (‘monitor promotions’); 6. Reviewed promotion procedures to identify indirect discrimination by age (‘review promotion procedures’); and 7. Reviewed relative pay rates by age (‘review pay rates’).4 The questionnaire therefore elicits evidence of whether organisations are adopting proactive measures to prevent and address age discrimination, above and beyond strict compliance with legislation. The questions also probe beyond whether organisations have adopted an age discrimination policy, which may just be a smokescreen or ‘empty shell’ for hiding poor organisational practice.5 By examining organisations’ policies and monitoring activities, the questions consider whether organisations are both professionalising equality measures (by adopting written procedures and policies) and routinising equality (by integrating practices into daily planning routines and activities).6 Taken collectively, the questions represent a holistic and fairly complete picture of organisational age-aware practices,7 and indicate which organisations are adopting a holistic approach to ageing. This reflects the concerns of the expert respondents in Chapter four, who identified a need for employers to adopt a more holistic, joined-up approach to their support of older workers.

4  These questions do not capture information regarding the DRA, the adoption of EJRAs in workplaces, or more specific information regarding the treatment of older workers. WERS was not designed with the EJRA or age discrimination in mind. As noted in Chapter 2 of this volume, this is one of the key challenges with research using government data sets: the topics selected and questions posed are those of interest, and acceptable, to government officials, potentially omitting key variables. It is clear that the government is aware of this limitation of the WERS data set: when reviewing the WERS survey instruments for use in WERS6, it was suggested during the consultation process that further questions could be introduced regarding older workers, particularly given that WERS already incorporated some questions on the recruitment of older workers and on monitoring recruitment, promotion and pay by age: BIS, ‘Report on the Consultation for the 6th Workplace Employment Relations Survey (WERS6)’ (August 2010) 14. However, it was ultimately decided that assessing the contribution of older workers and how working arrangements could be changed to keep older workers at work would entail additional questions which were not regarded as a high priority for WERS: ibid 15. Given that WERS is designed to provide a longitudinal data set for mapping the changing workplace over time, recent concerns regarding mandatory retirement fall less easily within its remit. 5  K Hoque and M Noon, ‘Equal Opportunities Policy and Practice in Britain: Evaluating the “Empty Shell” Hypothesis’ (2004) 18 Work, Employment & Society 481. 6  J Barnes and TF Burke, ‘Making Way: Legal Mobilization, Organizational Response, and Wheelchair Access’ (2012) 46 Law & Society Review 167, 174–75. 7  However, the absence of formal practices does not necessarily imply poor practice: H Barnes et al, ‘An Ageing Workforce: The Employer’s Perspective’, Report No 468 (Brighton, Institute for Employment Studies, 2009) vii. SMEs are more likely to adopt an informal approach to their workforce, rather than adopting formal measures. Therefore, the WERS questions may disproportionately favour larger organisations.

‘Compliance’ Approaches in UK Workplaces

 113

To obtain a holistic picture of organisational practice, I used a composite ‘score’ of organisational age-aware practices, created by recoding the presence of an age-aware practice with ‘1’ (or non-presence with ‘0’), and generating an overall ‘score’ with the sum of all recoded variables.8 This score is used as the dependent variable ‘age-aware practice score’ (AAPS), which ranges between 0, indicating no age-aware practices, and 7, indicating that all surveyed age-aware practices are in place. Composite scores allow researchers to distinguish between holistic and piecemeal organisational approaches9 and are more reliable than a single measure in evaluating organisational practices. Therefore, a composite score is well suited to answering the questions posed in this chapter. The AAPS variable may be described in two ways: (1) as count data or (2) as an ordinal scale*.10,11 To accommodate this ambiguity, this chapter presents the results of both count and ordinal analyses.12

II.  ‘Compliance’ Approaches in UK Workplaces The expert interviews reported in Chapter four noted the tendency of some UK employers to adopt a defensive, compliance-focused approach to the implementation of age discrimination laws. However, it remains unclear how widespread this compliance-focused approach is among employers, and how organisations are defining ‘compliance’ in practice. This can be explored further using the WERS data set. Figure 6 illustrates the spread of AAPS across the workplaces surveyed in WERS6. Of the 2,680 workplaces surveyed, only 25 received an AAPS of 7 out of 7 (1.0 per cent) and 484 workplaces (19.1 per cent) received an AAPS of 0. The median AAPS was 1, with a mean of 1.96, and 79.8 per cent of workplaces received 8  Detailed information regarding the recoding of the data set is available in A Blackham, ‘Extending Working Life for Older Workers: An Empirical Legal Analysis of Age Discrimination Laws in the UK’ (PhD Thesis, University of Cambridge, 2014). 9  Barnes et al (n 7 above) 80. 10  Terms identified with an asterisk are defined in the glossary of statistical terms and tests in Appendix 1 to this volume. 11  JS Long, Regression Models for Categorical and Limited Dependent Variables, Advanced Quantitative Techniques in the Social Sciences (Thousand Oaks CA, Sage Publications, 1997) 114; AC Cameron and PK Trivedi, Regression Analysis of Count Data, Econometric Society Monographs (Cambridge, Cambridge University Press, 1998) 9. 12  When seeking to make any inference regarding Great Britain, data used in the models that f­ ollow have (where possible) been weighted before analysis to make the sample properly representative of the population: see J Chaplin et al, ‘The Workplace Employment Relations Survey (WERS) 2004 Technical Report (Cross-Section and Panel Surveys)’ (November 2005) 1. This study uses the standard establishment weight and panel weight provided in the WERS6 data set, which can be used to produce workplace-level estimates from the management questionnaire. This corrects the sample bias in the survey, which over-represents large workplaces and certain industries: A Bryson, ‘Unions and Employment Growth in British Workplaces during the 1990s: A Panel Analysis’ (2004) 51 Scottish Journal of Political Economy 477, 484. However, for some statistical models, the weights prevent effective analysis. Therefore, this study explicitly notes where weights have or have not been used.

114 

Quantitative Analysis of WERS6

40.0%

Per cent of workplaces

30.0%

20.0%

10.0%

0.0% 0

1

2

3

4

5

6

7

AAPS

Figure 6:  Distribution of AAPS across Surveyed Workplaces, WERS6

an AAPS between 0 and 3. Thus, few workplaces have adopted holistic policies and practices that assist and encourage age diversity. While the composite AAPS variable is useful for obtaining a holistic picture of workplace age-aware practices, it is also informative to break down the variable and consider the adoption of each individual practice. Figure 7 depicts the percentage of respondents adopting each individual age-aware measure. The figure indicates that the adoption of a formal policy (by 77.7 per cent of workplaces) is far more common than the implementation of monitoring or review procedures (by between 35.5 per cent and 12.2 per cent of workplaces, depending on the measure). Further, adopting special procedures to encourage applications from older workers is uncommon, being undertaken by only 5.1 per cent of respondent workplaces. These statistics imply that organisations are adopting symbolic responses to age discrimination laws (in the form of formal equal opportunity policies) that ‘create a visible commitment to law’ but may have limited practical impact.13

13 LB Edelman, ‘Legal Ambiguity and Symbolic Structures: Organizational Mediation of Civil Rights Law’ (1992) 97 American Journal of Sociology 1531, 1542; see also L Dickens and M Hall, ‘Legal Regulation and the Changing Workplace’ in W Brown et al (eds), The Evolution of the Modern Workplace (Cambridge, Cambridge University Press, 2009) 349.

Age-Aware Workplaces

 115

100 90 80

77.7

70 60 50 35.5

40

31.8

30 17.3

20 10

19.2

5.1

12.2

0

Figure 7:  Per cent of Workplaces with Individual Age-Aware Practices, WERS6

This is consistent with the experts’ view in Chapter four that organisations are adopting a compliance-focused approach to age discrimination legislation, with few m ­ eaningful age-aware practices being implemented.

III.  Age-Aware Workplaces While most workplaces are adopting a compliance approach to the support of older workers, it is worth considering which (if any) employers are adopting ­proactive measures to address demographic ageing.

A.  Theoretical Considerations and Past Research Institutional theory posits that workplaces adopt equal opportunity practices as a means of ‘conforming to normative pressures from the wider society’.14 14  A Felstead et al, ‘Opportunities to Work at Home in the Context of Work-Life Balance’ (2002) 12 Human Resource Management Journal 54, 57; SJ Wood et al, ‘Family-Friendly Management in Great Britain: Testing Various Perspectives’ (2003) 42 Industrial Relations 221, 223.

116 

Quantitative Analysis of WERS6

Workplace­practice varies according to the need to maintain social legitimacy: the most visible organisations (large organisations, public sector organisations, and those with a union presence) will be compelled to conform to societal pressure by adopting equal opportunity practices.15 Therefore, we can predict that large organisations, those in the public sector, and those with a union presence will be more likely to have a high AAPS. The expert interviews reported in Chapter four also identified organisational size as influencing age-aware practices. Consistent with this prediction, previous studies of WERS data have found strong statistical relationships between the size of a workplace and the presence of equal opportunity policies. In examining the WERS5 data, Kersley and others found that having a formal equal opportunity policy was positively associated with the size of a workplace.16 A similarly strong relationship was found between workplace size and the presence of monitoring and review activities.17 Further, Barnes and others found that large workplaces were more likely to have a ‘formal pro-age orientation’ and a ‘strategic approach to an ageing workforce’, including through the adoption of age-related equal opportunity policies.18 Similarly, company size has been positively associated with the successful adoption and implementation of equality policies in surveys of EU businesses.19 The WERS data set provides detailed information regarding the number of employees at both a workplace and organisational level. According to institutional theory, workplaces belonging to a larger parent organisation will have higher visibility and experience stronger pressure to adopt equal opportunity practices.20 Further, while small workplaces may not have specialist HR staff, these roles may be filled at an organisational level. This may increase a small workplace’s organisational capacity and change its response to equality laws.21 It is therefore necessary to consider whether belonging to a large organisation influences the presence of age-aware practices at the workplace level. In examining WERS4 data from 1998, Hoque and Noon found that formal equal opportunity policies (including those addressing age) were more likely to be adopted in workplaces that were

15 

Felstead et al (n 14 above) 57; Wood et al (n 14 above) 223. Kersley et al, Inside the Workplace: Findings from the 2004 Workplace Employment Relations Survey (London, Routledge, 2006) 237. 17  ibid 248. 18  Barnes et al (n 7 above) 35, 80. 19  European Commission, ‘Continuing the Diversity Journey: Business Practices, Perspectives and Benefits’ (October 2008) 18. This may reflect a lack of knowledge or awareness about equality law in SMEs: a survey of 1,071 UK small employers found that most owner-managers had low self-assessed levels of knowledge of individual employment rights, and knowledge was positively associated with workforce size: R Blackburn and M Hart, ‘Employment Rights in Small Firms: Some New Evidence, Explanations and Implications’ (2003) 32 Industrial Law Journal 60, 61–62. See also M Winterbotham et al, ‘Fairness, Dignity and Respect in SME Workplaces’ (October 2015). 20  Felstead et al (n 14 above) 57; Wood et al (n 14 above) 223. 21  J Barnes and TF Burke, ‘The Diffusion of Rights: From Law on the Books to Organizational Rights Practices’ (2006) 40 Law & Society Review 493, 517. 16  B

Age-Aware Workplaces

 117

part of a larger organisation than in single, standalone workplaces.22 This finding was repeated in Kersley and others’ examination of the WERS5 data.23 A similarly strong relationship was found between being part of a wider organisation and the presence of monitoring and review activities.24 Therefore, this study uses a workplace’s organisational (rather than workplace) size in its analysis. Being in the public sector is also positively correlated with having equal opportunity practices. In addition to being more visible, public sector organisations are exposed more directly to equal opportunity regulatory requirements than private sector organisations, including through the PSED. Therefore, public sector organisations are likely to take more extensive steps to promote equal opportunities in the workplace than their private sector counterparts. In examining the WERS5 data, Kersley and others found that formal equal opportunity policies were almost universally present in the public sector, with 98 per cent of public sector respondents having a policy.25 This finding was repeated with the WERS6 data: 99 per cent of public sector workplaces had an equal opportunity policy in place in 2011.26 Public sector organisations were also more likely to have adopted age discrimination policies in 2004 than private sector organisations.27 A similarly strong relationship was found between being in the public sector and the presence of monitoring and review activities.28 Further, a Chartered Institute of Personnel and Development (CIPD) survey of 285 organisations found that only seven per cent of private sector organisations had sophisticated diversity management measures in place, compared with 34 per cent of public sector organisations.29 Thus, it is likely that public sector organisations will adopt more age-aware practices (and, therefore, have a higher AAPS). The presence of trade unions in a workplace may also promote equality by extending the scope of bargaining and consultation to embrace equality issues30 and increasing external scrutiny of the organisation.31 As ‘defenders of employees’ rights’, trade unions might act as ‘champions of sound equal opportunities practice’.32 In examining WERS4 data, Hoque and Noon found that formal equal opportunity policies (including those addressing age) were more common in workplaces with a recognised union than in those without.33 Similarly, in 22 

Hoque and Noon (n 5 above) 488. Kersley et al (n 16 above) 239. ibid 248. 25  ibid 239. 26  van Wanrooy et al (n 3 above) 116. 27  E Parry and S Tyson, ‘Organizational Reactions to UK Age Discrimination Legislation’ (2009) 31 Employee Relations 471, 483. 28  Kersley et al (n 16 above) 248. 29  CIPD, ‘Survey Report: Diversity in Business: A Focus for Progress’ (London, March 2007) 5. 30 T Colling and L Dickens, ‘Selling the Case for Gender Equality: Deregulation and Equality ­Bargaining’ (1998) 36 British Journal of Industrial Relations 389, 394, 405; H Bewley and S Fernie, ‘What Do Unions Do for Women?’ in HF Gospel and S Wood (eds), Representing Workers: Trade Union Recognition and Membership in Britain (London, Routledge, 2003) 116–17. 31  Felstead et al (n 14 above) 57. 32  M Noon and K Hoque, ‘Ethnic Minorities and Equal Treatment: The Impact of Gender, Equal Opportunities Policies and Trade Unions’ (2001) 176 National Institute of Economic Review 105, 106–7. 33  Hoque and Noon (n 5 above) 488. 23  24 

118 

Quantitative Analysis of WERS6

e­ xamining the WERS5 data, Kersley and others found that almost all workplaces with at least one recognised union had a formal equal opportunity policy in place.34 A similarly strong relationship was found between unionisation and the presence of monitoring and review activities.35 Further, Barnes and others found that unionised workplaces were more likely to have a ‘formal pro-age orientation’ and ‘strategic approach to an ageing workforce’, including through the adoption of age-related equal opportunity policies.36 Among the subset of public limited companies (PLCs), those listed on a stock exchange may be subject to additional monitoring and pressure from shareholders to adopt equal opportunity practices. Listed companies may also be more sensitive to reputational effects, increasing the ‘business case’ for adopting best practice in equal opportunity matters.37 This is consistent with institutional theory: the presence of shareholders increases company visibility and the need to respond to societal ­pressure.38 In analysing the WERS5 data, Deakin and others found workplaces with a stock market listing were significantly more likely to have formal equal opportunity policies in place that mentioned gender.39 However, there was no significant relationship between stock market listing and gender monitoring and reviewing activities. Therefore, while stock market listing made firms more likely to make a formal commitment to gender equality, it did not result in workplaces following through on their formal commitment by monitoring and reviewing equality outcomes. The expert interviews reported in Chapter four also identified dominant employee job or occupational group as a factor influencing the adoption of ageaware practices. There is limited existing evidence regarding whether the largest occupational group in a workplace affects its approach to age-aware practices: this is not a predictor that has featured in past WERS analyses. However, largest occupational group differs by workplace sector: in the WERS5 data set, professionals were most likely to be found in the education sector; elementary or routine occupations in the hotel and restaurant sector; and process, plant and machine operators were most likely to be the largest occupational group in manufacturing workplaces.40 Therefore, largest occupational group may act as a proxy for workplace sector, which is not included in the publicly available WERS6 d ­ ata set. In WERS5, workplaces in public administration (100 per cent), education (99 per cent) and financial services (96 per cent) were most likely to have an equal opportunity policy, and hotels and restaurants (50 per cent) and manufacturers (52 per 34 

Kersley et al (n 16 above) 239. ibid 248. 36  Barnes et al (n 7 above) 35, 80. 37 S Deakin et al, ‘Gender Inequality and Reflexive Law: The Potential of Different Regulatory Mechanisms for Making Employment Rights Effective’, Working Paper No 426 (Centre for Business Research, University of Cambridge, 2011) 1, 21. 38  Felstead et al (n 14 above) 57; Wood et al (n 14 above) 223. 39 S Deakin et al, ‘Gender Inequality and Reflexive Law: The Potential for Different Regulatory Mechanisms’ in L Dickens (ed), Making Employment Rights Effective: Issues of Enforcement and ­Compliance (Oxford, Hart Publishing, 2012) 131–32. 40  Kersley et al (n 16 above) 24–25. 35 

Age-Aware Workplaces

 119

cent) were least likely.41 Given the concentration of professionals in the education sector, it is likely that workplaces with professionals as the largest occupational group will have more age-aware practices in place. Further, workplaces with more process, plant and machine operators (like those in manufacturing) or elementary or routine occupations (like in the hotel and restaurant industry) may be less likely to have age-aware practices in place.

B.  Bivariate Analysis To consider the impact of organisational characteristics on the adoption of ageaware practices, we should consider how organisation type (public or private), organisation size, union presence (percentage of employees who are members of a union or staff association) and listing on a stock exchange are related to AAPS.42 Figures 8 to 11 illustrate the distribution of individual age-aware practices according to these variables. These figures indicate that public sector organisations, large organisations, workplaces with a union presence and listed companies are more likely to adopt age-aware measures. Bivariate correlations* may also be used to examine the relationships between the variables. The relationship between AAPS (which is a skewed categorical variable), grouped organisation size43 and grouped union density44 may be scrutinised using a Spearman’s Rank Order correlation*. Using this test, there was a moderate to large positive correlation between AAPS and grouped organisation size, which was statistically significant (rs = .44, p < .01), and a moderate statistically significant positive correlation between AAPS and grouped union density (rs = .38, p < .01). A Mann–Whitney U test* was employed to examine the relationship between AAPS and the dichotomous independent variables (organisation type and listing). This test showed a statistically significant difference in the distribution of AAPS by organisation type (z = –18.25, p < .001) and listing (z = –3.32, p = .001). Finally, a chi-square test of independence* was performed to examine the relationship between AAPS and organisation type, listing, union presence and largest occupational group.45 The relationship between these

41 

ibid 238. These variables are defined in detail in Blackham (n 8 above). of employees employed at the organisation, grouped as: small organisations (< 50 employees) = 1; medium organisations (50 to 249 employees) = 2; large organisations (≥ 250 employees) = 3. 44  Percentage of employees at the workplace who are members of a trade union or staff association, grouped as: 0% = 0 (nil); > 0 to 10% = 1 (very low); > 10 to 25% = 2 (low); > 25 to 50% = 3 (medium low); > 50 to 75% = 4 (medium high); > 75 to 100% = 5 (very high). 45  Largest non-managerial group of employees at the workplace. Occupational groups included: 1: Professional occupations; 2: Associate professional and technical occupations; 3: Administrative and secretarial occupations; 4: Skilled trades occupations; 5: Caring, leisure and other personal service occupations; 6: Sales and customer service occupations; 7: Process, plant and machine operatives and drivers; 8: Routine occupations. 42 

43 Number

120 

Quantitative Analysis of WERS6

100 90 80 70 60 50 40 30 20 10

Public Private

0

Figure 8: Per cent of Workplaces with Individual Age-Aware Practices by Organisation Type, WERS6 100 90 80 70 60 50 40 30

Small

20

Medium

10

Large

0

Figure 9: Per cent of Workplaces with Individual Age-Aware Practices by Organisation Size, WERS6

Age-Aware Workplaces 100 90 80 70 60 50 40 30 20 10 0

 121

No union presence Union presence

Figure 10: Per cent of Workplaces with Individual Age-Aware Practices by Union Presence, WERS6

100 90 80 70 60 50 40 30 20 10 0

Not listed Listed

Figure 11:  Per cent of PLCs with Individual Age-Aware Practices by Listing, WERS6

122 

Quantitative Analysis of WERS6

variables and AAPS was significant: for organisation type: χ2 (7, N = 2538) = 353.40, p < .01; for listing: χ2 (6, N = 389) = 17.51, p < .01; for union presence: χ2 (7, N = 2339) = 394.81, p < .01; and for largest occupational group: χ2 (49, N = 2538) = 378.39, p < .01. Therefore, bivariate correlations and descriptive statistics indicate that ­organisational characteristics are associated with age-aware practices: consistent with institutional theory, public sector and large organisations, and workplaces with a union presence appear more likely to adopt age-aware practices. Workplaces with a union presence also appear significantly more likely to adopt monitoring and review practices. However, the difference is less pronounced between listed and unlisted PLCs. These findings are worth considering further: do all ‘best practice’ workplaces reflect these organisational characteristics? Further, is there some relationship between these characteristics: for example, are large or public organisations more likely to have a union presence? Government statistics indicate that public sector organisations have far higher union density than private sector organisations: in 2013, 14.4 per cent of employees in the private sector were trade union members, compared with 55.4 per cent in the public sector.46 This is supported by bivariate correlations between the variables, which indicate a significant relationship between grouped organisation size and grouped union density (Spearman’s rho*: rs = .43, p < .01); and organisation type and union density (Pearson’s r*: r = –.58, p < .01).47 Thus, we need to ask whether organisational characteristics cancel each other out. These questions can be addressed using multivariate statistical techniques.

C.  ‘Best Practice’ Organisations The influence of organisational characteristics on age-aware practices may be considered further by examining the ‘best’ workplaces in the WERS data set (that is, those that achieved an AAPS of 7 out of 7). In 2011, 25 workplaces received an AAPS of 7 out of 7. Of these, 8 (or 32 per cent) were in the private sector, 6 (24 per cent) had no union presence, and 7 (28 per cent) were part of a small or medium-sized organisation. It therefore appears that best practice is not fully tied to organisational characteristics. Table 5 breaks this down further according to organisation type.

46 

BIS, ‘Trade Union Membership 2013’, Statistical Bulletin (May 2014) 5. Given the significant correlation between organisation type and union density, the variables must be considered in separate statistical models. 47 

Age-Aware Workplaces

123

Table 5: ‘Best Practice’ Workplaces by Organisational Characteristics, WERS6 (n = 25) Union density >50%

No union presence

Large organisation

Professionals/ administrative staff as core workforce

Overall

Public

7

1

13

16

17

Private

0

5

5

5

8

Total

7

6

18

21

25

Based on this limited sample, the unifying characteristic of ‘best practice’ workplaces is the largest occupational group: 21 of the 25 workplaces had professionals, associate professionals or administrative staff as their core workforce. No ‘best practice’ workplaces were PLCs, making listing a non-issue.

D. Multivariate Models of the Relationship between Organisational Characteristics and AAPS To examine further the relationship between organisational characteristics and the presence of age-aware practices, multivariate statistical techniques can be used to control for the interaction between independent variables. This study used both an ordered probit model*, which is designed to analyse ordinal outcome variables;48 and a negative binomial model*, which examined AAPS as a form of count data.49

(i) Ordinal Models Ordered probit regression provides an appropriate framework for estimating the independent effects of multiple explanatory variables on AAPS. The results of fitting the basic probit equation are demonstrated in Table 6.50 In Ordinal Model 151 there was a statistically significant relationship between organisation size and AAPS, with the odds of a small organisation being placed 48 

See Long (n 11 above) 114. Poisson loglinear model* was initially applied to the unweighted data. However, this model indicated that the data were over-dispersed (estimates for the scale parameter were greater than 1.0, being 1.47 for the deviance parameter and 1.45 for the Pearson dispersion statistic). This suggests that the use of an over-dispersed model, like the negative binomial model, was more appropriate for these data: JM Hilbe, Negative Binomial Regression (Cambridge, Cambridge University Press, 2007) 45. 50  Largest occupational group cannot be included in an ordered probit regression, as the model assumes that the variable categories are ordered (which largest occupational group is not). 51  This model used weighted data. The overall pseudo-R2 for this model was given as between 22.3% (Nagelkerke) and 9.0% (McFadden). The likelihood ratio test equalled 43.56 with a chi-square p value of < .05 (3 df  *). This means that adding the independent variables to the model significantly improved the fit compared to the predictions from the ‘empty’ model. The test of parallel lines was 49  A

124

Quantitative Analysis of WERS6

Table 6: Regression Coefficients for Grouped AAPS (Ordered Probit Regression) Ordinal Model 1

Ordinal Model 2

(PLCs) Regression Exponentiated Regression Exponentiated coefficients coefficients coefficients coefficients Organisational status Public

.64

1.90

Union presence No presence

–.36**

.70**

–.07

.93

Listing Not listed Grouped organisation size Small

–.99***

.37***

Medium

–.22

.81

–1.30*** –.18

.27*** .83

***Statistically significant at the .001 level; ** at the .01 level; * at the .05 level.

in a higher AAPS category being .37 to 1 compared with those for a large organisation. While there was no statistically significant difference between medium and large organisations, organisation size was still significant overall (Wald χ2 = 14.84, p < .005, 2 df ). However, there was no statistically significant relationship between organisation type and AAPS. When union presence was substituted for organisation type, there was a statistically significant relationship between organisation size and AAPS (Wald χ2 = 274.58, p < .001, 2 df ) and union presence and AAPS (Wald χ2 = 141.96, p < .001, 1 df ). The odds of a workplace with no union presence being placed in a higher AAPS category were .54 to 1 compared with the odds for a workplace with a union presence. Ordinal Model 2 considered only PLCs.52 In that model there was a statistically significant relationship between organisation size and AAPS, with the odds of a small PLC being placed in a higher AAPS category being .27 to 1 compared

non-significant (p > .05), indicating that the assumption that the effect of the predictor variables was the same across all levels of the outcome variable was not violated. Collinearity diagnostics run on the model indicated high tolerance and a low variance inflation factor for each variable, indicating that the assumption regarding multicollinearity held for this model. Therefore, this model is a good fit for the data. 52  This model used unweighted data. When applied to weighted data, the model did not significantly improve the fit compared to the predictions from the ‘empty’ model and the predictor variables were not significant. The overall pseudo-R2 for this model was given as between 17.4% (Nagelkerke) and 6.5% (McFadden). The likelihood ratio test equalled 131.56 with a chi-square p value of < .001 (4 df ) and the test of parallel lines was non-significant (p > .05). Collinearity diagnostics run on the model indicated high tolerance and a low variance inflation factor for each variable.

Age-Aware Workplaces

 125

with those for a large PLC. While there was no statistically significant difference between medium and large companies, organisation size was still significant ­overall (Wald χ2 = 32.41, p < .001, 2 df ). There was no statistically significant relationship between listing and AAPS. That said, there was a statistically significant relationship between union presence and AAPS: the odds of a PLC with no union presence being placed in a higher AAPS category were .70 to 1 compared with the odds for a PLC with a union presence.

(ii)  Count Models These results were explored further by modelling AAPS as a form of count data. A negative binomial parametric model can be used to accommodate over-dispersed count data53 as it allows the conditional variance of the data to exceed the conditional mean (unlike the Poisson model).54 Unweighted data55 were fitted to a negative binomial model, using AAPS as the dependent count variable. The negative binomial regression coefficients for each of the predictor levels in these models are presented in Table 7. In Count Model 1,56 a statistically significant relationship was found between AAPS and organisation size, organisation type and largest occupational group. The score rate for public workplaces was 1.36 times the rate for the reference group (private workplaces), holding the other variables constant, indicating that organisation type has a statistically significant correlation with AAPS. Further, the score rate for small organisations was less than half the rate for large organisations, indicating that organisation size also has a statistically significant relationship with AAPS. The score rate for workplaces where the largest occupational group consisted of professionals was 1.88 times the rate for the reference group (routine occupations). The rate for process, plant and machine operatives and

53 

Cameron and Trivedi (n 11 above) 71. Long (n 11 above) 218. One possible derivation of the model is that there is a degree of dependence in the data, with the occurrence of an event likely to influence later occurrences: see Cameron and Trivedi (n 11 above) 71. This is sometimes known as ‘contagion’: Long (n 11 above) 236. This could be the case with AAPS: adopting one age-aware practice may influence whether an organisation adopts other age-aware practices. 55 Neither the Poisson nor the negative binomial models could be run with organisational weightings, suggesting some issues with the weights themselves. 56  The log likelihood for the model was –4212.06. The deviance associated with the model was 2669.66. Dividing the deviance by its degrees of freedom yields a measure of the goodness of fit of the model: here that scaled deviance was 1.06. The dispersion coefficient was .10, and the coefficient’s 95% confidence interval was greater than zero (ranging from .07 to .14), suggesting that a negative binomial model was more appropriate than the Poisson for modelling these data. The likelihood ratio chi-square was 900.83 to 10 degrees of freedom (p < .001), indicating that the overall model was a significant improvement compared to a model without any predictors. However, the Akaike information c­ riterion* (AIC) score was 8448.12, indicating that the model was still not a good fit for the data. Each of the predictors was statistically significant as a whole in this model: grouped organisation size (Wald χ2 = 354.25, p < .001, 2 df ); organisation type (Wald χ2 = 75.58, p < .001, 1 df ); and largest ­occupational group (Wald χ2 = 179.37, p < .001, 7 df ). 54 

126 

Quantitative Analysis of WERS6

Table 7:  Regression Coefficients for AAPS (Negative Binomial Regression) Count Model 1 Regression coefficient

Exponentiated 95% Wald confidence coefficient interval for (Exp(B)) Exp(B) Lower

Upper

Grouped organisation size Small

–.99***

.37***

.33

.41

Medium

–.32***

.73***

.66

.80

.31***

1.36***

1.27

1.46

Professional

.63***

1.88***

1.67

2.12

Associate professional and technical

.56***

1.75***

1.53

2.00

Administrative and secretarial

.56***

1.75***

1.53

1.99

Skilled trades

.16

1.17

.97

1.40

Caring, leisure, other personal service

.37***

1.45***

1.26

1.67

Sales and customer service

.12

1.13

.98

1.31

Process, plant and machine operatives and drivers

.14

1.15

.98

1.35

Organisation type Public Largest occupational

group1

Count Model 2 Regression coefficient

Exponentiated 95% Wald confidence coefficient interval for (Exp(B)) Exp(B) Lower

Upper

Grouped organisation size Small

–.91***

.40***

.36

.45

Medium

–.26***

.77***

.70

.86

–.49***

.61***

.56

.66

Union presence No union

***Statistically significant at the .001 level; ** at the .01 level; * at the .05 level. Notes 1 Reference group: routine occupations.

Age-Aware Practices Over Time

 127

drivers was not significantly different to that for routine workers. Therefore, ­occupational groups had a statistically significant correlation with AAPS. In Count Model 2 union presence was substituted for organisation type.57 In this model, a statistically significant relationship was found between organisation size and AAPS. Having no union members at the workplace reduced the score rate to .61 of the rate for workplaces with union members. A count model was re-run with only PLCs to consider the relationship between listing and AAPS.58 In that model, being a small organisation (B = –1.21, p < .01) and not having a union presence (B = –.22, p < .05) had a statistically significant relationship with AAPS. Organisation size (Wald χ2 = 23.60, p < .001, 2 df ) and union presence (Wald χ2 = 5.52, p < .05, 1 df ) were statistically significant as predictors overall. There was no statistically significant relationship between listing and AAPS. This indicates that, as in previous studies, there is no statistically significant relationship between listing and the adoption of holistic age-aware practices.

IV.  Age-Aware Practices Over Time While it appears that there is a significant relationship between organisational characteristics and age-aware practices, it is also worth exploring whether organisational practices change over time. Kersley and others found an overall increase in the number of organisations that had a formal written equal opportunity policy between 1998 and 2004, increasing from 64 per cent in WERS4 to 73 per cent in WERS5.59 This finding was repeated in the panel survey—the proportion of continuing organisations with a policy increased from 68 per cent in 1998 to 82 per cent in 2004, indicating behavioural change.60 Further, policies in 2004 covering criteria such as age were ‘distinctly more prevalent’ than in 1998.61 Therefore, we

57  The log likelihood for the model was –3945.11. The deviance associated with the model was 2442.61, with a scaled deviance of 1.05. The dispersion coefficient was .18, and the coefficient’s 95% confidence interval was greater than zero (ranging from .14 to .22). The likelihood ratio chi-square was 634.27 to three degrees of freedom (p < .001). However, the AIC score was 7900.22. Each of the predictors was statistically significant as a whole in this model: grouped organisation size (Wald χ2 = 243.16, p < .001, 2 df ); and union presence (Wald χ2 = 143.76, p < .001, 1 df ). 58  The log likelihood for the model was –574.15. The deviance associated with the model was 352.56, with a scaled deviance of .97. This indicates that the model was under-dispersed. The dispersion coefficient was .07, and the coefficient’s 95% confidence interval was greater than zero (ranging from .02 to .25), suggesting that a negative binomial model was more appropriate than the Poisson for modelling these data. The likelihood ratio chi-square was 51.18 to four degrees of freedom (p < .001), indicating that the overall model was a significant improvement compared to a model without any predictors. However, the AIC score was 1160.31, indicating that the model was still not a good fit for the data. 59  Kersley et al (n 16 above) 237. 60  ibid 237. 61  ibid 240; J Walsh, ‘Equality and Diversity in British Workplaces: The 2004 Workplace Employment Relations Survey’ (2007) 38 Industrial Relations Journal 303, 304–5.

128 

Quantitative Analysis of WERS6 45 40 35 30 25

2004

20

2011

15 10 5 0

0

1

2

3

4

5

6

7

Figure 12: AAPS by Year (Per cent of Workplaces), WERS5 (n = 989) and WERS6 (n = 924)1 Notes: 1 The number of workplaces with these details missing is significantly higher in 2011 than in 2004. It is possible that this will be corrected over time, as the data continue to be checked and revised.

can predict that AAPS will improve over time, particularly given the passage of the Regulations in 2006. To explore the change in AAPS over time, this study used WERS6 panel data to compare workplaces’ AAPS in 2004 with those in 2011.62 Figure 12 represents the panel workplaces’ AAPS in 2004 and 2011. A visual inspection of Figure 12 strongly suggests that AAPS improved over time, with a smaller percentage of workplaces having an AAPS of 0 in 2011, and a higher percentage having an AAPS of 3 or above.63 The mean AAPS for the panel workplaces increased from 1.56 in 2004 to 2.04 in 2011, though the median and mode remained at 1. In 2011, 78.4 per cent of respondents had an AAPS between 0 and 3, down from 86.3 per cent in 2004. Therefore, it appears that AAPS increased slightly over time.

62  At the time of analysis, the WERS6 panel data set had not been publicly released. Therefore, it was necessary to manually create a panel data set by merging the WERS6 and WERS5 data sets, using the WERS5 unique identifiers as a cross-referencing key. 63  This may be linked to ‘survival bias’, particularly if workplaces with age-aware practices in 2004 (or who developed age-aware practices between 2004 and 2011) were more likely to be in existence in 2011, and therefore disproportionately represented in the panel data: see Bryson (n 12 above). Sixteen per cent of workplaces surveyed in 2004 had closed down in 2011, with smaller and younger workplaces more likely to fail: van Wanrooy et al (n 3 above) 26, 27–28. However, there is (as yet) no obvious link between AAPS and business survival, though there may be some link between equality policies and business performance: see K Monks, ‘The Business Impact of Equality and Diversity: The International Evidence’ (Dublin, The Equality Authority, July 2007). Therefore, survival bias appears less relevant to this study than research on other organisational characteristics.

Age-Aware Practices Over Time

 129

A paired sample t-test* was conducted to compare AAPS in 2004 and 2011. This test is particularly useful for conducting pre- and post-test evaluations when a sample has been exposed to some intervention (here, the passing of age discrimination laws). There was a significant difference in the scores for 2004 (mean = 1.56, standard deviation = 1.64) and 2011 (mean = 2.04, standard deviation = 1.83); paired t (924) = –7.59, p < .001. This indicates that the difference of means between AAPS in 2004 and 2011 is statistically significant: there is a statistically significant relationship between year and AAPS, and AAPS was significantly higher in 2011 than in 2004. A chi-square test of independence was used to examine further the relationship between AAPS in 2004 and 2011. The relationship between these variables was significant: χ2 (49, N = 924) = 228.66, p < .01. Therefore, while AAPS increased over time, a workplace’s AAPS in 2011 was still associated with their AAPS in 2004. It is also informative to examine the shift in AAPS across the different score levels. Table 8 demonstrates how AAPS changed between the two time points. No workplace with an AAPS of 0 in 2004 had an AAPS of 7 in 2011. However, the five workplaces with an AAPS of 7 in 2004 had a lower AAPS in 2011. It is therefore necessary to consider further how AAPS has changed over time, and whether organisational characteristics are associated with the change. Figure 13 illustrates the change in individual age-aware practices between 2004 and 2011. It appears that more workplaces had adopted a formal policy and monitoring and review activities in 2011. However, the use of procedures to encourage applications from older workers had actually declined from 2004 to 2011. This may reflect the broader economic conditions facing workplaces in 2011. It may also account for some of the ‘best’ workplaces in 2004 achieving a lower score in 2011.

Table 8:  Cross-Tabulation of AAPS between WERS5 and WERS6 (Panel Data) 2011 2004

0

1

2

3

4

5

6

7

Total

0

91

109

25

17

6

7

3

0

258

1

41

153

46

31

27

21

20

1

340

2

11

48

22

13

10

10

10

3

127

3

3

28

8

11

9

7

7

1

74

4

2

11

3

11

4

3

8

3

45

5

2

8

3

10

6

6

6

0

41

6

1

7

4

3

5

6

7

1

34

7

0

1

1

0

2

1

0

0

5

151

365

112

96

69

61

61

9

924

Total

130 

Quantitative Analysis of WERS6

90 80 70 60 50 2004

40

2011

30 20 10 0 Formal policy

Encourage applications

Monitor Review Monitor recruitment recruitment promotions procedures

Review pay Review rates promotion procedures

Figure 13: Valid Per cent of Workplaces with Individual Age-Aware Practices by Year (Panel Data)

V.  Workplaces that Adopt Age-Aware Practices Over Time Institutional theory posits that organisations with higher visibility are more likely to be compelled by societal pressure to adopt equal opportunity practices.64 Legal regulations are a significant form of societal pressure. Thus, visible organisations (including those that are large, in the public sector, with a union presence or listed) would be more likely to adopt age-aware practices following the introduction of the Regulations. Therefore, we can anticipate that these workplaces will have adopted more age-aware practices in 2011 than in 2004.

A.  Ordinal Models To examine the relationship between organisational characteristics and AAPS over time, an ordered probit model was posed with organisational characteristics and grouped AAPS65 at 2004 as the predictor variables and the grouped 2011 AAPS 64  65 

Felstead et al (n 14 above) 57; Wood et al (n 14 above) 223. AAPS is grouped as: score 0 = 0; score 1 = 1; score 2 = 2; score 3 = 3; score 4 to 7 = 4.

Workplaces Adopting Age-Aware Practices

 131

Table 9:  Regression Coefficients for Grouped AAPS (Ordered Probit Regression) (Panel Data) Panel Ordinal Model 1

Panel Ordinal Model 2

Regression Exponentiated Regression Exponentiated coefficients coefficients coefficients coefficients Grouped organisation size Small

–.85***

.43***

–.67***

.51***

Medium

–.39**

.68**

–.34**

.71**

–.61***

.54***

Organisation type Public

.54***

1.72***

Union presence No presence 2004 AAPS 0

–1.06***

.34***

–1.08***

.34***

1

–.63***

.53***

–.60***

.55***

2

–.55***

.58***

–.55***

.58***

3

–.35*

.70*

–.30

.74

***Statistically significant at the .001 level; ** at the .01 level; * at the .05 level.

as the dependent variable. In using the 2004 AAPS as a predictor variable, this conditional change model* accommodates the interaction between having ageaware practices in 2004 and 2011. The model used organisation size, type, union presence and listing at 2004 as predictor variables, as it was hypothesised that these variables would impact on the practices adopted (or not adopted) in 2011.66 The results of fitting the basic probit equation are presented in Table 9. Panel Ordinal Model 167 found that organisation size (Wald χ2 = 78.21, p < .001, 2 df) and organisation type in 2004 were statistically significant predictors of AAPS in 2011. However, a workplace’s 2004 AAPS was also a significant predictor:

66 SW Menard, Logistic Regression: From Introductory to Advanced Concepts and Applications (Newbury Park CA, SAGE, 2010) 268–69. 67  This model used unweighted data. When applied to weighted data, the only significant predictor variables were a 2004 AAPS of 0 (with a regression coefficient of –1.06, statistically significant at the .05 level) and being a small organisation (regression coefficient of –.69, statistically significant at the .01 level). Organisation size was significant overall (Wald χ2 = 6.98, p < .05, 2 df ) but 2004 scores were not (p > .05). Using unweighted data, the overall pseudo-R2 for this model was given as between 29.8% (Nagelkerke) and 11.2% (McFadden). The likelihood ratio test equalled 614.76 with a chi-square p value of < .001 (7 df ). The test of parallel lines was non-significant (p > .05). Collinearity diagnostics run on the model indicated high tolerance and a low variance inflation factor for each variable.

132 

Quantitative Analysis of WERS6

the odds of a workplace with an AAPS of 0 in 2004 being placed in a higher age score category in 2011 were .34 to 1 compared with the odds of an organisation with an AAPS of 4+ in 2004. Overall, previous AAPS was a significant predictor of future AAPS (Wald χ2 = 70.80, p < .001, 4 df ). Similar results emerged from Panel Ordinal Model 2, which substituted union presence for organisation type:68 organisation size (Wald χ2 = 43.81, p < .001, 2 df ); previous AAPS (Wald χ2 = 72.66, p < .001, 4 df ); and union presence (Wald χ2 = 51.72, p < .001, 1 df ) were significant predictors overall. When re-run with only PLCs, an ordered probit model was not a good fit for the data.69

B.  Count Models The data were also fitted to a negative binomial count model, using 2011 AAPS as the dependent count variable and testing a range of independent variables, including AAPS and organisational characteristics as at 2004. The negative binomial regression coefficients for each of the predictor levels in this model are presented in Table 10. In the Panel Count Model,70 2004 AAPS was a significant predictor overall (Wald 2 χ = 77.60, p < .001, 7 df). Similar results were obtained by running the model with union presence rather than organisation type: previous age scores were significant overall (Wald χ2 = 76.12, p < .001, 7 df), but not at any individual level. Union presence was also statistically significant (regression coefficient = –.48, p < .001; Wald χ2 = 52.43, p < .001, 1 df), as was organisation size (Wald χ2 = 34.93, p < .001, 2 df). When repeated with the subset of PLCs to examine the relationship between listing and AAPS, the model fit was not significant.71

68  This model used unweighted data. The overall pseudo-R2 for the model was given as between 30.5% (Nagelkerke) and 11.5% (McFadden). The likelihood ratio test equalled 603.91 with a chisquare p value of < .001 (7 df  ). The test of parallel lines was non-significant (p > .05). Collinearity diagnostics run on the model indicated high tolerance and a low variance inflation factor for each variable. 69  The likelihood ratio test equalled 25.67 with a chi-square p value of > .05 (7 df  ). This means that adding the independent variables to the model did not significantly improve the fit compared to the predictions from the ‘empty’ model. Further, convergence could not be obtained. 70  This model used unweighted data. When applied to weighted data, the model did not significantly improve the fit compared to the predictions from the ‘empty’ model and the predictor variables were not significant. The log likelihood for the model was –1571.58. The deviance associated with the model was 964.26, with a scaled deviance of 1.06. The dispersion coefficient was .11, and the ­coefficient’s 95% confidence interval was greater than zero (ranging from .06 to .18). The likelihood ratio chi-square was 284.72 to 10 degrees of freedom (p < .001). However, the AIC score was 3167.17. Each of the predictors was statistically significant as a whole in this model: 2004 age score (Wald χ2 = 77.60, p < .001, 7 df  ); grouped organisation size (Wald χ2 = 63.29, p < .001, 2 df  ); organisation type (Wald χ2 = 43.36, p < .001, 1 df  ). 71  The likelihood ratio chi square was 13.63 with a p value of > .05 (9 df  ). This means that adding the independent variables to the model did not significantly improve the fit compared to the predictions from the ‘empty’ model.

Workplaces Adopting Age-Aware Practices

 133

Table 10:  Regression Coefficients for AAPS (Negative Binomial Regression) (Panel Data) Panel Count Model Regression coefficient

Exponentiated coefficient (Exp(B))

95% Wald confidence interval for Exp(B) Lower

Upper

Grouped organisation size Small

–.67***

.51***

.43

.60

Medium

–.21*

.81*

.67

.98

1.44***

1.29

1.60

Organisation type Public

.36***

2004 AAPS 0

–.70*

.50*

.27

.90

1

–.33

.72

.40

1.28

2

–.26

.77

.43

1.39

3

–.15

.86

.47

1.56

4

.10

1.10

.60

2.02

5

–.01

.99

.54

1.82

6

.12

1.12

.61

2.07

***Statistically significant at the .001 level; ** at the .01 level; * at the .05 level.

From this analysis, we can draw three broad conclusions. First, while AAPS increased slightly across all workplaces between 2004 and 2011, most workplaces still had few age-aware practices in place. Despite legal regulation, there has been limited change in age-aware practices over time. Secondly, and relatedly, workplaces are more likely to keep practices that are already in place than to adopt new age-aware practices: AAPS in 2004 was a significant predictor of 2011 AAPS in all models. Therefore, a workplace’s previous age-aware measures are a strong predictor of what practices they will have in the future. This implies that organisational change is limited, as past behaviour predicts future behaviour. This is consistent with theories of organisational inertia,72 path dependence, and lock-in.73 72  That is, the tendency of organisations to continue on their existing paths, often at the expense of adapting to their changing environment: see L Godkin and S Allcorn, ‘Overcoming Organizational Inertia: A Tripartite Model for Achieving Strategic Organizational Change’ (2008) 8 Journal of Applied Business and Economics 82. 73 That is, ‘a stochastic process which obtains under two conditions (contingency and self-­ reinforcement) and causes lock-in in the absence of exogenous shock’: J-P Vergne and R Durand, ‘The Missing Link between the Theory and Empirics of Path Dependence: Conceptual Clarification, Testability Issue, and Methodological Implications’ (2010) 47 Journal of Management Studies 736, 741; cf R Garud et al, ‘Path Dependence or Path Creation?’ (2010) 47 Journal of Management Studies 760. Breaking an organisational path, and overcoming a practice’s historical roots, requires ­disrupting

134 

Quantitative Analysis of WERS6

In essence, organisational history matters.74 Thirdly, both count and ordinal ­models show significant relationships between organisation size, type and union presence in 2004 and AAPS in 2011: public sector organisations, large organisations and those with a union presence in 2004 were significantly more likely to have a high AAPS in 2011 than small or medium organisations, private sector organisations or those without a union. It was not possible to construct either an ordinal or count model to compare listing in 2004 with AAPS in 2011.

VI. Conclusion The results from the statistical analyses both corroborate and extend the findings from previous chapters. First, the results support the experts’ views in Chapter four that UK workplaces are adopting a compliance-focused approach to age discrimination legislation. Among WERS respondents, ‘compliance’ appears to consist of the adoption of a formal equal opportunity policy mentioning age. However, for most workplaces, it does not entail any monitoring, review or proactive measures to respond to demographic ageing. This reinforces the view that UK age discrimination laws are failing to encourage proactive employer responses to an ageing workforce. The comparative study in Chapter six considers how a broader and more ‘proactive’ role for government and legislation might encourage less compliance-focused organisational responses to ageing. Secondly, the results of this chapter indicate that organisational characteristics have some relationship with the adoption of age-aware practices: organisation size and type, union presence, and largest occupational group may be linked with the adoption of good practice. However, the ‘best’ WERS workplaces do not uniformly reflect these organisational characteristics. The ambiguity in these results, and the difficulties linking organisational characteristics with age-aware practices, supports Dickens and Hall’s argument that legal impact is mediated by internal and external organisational factors (such as market context, management style, organisational culture and social relations within the firm) rather than organisational size specifically.75 These factors are explored in more detail in the organisational case studies presented in Chapter seven. Thirdly, while organisational approaches to ageing appear to have improved slightly since the introduction of age discrimination legislation in 2006, it is

its self-reinforcing pattern and consciously reflecting on the path and its drivers: J Sydow et al, ‘Organizational Path Dependence: Opening the Black Box’ (2009) 34 Academy of Management Review 689, 702. Alternatively, it could be achieved via an exogenous shock: Garud et al, ibid, 760. 74  Sydow et al (n 73 above) 690; see also L Dickens and M Hall, ‘Fairness—up to a Point. Assessing the Impact of New Labour’s Employment Legislation’ (2006) 16 Human Resource Management Journal 338, 350. 75  Dickens and Hall (n 74 above) 349–50.

Conclusion

 135

i­mpossible to attribute causally any shift to legislative change using statistical ­analysis. Further, past age-aware practices are strongly predictive of present practices. Therefore, we must consider further how and why organisational ­ change occurs. This is taken up in Chapter seven, which explores the role of law in ­organisational change. Finally, this analysis may unwittingly diminish the actions and activities of SMEs: the absence of formal practices does not necessarily imply poor practice.76 SMEs are more likely to adopt an informal approach to their workforce, rather than adopting formal measures. Therefore, the WERS questions may disproportionately favour larger organisations. At the same time, a 2015 study for the EHRC found that SMEs were often over-confident regarding their equality measures, and had limited understanding of their legal obligations in this area.77 Thus, this requires further qualitative examination at a workplace level, as reported in ­Chapter seven.

76  77 

Barnes et al (n 7 above) vii. Winterbotham et al (n 19 above) pt 14.3.

6 A Comparative Perspective— The Case of Finland Age discrimination laws in the UK have significant limitations in their d ­ rafting and practical impact. As we have seen in previous chapters, most employers are adopting a ‘compliance-focused’ approach to the ageing workforce. UK laws provide few incentives for employers to respond proactively to demographic change. Instead, innovative practice may be ‘punished’ by legal challenges due to unclear laws around positive action. Given the limitations of the UK approach, it is ­important to consider how governments and employers could more proactively address demographic ageing. To this end, this chapter presents a comparative examination of age discrimination laws in Finland. The Finnish experience is relevant to the UK for three key reasons. First, Finland encountered the challenges of an ageing population earlier than other European countries,1 compelling policy-makers to respond to demographic change. Between 1990 and 2012, individuals over the age of 65 increased from 13.5 per cent to 18.1 per cent of the Finnish population.2 Finland’s old-age dependency ratio is projected to increase from 25 per cent in 2000 to 43 per cent in 2025 (compared with the OECD average of 33 per cent in 2025).3 Perhaps as a result, Finland also has the longest history of legal intervention in age discrimination in the EU.4 Secondly, Finland’s labour market interventions relating to older workers are regarded as remarkably successful:5 the National Programme for Ageing Workers (NPAW) (discussed further below) increased the employment rate for older workers by 25 per cent between 1998 and 2002, with the median

1 S Ilmakunnas and M Takala, ‘Promoting Employment among Ageing Workers: Lessons from ­Successful Policy Changes in Finland’ (2005) 30 Geneva Papers on Risk & Insurance 674, 686. 2 Statistikcentralen, ‘Tilastokeskus—Population’ (14 December 2012): www.stat.fi/tup/suoluk/ suoluk_vaesto_en.html. 3  OECD, ‘OECD Recommends Finland to Do More to Help Older People Stay in Work’ (OECD Employment Policies and Data, 9 March 2014): www.oecd.org/els/emp/oecdrecommendsfinlandtodomoretohelpolderpeoplestayinwork.htm. In contrast, the ratio in the UK is projected to reach 38.5% in 2050, up from 24.4% in 2000: OECD, Society at a Glance: 2005 OECD Social Indicators (OECD Publishing, 2005) 27. 4  N Adnett and S Hardy, ‘The Peculiar Case of Age Discrimination: Americanising the European Social Model?’ (2007) 23 European Journal of Law & Economics 29, 35. 5 H Sarfati, ‘Social Dialogue: A Potential “Highroad” to Policies Addressing Ageing in the EU ­Member States’ (2006) 59 International Social Security Review 49, 63.

Finland—A Comparative Perspective

 137

Finnish exit age rising from 58.2 in 1996 to 61.4 in 2001.6 Thirdly, Finland has adopted positive, proactive measures to respond to an ageing workforce, which demonstrate how governments and employers can collaboratively and proactively address demographic ageing. Finland is therefore an interesting and worthwhile comparator for other EU countries. Of course, the situation in the UK could be compared with that in many other countries, instead of (or in addition to) Finland. For example, as I have discussed elsewhere,7 the USA and Australia, which have effectively abandoned mandatory retirement ages, are interesting comparators for examining the effects of removing a DRA. At the same time, given that these countries are not bound by the Framework Directive, a comparative analysis of Finland can offer additional insights regarding how Member States respond to EU interventions. Further, extensive research has already been undertaken comparing age discrimination laws in the UK with those in the USA8 and, to a lesser extent, Australia.9 In contrast, the situation in Finland has not been subjected to rigorous comparative research. Thus, by focusing on Finland as a comparator country, this book makes an innovative contribution to the age discrimination literature. In this chapter, then, I draw on explanatory comparative legal analysis of ­Finnish law, semi-structured expert interviews conducted in June 2013 with 13 Finnish experts (variously representing the Finnish government, trade unions, employer bodies and academics),10 and secondary statistical analysis, to demonstrate how Finnish laws embody a more proactive, collective and employer-focused approach 6 

ibid 64. A Blackham, ‘Managing without Default Retirement in Universities: A Comparative Picture from Australia’ (2015) 35 Legal Studies 502. 8  See, eg, A-M Guillemard, ‘The Trend towards Early Labour Force Withdrawal and the Reorganisation of the Life Course: A Cross-National Analysis’ in PA Johnson et al (eds), Workers Versus Pensioners: Intergenerational Justice in an Ageing World (Manchester, Manchester University Press, 1989); B Casey, ‘Redundancy and Early Retirement: The Interaction of Public and Private Policy in Britain, G ­ ermany and the USA’ (1992) 30 British Journal of Industrial Relations 425; M Cooke, ‘Policy Changes and the Labour Force Participation of Older Workers: Evidence from Six Countries’ (2006) 25 Canadian Journal on Aging/La Revue canadienne du vieillissement 387; S Bisom-Rapp and M Sargeant, ‘Increasing the Employment Rate of Older Workers’ (2011) 27 International Journal of Comparative Labour Law and Industrial Relations 301. 9 See, eg, P Taylor et al, ‘Mature Age Employment: Recent Developments in Public Policy in Australia and the UK’ (2000) 19 Australasian Journal on Ageing 125; K Riach, ‘Older Workers: Learning from Three International Experiences’ (2006) 5 Social Policy and Society 551; S Bisom-Rapp et al, ‘Decent Work, Older Workers and Vulnerability in the Economic Recession: A Comparative Study of Australia, the United Kingdom, and the United States’ (2011) 15 Employee Rights & Employment Policy Journal 43. 10  Interviewees included representatives from: Finnish Institute of Occupational Health (Työterveyslaitos (TTL)), Aalto University, Finnish Centre for Pensions (Eläketurvakeskus (ETK)), Finnish Funding Agency for Technology and Innovation (Tekes), University of Helsinki, Ministry of Social Affairs and Health (Sosiaali-ja terveysministeriö (STM)), Finnish Confederation of Professionals (Toimihenkilökeskusjärjestö (STTK)), Central Organisation of Finnish Trade Unions (Suomen Ammattiliittojen Keskusjärjestö (SAK)), Confederation of Finnish Industries (Elinkeinoelämän keskusliitto (EK)), ­Ministry of Employment and the Economy (Työ-ja elinkeinoministeriö (TEM)), and KT Local Government Employers (KT Kuntatyönantajat). As in Chapter 4 of this volume, Finnish interviewees have been broadly classified as a representative of government or a non-departmental government agency 7 

138 

Finland—A Comparative Perspective

to preventing age discrimination than those in the UK. I identify a number of lessons from the Finnish experience that may inform UK law and policy-making. This study was conducted in English, drawing on English translations of Finnish legal materials (which are readily available online). Linguistic and cultural challenges or misunderstandings were addressed by triangulating comparative legal analysis with qualitative expert interviews, and by spending extended periods in Finland and Sweden in 2013–14 to develop my understanding of Nordic legal culture and the specific laws in question. Given that ‘high-sounding values’ and laws can operate differently in practice, the expert interviews and visits to Finland provided me with additional insights into the practical operation of the laws concerned, and acted as an appropriate check on legal and political rhetoric.11 Comparative researchers are often criticised for undertaking analysis within a frame of reference dictated by their home jurisdiction,12 thereby over-emphasising the similarities between countries, rather than emphasising both similarities and differences.13 As Örücü notes, the purpose of comparative research ‘is not to search particularly for similarity or difference but to observe [and, potentially, explain] what is actually there’.14 With this in mind, in this chapter I seek to acknowledge both similarities and differences between the UK and Finland as they emerge.

I.  Finnish Legal Provisions The Framework Directive forms a common legislative foundation for anti-­ discrimination law in the UK and Finland. However, the way Finland has approached its legislative obligations contrasts markedly to the UK approach.

A. Non-Discrimination Finnish age discrimination legislation has been in place since well before the Framework Directive.15 The Finnish Constitution provides: ‘No one shall, (Government), academia (Academic) or a non-governmental organisation (NGO); and allocated a number with the letter F (for Finland) prefixed. So the first government representative is identified as Government F1; the fourth non-governmental organisation representative as NGO F4. 11  D Nelken, ‘Comparative Law and Comparative Legal Studies’ in E Örücü and D Nelken (eds), Comparative Law: A Handbook (Oxford, Hart Publishing, 2007) 34. 12  M Adams, ‘Doing What Doesn’t Come Naturally: On the Distinctiveness of Comparative Law’ in M van Hoecke (ed), Methodologies of Legal Research: What Kind of Method for What Kind of Discipline?, European Academy of Legal Theory (Oxford, Hart Publishing, 2011) 234. 13  Nelken (n 11 above) 39; Adams (n 12 above) 234. 14  E Örücü, ‘Developing Comparative Law’ in E Örücü and D Nelken (eds), Comparative Law: A Handbook (Oxford, Hart Publishing, 2007) 50. 15  J Arrowsmith and M Hall, ‘Industrial Relations and the Ageing Workforce: A Review of Measures to Combat Age Discrimination in Employment’ (European Industrial Relations Observatory On-line, 28 October 2000): www.eurofound.europa.eu/eiro/2000/10/study/tn0010201s.htm.

Finnish Legal Provisions

 139

without an acceptable reason, be treated differently from other persons on the ground of … age’.16 By defining discrimination as differential treatment without an acceptable reason, the Constitution raises the possibility that direct discrimination may be justified.17 This is further developed in the Employment Contracts Act (55/2001) (Finland), which prohibits ‘any unjustified discrimination [by employers] against employees on the basis of age’.18 Employers must also ‘treat employees equally unless there is an acceptable cause for derogation deriving from the duties and position of the employees’.19 These older acts prohibit age discrimination in ‘rather general terms’20 and do not reflect the terms of the Framework Directive. The Non-Discrimination Act (21/2004) (Finland) (‘2004 Act’) was introduced to implement the Framework Directive in Finnish law and to ‘foster and safeguard equality and enhance the protection provided by law to those who have been discriminated against’.21 The 2004 Act was repealed and replaced in 2014 with the passage of the Non-Discrimination Act (1325/2014) (Finland) (‘2014 Act’), which came into force on 1 January 2015 (after the date of the fieldwork in this study). In the discussion that follows, reference is made to both Acts, though the 2004 Act is most relevant given the timing of this study. The 2004 and 2014 Acts prohibit discrimination on the basis of age.22 The 2004 Act prohibited discrimination in relation to: —— conditions for access to self-employment or means of livelihood; —— recruitment conditions, employment and working conditions, personnel training and promotion; —— access to training and vocational guidance; and —— membership and involvement in work-related organisations.23 The 2014 Act is framed more broadly, to prohibit discrimination in ‘public and private activities’.24 Under the 2004 Act, authorities were required to seek to: purposefully and methodically … foster equality and consolidate administrative and operational practices that will ensure the fostering of equality in preparatory work and decision-making [and] alter any circumstances that prevent the realization of equality.25

‘Authorities’ were defined broadly to include central and local government ­authorities; independent bodies governed by public law; and societies governed 16 

Finnish Constitution ch 2, s 6. R Hiltunen, ‘Report on Measures to Combat Discrimination: Directives 2000/43/EC and 2000/78/ EC—Country Report 2012 Finland’ (European Network of Legal Experts in the Non-discrimination Field, 2013) 72–73. 18  Employment Contracts Act (55/2001) (Finland) ch 2, s 2. 19 ibid. 20  Hiltunen (n 17 above) 5. 21  2004 Act s 1. 22  2004 Act s 6; 2014 Act s 8. 23  2004 Act s 2. 24  2014 Act ss 2, 8. 25  2004 Act s 4. 17 

140 

Finland—A Comparative Perspective

by public law, individual actors and non-incorporated state enterprises when ­discharging public administrative functions.26 The 2014 Act alters these provisions. Authorities must now ‘take necessary measures to promote the realisation of equality’, with measures being ‘effective, expedient and proportionate, taking into account the authorities’ operating environment, resources and other circumstances’.27 Authorities must also ‘evaluate the realisation of equality in their activities’,28 and put in place a plan for necessary measures for the promotion of equality.29 The definition of ‘authority’ has also been altered slightly.30 Even more significantly, the 2014 Act extends responsibility for the promotion of equality to employers, as well as public authorities. Section 7(1) of the 2014 Act says: The employer must assess the realisation of equality in the workplace and, taking into account the needs of the workplace, develop the working conditions as well as the methods complied with in the selection of personnel and in making decisions concerning the personnel. These measures shall be effective, expedient and proportionate, taking into account the operating environment, resources and other circumstances.

Employers who regularly employ over 30 staff must also have a plan for ‘necessary measures for the promotion of equality’.31 These measures and their effectiveness must be discussed with staff or their representatives.32 Finally, staff representatives have a right to access information on actions the employer has taken to promote equality in the workplace.33 Different treatment on the grounds of age was allowed under the 2004 Act where it had ‘a justified purpose that is objectively and appropriately founded and derives from employment policy, labour market or vocational training or some other comparable justified objective’ or where it related to qualification for retirement or invalidity benefits.34 The 2004 Act also allowed ‘justified different treatment, in due proportion, that is founded on a genuine and decisive requirement relating to a specific type of occupational activity and the performance of said activity’.35 Unlike the Framework Directive, section 7 did not require actions to be ‘appropriate and necessary’. However, proportionality is a general principle of the Finnish legal system, and should automatically be taken into account when applying the 2004 Act.36 That said, Hiltunen argues that the law would be clearer if the

26 ibid. 27 

2014 Act s 5(1).

28 ibid. 29 

ibid s 5(2). ibid s 4(1). 31  ibid s 7(2). 32 ibid. 33  ibid s 7(3). 34  2004 Act s 7(3). 35  ibid s 7(2). 36  Hiltunen (n 17 above) 28. 30 

Finnish Legal Provisions

 141

2004 Act had incorporated an express requirement that exceptions be ‘appropriate and necessary’.37 The 2014 Act narrows these exceptions, and brings them more explicitly in line with EU law. Now, different treatment does not constitute discrimination where: —— it is based on legislation, has an acceptable objective, and the measures to attain the objective are proportionate; or —— it is not based on legislation, but has an acceptable aim ‘in terms of basic and human rights’, and the measures to attain the objective are proportionate.38 Different treatment in employment can also be justified if ‘founded on genuine and determining requirements concerning the type of occupational tasks and their performance’ and the treatment is proportionate to achieve the legitimate objective.39 Finally, different treatment based on age can be justified if it ‘has an objectively and appropriately justified employment policy objective or an objective concerning the labour market’, or if it is related to age limits for qualification for retirement or invalidity benefits.40 The 2004 Act explicitly did not prevent positive discrimination ‘aimed at the achievement of genuine equality’ so long as it was appropriate to its objective.41 Further, the 2004 Act allowed ‘a procedure based on an equality plan, and intended to implement the intention of this Act in practice’.42 The 2014 Act is framed slightly differently, providing: ‘Proportionate different treatment that aims to promote de facto equality, or to prevent or remove the disadvantages attributable to discrimination, does not constitute discrimination’.43 With the passing of the 2004 Act, and now the 2014 Act, Finnish law reflects ‘a certain dualism’ between older acts and more recent legislation following the Framework Directive.44 However, preparatory works that guide legislative interpretation indicate that the law should be interpreted in accordance with the wording of the Framework Directive and the case law of the CJEU.45 This has alleviated any discrepancy between the acts.46 Under the 2004 Act, a court could award up to €16430 in compensation for suffering as a result of discrimination, or more where special cause existed.47 This limit was removed by the 2014 Act, which now provides for compensation which is ‘equitably proportionate to the severity of the act’.48 A court may also 37 

ibid 72–73. 2014 Act s 11. 39  ibid s 12(1). 40  ibid s 12(2). 41  2004 Act s 7. 42  ibid s 7(1). 43  2014 Act s 9. 44  Hiltunen (n 17 above) 5. 45  ibid 9. 46 ibid. 47  2004 Act s 9. 48  2014 Act s 24(1). 38 

142 

Finland—A Comparative Perspective

amend discriminatory contractual terms or declare a contract or any part of it to be void.49 Under the 2004 Act, the Discrimination Board (a tribunal for resolving discrimination matters) could prohibit specific discriminatory conduct and impose a conditional fine.50 The 2014 Act shifts this power to a new National NonDiscrimination and Equality Tribunal, which is also empowered to confirm conciliated settlements between parties and issue statements on the interpretation of the 2014 Act.51 Finally, the 2014 Act gives powers to a new Non-Discrimination Ombudsman, who is able to provide assistance to claimants, assist in the promotion of equality by authorities and employers, give general recommendations to prevent discrimination, take action to resolve a matter relating to the 2014 Act, and give reasoned opinions to prevent actions contrary to the 2014 Act.52 This may significantly assist claimants in enforcing the 2014 Act. Individuals may also seek damages under the Employment Contracts Act (55/2001) (Finland) if they incur loss due to an employer ‘intentionally or through negligence [committing] a breach against obligations arising from the employment relationship’ or the Act, including the obligation to treat employees equally.53 Discrimination in employment is also subject to criminal sanctions, with employers liable to a fine or imprisonment for up to six months for discriminating in recruitment or in employment ‘without an important and justifiable reason’.54 The criminalisation of discrimination in employment implies that discrimination is a social wrong, as well as an individual concern, and raises the apex of the ‘enforcement pyramid’ for equality law, potentially improving legal enforcement and compliance (see the discussion in Chapters two and three). Occupational safety and health authorities are responsible for supervising the prohibition of discrimination in employment.55 Authorities may receive ­communications from employees, carry out inspections, and report cases of ­probable discrimination to a public prosecutor.56

B.  Mandatory Retirement Finland has a DRA of 68 for the general workforce. Under the Employment Contracts Act (55/2001) (Finland) chapter 6, section 1a, employment relationships for employees other than civil servants are terminated without notice at the end of the 49 

2004 Act s 10; 2014 Act s 25. 2004 Act s 13. 51  2014 Act s 21. 52  ibid s 19. 53  Employment Contracts Act (55/2001) (Finland) ch 12, s 1. 54  Criminal Code of Finland (39/1889) (Finland) ch 47, s 3. Though this is rarely used: see below. However, in the context of sexual orientation, see Supreme Court of Finland, KKO: 2015:41 (10 June 2015). 55  2004 Act s 11; 2014 Act s 22. 56  See Act on Occupational Safety and Health Enforcement and Co-operation on Safety and Health at Workplaces (44/2006) (Finland). 50 

Finnish Legal Provisions

 143

month in which the employee turns 68, unless the employer and employee agree to continue the relationship, including on the basis of a fixed-term ­extension.57 Similar arrangements are in place for civil servants and municipal workers.58 Employers and employees may agree to a different retirement age, either in an employment contract or through collective agreement. However, negotiated retirement ages must comply with the 2004 or 2014 Acts. Contractual provisions may also be ‘adjusted or ignored’ under the Employment Contracts Act (55/2001) (Finland) chapter 10, section 2 if they are ‘contrary to good practice or otherwise unreasonable’. According to Hiltunen, many employers have adopted internal rules relating to retirement ages, and employers and employees often agree to include these rules in employment contracts.59 There have been few legal challenges to retirement ages in Finland. Since 1 January 2005, the Finnish pension system has used a ‘flexible retirement age’ of between 63 and 68. Individuals may draw their pension from age 63, though financial incentives encourage individuals to remain in work until the age of 68. Most employees choose to retire when they become entitled to a pension at the age of 63, well before they reach the DRA of 68.60 Indeed, in 2012 the ‘effective labour force exit age’ for Finnish men was 61.8 years, and for women 61.9 years.61 The DRA of 68 is therefore unlikely to be a contentious issue in Finland, as few employees remain in the workplace until that age.62 However, one expert noted the inherent contradiction of attempting to extend working lives while still allowing employees to be dismissed without reason at the age of 68 (Academic F1). To achieve a deeper cultural change, the expert believed that Finland must reconsider its upper retirement age (Academic F1).63

C.  Positive Responsibilities In addition to the negative prohibition of age discrimination, Finnish law makes provision for positive action by employers, including via the duty on authorities to ‘take necessary measures to promote the realisation of equality’;64 and by allowing ‘[p]roportionate different treatment that aims to promote de facto equality, or to prevent or remove the disadvantages attributable to discrimination’.65 57 

See further Hiltunen (n 17 above) 12. See further, ibid 77. ibid 78. 60  N Barr, ‘The Pension System in Finland: Adequacy, Sustainability and System Design’ (Finnish Centre for Pensions, 2013) 56–59. 61  OECD Older Workers Scoreboard. The Scoreboard calculates the effective labour force exit age as the weighted average of the exit ages of each five-year age cohort. 62  This hypothesis is supported by the Finnish organisational case studies, discussed in Chapter 7 of this volume. 63  This echoes concerns in the UK about the symbolic weight of allowing EJRAs: see Chapters 3 and 4 of this volume. 64  2014 Act s 5(1); cf 2004 Act s 4. 65  2014 Act s 9; cf 2004 Act s 7. 58  59 

144 

Finland—A Comparative Perspective

Companies in Finland also have positive statutory responsibilities towards the ageing workforce. For example, undertakings with at least 20 employees are required to prepare annual personnel and training objectives to maintain and improve the occupational skills of their employees. These plans and objectives must be prepared co-operatively with employee representatives and take the ‘particular needs’ of ageing employees into account.66 Finnish employers also have responsibility for protecting and maintaining workers’ wellbeing and working capacity. Under the Occupational Health Care Act (1383/2001) (Finland), Finnish employers must arrange occupational health care for their employees at the employers’ expense67 to ‘prevent and control health risks and problems related to work and working conditions and to protect and promote the safety, working capacity and health of [their] employees’.68 Employers must develop and revise annual occupational health care action plans.69 Decisions regarding occupational health care must be made in co-operation with employees or their representatives.70 Failure to comply with the Act may result in prosecution and a fine.71 The Act embodies the broader Finnish concern with promoting wellbeing, maintaining work ability72 and workplace health promotion,73 reflecting Finland’s strong tradition of occupational health and safety (NGO F3).74 As noted above, the 2014 Act has strengthened Finnish employers’ obligations in relation to equality. Authorities and all employers with more than 30 employees must now draw up a plan for the promotion of equality, including on the grounds of age, to analyse the state of equality and measures needed for improvement. Under previous legislation, plans only needed to be drawn up by authorities in relation to racial discrimination.75 Plans must also be developed in consultation with staff or staff representatives, increasing discussion and negotiation around equality issues. This reform places further weight on employers and workplaces to support and encourage diversity and equality. 66  Act on Co-operation within Undertakings (334/2007) (Finland) s 16; see also Adnett and Hardy (n 4 above) 35. 67  However, employers may receive compensation for the costs incurred: Occupational Health Care Act (1383/2001) (Finland) s 9. 68  ibid s 4. 69  ibid s 11. 70  ibid s 8. 71 ibid s 23. On occupational health, see also Government Decree on the Principles of Good ­Occupational Health Care Practice, the Content of Occupational Health Care and the Qualifications of Professionals and Experts (1484/2001) (Finland); and Act on Occupational Safety and Health Enforcement and Co-operation on Safety and Health at Workplaces (44/2006) (Finland). 72  See further text surrounding n 111 to 122 below. 73  See T Maltby, ‘Extending Working Lives? Employability, Work Ability and Better Quality Working Lives’ (2011) 10 Social Policy and Society 299. 74  This is also reflected in a long series of government programmes to promote work ability and wellbeing, including: FinnAge Respect for the Ageing Programme; Fitness for All Ages; the National Productivity Programme; National Programme for Ageing Workers; National Well-being at Work ­Programme; VETO Pull to Working Life/Well-being at Work and Extending Working Life; TYKE/ TYKES; Forum for Well-being at Work; Working Life Development Strategy to 2020; Socially Sustainable Finland 2020—Strategy for Social and Health Policy. 75  See further Hiltunen (n 17 above) 6.

Practical Implementation

 145

D.  Financial Incentives In addition to positive obligations on employers, Finnish law provides financial incentives to encourage employers to support older workers. For example, Finnish employers are required to pay disability premiums as part of the statutory pension scheme. Under the scheme, employers are divided into 11 different premium groups based on their disability claims over the last two years: the more employees who claim a disability benefit, the higher the employer’s group will be, incurring more costly premiums. Employers therefore have a financial incentive to promote occupational health and wellbeing for older workers, as well as statutory obligations. According to Raisanen and Hori, this has been the most effective policy measure aimed at reaching employment targets for older workers.76 This model of positive financial incentives is consistent with an experimentalist architecture of reflexive law (see Chapter two).

II.  Practical Implementation At the EU level, Finland is promoted as an exemplar of government and legislative programmes for encouraging the employment of older workers. However, the Finnish government’s focus has shifted away from older workers in recent years: reorienting from age to wellbeing and management more broadly (Government F4), from age management to generation management (Government F5, NGO F3),77 and from age discrimination to equal opportunity generally ­(Government F3). The implementation of programmes specifically for older workers is now ‘seen as old fashioned’ (Government F4), with the growing complexity of work and new challenges facing organisations ‘asking something more’ of government policies (Government F4, NGO F3). This change may reflect a maturing of government policies (Government F3, F4), making the practical implementation of age discrimination legislation less of a priority than in recent years. That said, there is still significant work to be done in Finland. The EU’s Country Specific Recommendations for 2013–14 recommended that Finland: Take further steps to increase the employment rate of older workers, including by improving their employability and reducing early exit pathways, increasing the

76  H Raisanen and H Hori, Employment Policies in Two Ageing Societies: Japan and Finland Compared (Government Institute for Economic Research VATT, 2008) 34. More blatant financial incentives have also been used to encourage the employment of older workers: between 2006 and 2010, low-wage subsidies were offered to support the employment of older workers. However, take up of the subsidies was mostly in the public sector. 77  See further n 124 below.

146 

Finland—A Comparative Perspective

e­ ffective retirement age by aligning the retirement age or pension benefits to changes in life expectancy.78

Similarly, the Recommendations for 2014–15 recommended that Finland: Improve the use of the full labour force potential in the labour market, including by improving the employment rate and the employability of older workers, and increasing the effective retirement age, by reducing early exit pathways and aligning the retirement age or pension benefits to changes in life expectancy.79

Therefore, ageing programmes should remain a national priority in Finland, and it is essential to consider their effectiveness in practice. Despite being lauded at the EU level, the majority of Finnish experts focused on the limitations of the Finnish national ageing programmes, rather than recognising their benefits.

A.  Enforcement of Age Discrimination Legislation The Non-Discrimination Act (21/2004) (Finland) has been in force since early 2004. However, despite this history, there have been very few cases brought under its provisions, and the majority of claims under the Act have been unsuccessful.80 Unlike in the UK, enforcement through individual litigation is regarded as a ‘last resort’ in Finland. Instead, age discrimination laws are more commonly supervised and enforced by occupational safety and health authorities,81 via negotiations between unions and employers where actions do not comply with a collective agreement, or by the police.82 This collectivist approach to enforcement reflects the ‘sense of common responsibility’83 typical of the Nordic communitarian legal

78  Council Recommendation of 9 July 2013 on the National Reform Programme 2013 of Finland and delivering a Council opinion on the Stability Programme of Finland, 2012–2017 [2013] OJ C217/24, para 3. 79  Council Recommendation of 8 July 2014 on the National Reform Programme 2014 of Finland and delivering a Council opinion on the Stability Programme of Finland, 2014 [2014] OJ C247/127, para 3. Similar provisions were included in the 2015–16 Recommendations: see Council Recommendation of 14 July 2015 on the 2015 National Reform Programme of Finland and delivering a Council opinion on the 2015 Stability Programme of Finland [2015] OJ C272/14. In contrast, no mention was made of the ageing workforce in the UK Country Specific Recommendations for 2013–14, 2014–15 or 2015–16. 80  See, eg, L Koto and P Viljakainen, ‘Finland’ in N ten Bokum and P Bartelings (eds), Age Discrimination Law in Europe (Deventer, Kluwer Law International. 2009) 112; Government F6. Though see KHO: 2011:107, where the Supreme Administrative Court ruled that the Decree on Police Administration requiring civil servants to resign at the age of 63 was in violation of the prohibition of age discrimination in the 2004 Act and the Framework Directive; and KHO: 2001:38, where the Court held that there was no substantial and acceptable reason for differential treatment on the basis of age in relation to redundancy measures. Anecdotal evidence also indicates that the 2004 Act was often raised as an alternative ground in cases regarding unfair dismissal (Government F6). 81  Ministry of Labour, Regulation of Discrimination and Working Life (Ministry of Labour, 2003) 4. 82  Ministry of Labour, For Equality, against Discrimination (Ministry of Labour, 2004) 26. 83  M Ahtisaari, ‘Finland’s Leap Forward’ (1999) 8 Presidents and Prime Ministers 7, 7.

Practical Implementation

 147

tradition84 and the broader Finnish policy emphasis on collective wellbeing at work.85 However, the lack of individual litigation has made the emergence of case law on discrimination issues ‘slow and sometimes selective’.86 This is inhibiting legal clarity in Finland, and government materials provide limited additional guidance to address the lack of judicial direction.87 Therefore, like in the UK, Finnish age discrimination law remains plagued by legal uncertainty. Legal clarity is essential if employers are to avoid costly legal mistakes, especially in relation to the implementation of positive measures (Government F6). These challenges may be addressed in part by the 2014 Act, which gives powers to the new Non-Discrimination Ombudsman to: provide assistance to claimants, assist in the promotion of equality by authorities and employers, give general recommendations to prevent discrimination, take action to resolve a matter relating to the Act, and give reasoned opinions to prevent actions contrary to the Act.88 This may significantly assist claimants in enforcing the 2014 Act, and may help to provide further guidance and support for employers in applying the Act. Further, the new National Non-Discrimination and Equality Tribunal is empowered to issue statements on the interpretation of the Act,89 which may help to promote legal clarity. Thus, the 2014 reforms to the legislative framework may help to advance individual enforcement and the implementation of age discrimination legislation in Finland. However, while noting the potential of the 2014 Act, it is debatable whether these new powers will be used in practice. Indeed, even existing collectivist enforcement mechanisms are used infrequently to address age discrimination in Finland. In 2010, the Southern Finland Division (a regional occupational health and safety authority that monitors 45.9 per cent of Finnish wage earners) received 17 complaints of age discrimination out of a total of 152 discrimination cases.90 Across all regional divisions, requests for information91 were sent to employers in only 13 cases of alleged age discrimination in 2010, out of a total of 113 cases.92 In relation to discrimination more broadly, 368 reports of labour discrimination were received by the Finnish police between 2005 and 2010; of these, 104 cases were

84  A Pylkkänen, ‘Transformation of the Nordic Model: From Welfare Politics to Gendered Rights’ (2007) 19 Canadian Journal of Women and the Law 335, 336. 85  See H Piekkola, ‘Nordic Policies on Active Ageing in the Labour Market and Some European Comparisons’ (2006) 58 International Social Science Journal 545. 86  Hiltunen (n 17 above) 6. 87  ibid 13. 88  2014 Act s 19. 89  ibid s 21. 90  Hiltunen (n 17 above) 90. 91 A request for information generally indicates that the inspector has reason to believe ­discrimination has occurred and has started to investigate the matter: L Larja et al, ‘Discrimination in the Finnish Labor Market: An Overview and a Field Experiment on Recruitment’, Employment and Entrepreneurship 16/2012 (Ministry of Employment and the Economy, May 2012) 100–101. 92  ibid 102.

148 

Finland—A Comparative Perspective

heard by the District Courts and 32 cases were heard by the Courts of Appeal.93 None of these cases related to age.94 More generally, between 1995 and 2006 there were no reported cases on the use of penal sanctions in Finnish discrimination cases.95 Discrimination may instead be addressed in Finland via negotiations between unions and employers at the local level.96 Enforcement through trade unions or works councils often leads to high levels of compliance and enforcement at a macro or societal level.97 However, there is no firm evidence of this occurring in Finland in relation to age discrimination. Indeed, equality measures are mostly contained within and advanced by legislation, rather than being negotiated via collective bargaining.98 The limited enforcement of age discrimination laws in Finland is problematic, and it remains to be seen whether the 2014 Act is able to remedy these challenges.

B.  Involvement of the Social Partners Finland has adopted a co-operative tripartite approach to implementing the ­Framework Directive and extending working life, involving negotiations between government, employer associations and trade unions.99 This reflects Finland’s long tradition of tripartite co-operation in developing and implementing labour market reforms.100 Finland is a ‘heavily organised country, heavily unionised’ and governments ‘listen with [a] very sensitive ear’ to the social partners (Government F1). The Finnish social partners are seen as more collaborative and 93 

ibid 108. ibid 109. 95  C Waaldijk and M Bonini-Baraldi, Sexual Orientation Discrimination in the European Union: National Laws and the Employment Equality Directive (The Hague, TMC Asser Press, 2006) 134. This limited use may reflect broader challenges with using criminal sanctions in discrimination cases, which are on the whole ‘remarkably underuse[d]’: ibid. This may be due to the higher burden of proof in criminal matters (and the inappropriateness of shifting the burden of proof, unlike in civil cases), the higher psychological cost of criminal proceedings, the greater separation of criminal proceedings from individual citizens, and the greater potential for political control of prosecutions: G Moon, ‘Race and Employment Directives: Remedies’ (The Fight against Discrimination in Daily Practice (Seminar), Trier, 4–5 June 2004) 3; Waaldijk and Bonini-Baraldi 134. As a result, some Member States view the criminal law as being ‘of limited use’ in discrimination matters: ibid 134. However, see Supreme Court of Finland, KKO: 2015:41 (10 June 2015), which related to penal sanctions for sexual orientation discrimination. 96  Ministry of Labour (n 82 above) 26. 97  J Malmberg, ‘Effective Enforcement of EC Labour Law: A Comparative Analysis of Community Law Requirements’ (2004) 10 European Journal of Industrial Relations 219, 223. 98 See, eg, General Agreement between TT and SAK for the Technology Industry (24.10.2011– 31.10.2013) ch 1. 99  Hiltunen (n 17 above) 5. This ‘collaborative’ process has been criticised for excluding NGOs and other bodies from the negotiations, limiting the ability of civic organisations to contribute to and influence the change process: ibid. 100  Hiltunen (n 17 above) 5. 94 

Practical Implementation

 149

less ­antagonistic than those in other countries (Government F3), allowing changes to be more readily accepted and implemented and ensuring important issues are raised at the policy development stage (Government F6). This is one of the key strengths of the Finnish system (Government F3).101 Reforms to support older workers are invariably devised, introduced and implemented in close collaboration with the social partners, who are key members of management groups for implementing government ageing programmes.102 ­Further, the social partners are responsible for implementing and developing aspects of ageing policy: in 2011, the Finnish government introduced a requirement that the social partners reach a common understanding that would lead to a long-term approach to extending working lives. In March 2012, the government and social partners concluded a strategy to help Finnish working life become the best in Europe by 2020 (‘Finnish Working Life Development Strategy’), which included an agreement on extending working life. The government and social partners will work collaboratively to implement the strategy as part of a broad national co-operation project in workplaces. Extensive tripartite co-operation in Finnish ageing policy has contributed to the overall success of the policies. Consensus between the social partners regarding the importance of extending working life has assisted significantly with the implementation of reform (NGO F3). Securing the agreement of the social partners before introducing legal reform means changes come to reflect a paradigm shift in Finnish thinking, rather than reflecting purely legal change.103 Further, Sarfati argues: across Europe, in countries where comprehensive reform projects were publicly debated and social partners closely involved in the reform formulation and implementation, they were successfully adopted and carried out. By contrast, where governments tried to unilaterally impose reforms, the projects invariably failed or only partially addressed the major issues at stake. It is therefore argued … that social dialogue, in the broadest sense, constitutes the ‘highroad’ for achieving consensus on feasible and acceptable reforms that deal with demographic ageing.104

101 However, this collaborative approach may be in jeopardy: some experts viewed economic ­ evelopments as upsetting the relationship between the social partners, variously noting the growth of d ‘mistrust and misunderstanding’ during these ‘tough times’ (NGO F2), the development of a ‘stalemate nowadays in the co-operation, especially at a central level’ (Government F1; similarly, Government F4) and the lack of a ‘common will’ between the social partners (Government F1). However, there is still a consensus among the social partners that wellbeing and productivity go ‘hand in hand’ (NGO F4) and there remains a ‘good spirit and good atmosphere’ between the social partners regarding wellbeing programmes, which make good economic sense for both unions and employers (NGO F2). Therefore, the collaborative approach is likely to endure in at least some areas. 102 R Arnkil et al, ‘The National Programme on Ageing Workers: Evaluation’ (Reports of the ­Ministry of Health and Social Affairs 2002/5, 2002) 9. For example, the social partners were included in the VETO management group and the NPAW advisory board (n 74 and text to n 6 above). 103  Ilmakunnas and Takala (n 1 above) 677. 104  Sarfati (n 5 above) 50.

150 

Finland—A Comparative Perspective

Similarly, the quality of regulation and its fit with existing practice increases if those affected are incorporated into the decision-making and implementation process.105 However, the need to achieve consensus can slow the pace of reform (Government F6), result in more limited measures being adopted106 and make legislation more difficult to understand and apply (Government F6). Indeed, objections from the social partners (and the threat of a general strike) prevented the Finnish government from unilaterally increasing the national ‘personal retirement age’ from 63 to 65 in 2009.107 Following two weeks of negotiations, the government and social partners committed to increasing the average age at which people leave the workplace by three years by 2025.108 This negotiated consensus between government and the social partners is a weaker, slower agenda for change than the government’s original proposal. The Finnish social partners also play a significant role in effecting change at a local level, particularly by educating employers about measures to accommodate and support an ageing workforce (Government F3). However, the experts identified a disjuncture between the policies and rhetoric adopted by the social partners at the national level, and practices adopted at the local level (Government F1), which can result in trade unions being too passive at a local level (NGO F1). This may limit the social partners’ role in encouraging change. That said, the social partners are strongly influencing employer conduct in some sectors: for example, the Finnish technology industry has adopted a project focused on promoting work wellbeing and best practice in extending working life, entitled ‘Good Work— Longer Career’. The programme includes 60 companies and over 5,250 workers.109 Thus, the involvement and active participation of the social partners at the local level varies significantly by industry and region.

C.  Implementation by Employers As in the UK, the implementation of age discrimination legislation in Finland varies significantly by company and sector (Government F1, F2, NGO F2, F5). While the Finnish experts saw implementation as ‘fairly good’ in many workplaces (NGO F3), and some employers are doing a ‘very good job’ at supporting

105  S Fredman, Discrimination Law, Clarendon Law Series (Oxford, Oxford University Press, 2002) 181; L Dickens and M Hall, ‘Fairness—up to a Point. Assessing the Impact of New Labour’s Employment Legislation’ (2006) 16 Human Resource Management Journal 338, 348. 106  See, eg, L Dickens, ‘The Road is Long: Thirty Years of Equality Legislation in Britain’ (2007) 45 British Journal of Industrial Relations 463, 467. 107  P Jokivuori, ‘Finland: Social Partners’ Involvement in Pension Reform in the EU’ (Eurofound, 4 October 2013): www.eurofound.europa.eu/eiro/studies/tn1304032s/fi1304031q.htm. 108 ibid. 109 See further Signing Minutes of the Collective Agreement between the Federation of Finnish Technology Industries and the Finnish Metalworkers’ Union (27 October 2011) ss 9–10.

Practical Implementation

 151

older workers (Government F3), it is still ‘far off the optimum’ (NGO F3). Indeed, ageing p ­ rogrammes are having only limited impact in the private sector, particularly among small companies (Government F2, NGO F5). While the need to extend working life is generally accepted by trade unions and employer associations, workplace managers are often not putting meaningful change into place (NGO F2). Many employees still experience different levels of training on the basis of age, and the majority of workplaces are not actively encouraging older workers to remain in employment (NGO F4). However, some employers are modifying work for their older employees, and some companies are bringing these issues onto the national agenda (NGO F5). The Finnish government has introduced a series of programmes since 1990 to promote better employer practice and extend working lives, including through: targeted information, training and dissemination of good practice; support for workplace development projects; strategic funding; and the creation of networks to connect experts, researchers, service providers and workplaces. In particular, the TYKES Workplace Development Programme, which ran from 1996 to 2010, sought to develop collaborative and innovative solutions to improve the quality of working life, by working with managers, researchers, authorities and the social partners to disseminate information and expertise, provide support for development projects, and promote co-operation. Further, recognising the particular challenges faced by small private employers, the ‘Carrot’ project (which ran from 2000 to 2003) analysed existing practice and promoted age management practices in small and mid-sized companies.110 The government agency Tekes also provides public funding for research, development and innovation projects that improve the quality of working life and worker wellbeing, including through the development of work organisations. Therefore, public funding is available for employer capacity building where it will support and enhance worker wellbeing. Dissemination of good practice also occurs through the employer-focused Government Leadership Development Network (Government F5) and the Age Management Training Consortium, a co-operative network designed to share ideas between organisational trainers (Government F5). Through these government programmes, Finnish researchers have developed and tested two key models to help employers support an ageing workforce, and promoted their adoption by Finnish workplaces. First, work ability is concerned with the physical, mental and social capacities of an individual to manage their role in the workplace,111 balancing personal factors with the demands of the job itself.112 Work ability encompasses individual characteristics such as education,

110  Warwick Institute for Employment Research (WIER) and Economix Research & Consulting, ‘Ageing and Employment: Identification of Good Practice to Increase Job Opportunities and Maintain Older Workers in Employment: Final Report’ (March 2006) 193. 111  J Ilmarinen and K Tuomi, ‘Work Ability of Aging Workers’ (1992) 18 Scandinavian Journal of Work, Environment & Health 8, 8. 112  J Ilmarinen, ‘Aging Workers’ (2001) 58 Occupational and Environmental Medicine 546, 548.

152 

Finland—A Comparative Perspective

Figure 14:  Work Ability House

knowledge, skill, experience and motivation,113 and targets four key areas: work demands and the environment; work organisation and the work community; workers’ health and functional capacity; and workers’ professional competence.114 An individual’s working ability can vary over time, both with changing human competencies and the evolving nature of work.115 Policies to enhance work ability embrace areas such as training and professional development, occupational health and safety, workplace (re)design, employee engagement and autonomy, and physical and mental wellbeing. To illustrate how work ability operates, Ilmarinen has developed the work ability ‘house’, where the multiple floors of the building represent the various factors influencing work ability (see Figure 14).116 Promoting work ability is thought to lower absence rates and disability costs, increase productivity, improve management, secure competent personnel, and 113  Ilmarinen and Tuomi (n 111 above) 8; see also J Ilmarinen et al, ‘New Dimensions of Work Ability’ (2005) 1280 International Congress Series 3. 114  K Tuomi et al, ‘Promotion of Work Ability, the Quality of Work and Retirement’ (2001) 51 Occupational Medicine 318, 318. 115  Ilmarinen, ‘Aging Workers’ (n 112 above) 548. 116  J Ilmarinen, Towards a Longer Worklife!: Ageing and the Quality of Worklife in the European Union (Helsinki, Finnish Institute of Occupational Health, 2005) 132–33.

Practical Implementation

 153

improve individual quality of life and wellbeing, both during work and during retirement.117 Improved work ability is associated with enjoyment in staying in work, fewer thoughts of retirement118 and lower levels of later disability.119 Therefore, work ability promotion is seen as a positive measure to reduce premature retirement and extend working lives.120 Activities to promote work ability are common in Finnish workplaces121 and focus on four key areas: adjustments needed in the physical work environment (such as ergonomics and adjustments to physical demands); adjustments needed in the psycho-social work environment (including effective age management and flexible working); health and lifestyle promotion (particularly exercise); and updating professional skills.122 Secondly, age management is seen as a key means of improving the psycho-social work environment within a work ability approach. Age management involves the consideration of age-related factors in the daily management, design and organisation of individual work tasks and the work environment.123 Good age management tends to include: a positive attitude toward ageing; support for team and group work; finding individual solutions which fit with changing strengths and weaknesses during ageing; and open-minded communication within organisations.124 Despite the Finnish government’s extensive research and development activities, the experts expressed concern that best practice and workplace innovation projects are not reaching the majority of employers (Government F1, F4, NGO F3). Companies are not aware of best practice and are not using information that is available (NGO F3). This reflects a ‘knowing-doing’ gap: while there is a significant amount of information available to Finnish employers, problems are experienced in accessing and implementing it.125 Indeed, as the amount of information increases, the pile of actual doing remains distant from the mountain of knowing, and the mountain seems to become more and more unapproachable because of an overload of information. The gap is widening. The pile of knowing also grows at a more rapid rate than doing, so doing is increasingly lagging behind.126 117  J Ilmarinen and J Rantanen, ‘Promotion of Work Ability during Ageing’ (1999) 36 American Journal of Industrial Medicine 21, 23. 118  Tuomi et al (n 114 above) 322. 119  Ilmarinen and Tuomi (n 111 above) 9; K Tuomi et al, ‘Summary of the Finnish Research Project (1981–1992) to Promote the Health and Work Ability of Aging Workers’ (1997) 23 Scandinavian ­Journal of Work, Environment & Health 66, 68; Ilmarinen and Rantanen (n 117 above) 21. 120  Tuomi et al (n 114 above) 318. 121 ibid. 122  Ilmarinen and Rantanen (n 117 above) 22. 123 Ilmarinen, Towards a Longer Worklife! (n 116 above) 120. 124  Ilmarinen and Rantanen (n 117 above) 22. As noted above, ‘age management’ may now be superseded by ‘generation management’, which arguably replaces a focus on older workers with an emphasis on workers of all generations. Indeed, existing age management approaches are criticised for their ‘implicit devaluation of the young and middle generation’: R Karazman, ‘Employment and Labour Market Policies for an Ageing Workforce and Initiatives at the Workplace: National Overview Report: Austria’ (Dublin, Eurofound, 2007) 2. 125  WIER and Economix Research & Consulting (n 110 above) 85. 126  ibid 189.

154 

Finland—A Comparative Perspective

The ‘knowing-doing’ gap is likely to compound over time, making communication and dissemination even more challenging. To address the gap, researchers and organisations can engage in co-creation and co-problem solving,127 as was done with the TYKES Workplace Development Programme. While this works ­‘somewhat’,128 the Finnish experience suggests that co-generated solutions do not then disseminate to other employers (Government F1, F4, NGO F3). Government programmes therefore do not ‘go into the deep roots of labour markets’ (Government F4). While the best organisations adopt measures as a result of government programmes, and some companies are progressive, ‘very few companies follow their path’ (Government F1, F4). The impact of workplace development programmes therefore ‘remains quite modest’, having only indirect influence on the broader workplace climate (Government F1). This reflects a broader difficulty with mainstreaming innovative practice (Government F1). Despite these challenges, the experts still firmly believed that employers could encourage older employees to remain in the workplace by systematically taking age into account (Government F1) and adopting a more mature approach to HR management (Government F4, NGO F1), thereby implementing and advancing government objectives (NGO F3). According to one respondent: ‘the real thing happens in the companies. So the company level is the most important’. (NGO F3; similarly, NGO F4) According to the experts, if one group should take the lead to address the ageing workforce, it should be companies: employers are in the best position to make positive changes to working life to ensure that workers can (with adequate health), will (with adequate motivation) and may (with adequate jobs) work longer (NGO F3).129 Managers therefore need the competence to think about how age might impact upon their organisational practices and how older workers could best be supported (NGO F3). At the same time, there are a number of obstacles to achieving this, including a lack of time and money and limited job descriptions for line managers (NGO F3).

D.  Attitudinal Change The Finnish government has made a concerted effort to change attitudes towards older workers (NGO F3). By raising awareness of age and ageing and the strengths of different age groups, government programmes have (at least to some extent) changed attitudes and created a new way of thinking about older workers (NGO F3).130 At the conclusion of the NPAW in 2002, the majority of those surveyed (59 per cent) felt that general attitudes towards older workers had improved

127 

ibid 190.

128 ibid. 129 

See also J Ilmarinen, ‘Promoting Active Ageing in the Workplace’ (European Agency for Safety and Health at Work, 2012) 6. 130  See Ilmakunnas and Takala (n 1 above) 676.

Practical Implementation

 155

over the duration of the programme.131 There is now less age discrimination in Finland following governmental interventions (Government F1). However, these results are far from overwhelming: according to Arnkil and others, ‘[a]t this stage, it is impossible to report any change (in one direction or the other) in age-based discrimination. It is clear, however, that the problem still exists’.132 Further, attitudinal change during the NPAW may have been the result of a ‘strong economic upturn’ rather than government programmes.133 Indeed, this whole process of change has been underscored by a growing scarcity of workers (NGO F1): attitudes have improved as ‘prime age’ workers become scarcer in recruitment (Government F1, F6, NGO F1). Demographic and economic trends may therefore be a stronger catalyst for attitudinal change than policy interventions.134 However, Sarfati still uses the Finnish case as evidence that age discrimination can decline.135 Attitudes towards older workers remain an issue in Finland (Academic F1, F2, Government F1, NGO F1, F3) and are still ‘largely negative’.136 Attitudinal change takes time (NGO F2): employees are still resistant to working longer and feel entitled to a pension income (NGO F1, F2),137 and employers have limited willingness to let their employees continue in work and often do not recognise the practical importance of longer work lives (Government F2). These concerns are consistent with survey evidence: according to a 2011 Finnish study, 54 per cent of employers think it is very or somewhat important to extend working careers; however, only 33 per cent of employers think that working careers will be prolonged at their workplace.138 The majority of employers therefore support the rhetoric of extending working lives, but few are taking action to achieve the goal in practice (Academic F1). Further, a 2011 survey of Finnish employees who retired in 2010 found that 11 per cent of respondents would have liked to continue at work, but believed their employer would not have accepted them continuing.139 Finally, employers’ support for employees working until the retirement age has remained constant since 2004,140 demonstrating limited change in employers’ attitudes over time. According to one expert, age discrimination continues to be seen as ‘natural’ in Finland, and ageing is still regarded as a social problem rather than an equality issue (Academic F1). As a relatively homogenous country, equality issues are 131  Arnkil et al (n 102 above) 14; see also I Virjo, The Ageing and the Labour Market in the Nordic Countries: A Literature Review (University of Tampere, 2004) 23. 132  Arnkil et al (n 102 above) 17. 133  Virjo (n 131 above) 25. 134  C-G Lindén, ‘Older Workers: A Mental Change for Finland’, Nordic Labour Journal, 1 March 2006. 135  Sarfati (n 5 above) 64. 136  Virjo (n 131 above) 25. 137  cf Government F1, F2 who felt that employees’ expectations of their retirement age had increased slightly in line with pension reforms. 138 E Tuominen, ‘Flexible Retirement Age in Finland: The Evaluation of the Finnish Flexible Retirement Scheme in Light of Employer and Employee Surveys’, Working Paper 3/2013 (Helsinki, Finnish Centre for Pensions, 2013) 23. 139  ibid 18. 140  ibid 22.

156 

Finland—A Comparative Perspective

still relatively novel in Finland (Government F6). Through the enactment of legal regulations and the actions of trade unions, individuals are becoming more aware of the issues around ageing and age discrimination (Government F6, NGO F3). However, there is still limited individual awareness of work adjustments that could be made for older workers (NGO F3) or legal rights if discriminatory action occurs (Academic F1). To achieve further attitudinal change, experts variously suggested utilising the business case for change (Government F3, NGO F2);141 conducting training for employers and other groups, which ‘is still powerful’ at achieving change (NGO F3); and providing organisational leaders and line managers with new competencies to manage an ageing workforce (NGO F3). According to one expert, systematic, coordinated training is most likely to effect the necessary changes to support older workers (NGO F3). To some extent these training programmes are already in place in Finland: the Finnish Institute of Occupational Health (FIOH) conducts an Age Power at Work programme which trains managers, HR professionals, adult education institutions, occupational health service providers and unions to encourage good HR management in the context of age. The programme seeks to raise and address age stereotypes using an evidence-based approach and encourages organisations to create an age plan for their workforce (Government F5).142 The programme is conducted both as an open course, with public enrolments, and in tailored sessions for individual companies and international governments (Government F5). The impact of the programme was, at the time of this study, being evaluated (Government F5) and it remains unclear whether the programme reaches employers most in need of help: as employers enrol themselves in the programme, individuals who do not self-identify ageing as an issue are unlikely to participate (Government F5). However, occupational health staff who attend the programme often influence the workplaces where they are deployed, and are frequently involved as expert advisors in company wellbeing programmes (Government F5). Further, insurers regularly run training programmes for the companies they insure, extending the programme’s reach (Government F5). At the individual level, FIOH was, at the time of study, piloting a programme to address worker attitudes called ‘58 and Beautiful Exit Plan’. The aim of the programme is to encourage individuals to create a workforce exit plan in advance, and encompasses education on a manager’s role in supporting work ability, competence and successful ageing; the economy of retirement and pensions; and creating a meaningful life in retirement ­(Government F5).

141  Some employers have moved beyond requiring a business case for change, having now internalised an understanding of the benefits of a diverse workforce (Government F3). For these employers, a financial return on their investment in diversity ‘doesn’t matter’ (Government F3). See further Chapter 7 of this volume regarding the drivers of organisational change. 142  This would now have legislative weight and backing, given the need to have an equality plan in place under the 2014 Act.

Comparative Analysis: The UK and Finland

 157

In sum, age discrimination laws in Finland reflect a strongly collectivist ­ rientation in how they are drafted, enforced and implemented, which stands in o marked contrast to the UK’s individualistic focus. This collectivist approach is also reflected in the experts’ attitudes towards law. Echoing the UK, some Finnish experts viewed law as specifying minimum levels of behaviour (NGO F1) as a ‘bottom line’ (Government F3) and basis for action (Government F6). However, other respondents went beyond this to talk about the symbolic ‘power’ of law (NGO F3): if you change the law, something will happen, though ‘what will actually happen is a different thing, and it’s hard to quantify the impact of … legislative change’ (Government F2). Law serves as a ‘strong anchor’ and provides tools and concepts for employers and other bodies (NGO F3). Law is therefore leading change and changing attitudes in Finnish society (Academic F1): while ‘raising awareness is great’, ‘if you want to change something, you need law’ (Academic F1). Indeed, law is necessary to encourage behavioural change by private companies, who are unlikely to act voluntarily (Academic F1, NGO F2).143 This reflects the broader view of law and legal impact held by the Finnish experts.

III.  Comparative Analysis of the UK and Finland Drawing on the material above, this section considers the similarities and enduring differences between the two jurisdictions, and provides commentary and statistical analysis of the impact of the two different approaches.

A.  Differences between the Jurisdictions Age discrimination laws in the UK and Finland demonstrate significant doctrinal differences. While both countries operate within the structure of the Framework Directive, the jurisdictions exhibit fundamental differences in the focus and execution of their age discrimination laws. This is consistent with Bell’s analysis of the extent to which EU Directives have encouraged the convergence of antidiscrimination law: while Directives may have encouraged Member States to take action in the equality field, ‘deeper scrutiny suggests that national models have not withered away under the influence of EU law. Europeanisation may have modified national practices, but there is still ample evidence of local diversity’.144 This is 143  For example, experts argued that there needed to be legislation requiring employers to pay disability pensions: without this economic incentive, employers would be less proactive in securing employee health and wellbeing (NGO F2). 144  M Bell, ‘The Implementation of European Anti-Discrimination Directives: Converging towards a Common Model?’ (2008) 79 Political Quarterly 36, 43.

158 

Finland—A Comparative Perspective

consistent with an experimentalist approach to EU governance:145 in this framework, national diversity provides scope for experimentation and the transmission of ideas between jurisdictions. Thus, acknowledging the differences between the UK and Finland provides scope for mutual learning and improvement. In brief, laws in the UK and Finland diverge in eight key areas. Four of these differences relate to the doctrinal provisions in place in each jurisdiction. First, the Finnish legislation reflects ‘a certain dualism’, with a general prohibition of discrimination in older acts, and more recent legislation following the Framework Directive.146 In contrast, the UK has no real history of age equality laws, and the terms of its legislation closely mirror the Framework Directive.147 Secondly, the Finnish legislation imposes a broad duty on authorities and employers to ‘promote the realisation of equality’, whereas the more limited duty in the UK requires that authorities (and not employers) merely have ‘due regard’ to the need to advance equality. Thirdly, while Finnish legislation establishes a DRA of 68 for the Finnish workforce, the UK abolished its national DRA in 2011, though employers may still justify mandatory retirement in limited circumstances. Fourthly, Finnish age discrimination legislation is enforced by occupational safety and health authorities, individual claims and the police, with age discrimination being regarded as both a civil and criminal wrong. In the UK, while the EHRC may take enforcement action in age discrimination matters, the ‘weight of enforcement’ has fallen on individual complainants and enforcement in the civil courts.148 The remaining four areas of difference relate to the execution and underlying foci of the national laws. First, Finnish laws and policies place far more emphasis on the role of employers in achieving change, and are significantly more proactive at providing legal penalties and incentives to achieve the government’s agenda at the local level.149 However, while Finnish employers are expected to do a lot more for the ageing workforce than those in the UK, they are not doing it on their own—they are supported by a significant network of government agencies and social partner institutions, financial assistance and extensive research and scholarship. This support structure is lacking in the UK. Secondly, Finnish law reform is grounded in tripartite collaboration with the social partners, meaning legal changes gain broad acceptance before they are adopted. In contrast, while the UK government undertakes consultation on draft legislation, this is not designed to achieve consensus, and it is clear that the government retains the 145  See CF Sabel and J Zeitlin, ‘Learning from Difference: The New Architecture of Experimentalist Governance in the EU’ (2008) 14 European Law Journal 271 and Chapter 2 of this volume. 146  Hiltunen (n 17 above) 5. 147 As Sargeant notes, ‘it is … impossible to know whether the UK Government would have ­progressed to [legislative intervention] without the need to transpose the Framework Directive’: M Sargeant, ‘The Employment Equality (Age) Regulations 2006: A  Legitimisation of  Age  Discrimination in  Employment’ (2006) 35 Industrial Law Journal 209, 214. 148  Dickens (n 106 above) 475. 149  However, this only extends so far: laws still recognise the need for space for local negotiation and adaption, to allow employers to adapt provisions to their particular needs and culture (NGO F2, F3).

Comparative Analysis: The UK and Finland

 159

last say in how laws should be designed, adopted and implemented.150 Thirdly, the Finnish focus on workplace occupational health and ‘work ability’ is firmly at odds with the UK emphasis on ‘employability’ and adapting individuals to the needs of the workplace.151 Finally, ‘equality’ in Finland is focused on achieving social cohesion and the even distribution of resources, rather than an individual right not to be discriminated against.152 The Finnish notion of equality is grounded in substantive equality of opportunity and equality of outcomes, in contrast to the UK’s focus on equal treatment and formal equality of opportunity (see Chapter one). That said, Finnish government policy is increasingly moving towards a focus on equality of opportunity: ‘Equality is a fundamental right. In a just society, everyone has equal opportunities in relation to access to education, employment and different services’.153 A key challenge for contemporary Finnish discrimination law is reconciling a traditional collectivist focus with the new push to achieve individual equality of opportunity.154 Therefore, like in the UK, ‘equality’ in Finland remains a contested concept. At the same time, Finnish policy is still more focused on substantive equality than that in the UK.155 The countries also diverge in their views of the role of law and government ­intervention. Unlike in the UK, Finnish experts emphasised the importance of effective legal implementation, and the broad role of government in leading and achieving change. Legal change ‘won’t help much’ (Government F1) without the help of development work, case studies, networks and training (NGO F1). ­According to one respondent, a ‘carrot is always a better means … than [a] stick … ­legislative changes alone won’t help without a more comprehensive policy approach’ (­Government F1). Finnish experts (including government r­ epresentatives) were far more supportive of government intervention than experts in the UK.

150 

See Sargeant (n 147 above) 218–19. See Maltby (n 73 above). 152 Pylkkänen (n 84 above) 340. This notion of equality endures in Finland, despite Finnish legislation increasingly reflecting an individualised, Anglo-Saxon ‘rights-based model’ of equality: E-M Svensson et al, ‘Introduction: Nordic Feminist Legal Studies at a Crossroads’ in E-M Svensson et al (eds), Nordic Equality at a Crossroads: Feminist Legal Studies Coping with Difference (Aldershot, Ashgate, 2004) 3; Pylkkänen (n 84 above) 340–41. 153  Sisäasiainministeriö, ‘Welcome to Equality.fi!’ (yhdenvertaisuus.fi, 2013): www.yhdenvertaisuus. fi/welcome_to_equality_fi/. 154  Svensson et al (n 152 above) 3. 155  That said, Svensson et al argue that Nordic notions of equality (including that in Finland) are ‘based on sameness to such a degree that it leaves little room … to recognize and appreciate difference’: ibid 4; see also J Kvist et al, ‘Changing Social Inequality and the Nordic Welfare Model’ in J Kvist et al (eds), Changing Social Equality: The Nordic Welfare Model in the 21st Century (Bristol, Policy Press, 2012) 5. Building discrimination law on a foundation of ‘similarity and assimilation’ and ‘equality as sameness’ has thwarted proactive attempts to achieve equality of results and undermines the need to promote equality more generally: Svensson et al (n 152 above) 4. Therefore, the Finnish notion of ‘equality’ may undermine equality of results and substantive equality of opportunity, rather than promote it. 151 

160 

Finland—A Comparative Perspective

B.  Similarities between the Jurisdictions While the countries demonstrate significant doctrinal differences, they also exhibit marked similarities. Both jurisdictions have limited legal clarity and few cases to illuminate their legal provisions, particularly in the area of positive action. Further, both jurisdictions have encountered difficulties with effecting attitudinal change towards older workers, and have struggled to disseminate knowledge and good practice to employers. While the Finnish government has been more proactive at introducing programmes and interventions to achieve these ends, there is limited evidence of success in both jurisdictions. The countries are also united in their current de-prioritising of older workers. Despite Finland’s early leadership in this area, the experts noted that there is less of a concerted focus on ageing in current social programmes (Government F1, F6), with a shift from age-specific programmes to general programmes (Government F4). In an era of government cutbacks (Government F4), choices have had to be made, and youth and race discrimination have been prioritised as more pressing issues than old-age discrimination (Government F4, F6).156 The experts felt there was therefore ‘not so keen [an] interest’ in ageing workers now (Government F4).157

C.  Evaluating the Different National Approaches It is difficult to evaluate the success of national interventions, particularly where programmes are designed to change attitudes toward older workers.158 While most Finnish programmes have complex monitoring and evaluation systems in place to assess their effectiveness,159 there are few evaluation processes in place in the UK. Therefore, the only viable way of comparing the different approaches at a macro level is to use statistical evidence of the extent to which the countries now offer opportunities for older workers to work, offer quality employment for older workers and promote equality for older workers.

156  Broader economic trends may also inhibit government initiatives: the majority of respondents noted that it was harder to achieve change in a bad economy (Governments F1, F4, NGOs F2, F4, F5), with previous improvements in the employment rate for older workers having been achieved in a good economy (Government F1). While the Finnish economy is not struggling as much as other European countries, it was still regarded as having ‘some worrying signs’ (NGO F2). As a result, managers are more focused on external factors, rather than focusing internally on their management practices and workforce needs (NGO F3). In contrast, one respondent felt that a bad economy was a good impetus for employers to ‘work better’ (NGO F1). 157 However, this may also reflect a ‘maturation’ of government policies to focus less on age: see further Chapter 9 of this volume. 158  See, eg, Ilmakunnas and Takala (n 1 above) 677. 159  Arnkil et al (n 102 above) 9. For example, the NPAW had an expert group appointed to monitor its implementation in areas such as employment and unemployment rates, calculated retirement age, work ability, vocational skills and workplace health promotion: ibid. Annual reports were published to report on the findings: ibid.

Comparative Analysis: The UK and Finland

 161

However, these statistics should be used with caution. The different pension arrangements in the two countries may significantly influence individual behaviour, skewing national statistics. While UK pensions are inadequate for many individuals, with a net pension replacement rate for an average earner of 48 per cent, Finnish workers are regarded as having very generous pension entitlements, with an average replacement rate of 64.8 per cent.160 Inadequate pension benefits in the UK may force individuals to remain in employment, driving up participation rates for older workers, whereas Finnish workers may have more financial choice regarding retirement. Therefore, it is important to consider both employment rates for older workers and the quality of work being undertaken. It is also difficult to attribute causal effect to legal interventions. The relationship between law and social change is far from clear, and there is no systematic account in the literature of the relationship between law and social reform.161 Further, quantitative data analysis poses a ‘fundamental problem of causal ­inference’.162 Therefore, caution is needed when using statistical evidence to ­evaluate the ­practical impact of different legal responses to age discrimination.163

(i)  Creating Opportunities for Older Workers to Work There is a general consensus in the academic literature that Finnish reforms have helped to extend working life.164 Figures 15 and 16 indicate that average effective retirement ages have increased significantly in both Finland and the UK since 1990. However, Finland still lags behind the UK, particularly in the average effective retirement age for men (in 2007–12, 61.8 years for Finnish men and 63.7 years for men in the UK).165 This indicates that fewer Finnish male workers are remaining in the workforce into old age, increasing the country’s dependency ratio and the demand on pension systems. However, employment rates for older workers indicate that Finland may be gaining ground on the UK. While the rates in both countries have increased ­significantly since 2002, the overall employment rate in Finland for 55–64 year olds now (just) surpasses that in the UK, rising from 47.8 per cent to 58.2 per cent between 2002 and 2012. While this is a significant increase, it is mostly confined to those aged between 55 and 59: the rate for 60–64 year olds is still lower than

160  OECD, ‘Pensions at a Glance 2011: Retirement-Income Systems in OECD and G20 Countries’ (OECD, 2011) 125. 161  R Cotterrell, The Sociology of Law: An Introduction (London, Butterworths, 1984) 68. 162  L Epstein and AD Martin, ‘Quantitative Approaches to Empirical Legal Research’ in P Cane and HM Kritzer (eds), The Oxford Handbook of Empirical Legal Research (Oxford, Oxford University Press, 2010) 903. 163  See L Dickens and M Hall, ‘Legal Regulation and the Changing Workplace’ in W Brown et al (eds), The Evolution of the Modern Workplace (Cambridge, Cambridge University Press, 2009) 345. 164  Sarfati (n 5 above) 64. 165  It is also arguable that this extension in working life is the result of early exit pathways being restricted or closed entirely, rather than reflecting better opportunities for older workers.

162 

Finland—A Comparative Perspective

65.0 64.0 63.0 62.0 61.0

United Kingdom Finland

60.0 59.0

57.0

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

58.0

Figure 15: Average Effective Retirement Age for Men, UK and Finland, 1990–2012 (Source: OECD)

64.0 63.0 62.0 61.0 United Kingdom 60.0

Finland

59.0

57.0

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

58.0

Figure 16: Average Effective Retirement Age for Women, UK and Finland, 1990–2012 (Source: OECD)

Comparative Analysis: The UK and Finland

 163

Table 11:  Employment Rates for Older Workers, UK and Finland, 2002–12 (Source: OECD Older Workers Scoreboard, 2012) Finland

UK

2002

2007

2012

2002

2007

2012

Employment rate 55–64 (per cent of the age group)

47.8

55.0

58.2

53.1

57.3

58.1

 55–59

65.0

67.8

74.0

65.4

68.9

70.8

 60–64

26.1

39.1

42.9

37.9

44.6

45.3

 65–69

5.3

9.7

12.5

12.3

15.2

19.5

Gender gap in employment, 55–64 (ratio men/women)

1.02

1.00

.95

1.39

1.35

1.28

Retention rates after 60 (per cent of employees t-5)

46.8

47.2

50.2

40.5

45.2

43.9

Hiring rate, 55–64 (per cent of employees t-1)

6.4

6.5

6.5

7.9

7.0

5.6

in the UK, and the rate for 65–69 year olds lags significantly behind (see further Table 11).166 Retention rates for employees after the age of 60 are significantly higher in Finland than in the UK, implying that employees in Finland are more likely to be able and willing to remain in work with the same employer. This indicates that the job market for older workers in the UK is more casualised than that in Finland (see further below). The gender gap in employment is also dramatically lower in Finland, though this may reflect broader social differences in the two countries, and higher female participation rates across the life course in Finland. Finally, the hiring rate for older workers in Finland now surpasses that in the UK, though it is impossible to divorce the effect of government programmes from broader economic trends and the education and career patterns of different cohorts.167 Despite this, the significant improvements in employment rates in Finland, and high retention rates for older workers, imply that government programmes have been at least partly successful.

(ii)  Improving the Quality of Employment for Older Workers It is difficult to examine the quality of employment available to older workers at a macro level. While longitudinal surveys have been undertaken on the work ­experiences of older workers in the EU, neither the UK nor Finland participated.168 Therefore, limited comparative data are available on employment quality. 166 

This may reflect the different state pensionable ages in the UK and Finland. WIER and Economix Research & Consulting (n 110 above) 66. 168  See the Survey of Health, Ageing and Retirement in Europe: www.share-project.org/. 167 

164 

Finland—A Comparative Perspective

Table 12:  Job Quality Indicators for Older Workers, UK and Finland, 2002–12 (Source: OECD Older Workers Scoreboard, 2012) Finland

UK

2002

2007

2012

2002

2007

2012

Incidence of part-time work, 55–64 (per cent of total employment)

20.7

16.1

16.8

29.5

28.3

29.1

Incidence of temporary work, 55–64 (per cent of employees)

6.2

6.7

6.9

5.6

5.0

5.1

20.7

17.4

17.3

18.8

18.6

20.3

Incidence of selfemployment, 55–64 (per cent of total employment) Full-time earnings, 55–59 relative to 25–29 (ratio)

1.25

1.28

1.27

1.16

1.23

1.28

The OECD Older Workers Scoreboard provides some data regarding the prevalence of part-time and temporary work and self-employment among older workers, and the ratio of full-time earnings of 55–59 year olds relative to those of 25–29 year olds (see Table 12). However, these data are ambiguous: while parttime and temporary work can indicate instability and precariousness of employment, older workers often wish to work part-time, making the prevalence of part-time work an indication of quality work opportunities.169 These data are impossible to interpret accurately without accompanying evidence of older workers’ preferences. Additional evidence of job quality may be derived from underemployment170 statistics for older workers. Part-time older workers have significantly less underemployment in Finland than in the UK (see Table 13). However, both the UK and Finland have underemployment rates for older part-time workers that are lower than the rates for the general part-time worker population. Further, as demonstrated by Figure 17, the growth in underemployment for UK older workers appears to mirror the increase in underemployment in the general workforce (and, therefore, perhaps relates to broader economic trends). 169  See D Smeaton et al, ‘Older Workers: Employment Preferences, Barriers and Solutions’, Research Report 43 (EHRC, 2009) 31. For further discussion of this point, see A Blackham, ‘Rethinking Working Time to Support Older Workers’ (2015) 31 International Journal of Comparative Labour Law and Industrial Relations 119. 170  ‘Underemployment’ is defined by Eurostat as ‘persons … working part-time who wish to work additional hours and are available to do so’.

Comparative Analysis: The UK and Finland

 165

Table 13:  Per cent Underemployment for Part-Time Workers aged 55–74, UK and Finland, 2008–12 (Source: Eurostat Labour Force Survey, 2013) 2008

2009

2010

2011

2012

Finland

1.7

2.1

2.1

1.9

2.0

UK

2.7

3.3

3.7

4.2

4.3

7.0

Per cent of active population

6.0 5.0 UK: 15–74 year olds

4.0

UK: 55–74 year olds Finland: 15–74 year olds

3.0

Finland: 55–74 year olds 2.0 1.0 0.0 2008

2009

2010

2011

2012

Figure 17: Per cent Underemployment for Part-Time Workers aged 55–74 and the General Part-Time Working Population, UK and Finland, 2008–12 (Source: Eurostat Labour Force Survey, 2013)

The quality of employment on offer for older workers in the UK and Finland may also be discerned from individual workers’ reasons for retiring. The 2011 Special Eurobarometer on Active Ageing asked respondents to identify, for a range of factors, ‘how important you think it is for why people aged 55 and over might stop working’. The results for this question are presented in Table 14.171 There is a significant association between country and perceptions of the influence of each of the factors on decisions to retire. Surprisingly, workplaces not being adapted to the needs of older workers was seen as being far more influential in Finland than in the UK (χ2(4) = 89.067 (p < .001), V = .211 (p < .001)), despite the Finnish focus on occupational health and work ability. This may reflect a greater

171 

Two additional factors, relating to age discrimination, are presented below.

166 

Finland—A Comparative Perspective

Table 14:  Factors Perceived as Influencing Retirement Decisions, By Country (All Respondents) (Source: 2011 Special Eurobarometer on Active Ageing) Factor

Per cent believe important

Pearson Chi-Square (χ2)

Cramer’s V Coefficient (V)*

UK n = 1004

Finland n = 1005

Workplaces are not adapted to the needs of people aged 55 and over

50.9

64.2

89.067***

.211***

People aged 55 and over often lack the skills for the modern workplace

60.8

58.6

49.157***

.156***

Pension and tax systems mean that it is not beneficial to continue working

61.9

43.9

215.327***

.327***

There are not enough opportunities to gradually retire by progressively reducing one’s working hours

72.5

69.6

88.738***

.210***

Many people retire early because they want to retire at the same time as their partner

65.4

59.2

50.023***

.158***

Care obligations towards grandchildren or other relatives prevent people from continuing to work

67.3

34.0

343.878***

.414***

*** Significant at p < .001 to 4 df.

expectation that workplaces should be adapted to the needs of older workers. The factors that had the strongest association between country and respondents’ perceptions were those unrelated to the quality of work, namely pension systems and family care obligations (both seen as more important in the UK). These data therefore provide limited evidence of the quality of work opportunities available in each country.

(iii)  Achieving Equality for Older Workers Finnish interventions are thought to have eliminated most openly discriminatory behaviour in employment, though the impact on more subtle age discrimination

Comparative Analysis: The UK and Finland

 167

Table 15:  Personal Experience and Witnessing of Old Age Discrimination in Employment, By Country (Source: 2011 Special Eurobarometer on Active Ageing) Experience of age discrimination

Personally discriminated against at work Witnessed at work

Per cent experienced UK n = 1004

Finland n = 1005

4.38

5.77

13.4

13.2

Pearson Chi-Square (χ2)

2.010 .020

* Significant at p < .05 to 1 df.

and the use of age in employer decision-making is less clear.172 That said, other studies have failed to find a persuasive link between government interventions and the prevalence of age discrimination.173 The 2011 Special Eurobarometer on Active Ageing also investigated the prevalence of age discrimination in employment within Member States. Respondents were asked: In the last two years, have you either been a victim of … discrimination because of your age yourself [because you were perceived to be too old] or have you witnessed age discrimination [because someone else was perceived to be too old] in [the workplace or when looking for work]?

This question is of particular interest as it explicitly asks about old age ­discrimination in employment, rather than investigating age discrimination more broadly.174 Table 15 presents the results for this question for the UK and Finland. There is no statistically significant relationship between country and the experience of age discrimination at work, either from personal experience or from witnessing ­discrimination against others. It is therefore difficult to argue that ­different ­legislative measures have a significant association with the prevalence of age ­discrimination in employment.

172 C O’Cinneide, ‘Comparative European Perspectives on Age Discrimination Legislation’ in S Fredman and S Spencer (eds), Age as an Equality Issue: Legal and Policy Perspectives (Oxford, Hart Publishing, 2003) 201. 173  See AC D’Addio et al, ‘Population Ageing and Labour Markets’ (2010) 26 Oxford Review of Economic Policy 613, 627–28; OECD, ‘Pensions at a Glance 2011’ (n 160 above) 70. 174  That said, the survey also had a number of limitations. First, it did not specify what constituted ‘too old’ (such as by defining a specific age group), leaving the boundaries of what was ‘too old’ to respondents’ subjective interpretation: Larja et al (n 91 above) 43. Further, extensive discussion of ageing and age discrimination in the EU may have cultivated ‘age sensitive’ respondents who were more likely to identify age discrimination when surveyed: ibid. Finally, as a ground of discrimination, age will be well represented in the Eurobarometer sample: everyone has an age. Therefore, higher incidences of age discrimination are likely: ibid 50. As both the UK and Finland were surveyed using the same instrument, these issues are less significant for the purposes of this chapter.

168 

Finland—A Comparative Perspective

Table 16:  Factors Perceived as Influencing Retirement Decisions, By Country (All Respondents) (Source: 2011 Special Eurobarometer on Active Ageing) Factor

Per cent believe important

Pearson Cramer’s V Chi-Square (χ2) Coefficient (V)

UK Finland n = 1004 n = 1005 As people get older they are more likely to be excluded from training in the workplace

71.9

65.6

101.441***

.225***

Older workers are not viewed positively by employers

68.2

67.1

70.843***

.188***

*** Significant at p < .001 to 4 df.

The Eurobarometer survey also examined perceptions of the impact of discriminatory behaviour on individual decisions to retire, as presented in Table 16. There is a significant association between country and perceptions of the influence of negative employer attitudes on decisions to retire (χ2(4) = 70.843 (p < .001), V = .188 (p < .001)), suggesting a moderate substantive relationship between the variables. Similarly, there is a significant association between country and perceptions of the influence of older employees being excluded from training on decisions to retire (χ2(4) = 101.441 (p < .001), V = .225 (p < .001)), suggesting there is an even stronger substantive relationship between the variables. Exclusion from training may represent another manifestation of age discrimination in employment. While there is no statistically significant relationship between country and ­personal experiences of age discrimination, there is a significant relationship between country and perceptions of the influence of age discrimination on decisions to retire. It is arguable that individual perceptions of others’ reasons for retiring may depict inaccurately the actual prevalence of age discrimination, p ­ articularly when the question was posed to all survey respondents, not just those who have retired. However, when the data are reanalysed to focus only on respondents who self-identified as retired or unable to work through illness, s­ imilar s­ tatistical relationships remain, as demonstrated by Table 17. It therefore appears that discriminatory behaviour is less influential on individual decisions to retire from the labour market in Finland than in the UK. However, the Eurobarometer data represent a cross-sectional snapshot, and cannot be used to indicate changes in age discrimination over time. It is therefore difficult to assess whether government interventions have made any significant difference to the prevalence of age discrimination in employment. This reflects the limitations of statistical analysis in this area.

Conclusion

 169

Table 17:  Factors Perceived as Influencing Retirement Decisions, By Country (Respondents who Self-Identify as Retired or Unable to Work due to Illness) (Source: 2011 Special Eurobarometer on Active Ageing) Factor

Per cent believe important

Pearson Cramer’s V Chi-Square (χ2) Coefficient (V)

UK n = 383

Finland n = 388

As people get older they are more likely to be excluded from training in the workplace

71.5

67.5

47.091***

.247***

Older workers are not viewed positively by employers

70.8

69.8

22.032***

.169***

*** Significant at p < .001 to 4 df.

IV. Conclusion The situation in Finland provides a number of telling lessons for the UK. In particular, it demonstrates that it is possible for legislation and government to introduce positive, proactive measures to respond to an ageing workforce. It also illustrates the utility of focusing on workplaces and employers as a means of supporting and protecting older workers. In all, Finland demonstrates that there is far more scope for legal and governmental intervention in the UK in this area. While Finland is seen as a ‘success story’ in managing demographic change, this chapter has demonstrated the enduring implementation challenges still being encountered in Finland, particularly in securing attitudinal change, facilitating effective enforcement and bridging the ‘knowing-doing’ gap. Further, it is unclear how transferable or replicable these initiatives would be in the UK.175 Finnish interventions are financed through high levels of taxation and public funding, which is generally antithetical to UK governments. In a period of austerity, it will be difficult to persuade any government to invest heavily in public programmes of this nature. Further, Finland has a very different industrial relations context to that in the UK, particularly in relation to collective bargaining and the involvement of the social partners.176 Given the decline of UK trade unions and the limited involvement of the social partners in UK policy formation, it is unlikely that a similarly collaborative approach to law reform could be adopted. Finally, Finnish

175 

Sarfati (n 5 above) 65.

176 ibid.

170 

Finland—A Comparative Perspective

workers generally have high levels of formal education (NGO F1, Government F1, F2), meaning it is easier to encourage skill development and less intensive ways of working. This is not universally the case in the UK.177 In addition to issues of transferability, it is difficult to make a positive case for the success of the Finnish reforms using available statistical evidence. While ­Finland now surpasses the UK in the overall employment rate for 55–64 year olds, has higher retention rates for employees after the age of 60, and discriminatory behaviour is less influential on individual decisions to retire, there are at least an equal number of statistical measures that indicate that UK laws are more s­ uccessful at improving workforce participation for older workers. At the same time, it is impossible to causally link government programmes with national statistical outcomes. It is therefore necessary to consider qualitative means of ­evaluating age discrimination measures, to determine whether some national approaches are more successful at extending working lives than others. This is taken up in Chapter seven, which considers the impact of different legislative frameworks on employer practice through organisational case studies in the UK and Finland.

177  See U Hotopp, ‘The Employment Rate of Older Workers’, Labour Market Trends (ONS, February 2005) 73, 76.

7 Explaining the Quantitative Findings—Organisational Case Studies Employers have an essential role to play in managing the implications of ageing, given they are in the best position to promote equality and achieve structural change at the workplace level.1 Indeed, experts in both the UK and Finland (as reported in Chapters four and six respectively) felt that organisations should lead the national response to the ageing workforce. To respond effectively to population ageing, employers must adopt more holistic, joined-up and coherent approaches to supporting older workers.2 However, the expert interviews reported in Chapter four and the quantitative analysis in Chapter five demonstrate that UK age discrimination laws are not encouraging proactive or holistic employer responses to an ageing workforce. Therefore, we need to consider how organisations can be encouraged to engage actively with demographic ageing and older workers, and what role law could play in achieving this end. This chapter presents five organisational case studies of ‘best practice’ employers of older workers in the UK and Finland to consider the role of law and age discrimination legislation in effecting organisational change, and draw out the impact of different organisational structures and national laws in promoting best practice. In particular, I address the following questions: —— Why do organisations adopt positive or proactive approaches to the ­employment of older workers? —— How can law encourage or support positive change at the organisational level? I begin with a discussion of what constitutes ‘best practice’ in relation to older workers and findings from the literature on the drivers of best practice in

1  S Fredman, Discrimination Law, Clarendon Law Series (Oxford, Oxford University Press, 2002) 176; S Fredman, ‘Breaking the Mold: Equality as a Proactive Duty’ (2012) 60 American Journal of ­Comparative Law 265, 271. 2  See T Weber and M Barton, ‘Employers’ Practices for Active Ageing: Final Synthesis Paper of the European Employers’ Organisations Project on Age Management Policies in Enterprises in Europe’ (10 December 2012) ii; Chapter 4 of this volume.

172 

Organisational Case Studies

­ rganisations. I then present the results of the case studies, and compare and o contrast the results with emerging theories in the literature. In this way, I use organisational case studies to expand and generalise existing theories on how legal structures influence organisational practice.3

I.  Why Organisations Adopt ‘Best Practice’ A number of studies have examined employer best practice in retaining, recruiting and deploying older workers. Policies that are highlighted in these studies include: the availability of flexible work and retirement options, including part-time work and bridge employment; ‘age friendly job design’, including adjustments for occupational health and safety; the availability of training for older workers; recruiting workers of all ages; not using a retirement age; and not using age as a criterion for selection in redundancy.4 Similar ideas emerge in the literature as options for increasing the retention of older employees.5 Despite extensive research describing what ‘best practice’ looks like, there is limited literature examining how and why organisations adopt these measures.6 There is also limited understanding of how organisations respond to legal change.7 However, the literature suggests that law plays at least some role in promoting organisational change and good practice: in surveys of HR managers about changes to their age discrimination policies, over 90 per cent of respondents identified legislation as an important or very important driver of change.8 This is supported by a 2007 study by the CIPD, where 32 per cent of respondents identified ‘legal pressures’ as the key driver of diversity management in their o ­ rganisation.9 Therefore, age discrimination legislation has ‘created strong ­coercive pressure’ on 3 

RK Yin, Case Study Research: Design and Methods, 5th edn (Thousand Oaks CA, SAGE, 2013) 21. B Casey et al, ‘Human Resources Strategies and the Third Age: Policies and Practices in the UK’ ­51–54, and P Taylor and A Walker, ‘The Employment of Older Workers in Five European Countries’ 25–35, both in Age and Employment: Policies, Attitudes and Practice (London, Institute of Personnel Management, 1993); K Bunt et al, ‘Age Positive Business Case Studies’ (2005) 11–12; W Loretto et al, ‘Older Workers and Options for Flexible Work’, EOC Working Paper No 31 (EOC, 2005) 64; S Bond et al, ‘Integration in the Workplace: Emerging Employment Practice on Age, Sexual Orientation and Religion or Belief ’, Research Report 36 (EHRC, 2009). 5  A Griffiths, ‘Healthy Work for Older Workers: Work Design and Management Factors’, D Hirsch, ‘Sustaining Working Lives: The Challenge of Retention’, W Loretto et al, ‘Flexible Work and Older Workers’, and C Phillipson, ‘Is Extending Working Life Possible?: Research and Policy Issues’, all in W Loretto et al (eds), The Future for Older Workers: New Perspectives (Bristol, Policy Press, 2009). 6  E Parry and S Tyson, ‘Organizational Reactions to UK Age Discrimination Legislation’ (2009) 31 Employee Relations 471. 7  LB Edelman, ‘Legal Ambiguity and Symbolic Structures: Organizational Mediation of Civil Rights Law’ (1992) 97 American Journal of Sociology 1531, 1532. 8  Parry and Tyson (n 6 above) 480. 9  CIPD, ‘Diversity in Business: A Focus for Progress’, Survey Report (March 2007) 6. 4 

Why Organisations Adopt ‘Best Practice’

 173

organisations to adopt age discrimination policies.10 That said, the role of law may be limited in some organisations, particularly where practices are already ‘ahead of the game’.11 Even in these companies, legal provisions can give a positive impetus to existing actions and provide senior management with an additional justification for e­ stablished programmes.12 While law may be a driver of change, weak or broadly worded legal provisions give significant scope for employers to determine what practices they will introduce (see also the discussion of ‘compliance’ in Chapter four and experimentalism in Chapter two). Edelman argues that organisations ‘mediate’ the impact of equal opportunity law by collectively constructing the meaning of legislative ­compliance.13 Given most organisational responses are never challenged in court, the ‘collective response to law becomes the de facto construction of compliance’:14 firms become ‘states unto themselves’15 and courts merely vet organisational ­initiatives.16 Therefore, while companies must conform with equal opportunity laws to gain ‘legitimacy’, managers negotiate the meaning of ‘compliance’ to preserve and maximise their managerial prerogative.17 Companies are therefore likely to adopt symbolic responses (such as formal equality policies) that ‘create a visible commitment to law’ but have limited practical impact.18 At the same time, Croucher and Kelliher argue that employers have discretion to adopt one of four responses to weak legislation: ignore the law; follow the letter, but not the spirit of the law; ‘embrac[e] the spirit of the legislation’ and put policies in place to achieve the broad policy aim; or go beyond the intention of the legislation.19 The question, then, is why workplaces adopt different approaches to equality law or different constructions of ‘compliance’; and why and how some employers (but not others) come to adopt best practice measures. The literature identifies

10 

Parry and Tyson (n 6 above). Colgan, ‘Equality, Diversity and Corporate Responsibility: Sexual Orientation and Diversity Management in the UK Private Sector’ (2011) 30 Equality, Diversity and Inclusion: An International Journal 719, 728. 12  R Croucher and C Kelliher, ‘The Right to Request Flexible Working in Britain: The Law and Organisational Realities’ (2005) 21 International Journal of Comparative Labour Law and Industrial Relations 503, 520. 13  Edelman (n 7 above) 1532. See also F Dobbin and EL Kelly, ‘How to Stop Harassment: Professional Construction of Legal Compliance in Organizations’ (2007) 112 American Journal of Sociology 1203. Dobbin is more specific, arguing that US personnel managers have ‘created a legal code internal to the corporation’ through the creation of ‘wave after wave of equal opportunity innovations’: F Dobbin, Inventing Equal Opportunity (Princeton NJ, Princeton University Press, 2011) 2. This specifically attributes the process of defining compliance to personnel managers, rather than organisations more broadly. 14  Edelman (n 7 above) 1568. 15  Dobbin (n 13 above) 2. 16  ibid 12. 17  Edelman (n 7 above) 1535. 18  ibid 1542; see also L Dickens and M Hall, ‘Legal Regulation and the Changing Workplace’ in W Brown et al (eds), The Evolution of the Modern Workplace (Cambridge, Cambridge University Press, 2009) 349. 19  Croucher and Kelliher (n 12 above) 510. 11  F

174 

Organisational Case Studies

six overlapping factors that may encourage proactive organisational responses to equality frameworks. First, a prevailing organisational culture that values diversity may promote a more proactive response to legal regulation.20 Organisations are more likely to adopt and integrate legal changes where reforms and legislative objectives reflect pre-established organisational practices21 and values. Further, companies with a strong diversity culture are more likely to adopt ‘an internally oriented assimilation strategy aimed at customized change’ that applies new laws conscientiously.22 In contrast, companies with a weak diversity culture may ‘go for an externally oriented strategy involving minimum compliance’.23 The values of an organisation (such as inclusiveness, social awareness, corporate social responsibility and social justice) therefore influence the adoption of age discrimination policies24 and may make equality an organisational priority, spurring more concrete managerial responses.25 Organisations are more likely to adopt good practice where they acknowledge and internalise social justice arguments by integrating them into their organisational culture. This reinforces the argument that organisational history matters (see Chapter five): the strong effect of corporate culture on organisational action creates a link between past and present responses to new societal norms.26 Secondly, business needs and the ‘business case’ may promote a more proactive organisational response.27 Where business needs and competitive pressures ‘fit’ with diversity measures, organisations are more likely to adopt and integrate diversity strategies into their organisational processes.28 In surveys of HR managers about changes to their age discrimination policies, skills shortages were an important driver of change for over 80 per cent of respondents.29 This prompted Parry and Tyson to conclude that: ‘Changes to policies may be due as much to a good business need to find scarce labour, and to deal with the threats of demographic change, as to a coercive response to the power of the state’.30 Similarly, in the 2007 CIPD study, respondents identified business factors driving diversity management as including the ‘business case’ (17 per cent of respondents), ­recruiting and retaining the best talent (12 per cent) and being an employer

20  See J Saari et al, ‘How Companies Respond to New Safety Regulations: A Canadian Investigation’ (1993) 132 International Labour Review 65, 65. 21 L Dickens and M Hall, ‘Fairness—up to a Point. Assessing the Impact of New Labour’s ­Employment Legislation’ (2006) 16 Human Resource Management Journal 338, 348. 22  See Saari et al (n 20 above) 66. 23  See ibid. 24  CIPD (n 9 above) 7; Parry and Tyson (n 6 above) 483–84; Colgan (n 11 above) 727, 729. 25  See M Armstrong-Stassen, ‘Human Resource Practices for Mature Workers—and Why Aren’t Employers Using Them?’ (2008) 46 Asia Pacific Journal of Human Resources 334, 343. 26  F Dobbin et al, ‘You Can’t Always Get What You Need: Organizational Determinants of Diversity Programs’ (2011) 76 American Sociological Review 386, 395; see also Dickens and Hall (n 21 above) 350. 27  Colgan (n 11 above) 728. 28  Dickens and Hall (n 21 above) 349. 29  Parry and Tyson (n 6 above) 480. 30  ibid 472.

Why Organisations Adopt ‘Best Practice’

 175

of choice (15 per cent).31 Therefore, market pressures may be more important than the terms of legislation in effecting change, particularly in tight labour markets.32 Indeed, McNair and others found that organisations only translate positive attitudes towards older workers into concrete actions when they ‘experience serious labour problems’ and, even then, will only act if they make the connection between skills gaps and shortages and the recruitment or retention of older workers.33 While the ‘business case’ may be a driver of positive change, this can be problematic: where driven by a profit motive, diversity strategies can be (and are) abandoned if market conditions change.34 Companies driven by the business case may have a more fragile commitment to diversity than those driven by cultural or other factors. Further, employers are often unaware of or complacent about demographic trends,35 meaning business needs are unlikely to promote proactive responses to ageing. Thirdly, organisational leadership may be a key driver of good practice. For example, commitment from a company’s board encourages the introduction of age discrimination policies36 and firms are more likely to adopt equality measures when managers advocate for them.37 While Meyer and Goes found that organisational leaders exerted less influence on innovative practice than organisational culture and decision-making processes, leaders could have a substantial impact if they championed a particular change or cause.38 Middle managers may also have significant influence on how equality laws are implemented in practice.39 Therefore, if there is a gap between senior management and middle managers, this may seriously affect the implementation of age-related practices. Further, a mismatch between senior leadership and a company’s individual employees may prevent individuals from making use of best practice policies.40 Many employees are not aware of their organisations’ policies or the availability of ‘best practice’ options,41

31 

CIPD (n 9 above) 6; see also Croucher and Kelliher (n 12 above). Croucher and Kelliher (n 12 above) 506, 520. 33  S McNair et al, ‘Employer Responses to an Ageing Workforce: A Qualitative Study’, Research Report No 455 (DWP, 2007) 128. 34  K Malleson, ‘Diversity in the Judiciary: The Case for Positive Action’ (2009) 36 Journal of Law and Society 376, 384. 35  W Loretto, ‘Work and Retirement in an Ageing World: The Case of Older Workers in the UK’ (2010) 5 Twenty-First Century Society 279, 283. 36  Parry and Tyson (n 6 above) 483–84. 37  Dobbin et al (n 26 above) 388. Dobbin et al also argue that managers are more likely to advocate for equality measures when they stand to benefit from the change: ibid. Therefore, female and black managers are more likely to be advocates for equality innovations. 38  AD Meyer and JB Goes, ‘Organizational Assimilation of Innovations: A Multilevel Contextual Analysis’ (1988) 31 Academy of Management Journal 897, 916, 918. 39  F McAndrew, ‘Workplace Equality: Turning Policy into Practice’ (November 2010) 12; see also McNair et al (n 33 above) 131. 40  McNair et al (n 33 above) 131. 41  S Vickerstaff et al, ‘Employers and the Management of Retirement’ (2003) 37 Social Policy & Administration 271, 279–82; S McNair and M Flynn, ‘The Age Dimension of Employment Practices: Employer Case Studies’, Employment Relations Research Series No 42 (DTI, June 2005) 9. 32 

176 

Organisational Case Studies

and ‘practically nobody’ makes use of organisational policies in this area.42 Therefore, while organisational leadership is important, it needs to penetrate from senior leaders to middle managers to individual employees to secure good practice. Fourthly, the national context in which a company operates may affect organisational responses to law. For example, in liberal market economies like the UK, employers are ‘expected to resist state legislative interference’ and ‘to work to maintain maximum freedom’.43 This differs from other national contexts: the Nordic communitarian legal tradition, which still prevails in Finland, regards individual rights as subservient to societal goals and norms.44 Given the enduring ‘sense of common responsibility’ in Finland, and the express rejection of ‘selfish ­individualism’,45 it is likely that employers will be less resistant to legislative ­intervention. Therefore, the history and national context of a company may affect its implementation of diversity policies.46 Fifthly, organisations may adopt proactive practices as a result of organisational ‘peer pressure’ or to compare ‘more favourably’ with their peers.47 Ollier-Malaterre and others describe this as a ‘benchmarking logic’, derived from pressure to model themselves on other successful organisations and adopt norms prescribed by professional associations and consultancies.48 Similarly, Dobbin argues that equal opportunity practices spread between US organisations via ‘a national network of [personnel] professionals’, meaning that organisations’ ‘internal corporate legal codes’ resemble each other.49 However, this does not explain why some organisations adopt better practices than others, unless this spread of good practice has not yet occurred in the UK.50 Sixthly, unions may play a significant role in promoting good practice.51 While trade unions have come to the equality agenda slowly and reluctantly, they are 42 A Wagner and U Leber, Early and Phased Retirement in European Companies: Establishment ­Survey on Working Time 2004–2005 (Dublin, Eurofound, 2007) 23–25; Loretto et al (n 5 above) 148. 43  Croucher and Kelliher (n 12 above) 509. 44  A Pylkkänen, ‘Transformation of the Nordic Model: From Welfare Politics to Gendered Rights’ (2007) 19 Canadian Journal of Women and the Law 335, 336. 45  M Ahtisaari, ‘Finland’s Leap Forward’ (1999) 8 Presidents and Prime Ministers 7, 7. 46  Colgan (n 11 above) 731. 47  A Ollier-Malaterre et al, ‘Looking up to Regulations, out at Peers or down at the Bottom Line: How Institutional Logics Affect the Prevalence of Age-Related HR Practices’ (2013) 66 Human Relations 1373, 1375. This is consistent with an experimentalist architecture, which focuses on peer review and scrutiny to secure change: see Chapter 2. 48  ibid. See also Dobbin et al (n 26 above) 388–89, 400; A Klarsfeld et al, ‘Social Regulation and Diversity Management: A Comparative Study of France, Canada and the UK’ (2012) 18 European ­Journal of Industrial Relations 309, 312. This could also be linked to business needs and pressure to be regarded as an ‘employer of choice’ (see above). 49  Dobbin (n 13 above) 11. 50 Dobbin et al also recognise that the spread of diversity programmes may stall well before ‘saturation’: Dobbin et al (n 26 above) 405. Therefore, organisational peer pressure should be considered along with internal pressure for change, such as organisational culture and leadership (discussed above). 51 See, eg, A Greene and G Kirton, ‘Trade Union Perspectives on Diversity Management’ in M Özbilgin (ed), Equality, Diversity and Inclusion at Work: A Research Companion (Cheltenham, Edward Elgar, 2009) 265–66.

Why Organisations Adopt ‘Best Practice’

 177

increasingly involved in campaigning and bargaining for equality.52 Bargaining may canvass issues such as pay and benefit discrimination, employment segregation, job access, work-life balance and harassment.53 Therefore, consistent with the findings in Chapter five, a union presence and collective bargaining may encourage employers to consider and address equality issues, thereby promoting best practice. In addition to factors that may encourage proactive responses to the ageing workforce, the literature identifies a number of barriers that may inhibit good practice. For SMEs, barriers to adopting good equality practices include: a small workforce, high employment of family members, limited time and resources, other organisational priorities, economic insecurity and no formal HR processes.54 ­Further, it is clear that not all employers are the same, and that the factors above will not weigh equally on each organisation. Baldwin identifies three types of employers, who will each react differently to legal change: —— well-intentioned and well-informed employers, who tend to be large companies; —— well-intentioned, ill-informed employers, who tend to be SMEs; and —— ill-intentioned, ill-informed employers.55 For those who are well-intentioned, not complying with rules is often the result of limited knowledge or limited ability (or inclination) to process the information needed for compliance. Those who are ill-intentioned fail to comply because rules have not been effectively enforced.56

52 G Kirton and A Greene, The Dynamics of Managing Diversity: A Critical Approach, 3rd edn (Oxford, Elsevier, 2010) 180; see also G Healy, ‘Business and Discrimination’ in RD Stacey (ed), Strategic Thinking and the Management of Change: International Perspectives on Organisational Dynamics (London, Kogan Page, 1993) 178–79. That said, unions are still criticised for being unrepresentative (‘male, pale and stale’), which may impair their ability to advocate effectively on equality matters: see Kirton and Greene at 189. Indeed, unions face their own internal equality issues: see ibid 189–92. 53  See Kirton and Greene (n 52 above) 182–83. 54  European Commission, ‘Continuing the Diversity Journey: Business Practices, Perspectives and Benefits’ (Luxembourg, October 2008) 18. 55  R Baldwin, ‘Why Rules Don’t Work’ (1990) 53 Modern Law Review 321, 324. 56  ibid 329. Baldwin’s approach may be compared with Kelman’s three levels of change (compliance, identification and internalisation: see Chapter 4), which places more emphasis on organisational culture than organisational ignorance and goodwill: HC Kelman, ‘Compliance, Identification, and Internalization: Three Processes of Attitude Change’ (1958) 2 The Journal of Conflict Resolution 51, 52–53. Kagan and Scholz take Baldwin’s categorisation a step further by identifying three ‘images’ of organisations to explain why firms fail to comply with law. These include: (1) firms as ‘amoral calculators’, where non-compliance is the result of economic calculation; (2) firms as ‘political citizens’, where non-compliance is the result of ‘principled disagreement’; and (3) firms as ‘organizationally incompetent’, where non-compliance stems from organisational failure: RA Kagan and JT Scholz, ‘The “Criminology of the Corporation” and Regulatory Enforcement Strategies’ in K Hawkins and JM Thomas (eds), Enforcing Regulation, Law in Social Context (Boston MA, Kluwer-Nijhoff, 1984) 67–68.

178 

Organisational Case Studies

Drawing on Healy,57 Kirton and Greene also posit four types of organisation that embody different approaches to the equality agenda: —— the negative organisation, which lacks an equality policy and may not comply with the law; —— the minimalist/partial organisation, which declares itself to be an equal opportunity employer but has a narrow business case orientation and ­implements few formal measures to achieve equality; —— the compliant organisation, which fulfils its legal obligations via a narrow business case orientation and formal policies; and —— the comprehensive proactive organisation, which adopts a business and social justice case for equality, and develops, implements and measures best practice initiatives.58 However, these categorisations do not distinguish between different types of best practice employers, or the varieties of best practice. Further, they do not help to explain how law and other factors may promote best practice.

II.  Overview of the Case Studies To advance our understanding of why employers adopt best practice measures, I conducted explanatory case studies with five organisations in the UK and F ­ inland that had been identified as ‘best practice’ employers in relation to older workers.59 Cases were selected to reflect different organisational structures, which were predicted to influence the workplaces’ response to older workers: of the five organisations, four were large enterprises, with the other being a small undertaking; two were in the private sector, two were local councils, and one was a hybrid (former government) entity (see further Chapter five). The organisations selected for the case studies represent critical cases for studying the change process, particularly given the organisations’ proactive commitment to supporting older workers.60 Thus, examining these cases contributes to better understanding and theorising about a larger collection of cases.61 The case studies comprised semi-structured interviews with HR directors, ­management (including middle managers), older workers and, where relevant,

57 

Healy (n 52 above) 182–83. Kirton and Greene (n 52 above) 205. 59 Including through the AARP International Innovative Employer Awards for age-friendly employment practices; Eurofound case studies; and the Employers Forum on Age awards. 60  MQ Patton, Qualitative Evaluation and Research Methods, 2nd edn (Thousand Oaks CA, Sage, 1990) 174. 61  R Stake, ‘Qualitative Case Studies’ in Y Lincoln and N Denzin (eds), Strategies of Qualitative Inquiry, 3rd edn (Thousand Oaks CA, SAGE, 2008) 123. 58 

Overview of the Case Study Organisations

 179

unions and members of equity committees.62 The interview data were triangulated with a documentary review of corporate policies and other documents (including committee minutes, reports and proposals) in each case study organisation.

III.  Overview of the Case Study Organisations A.  Case Study 1: UK Small Enterprise Case study organisation 1 (‘UK SME’) is a small UK engineering company that supplies forgings to the automotive industry and employs 21 workers. The organisation has previously been cited as a good practice employer for older workers and is listed as an Age Positive Employer Champion. I conducted interviews with two members of management, two office workers and two forge workers, with the forge workers being interviewed in a group interview. This was complemented by a review of organisational policies and previous studies of the company, including by viewing recordings of previous interviews and documentaries about the organisation. Three aspects of the company’s operations reflect good practice: first, the company operates without a retirement age63 and allows employees to work past the age of 65; secondly, it trains and recruits older workers64 (the most recently hired employee was aged 62); and, thirdly, it offers phased retirement and has effective succession planning measures in place.65 The company values employees of all ages: ‘The whole philosophy we adopt in employment terms is that whether you are 16 or 60 you have a part to play in our employment programme’.66 While the company has few or no formal diversity policies (UKSME2) (and struggled to find relevant policies when asked),67 the Company Rules contain an equal opportunity statement: ‘The Company aims to achieve equal opportunities for its employees and will not discriminate in respect of pay, gender, age, disability, sexual orientation or religious beliefs’. By recruiting and retaining older workers,

62  A classification scheme has been adopted to identify different respondents in each workplace (the UK SME ‘UKSME’, the UK large enterprise ‘UKLE’, the UK Council ‘UKCOU’, the Finnish council ‘FICOU’ and the Finnish company ‘FICO’): for example, the first UK SME respondent is identified with UKSME1; the second Finnish council respondent with FICOU2. 63  CIPD, ‘Retirement Practices—Making the Right Choice! An Employer’s Guide’ (DWP, 2010) 8. 64  DWP, ‘Good Practice in the Recruitment and Retention of Older Workers: Summary’, Research Report 303 (DWP, October 2001) 13. 65  DWP, ‘Employer Case Studies: Employing Older Workers for an Effective Multi-Generational Workforce’ (February 2013) 42. 66  DWP (n 64 above) 9. 67  This is not uncommon among SMEs: see M Winterbotham et al, ‘Fairness, Dignity and Respect in SME Workplaces’, Research Report 98 (EHRC, October 2015) ch 5.

180 

Organisational Case Studies

the company has been able to attract skilled employees in an industry where it can be hard to recruit staff, achieved very low staff turnover,68 and has found that older workers can support new workers.69 Further, the company does not appear to have experienced negative repercussions from abandoning fixed retirement ages: ‘We have no retirement age. There are no concerns about reduced progression opportunities and the process of training younger staff has been improved greatly by involving experienced, skilled, older workers as trainers on the company’s apprenticeship scheme’.70 While the company has been operating without a retirement age for a number of years, respondents noted that two older staff had been ‘retired’ around 18 months prior to the interview, though these employees were now working as consultants for the company and were ‘happy to retire’ (UKSME2).71 This is consistent with previous research, which has found that companies generally control who is able to work beyond retirement age.72 At the time of the study, the UK SME employed 21 workers, down from 70 in 200173 and over 400 in the 1970s (UKSME1).74 Following a turbulent corporate history in the 1980s and 1990s, and a management buyout in 2000, in the 2008 financial recession the automotive ‘world completely stopped’ (UKSME1), resulting in a loss of 80 per cent of the company’s business and putting severe financial pressure on the organisation. In response, the company made a number of staff redundant, particularly in office or administrative roles, and introduced flexible work practices such as banking hours; enforced holidays; two weeks on, one week off work arrangements; four day working weeks; asking employees to do tasks not in their job description; and pay cuts. This is consistent with previous research, which has found that companies adopted flexible work practices in preference to redundancy to manage the recession.75 The fact the company survived (where most others failed) was attributed to a ‘team effort’ (UKSME1), luck, and the fact that the company was ‘small and flexible’ (UKSME1). If the company had

68  According to management, ‘if [new employees] stay a week, likelihood is they’ll stay forever’ (UKSME1; similarly UKSME5). 69  Age Positive, ‘Age Isn’t an Issue: Employers’ Guide to Your 21st Century Workforce’ (Age Positive, 2008) 12; DWP (n 65 above) 43. 70  Quoted in CIPD (n 63 above) 8. 71 Rather than retirement, this may more accurately be described as redundancy with re-employment as a self-employed worker. Thus, ‘retirement’ may have been used to avoid paying any entitlement to statutory redundancy pay: see Employment Rights Act 1996, pt XI. 72  See V Beck, ‘Employers’ Use of Older Workers in the Recession’ (2013) 35 Employee Relations 257; V Beck and G Williams, ‘The (Performance) Management of Retirement and the Limits of Individual Choice’ (2015) 29 Work, Employment & Society 267. As one worker noted, if you are not a good worker and reach retirement age, management will ‘find something’ to enable them to let you go (UKSME6). However, the worker also noted that the company was open to employees staying at work beyond the age of 65 (UKSME6). 73  DWP (n 64 above) 23. 74  This reflects the broader decline in the UK forging industry over that period: in the mid-1970s, the UK produced 700,000 tonnes of forgings per year; in 2012, it produced only 70,000 tonnes (UKSME1). 75  See Beck (n 72 above) 262–65.

Overview of the Case Study Organisations

 181

been unionised, management felt that everything would have been more difficult (UKSME1). As one manager noted, ‘[we have] still got a business, but that’s all I can say’ (UKSME1). The company operates a day shift with 15 workers and a night shift with six workers. Most older workers are allocated a permanent shift, meaning they do not rotate from day to night shift, reducing the physical impact of shift work (which is more of an issue if a worker is rotating between shifts) (UKSME1). Of the 21 company employees, only two are aged under 30, with all remaining staff over 50 years of age, and four over 60. Company management consists of a General Manager, Finance Director and the company Chairman. When management were questioned as to why the company employed so many older workers, they noted that it had ‘just happened naturally’ and that the workforce demographic fluctuated with the financial wellbeing of the company: younger workers are more likely to be made redundant (UKSME1). Staff at the UK SME are employed on individual employment contracts that provide for one week’s notice of termination for forge workers76 and one month’s notice for office workers. The contracts also include a clause giving the company the right to ‘lay off employees without pay or to reduce your normal hours of work if in the view of the Company this is necessary’ with no less than 24 hours’ notice.77 There is no union presence in the company: while employees can be members of a union, management ‘don’t believe in unions anymore’, and believe they are more relevant for ‘bad employers’ (UKSME1).

B.  Case Study 2: UK Large Enterprise Case study 2 was of a large UK company (‘UK LE’)78 that employs nearly 90,000 employees, 73,000 of whom are based in the UK. The company is a former government utility that was privatised in the 1980s. The company has five main business groups, including a central ‘group function’ which is responsible for policies and practices across the organisation. Interviews were conducted with the company’s Director of Workforce Management and representatives from two different unions. This was complemented by a limited review of organisational policies, websites and secondary reports.79 The company operates in a rapidly changing context and in recent years the ‘business has changed dramatically’ (UKLE1). Change management is a constant

76  cf Employment Rights Act 1996, ss 86(1)(b), 86(3), which requires longer notice periods for employees with over two years’ service. 77  Though see Employment Rights Act 1996, s 148, which may mean that exercising this clause leads to a right to redundancy pay. 78  This company has asked not to be identified. As a result, references to secondary sources will not be provided and less detailed contextual information is offered. 79  Permission was sought (but not given) to speak with individual employees and review broader organisational policies.

182 

Organisational Case Studies

challenge for the group. As the business changes, many staff in redundant roles are put on the ‘redeployment route’, retrained and placed in new roles. Demographic change is also a ‘big’ challenge for the organisation, though the company declined to provide statistics relating to its employees’ age profile.80 The company notes that it has a significant number of people who are at the upper end [of the age spectrum]… who are 55 and over, 50 and over. We’ll get a number of people who are actually 60 and over as well. So that’s something we’ve been dealing with over the last few years (UKLE1).

The company’s workforce is highly unionised and aligned with two main trade unions. The company prides itself on its ‘record of stable industrial relations and constructive relationships’ with unions (UKLE1). This was echoed by the unions, which saw the relationship as being ‘very mature’ (UKLE2). The company’s employees are generally members of one of two pension plans: a defined benefit plan (with mostly older workers, and now closed to new staff); and a defined contribution plan (with mostly younger workers). General workforce measures that assist and support the company’s ageing workforce include: —— health awareness programmes, led by a Chief Medical Officer and Chief Medical Group, consisting of education, support, and fitness campaigns; —— equal access for all employees to performance support, coaching and training; —— retraining and reskilling programmes, including through the company’s ‘redeployment route’; —— a commitment to ‘treat[ing] everybody equally no matter what age’, including through inclusion policies and practices and an inclusion group; and —— flexible working arrangements: in 2011, around 80 per cent of the company’s workforce was working flexibly.81 The company has previously been recognised for its age-friendly employment practices. In the past, researchers have commented favourably on the company’s flexible retirement scheme, which allowed employees to work part-time or flexibly, take sabbatical leave, be seconded to a voluntary organisation, or gradually reduce their hours. At the time, the programme was lauded as a way of increasing the company’s effective retirement age. However, during the course of this case study, the company described this scheme as ‘basically a [voluntary] leaver scheme’ to be used during an economic downturn (UKLE1). The company removed its NRA in 2006. This led to a substantial increase in the number of staff working after the age of 60, from 25 workers to over 2,000 workers in three years. 80  However, according to secondary sources, in 2011 32% of the company’s UK employees were aged 50 or above. 81 However, during the interview the company implied that this figure has since declined. Further, the company noted that ‘flexible working’ mainly consisted of flexible start and finish times, not ­homeworking (UKLE1).

Overview of the Case Study Organisations

 183

C.  Case Study 3: UK Council Case study 3 was of a UK city council (‘UK council’).82 The council is responsible for streets, parks, car parks, refuse, housing and approving planning applications. The council employs around 1,000 staff. For this case study, an interview was conducted with the council’s head of HR and its union representative. This was complemented by an extensive review of organisational policies, websites, statistics and secondary reports. The council is very successful at retaining employees and has very low staff turnover: in March 2013, 42 per cent of staff had worked for the council for 10 years or more. However, there is ‘more dynamism than the statistics imply’ (UKCOU1), as staff frequently move and are redeployed to different internal positions. This long average job tenure has implications for change management, as many staff have an ‘expectation things will continue in [the same] way’ (UKCOU1). At the same time, employees have strong loyalty to the council as an employer: ‘people are very passionate about what they do and enjoy what they do’ (UKCOU1). Long job tenure does not necessarily mean the council’s workforce is ageing: many staff join the council at a young age, meaning the workforce profile is still relatively youthful. In March 2013, 20 per cent of the council’s staff were aged over 55. Over 30 per cent of staff were aged between 45 and 54, but only 12 staff members (or 1.14 per cent) were aged 65 and over. This age profile is not unusual for a local authority in the UK. As an employer, the council does not regard ageing as a major issue, as its workforce is broadly distributed across the 25–64 age band and is ‘continuing to renew within those bands’ (UKCOU1). Equality is a central concern of the council, both in relation to the services it provides to the community and the composition of its workforce. According to the council’s values statement, it: ‘aim[s] to eliminate prejudice and discrimination, and to promote good relations between different groups’. The council is working proactively to make its workforce more reflective of the population it represents, focusing particularly on increasing the numbers of young people, people with disabilities and people from black, Asian or minority ethnic communities it employs. A key challenge for the council is to make its jobs attractive to young people, and to promote working at the council as a ‘career’. The council offered apprenticeships to two young people in 2012–13. While older workers are not a target group for the council, in 2012–13 it employed 15 new employees aged between 55 and 64 (nearly 10 per cent of all appointments). Older staff are also encouraged to attend training, and 16 per cent of those attending training in 2012–13 were aged 55 and older. However, take up of training is ‘very much a personal choice’ (UKCOU1).83 Equalities training is provided to staff and managers, with all new staff required to attend. 82  The council has asked not to be identified. As a result, references to secondary sources will not be provided and less detailed contextual information is offered. 83  That said, the UK council’s union representative noted some practical issues in the take up of training opportunities, particularly for manual workers, as ‘there’s pressures on time for people to get

184 

Organisational Case Studies

The council undertakes comprehensive reporting on its workforce, including by monitoring disciplinary and grievance cases. These findings are reported to the council’s Equalities Panel (which has elected, public and staff representatives) and published publicly on the council’s website. The council also has a number of structures in place to promote equality. Most pertinently, the council’s Equalities Panel is tasked with leading and overseeing the council’s equalities work; the Equalities Champion, a member of the council’s senior management team, is responsible for ‘championing all equality areas’; and the Joint Equalities Group is responsible for developing, delivering and overseeing the council’s equality and diversity programme across employment, service delivery and community relations. The council also has an equality and diversity policy that explicitly mentions age. In 2013–14, the council was working to ensure that, as a minimum standard, their employment policies and practices were non-discriminatory and compliant with equality legislation. Further, equality impact assessments are now used in developing the council’s employment policies to ensure the proper consideration of equalities in decision-making processes. To complement this, the council has introduced training to help council staff carry out effective impact assessments. Despite these measures, in a 2011 staff survey, only 74 per cent of the council’s staff agreed that the council provided equality of opportunity regardless of age, 19 per cent neither agreed nor disagreed, and 7 per cent disagreed. This was the lowest positive response of all the grounds of discrimination surveyed. Further, only 54 per cent of staff aged over 61 agreed that the council provided equality of opportunity regardless of age. Prior to the introduction of the DRA, employees at the council ‘just worked’ regardless of age (UKCOU1). When the DRA was introduced, the council started actively managing the retirement process, and used the retirement procedures under the legislation. However, the council regarded the DRA as having ‘created all these barriers which can tie you up in knots’ (UKCOU1). Since 2011, the council has elected to remove its retirement age. As a consequence, there has been ‘some change’ (UKCOU1) as some employees choose to stay beyond the age of 65. However, only a few staff each year wish to stay on, creating only a ‘slight’ increase in the council’s age profile.84 The council now views working without a retirement age as ‘business as usual’ (UKCOU1). The council has made flexible working available to all staff for a ‘long time’, and is happy to accommodate career breaks and measures to promote work-life ­balance (UKCOU1). The council also supports flexible retirement and is ‘very open’ to part-time or job share arrangements for older workers (UKCOU1). For the council, flexible working, part-time work and flexible retirement have ‘definitely’ encouraged employees to work for longer, as they give staff ‘options to do things differently’ and to achieve work-life balance (UKCOU1). released for training’ and training courses were not ‘particularly proactive’ in providing workers with transferable skills (UKCOU2). 84 

This may also reflect the generous pension scheme in place at the council: see further below.

Overview of the Case Study Organisations

 185

D.  Case Study 4: Finnish Local Council Case study 4 was a metropolitan city council in Finland (‘Finnish council’). The council is responsible for providing a broad range of services, including social services and health care, public works (such as streets and parks), rescue services, education and early childhood education, water, energy, museums, transport, urban planning, sporting facilities, libraries, and a zoo. The council is governed by an 85-member City Council, which is elected every four years. Council administration is managed by a 15-member City Board, which is elected by the City Council for two-year terms. The Finnish council is divided into 35 City Departments or City Group Companies, in addition to the City Executive Office that encompasses the Administrative Division, Economic Development and Planning Division, Economic Development, Legal Services, Information Technology and Communications, and Human Resources. The council has 40,000 employees, and around 4,000 managers. This diversity means there are ‘so many different kinds of people working here … and … different things that you have to take account of ’ in planning (FICOU1), including the distinction between physical and ‘specialist’ work. Terms and conditions of employment vary markedly from department to department, and different departments experience significantly different HR challenges. For this case study, a group interview was conducted with three council representatives, including the Chief of Wellbeing at Work, a HR specialist and the legal counsel (who was also a member of the council’s central age management group). This was complemented by a review of organisational policies, websites and statistics where available in English. Ageing is a ‘very important issue’ for the council. In 2013, 42 per cent of the council’s permanent employees were aged over 50. That same year, 24 per cent of new permanent employees were also aged over 50. The council has previously been identified as a good practice employer for the ageing workforce: it has ‘developed holistic approaches for age management in a physically demanding work setting’85 and is ‘regarded as a trail-blazer for the municipal sector in Finland for these strategies’.86 In 2009, the council’s strategic programme was amended at the political level to include the objective of becoming ‘a model town for the management of different age groups’ (FICOU1). Funding was made available for pilot projects and initiatives across the council’s departments, and a central age management group was established to monitor and guide the initiatives. Over the course of the four-year strategy, activities included: —— awareness-raising activities (particularly from 2009–10) to familiarise the workforce with issues around age and ageing; 85 Warwick Institute for Employment Research (WIER) and Economix Research & Consulting, ‘Ageing and Employment: Identification of Good Practice to Increase Job Opportunities and Maintain Older Workers in Employment: Final Report’ (March 2006) 66. 86  ibid 159.

186 

Organisational Case Studies

—— creation of a website to share and model concrete examples of best practice across departments. It was hoped that departments and individuals could then adapt these ideas to their own work setting and ‘customise it to their own needs’ (FICOU3); —— initiation of over 10 pilot programmes tailored to the circumstances of ­individual departments, including: —— Good Age 60+, a one-week residential programme offered to employees aged 60 or over after every two years of service. The programme was aimed at increasing employees’ physical fitness and extending working lives, and encouraged employees to consider and discuss how they could remain at work for longer. —— the Active Ageing pilot, which allowed older workers undertaking heavy physical work to take paid days off to undertake health and fitness training. The programme was aimed at reducing the number of sick leave days taken by older workers and to ‘help ageing employees to cope at work’; —— the development of action plans for ageing and retirement by individual departments within the city (at the time of interview, 33 of 35 departmental action plans had been completed), as well as a strategic and city-wide plan; and —— lectures and meetings to allow departments to share and discuss pilot p ­ rojects and best practice. The programme was funded out of the council’s budget for workers’ w ­ ellbeing, and was allocated around €100,000 per annum. The Active Ageing pilot was funded separately, and cost around €200,000 per annum.87 The council also has a number of ongoing programmes that support and assist older workers, including: —— a formalised shadowing programme, where central funding is available to allow older workers and their replacements to be employed concurrently for between three and six months. This allows older workers to train their ­successor and impart tacit knowledge, facilitating succession planning; —— training for all managers in good management practice, including by ­requiring managers to complete an online test (approximately 2,000 of the council’s 4,000 managers had completed the test at the time of interview);

87  Specific programmes have also been implemented in individual council departments to respond to particular demographic challenges. The programmes introduced in the Finnish council’s Public Works Department (PWD)—which is responsible for planning, building and maintaining streets, parks and the council’s work facilities, and employs around 1,817 people (approximately 58% of which are blue-collar workers)—are particularly noteworthy: Eurofound, ‘City of Helsinki Public Works Department, Finland: Redeployment, Training and Development’ (Dublin, Eurofound, 24 May 2005): www.eurofound.europa.eu/areas/populationandsociety/cases/fi006.htm. With the support of the Finnish government’s Workplace Development Programme, a strategy was introduced that allowed individual workers and managers in the PWD to negotiate to: adjust working time on an individual

Overview of the Case Study Organisations

 187

—— internal lectures and educational sessions for managers on management issues (though not on age management); and —— requiring conversations about retirement intentions ‘in a positive way’ (FICOU1) in annual work discussions for all employees over 59 years of age. The council is also supportive of older workers who wish to work beyond the personal retirement age of 63: ‘we are really happy if someone continues working after that’ (FICOU3). However, in accordance with Finnish law, from the age of 68 ‘it’s not like your own decision to work … you need to … make a deal with your manager’ to continue working on a fixed-term contract (generally for six months to a year in duration) (FICOU3). While it is ultimately the supervisor’s decision as to whether an employee can keep on working, ‘of course we are supporting it whenever it’s possible’ (FICOU3). Although retirement ages are still in place at the council, this affects a very small minority of the workforce: most workers have retired by the age of 63, with an average workforce retirement age of 63.1 in 2012. As a result of the council’s ageing programmes, awareness of ageing issues is increasing, and various age-related programmes have resulted in employees having increased motivation to work and more positive feelings towards their employer: ‘“I think that my employer really respects me, because they give me something like this”’ (FICOU1). The programmes have also helped to extend working life: the council’s average retirement age increased by 2.5 years between 2009 and 2012. Further, the number of people working beyond the personal retirement age of 63 increased from 650 in 2010 to 773 in 2012. While it is difficult to establish a causal connection between the programmes and improved retention of older workers, the respondents were convinced of the link between the two: ‘this ageing management group, like it has had something to do with it, it must have’ (FICOU3). At the same time, some programmes that were hoped to generate cost savings (including by reducing levels of sick leave) have not proven as successful as had been hoped. In particular, the Active Ageing pilot has not generated the savings that were originally predicted, though there has been some small reduction in the level of sick leave taken by participants. However, this limited impact may be attributable to the small scale of the pilot (with only 50 employees) and the short time frame for evaluation and follow-up (two years). It is possible that the pilot will have longer-term benefits. Regardless, the programme clearly had benefits for employee motivation and morale: ‘the people, they were very enthusiastic about the programme, and now people are still calling me [to ask], “Could we start a pilot like this?”’ (FICOU1). basis; adjust earnings according to individual working time or tasks; provide one week of fitness leave each year for employees over the age of 45; provide ‘age leave’ of up to 20 days per year for employees over the age of 56 (depending on individual age, health, workload and so on); and make arrangements for a ‘smooth transfer’ to retirement, including through knowledge transfer activities and working as a trainer or on specific projects post-retirement: WIER and Economix Research & Consulting (n 85 above) 160. The programme aimed to support and promote employees’ physical and mental work capacity and to improve employee motivation through ‘planning of the individual work-life span’: Eurofound.

188 

Organisational Case Studies

E.  Case Study 5: Finnish Manufacturing Company Case study 5 was of a private Finnish food manufacturing company (‘Finnish company’). The company is family owned, and employs approximately 1,700 people across five locations in Finland and Estonia. Around three-quarters of the company’s employees are blue-collar workers, and some factories rely on physically demanding work, including heavy lifting. A group interview was conducted with three company representatives, including the Director of HR, a HR manager and an older worker. This was complemented by a review of organisational policies, websites, secondary reports and statistics where available in English. Ageing is not an issue for the company as a whole, but is affecting some smaller factories, where the average age of employees is around 47. Some factories may eventually experience a shortage of skilled labour as employees retire. The company is also experiencing difficulties in recruiting younger workers, as the work is regarded as ‘not very sexy for young people’, being ‘wet, dirty, cold, … noisy’, involves operating old machinery, and young people sometimes lack the physical fitness required of workers (FICO1). However, the company is very successful at retaining workers, with an average work career of 14.6 years for blue-collar workers and 16.4 years for white-collar workers. The family ownership culture also extends to the workforce: at one point, three generations of one family were working at a factory at the same time. Between 1986 and 1994, early disability pensions cost one of the company’s subsidiaries more than €3 million.88 In response, the company implemented comprehensive measures to improve workplace health, work ability and working conditions.89 These initiatives ‘[continue] as part of the company’s normal routine’.90 Since 2004, the company has also maintained a Senior Programme for older workers. As part of the programme, blue-collar workers aged over 55 with five years or more of service are encouraged to undertake individual developmental discussions and develop individual career plans, which may include a focus on rehabilitation, work content, work environment, physical fitness measures, work ability, training and education, and/or coaching.91 The programme also allows older workers to undertake a lighter workload without a reduction in pay. The programme was initially introduced at one factory, before being rolled out to all blue-collar employees. It is financed by the company, with indirect assistance from

88 Eurofound, ‘Ruoka-Saarioinen Oy, Finland: Comprehensive Approach’ (Eurofound, 29 October 2009): www.eurofound.europa.eu/areas/populationandsociety/cases/fi002.htm. See also n 108 below. 89  WIER and Economix Research & Consulting (n 85 above) 98; Eurofound (n 88 above). 90  Eurofound (n 88 above). 91  WIER and Economix Research & Consulting (n 85 above) 98; Eurofound (n 88 above).

Overview of the Case Study Organisations

 189

the pension institute,92 and run by the company’s health promotion group in ­co-operation with employee groups.93 Around 50 per cent of employees over the age of 55 participated in the programme in 2009,94 rising to 64 per cent in 2014 (or 178 of the 279 employees who were eligible) (FICO1).95 While there has been no systematic evaluation of the programme, the initiatives ‘were key factors’ in improving job satisfaction, extending working careers, and reducing turnover and sick leave.96 Since the implementation of the programme, the proportion of the company’s employees who are aged over 60 has increased from 3.5 per cent to 6.3 per cent of the workforce.97 The average retirement age has increased by around two years from a starting point of ‘way below 60 years’ (FICO1). The company has also witnessed ‘a big cultural change’ away from early retirement at age 58, with employees now aiming to retire at 63 (the ‘personal retirement age’, and the age at which a pension may be drawn without financial penalty) (FICO1).98 While changing statutory pension entitlements are likely to have driven this shift, it is probably also related to the company’s programme. The company has also noticed positive improvements in attitudes towards older workers and reduced age discrimination at work.99 According to the company, ‘the seniors, they are more valued, more appreciated nowadays’ (FICO1), and there is more age-integration across the workforce. Similarly, ‘inter-group relations have been improved. It cannot be measured, but it has … been noted’ (FICO1). The company’s early disability pensions and insurance costs are now lower than the industry average.100 The programme was awarded the Finnish National Prize for Innovative Practices in Employment and Social Policy in 2010.101 As in the Finnish council, mandatory retirement is not a major issue for the Finnish company: around 98 per cent of staff will retire by the age of 65, well before the statutory retirement age of 68. The more relevant challenge, then, is to extend working lives beyond the personal retirement age of 63.

92 

For example, the pension institute runs courses and assists with employee rehabilitation. Eurofound (n 88 above). 94 ibid. 95  The company also has a ‘positive attitude’ towards part-time retirement and shared work and paid sabbatical arrangements, and supports employees taking part-time pensions and temporary leave (which provides 90 to 359 days’ leave at a salary of 70% of the unemployment benefit): ibid. 96  WIER and Economix Research & Consulting (n 85 above) 98. 97  Sitra, ‘The Finnish Metalworkers’ Union, the Federation of Finnish Technology Industries and Ruoka-Saarioinen, Awarded for Achievements in Wellbeing at Work’ (Sitra, 7 October 2010): www.sitra.fi/en/news/working-life-prize/finnish-metalworkers-union-federation-finnish-technologyindustries-and. 98  See also A Pärnänen, ‘Does Age Matter in HR Decision Making? Four Types of Age Policies in Finnish Work Organizations’ (2012) 2 Nordic Journal of Working Life Studies 67, 77. 99  WIER and Economix Research & Consulting (n 85 above) 98. 100  Eurofound (n 88 above). 101  Sitra (n 97 above). 93 

190 

Organisational Case Studies

IV.  Case Study Findings A.  Drivers of Best Practice (i)  Business Factors Most organisations recognised that their age-aware practices were driven (at least to some extent) by their economic or practical self-interest: for example, for the UK LE, work practices were grounded in what was ‘good for the business’ (UKLE1). In the UK SME, there was a clear consensus that age policies were good for business, which was a key driver in their adoption: ‘[we] do what is right by the business’ (UKSME1). Employing older workers made fiscal and practical sense for the company. Older workers were highly valued and sought-after for their experience, particularly if they had previously worked ‘somewhere dirty’ in heavy engineering (UKSME1). Further, as the machines used by the company dated back to the 1960s, older workers were more likely to be familiar with the machines and how they operated (UKSME2; similarly UKSME5). The respondents recognised that while younger workers could be trained in these skills, they were also more likely to be ‘poached’ by other companies. As a member of management noted, employees in their 30s were likely to stay with the company for five to ten years, whereas workers over 55 were likely to remain with the company for the rest of their working lives: ‘no one will try to pinch’ them (UKSME1). The Finnish council also noted its desire to benefit from older workers’ know-how. Most organisations also recognised and valued the benefits of being seen as an ‘employer of choice’ for older workers, both in terms of beneficial publicity (Finnish company, UK SME) and in helping to recruit and retain staff (Finnish council, UK council). According to the Finnish council: It’s really important for [our] image as an employer to be flexible and that [we] be able to customise things for the workers, when we cannot really compete with the [salaries in the private sector] as we are a public employer. We need to have these other kinds of things that we can use [to attract and retain staff]. (FICOU3)

Similarly, the UK council prided itself on being ‘an attractive employer of all ages’ (UKCOU1).102 Finally, programmes in both the Finnish council and Finnish company were introduced to reduce sick-leave absences and early retirement due to ill health (and their associated financial costs): ‘it stemmed from a clear need that okay, these people, they are not coping well, we need to do something. Problem-based’ (FICO1). Thus, legislation and statutory financial incentives influenced organisational business factors. This illustrates the potential of an experimentalist architecture for effecting organisational change (see Chapter two). 102 

See also Weber and Barton (n 2 above) 6.

Case Study Findings

 191

While statutory financial incentives were absent in the UK, the UK council union representative acknowledged their potential impact: at present, there is ‘little incentive’ for employers to proactively respond to older workers’ needs: They don’t really need to do that because, you know, they’ll get people working for longer anyway … [employers are unlikely to respond proactively] unless there’s a real cost to them around ill health retirement. (UKCOU2)

(ii)  Organisational Leadership In both the Finnish and UK councils, political agendas and leadership were strong drivers of the councils’ age and equality initiatives. In the Finnish council, ageaware practices had been in place for a number of years. However, by introducing ageing as a strategic priority for the council, and allocating funding to the initiative, the politicians provided significant impetus for new and varied ageing programmes.103 When ageing became a high priority for the politicians, it elevated ageing as a priority across the organisation. Political support of the ageing agenda, and individual departments’ accountability to the political group, were strongly conducive to the programme’s success (and, more particularly, to the ability of HR to get 33 of 35 departments to complete ageing action plans): ‘They wouldn’t do it if we couldn’t say that the politic[ian]s are waiting for the plans’ (FICOU1). Equality issues were also high on the agenda of UK council members, ensuring equality matters had a ‘high profile’ across the council more broadly (UKCOU1). The political persuasion of the authority was a strong influence on the council’s practices. Similarly, for the Finnish company the engagement and leadership of its owners and managers were integral to the success of the programmes, particularly when combined with a consultative approach: The management was really enthusiastic about this, but it was done in co-operation, you know, with all levels, with workers, with management, with [middle managers]. And that was one point why it was a success. It was not given from, you know, the ivory tower, but it was made in co-operation with every party. (FICO1)

In contrast, the Finnish council viewed leadership and training of middle managers as a secondary issue, and something that should come after awareness-raising and concrete pilot programmes: leadership and their knowledge about these things [should be a priority]. But it’s not the first one … In a way, the ground or the soil has to be worked before that, and when the ground has been done, then it’s time for the leadership to step in. (FICOU1)104 103  The politicians introduced ageing as a strategic priority without any encouragement from the council’s HR function: ‘it was the politic[ian]s who put it there in the strategy paper’ (FICOU1). Apparently the politicians were aware of earlier ageing programmes the council had implemented, and some had relatives participating in the programmes. 104  This reflects the nature of the council’s management team, which includes around 4,000 individuals across numerous departments, a large proportion of which work in blue-collar or front-line service roles.

192 

Organisational Case Studies

(iii)  Organisational Culture Organisational culture and a culture of equality were also strong drivers of ageaware practices in the case study organisations. In the UK SME, management aimed to treat people ‘the way you want to be treated yourself ’ (UKSME1). While company management was ‘on the other side of the fence’ to workers, the workforce was small enough that management were still in tune with what the workforce needed (UKSME1). According to management, the business was ‘really just a big family’, where it was ‘all about working together’ (UKSME1).105 Similarly, the Finnish company viewed its programmes as being driven by its culture as a family-owned company: It’s also in our culture or values, we are a family-owned business. And our owners, they appreciate, you know, long careers … we need to help older people, you know, cope. So it’s a value or culture thing as well. (FICO1)

The company was keen to protect and support its workers: ‘we want our people to, you know, stay long, stay healthy, be happy’ (FICO1). The Finnish council was also driven by altruistic cultural factors: employing older workers and extending working lives were seen as part of its societal responsibility as a local authority (FICOU1). The importance of workplace culture was also echoed by the UK council, which had ‘deeply held cultural values’ that made dignity and respect important issues (UKCOU1). The UK council’s ‘culture of equality’ was seen as a key driver of organisational practice (UKCOU1). In contrast, the UK LE did not mention cultural or value-based drivers relating to its policies. However, the company’s union representatives noted that its culture was shifting away from its traditionally family-oriented approach, leading to a ‘feeling of betrayal’ among employees (UKLE2).

(iv)  Law and Legislation Legal provisions were seen as having limited impact on organisational practice for the UK SME and the two councils. According to the UK SME, legal changes had ‘no impact on the ground’ for the company (UKSME1, 2): ‘nothing [has] changed over the last few years’ that had ‘radically changed things’ (UKSME2). The councils (rightly) believed they were operating above and beyond what was required by law in this area: This is extra when you think about the legislation in Finland. We have all kinds of ­different and very variable and wide possibilities that the legislation gives us. [But] this 105  However, this ran both ways: while management would endeavour to accommodate individual needs (so long as they did not impact on the business), employees were also expected to accept some of the risk of the enterprise. Any disruption to the business would affect the company’s ability to pay the wage bill, and employees ‘can’t hide from that’ (UKSME1). Management felt there was an ‘acceptance’ by workers of the economic ‘reality out there’, which necessitated stricter working conditions and fewer benefits (UKSME1). For example, staff were not given the bank holiday in 2011 for the wedding of the Duke and Duchess of Cambridge as the company could not afford it (UKSME1).

Case Study Findings

 193

is something … extra to it. We have gone further, and we believe we need to. (FICOU3; similarly UKCOU1)

However, the organisations still regarded law as having some role to play in encouraging employers to adopt a more age-aware and flexible approach to their workforce. In Finland, the legislation provided ‘some pressure’ on employers to ‘become more flexible’ (FICOU3). At the same time, law could not solve all ­problems—it was also up to each workplace to adopt ‘good practices’ (FICOU3). Similarly, the UK council saw the abolition of the DRA as being ‘just another element in the mix’ leading to a change in procedures: its abolition was not seen as having a particularly ‘dramatic impact’ (UKCOU1). Ironically, it appeared that the introduction of the DRA had more of an impact on the council than its ultimate abolition, as the council started actively managing the retirement process using the legislative procedures. Thus, law and legislation can have unanticipated and potentially deleterious effects on organisational practice. Even the PSED106 had not dramatically changed the UK council’s organisational practices: the council was already fulfilling its responsibilities before the law was introduced, and had ‘always’ reported its workforce equality data to the Equalities Panel. At the same time, the level of detail in council reports had increased since the PSED came into force, and the duties provided increased visibility for council statistics. Indeed, the council believed it was ‘more sophisticated’ about workforce monitoring than private sector organisations because of the PSED (UKCOU1). Law may therefore play some role in promoting good organisational practice in the public sector, and legislation may have enhanced the council’s pre-existing practices.107 In contrast, for the Finnish company and UK LE, legal provisions were a significant driver of corporate practice. ‘Money reasons’ (FICO1) and financial penalties imposed by legislation were a significant incentive for the Finnish company to reduce illness and early retirement and were a key driver in the company’s adoption of its Senior Programme: ‘that’s also a fiscal reason to try to keep these people working until they’re 63’ (FICO1).108 Further, because of the substantial financial costs of sick leave and early retirement, the company did not feel the need to assess the financial costs or benefits of the Senior Programme: ‘we are 99 per cent sure that it means that we have less early retirements, and lower sickness absence. We can count what it costs, but it would be just an educated guess [and not worthwhile]’ (FICO1). 106  As a public authority, the UK council has a statutory duty to have due regard to equality issues and impacts in policy and decision-making (the ‘general equality duty’): see EqA s 149 and Chapter 3 of this volume. Further, it has a specific duty to publish information to demonstrate its compliance with the general equality duty: Equality Act 2010 (Specific Duties) Regulations 2011, SI 2011/2260, reg 2. 107  That said, while there was now more openness, the council suspected that ‘nobody’s watching’ (UKCOU1). 108  In Finland, large companies are required to pay a sick employee’s salary for up to three months before a state-paid social benefit is available, and must also pay pension insurance to fund early ­retirement due to illness: see Chapter 6 of this volume.

194 

Organisational Case Studies

Similarly, the UK LE closely monitored legislative requirements, and sought to anticipate and pre-empt legislative change in its employment policies: these changes don’t just arrive and we try and make sure, you know, we’re ready or ahead of the changes where we can be … the DRA … and changes to pensions … we’re trying to sort of stay ahead of the game on those. (UKLE1)

Rather than wait for legislative change, the company worked with the CBI and other groups to monitor the government’s mood, and sought to create HR ‘products’ that proactively met the government’s emerging agenda, managed the ‘impact’ of legislative change and ‘support people through those impacts, and also obviously, you know, keep the business successful’ (UKLE1). At the same time, the company viewed itself as having a proactive role in relation to legislation, and a responsibility to effectively implement legislation once it was passed: Things like legislation need to be embraced, you know. It’s not always perfect, but it needs to be embraced and it needs to be implemented once it’s there, you know, and if it’s not good for your business, then you need to … you may need to make representations about that, but you need to bring it in. (UKLE1)

Therefore, for the UK LE, legislation (or the prospect of legislation) had a significant impact on the company’s practices and programmes. However, the company also demonstrated a proactive, positive view of its own role in implementing legislative provisions. Law may also encourage organisations to improve their practices by subjecting employers to age discrimination claims.109 However, this did not appear to be a factor in the case study organisations. None of the organisations could recall being subject to an age discrimination claim: according to management of the Finnish company, ‘I’m quite sure that is not a problem for [us]’ (FICO1). Similarly, the UK SME had never been subject to a claim of age discrimination (UKSME2). While legal claims were a minimal driver of change in these organisations, this may reflect the difficulties of mobilising age discrimination law in this context: according to the UK LE’s union representatives, equality law is not supportive of individual litigants: ‘the legal system isn’t in favour of us working people at all’ (UKLE2); ‘you can bring a challenge, but it’s always on the basis of one particular case and where you[’ve] got the evidence and where the individual is willing to do it and wants to go through the absolute hell of an employment tribunal’ (UKLE3). Further, age discrimination is generally invisible: ‘it’s rare that people come to us and say, “We think it’s on [the] ground of age discrimination”, because you can 109  See R Blackburn and M Hart, ‘Employment Rights in Small Firms: Some New Evidence, Explanations and Implications’ (2003) 32 Industrial Law Journal 60, 63–64; LP Wooten and EH James, ‘When Firms Fail to Learn: The Perpetuation of Discrimination in the Workplace’ (2004) 13 Journal of ­Management Inquiry 23; EH James and LP Wooten, ‘Diversity Crises: How Firms Manage Discrimination Lawsuits’ (2006) 49 Academy of Management Journal 1103; J Barnes and TF Burke, ‘Making Way: Legal Mobilization, Organizational Response, and Wheelchair Access’ (2012) 46 Law & Society Review 167, 171–72, 189; Winterbotham et al (n 67 above) ch 7.

Case Study Findings

 195

never see it … they make sure nothing is ever in writing’ (UKLE3). Therefore, ‘it’s the last-ditch really to go to an employment tribunal’ (UKLE3): the legal side is [a] last resort. We try to deal with things if we can internally and to get deals and accommodations with the company … We think we can get better results ­internally. And I think that’s true. I think we do, in the main, get better results. (UKLE2)

At the same time, law remains a ‘backdrop’ to negotiations (UKLE3), and can be used as a tool in bargaining to achieve change. Reflecting on one particular case, the UK LE’s union representative noted: it was absolutely illegal. If it had become public, they would have been in [serious ­trouble] … We used it in the negotiation to say look, this is what you are doing and if that went outside the company, you’d be in the court at a most senior level. (UKLE2)

Therefore, while there was a noticeable absence of formal legal claims, law may still be used instrumentally to achieve organisational change, including through the collective bargaining process.

(v) Unions Despite this, unions appeared to play a minimal role in the adoption of best­ practice. While some organisations consulted unions in the design and implementation of their programmes, no organisation felt that unions had prompted the introduction of such practices. For the Finnish council, despite unions being ‘quite strong in the city’, and being ‘very interested in developing this age management thing’, their main interest was to ensure other employees were not disadvantaged or made to ‘suffer’ when older workers were given time off to attend health and wellbeing training (FICOU1). Unions were ‘not so active’ in relation to other ageing measures and programmes (FICOU1), though new measures were generally presented to and discussed with the unions before implementation. The Finnish company had a similar experience: ‘the union played no part in the beginning … this is not unions, this is just [an] employer/employee thing, our internal thing’ (FICO1). Indeed, in the Finnish company, the company’s initiatives actually drove union policies across the sector: as a result of the Senior Programme, the sectoral collective bargaining agreement was amended to impose a ‘moral ­obligation’ on all companies to take into account the physical needs of employees aged over 50. The company’s ‘encouraging example … has inspired the entire field to join in the campaign of extending working careers’110—‘we’re pioneers in that sense’ (FICO1). Therefore, the union was following (rather than leading) the company’s example. This is also reflected in the situation at the UK council: while the union’s agreement was sought on policies before implementation, and union representatives

110 

Sitra (n 97 above).

196 

Organisational Case Studies

occasionally came to equality meetings,111 most workforce practices and changes had been initiated by the council rather than the union. That said, the UK council union representative felt the union could have a strong influence in this area: ‘if we make a good case in terms of equality, it’s likely to be heard because it should be one of the areas that [the council and union] don’t disagree on’ (UKCOU2). At the same time, older workers were not a priority area for the union, which was more focused on recruiting and supporting young workers: ‘historically … our focus [has been] on getting the youngsters in and keeping them in because we haven’t had any issues with, you know, older workers’ (UKCOU2). Further, severe time constraints on union representatives112 meant they were unable to engage with equality issues as fully as they may have liked: ‘we used to meet up regularly and kind of review our equalities work, but unfortunately, I think when the reduction in facility time stopped, then I wasn’t kind of able to keep all the groups going’ (UKCOU2). This reveals broader constraints and limitations on the role of unions in advancing equality measures in the UK. The UK LE had a similar experience. While the workforce was highly unionised (with one union having 95 per cent density in the company), and the company worked with the unions on equality issues and conducted ‘genuine consultation’, [i]t would be wrong to say most things emerge from the unions, most things will emerge from us, or from legislation, you know, changing business needs, etcetera. And implementing those things, you know, there’ll always be a start point and most of the start points will be us. Then, how do we get to the implementation point will involve a number of stakeholders including the unions at various stages, and they may well influence and shape, you know, some of the ways and some of the things that are implemented. (UKLE1)

This contrasts sharply with the unions’ perceptions of their role and influence: we’ve always had a massive influence on those issues … [We] try to make sure [management] don’t forget, you know, the lower grades [in policy formation], you know, who actually are probably the most important people. (UKLE2) [We] try and ensure that the policy matches up to best practice, so there’s a sort of link between the two. We try and look externally and try and ensure actually that [the UK LE] or any other employer is actually matching up to that external best practice … it’s trying to make work as good as possible. (UKLE3)

In contrast to the UK LE’s minimalist view of union influence, the unions saw themselves as occupying a broad role as advocates for good practice, influencers of corporate policy and a check on corporate power. 111  Though, according to the union representative: ‘I must admit I haven’t been to [the meetings] for a little while’ (UKCOU2). This may reflect broader time constraints experienced by union representatives (see n 112 below). 112  The union representative was seconded from their role at the UK council to the union on a parttime basis to undertake and coordinate union activities. This secondment used to be on a full-time basis, but had been reduced to a part-time role, meaning ‘there’s a challenge now about, you know, how much we can do as opposed to how much we were able to do before’ (UKCOU2).

Discussion

 197

These findings may be compared with previous research on the role of unions in advancing equality, which has found that unions can play a significant role in promoting good practice;113 and the results in Chapter five, which identified a statistically significant link between unionisation and age-aware practices. The experiences of these case study organisations indicate that unions are playing a lesser role in age equality matters than might be anticipated (or, at least, that organisations perceive unions to be playing a lesser role).

V. Discussion A.  Drivers of Best Practice The results of the organisational case studies corroborate the theories and findings in the literature: business factors and organisational culture were key drivers of best practice in relation to older workers. However, organisational leadership proved to be more important than the literature suggested, and law had less impact than might have been expected. That said, for the Finnish company and UK LE, legislation (and proposed legislation) was a significant driver of best practice. Therefore, law may have different impacts on different organisation types. A union presence did not appear to have any significant impact in these cases: indeed, the UK SME had no union presence at all. At the same time, co-operation and consultation were seen as important in developing ageing programmes, though not in their initiation (FICO1). The organisations’ focus on what is ‘good for business’, particularly in the UK, reflects a broader shift towards a diversity approach and an emphasis on achieving business and organisational goals via employee diversity.114 Kirton and Green argue that this shift towards ‘diversity management’ has lessened the focus on social justice that has traditionally underpinned equality approaches, replacing it with a new and exclusive focus on the business case for diversity.115 However, the case study organisations were still concerned with social justice and values other than organisational efficiency, implying that they are still adopting an ‘equal opportunity’ approach to their workforce. The business case and business factors, while influential, were not the sole driver of good practice in these organisations. This is fortunate, given the limitations of a ‘business case’ approach to equality. As noted by the UK LE’s union representative: The thing is that being nice doesn’t have any nice bottom line effects that you can point to … there’s no hard financial value placed on things like employee engagement, things

113 

See, eg, Greene and Kirton (n 51 above) 265–66. Kirton and Greene (n 52 above) 127. 115 ibid. 114 

198 

Organisational Case Studies

like high levels of trust. You can’t put a financial value on it … And the problem is that when times are hard, all that fluffy [HR] stuff is seen as fluff. That’s the danger … there’s always this, you know, the business case for equalities, and they always say, ‘Well, the cost of recruiting somebody new is so much more than managing correctly with your existing staff ’. And actually, that’s a useless argument in the current context, because they want to get rid of the expensive old ones. (UKLE3)

If relying on the business case for equality, business and productivity pressure ‘kind of pushes the good policy to one side or trumps it’ (UKLE2), particularly in an economic downturn (UKLE3): ‘when times are hard, it really puts the values of an organisation to the test’ (UKLE3). Therefore, these results emphasise the limits of the business case for equality. While organisational culture was a key driver of best practice in the case study organisations, it was not easy to create a culture that valued diversity. According to the Finnish company: You cannot force it … it takes a long time to change a culture, and it has to start from the beginning, from the top of the organisation, and to show people that the seniors are appreciated, they are important. And it’s a slow change, but it can happen. (FICO1)

Effecting cultural change in organisations may therefore not be the easiest way to promote best practice. If governments wish to promote good practice in organisations, interventions should focus on: promoting organisational leadership (which may be a first step towards cultural change); effectively using legislation’s limited influence on organisational practice to achieve beneficial ends; and using incentives and sanctions to influence organisations’ perceptions of business factors. How this might be achieved in practice is considered further in Chapter eight.

B.  Impact of Organisational Characteristics on Best Practice Organisational characteristics strongly influenced the nature of the practices adopted by the case study organisations and the challenges they faced in their implementation. In contrast to the larger organisations, the UK SME adopted practices on an ad hoc or individually negotiated basis, with few formal policies or procedures and no overall strategy for workforce diversity. This is consistent with previous studies, which have found that SMEs tend to adopt an informal approach to flexible work practices.116 It also supports the argument that good practice does not necessarily require formal age policies and practices,117 raising further questions about the utility of the WERS data set in evaluating

116 V Fuertes et al, ‘Extending Working Lives: Age Management in SMEs’ (2013) 35 Employee ­Relations 272, 276. 117  See also K Cafferkey et al, ‘Human Capital in Malaysian SMEs: HR Practices, Uniqueness and Value’ in C Machado and P Melo (eds), Effective Human Resources Management in Small and Medium Enterprises: Global Perspectives (Hershey PA, IGI Global, 2014) 32.

Discussion

 199

­ rganisational practices in SMEs (see further Chapter five). However, there are o also risks with this ad hoc approach, particularly if SMEs lack awareness of their legal obligations. The challenges faced by the organisations also varied according to organisational characteristics. In particular, large employers experienced more issues with the roll-out of new programmes. The Finnish council faced substantial challenges in up-skilling and raising awareness across a vast and widely distributed workforce: It’s a challenge that we have so many managers to get that they have the knowledge that they need and all the skills they need to be good leaders … you can’t go to every workplace and say, ‘This is what we’re going to do now’. So maybe the information doesn’t always get through to every place and the practices also. (FICOU2)

To address this, the council relied heavily on online materials and exercises, which allowed managers to access information at a time and location that suited them. The council also felt that the creation of ageing action plans by each department was essential for cultivating local ‘buy-in’: through the planning process, local departments were required to engage with ageing issues and consider actively the implications of ageing for their workforce, and ‘it was actually only then when all these departments got motivated and understood this, because they had been so much involved’ (FICOU1). The council also experienced challenges with the coordination and monitoring of the various ageing projects and activities in each department, eventually requiring a full-time project coordinator to stay on top of what was happening: ‘it’s hard to know what’s going on in every workplace’ (FICOU2). While the UK LE also had a large and distributed workforce, unlike the Finnish council it desired ‘consistency’ across its various divisions, and worked closely with each of the lines of business to ‘look at the challenges we’ve got and try and come up with something that was flexible’, and could be tailored to individual business needs, but also acted as a ‘framework’ for the different divisions (UKLE1). That said, ‘things are quite complex in big organisations and understanding that things are quite complex and accepting the context of complexity is quite important’ (UKLE1). One respondent described the UK LE as ‘a huge ship’, meaning there is a ‘huge lag … [when] trying to turn it round’ (UKLE2). Therefore, organisational change becomes a difficult and long-term process. Union representatives also raised concerns that the UK LE’s central policies were not translating down to front-line management: in terms of [the UK LE’s] policies, very, very good, very progressive, hard to fault them … There’s always been that good policy base. In practice [things are] a bit more difficult. A lot of the equality stuff seems to stop at the managerial levels from our perspective and doesn’t always filter down to lower grades. (UKLE2)

The disjunction between the UK LE’s central leadership and local line managers meant that equality policies became lost in practice, and were not effectively integrated into the company’s daily routine: ‘[there is] a real genuine dilemma …

200 

Organisational Case Studies

between people who establish policy in a big company like [the UK LE] and those who actually carry out the operations … the disconnect … comes at that level’ (UKLE2). In some cases there was a sharp disjunction between organisational policy and employees’ lived reality, or a ‘mismatch between spin and substance’ (UKLE3). While the UK LE was firmly opposed to age discrimination, the organisation’s ‘very aggressive approach to performance management’ (UKLE3)118 and desire for ‘healthy churn’ and annual staff turnover created the impression that ‘the company … has started to instigate a plan to get rid of the old ones’ as ‘they’re the easy ones’ (UKLE2): time and time again there is very strong evidence of age bias [in performance management]. So older workers are more likely to get lower appraisal marks and less likely to get the higher appraisal marks … In some parts of the business, I wouldn’t say in all, but in some parts of the business, we are certain that there has been active targeting of older people. Because they’re more likely to go. They might be 58, 59, 60, coming up to [a] pension, they’re a soft target. (UKLE3)

Union representatives doubted this outcome was intended or desired by the UK LE’s management. Indeed, the central leadership team were genuinely shocked [when informed of practices at the local level]. They could not believe it. But that’s how it is being played out at the lower level … in order to get a delivery [of organisational targets]. (UKLE2)

The ‘unintended consequences of a punitive performance management culture’ (UKLE3) had ‘gate-crashed’ (UKLE2) and ‘overshadowed’ (UKLE3) the effectiveness of other policies that were ‘going in the right direction’, such as the push to improve occupational health (UKLE2). As recognised by one union representative, ‘[the UK LE] has done amazing stuff around equalities in the past, and it’s a bit of a shame that we’re going through this sort of dark night at the moment’ (UKLE3). Similar concerns about how change is effected in a large organisation were raised about the UK council, particularly in relation to how good policies are implemented and translated ‘down to the front line’ (UKCOU2): ‘The message is up there and it’s brilliant, but, you know, some of the things that happen at the coalface … sometimes, you know, there’s a problem’ (UKCOU2). Therefore, while large employers may have additional resources and knowledge that help with adopting best practice, the sheer size and diversity of the workforce may pose additional implementation challenges. The potential disjunction between policy and practice in large organisations is discussed by Barnes and Burke: ‘Even ostensibly committed and professionalized organizations can produce rights practices that are not routinized and thus lost in the shuffle of competing policies, goals, and factions

118  Quarterly performance reviews were required for all staff, and subtle pressure was imposed on managers to achieve a ‘performance distribution’ of employee appraisals on a bell curve (UKLE2, UKLE3).

Discussion

 201

that characterizes complex organizations’.119 This reinforces the need to compare what organisations say they do with the actual results they achieve.120 The challenges faced by organisations also differed according to the ­availability and generosity of employees’ pensions upon retirement. In the UK SME, most employee respondents had no or limited occupational pensions, and would be reliant on the state pension in retirement. Therefore, the employees were keen to continue working into old age. This raised issues for the organisation related to intergenerational fairness, performance management for older workers121 and succession planning. While the UK SME had abandoned a fixed retirement age, it still had significant concerns regarding how that would play out in practice.122 In contrast, organisations with a generous pension plan were less concerned about these issues. The UK council’s pension scheme ‘allows people to make choices’, and people generally choose to retire at 65 (when they are entitled to a defined benefit pension), with only a few electing to work beyond that age (UKCOU1). This simplifies workforce planning.123 Similarly, in the UK LE’s experience, ‘most people do want to [retire] … [mostly] between the age of 59 to 62’ (UKLE1). This was assisted by the company’s pension plan, which provided ‘a good level of income to people in their retirement’ (UKLE1).124 The UK LE also believed that many older workers ‘understand … that they’re also going to create opportunities at some point for other people that come in’ when they retire, making intergenerational fairness a personal responsibility (UKLE1). The company felt its role was to ‘create opportunities for young people to come into employment’ by ‘creat[ing] the environment for people to make the right decision at the upper end of their career’ and educating workers about their pension entitlements to allow them to make these ‘right decisions’ (UKLE1). Overall, the case studies suggest that different organisations need different approaches and different support to implement best practice, and that the nature of best practice can vary according to organisation type. This was also acknowledged by the respondents, who felt it essential that practices, programmes and an 119 

Barnes and Burke (n 109 above) 175. ibid 170. 121  That said, the UK council found that they rarely performance managed older staff members. In 2012–13, five council staff members were subject to disciplinary proceedings: all five were aged under 55. Given these figures, the council ‘would not make a correlation’ between age and performance issues (UKCOU1). 122  This was despite the company having previously stated that the abandonment of fixed retirement ages had not had any negative impact upon their business, as quoted in CIPD (n 63 above) 8. 123  That said, the UK council is also used to having a mixed workforce, and to managing employees ‘coming and going’ (UKCOU1). It has therefore developed the capacity to manage succession and workforce planning even in the face of demographic change. 124  However, one of the UK LE’s union representatives felt this would only be a temporary state of affairs, particularly given that reforms to the company’s pension scheme had shifted most younger workers to a defined contribution (‘DC’) scheme: ‘were we in a situation where everybody was in a DC scheme, then people would be far less willing to go … this is a bit of a time bomb. But since most of the people are younger, it’s like 20 to 30 years’ time [before it becomes an issue]’ (UKLE3). 120 

202 

Organisational Case Studies

organisation’s overall approach to age diversity be tailored to the individual needs and circumstances of the organisation: Every organisation has to think … what are our main issues here, what is most important for us, what we should develop? (FICOU1) Every workplace is different. You cannot just, you know, copy. You have to adapt the model, or whatever, to your workplace, you know, to find out the problems which need to be addressed at your place, and adapt the programme. (FICO1)

Therefore, a one-size-fits-all approach to age diversity is unlikely to be successful, and tailored and individualised support for organisations may be necessary.125

C.  Impact of National Context on Best Practice The national context of the case study organisations also influenced their approaches to best practice. In the UK, the approach of the case study organisations is best described as ‘age-neutral’, where good HR practice is applied to employees of all ages (UKCOU1).126 The UK council and LE saw themselves as ‘doing a lot to support people’ generally, and treated older workers like all other staff (UKCOU1). The UK organisations were concerned with not imposing artificial barriers on older workers’ participation in the workforce, rather than promoting and supporting the employment of older workers. As a result, while age was not a barrier to employment in the UK organisations, there were few proactive measures in place to support older workers or accommodate older workers’ particular needs.127 Indeed, the UK LE felt they could not adopt age-specific programmes: ‘what we’re very careful of is you can’t be very specific to groups. You have to kind of offer [programmes to] everybody’, and assume that individuals will ‘naturally select’ which programmes are relevant to them (UKLE1). This may reflect the legal uncertainty around positive action in the UK (see Chapter three) and the focus on formal equality of opportunity and equal treatment in UK legislation (see Chapter one). There was also limited focus on health and wellbeing in the UK organisations, though the UK LE had introduced a general health awareness programme. Indeed, the UK council union representative noted that the council could do more in relation to workplace health promotion and the management of stress: ‘promoting health in the workforce would be good. That’s something that we could do, you

125 

McAndrew (n 39 above) 17. See further Ollier-Malaterre et al (n 47 above) 1376. 127  Indeed, if older workers at the UK SME were unable to meet the needs of the workplace, they were unlikely to be kept on. While age did not matter ‘so long as you’re able-bodied’ and able to ‘climb all over the machines’ (UKSME1), there was limited scope to adjust work patterns for manual workers who were less physically able. 126 

Discussion

 203

know, the council could do fairly well, I think … health promotion is … not an expensive thing to organise’ (UKCOU2). In contrast, the Finnish organisations had adopted programmes geared to the specific needs of older workers, and were far more proactive in adjusting work to older workers’ changing needs (particularly in relation to health and physical capacity) and promoting and supporting older workers in employment. Indeed, age policies in the Finnish organisations were a ‘strategic part of human resources policy’.128 This reflects the broader Finnish concern with workplace health promotion and work ability (see Chapter six) and the national legislative framework. Finnish organisations were also far more likely to adopt age-specific measures. This stands in marked contrast to the UK organisations, which generally worked to be ‘age-neutral’ in their policies and practices. Therefore, consistent with the findings in the literature,129 and irrespective of differences due to organisational type, it appears that national context may have substantial implications for organisational practice.130 National legislative frameworks appear to be shaping and influencing employers’ definitions of ‘compliance’ (see Chapter four) and ‘best practice’. Despite the case study organisations disclaiming the importance of legal intervention, law appears to matter in shaping employers’ responses to ageing. More particularly, the UK’s focus on formal equality of opportunity and equal treatment is translating into minimalistic age-aware measures on the part of UK employers: even best practice employers in the UK are not adopting age-aware measures commensurate with those in Finnish organisations. While legal uncertainty may be one driver of this ‘compliance approach’ in the UK, and UK case study respondents noted some legal uncertainty (particularly around the use of EJRAs), they did not feel law reform was a priority in this area: There’s other laws, nothing to do with [the] DRA, that … I would choose before [the] DRA, to be honest with you. So I think, you know, we’ve broadly got something we can work [with] and work well, but we just need a little clarity around a couple of areas within it. (UKLE1)

This limited desire for legal change is consistent with the views of the expert respondents presented in Chapter four (see also Chapter eight). At the same time, the law as it stands is not encouraging proactive or age-aware measures in UK organisations. While the case study organisations exhibit marked national differences, the Finnish council’s focus on older workers appears to be shifting. Though the council’s programmes originally aimed to make it ‘a model town for the management of different age groups’, the 2013 council strategy aspired towards ‘managing multiplicity’, acknowledging the importance of issues of race, ethnicity and gender, as well as age. While the council previously had measures in place to address these

128 

Pärnänen (n 98 above) 83. See Colgan (n 11 above) 731. 130  See also Weber and Barton (n 2 above) i, ii. 129 

204 

Organisational Case Studies

other grounds of diversity, this new programme acts as an ‘umbrella’ (FICOU1) for its various programmes and initiatives, to create a more holistic approach: ‘the … city has like awakened up to see that it’s not just the age thing, but the sex thing and the origin thing. So it’s natural to widen out’ (FICOU3). Age is no longer specified as a particular area of concern for the council’s programmes, indicating a shift of focus and emphasis: it is ‘not only about age’ (FICOU2). While thinking about older and younger workers as discrete groups was a way to conceptualise the problem initially, ultimately it was important to ‘count the needs of all workers’ and consider the many factors that influence an individual’s work career. Therefore, the approaches in the UK and Finland may converge over time to a more ­general focus on ‘equalities’. Again, this supports the Finnish experts’ idea that a more ‘mature’ approach to equality does not focus on older workers specifically (see Chapter six). However, unlike in the UK, age still remains a relevant consideration in Finland, being one element of a life-course approach.131 While the case studies demonstrate organisational and national differences, a recurring theme in the interviews was that successful programmes had become ‘business as usual’ (UKCOU1): ‘this has been quite like a normal thing to do … we don’t put too much emphasis [on it], it’s just the way we do things’ (FICO1). Few organisations viewed their practices as being particularly innovative: ‘I don’t think we do anything which you don’t normally do’ (UKSME1).132 Further, viewing the programmes as normal practice was seen as a sign of success: ‘one point of measuring if it’s working is that … it’s a normal way of doing things. It’s not, you know, anything special. It’s just the way we do things … It’s integrated in our operations’ (FICO1). Therefore, successful good practice must become embedded and routinised into an organisation’s day-to-day operations.133

VI. Conclusion The organisational case studies corroborate the findings and theories from previous chapters: small organisations are less likely to have formal age diversity processes in place (though may still be good practice employers); Finnish laws and financial incentives encourage a more proactive and holistic response by employers to demographic ageing; and attitudinal and cultural change remain ongoing challenges in organisations. The case studies also illustrate the barriers that may inhibit a proactive organisational approach to age diversity in the UK. While the

131  See further T Hareven and K Adams (eds), Ageing and Life Course Transitions: An Interdisciplinary Perspective (London/New York, Tavistock Publications, 1982); S Hunt, The Life Course: A ­Sociological Introduction (Basingstoke/New York, Palgrave Macmillan, 2005). 132  The exception was the Finnish council, which saw itself as being a leading example of good practice to other councils. 133  Barnes and Burke (n 109 above) 174–75.

Conclusion

 205

Finnish organisations were committed to their ageing programmes, they were significantly costly (and only made financially viable by the existence of legislative penalties), required close coordination in large and dispersed workforces, and needed strategic ‘buy-in’ from organisational leadership. It appears unlikely that UK employers will adopt these programmes without some additional incentive or impetus. That said, the case studies also indicate where UK government interventions can make a difference and encourage best practice. Consistent with the findings in Chapter five, legal change appears to have most impact on large private employers, who seem highly responsive to government programmes and legislation. While the PSED did not appear to influence the behaviour of best practice public sector organisations, it may bring equality issues to the forefront of organisational practice. At the same time, large employers are likely to encounter particular difficulties in the implementation and roll-out of change. Conversely, small organisations are less likely to be responsive to legal intervention, and appear more concerned with business needs (and remaining a going concern). Further, small organisations are less likely to provide workplace pensions for their employees,134 exposing small employers to issues of performance management, succession planning and inter-generational fairness. Therefore, government interventions need to recognise the diverse and specific organisational contexts of the workplaces they hope to influence. This may require tailored, context-driven interventions that acknowledge employers’ diverse contexts and challenges. Finally, while there are many drivers of organisational best practice, governments may find it difficult to influence internal organisational cultures and business factors. However, law and government intervention can encourage organisational leadership in this area, or be used to re-balance workplaces’ assessments of business factors (such as through the use of financial incentives or sanctions). This is explored further in Chapter eight, which uses the Delphi method to consider how these strategies might be implemented in practice. Further, it appears that national context and domestic legislative frameworks have a strong association with organisational definitions of compliance and best practice: law does matter.

134  J Forth et al, ‘Employers’ Pension Provision Survey 2011’, Research Report No 802 (DWP, 2012) 20. However, this may change with pension auto-enrolment.

8 Constructing Future Scenarios—The Delphi Method The doctrinal research, expert interviews, comparative analysis and organisational case studies discussed in previous chapters have flagged legal and policy reforms that may improve and encourage age-aware practices in the UK. However, it remains unclear whether these reforms are important and practicable in the UK context. To develop these proposals for change further, this chapter presents the results of expert discussions conducted via the Delphi method. Using this method, the UK experts were asked to consider a range of scenarios derived from previous chapters that could help the government and employers to promote the employment and retention of older workers. While most scenarios were regarded as impracticable under a deregulatory governmental agenda, the respondents regarded some changes as being important, desirable and practicable. Given this consensus, these reforms should be seriously considered and vigorously pursued.

I.  The Delphi Survey The Delphi method offers an effective means of achieving reliable consensus between different expert views and developing and analysing future policy ­scenarios (see Chapter two). The policy Delphi adopted in this study was a multistage process with two asynchronous rounds of online surveys. The p ­ rocess involved: —— —— —— ——

initial measurement of opinions (Round 1); data analysis and feedback to participants; design of a subsequent questionnaire based on initial responses; and second measurement of opinions (Round 2).1

1  MK Rayens and EJ Hahn, ‘Building Consensus Using the Policy Delphi Method’ (2000) 1 Policy, Politics, & Nursing Practice 308, 309.

The Delphi Survey

 207

Figure 18:  The Delphi Process

Between rounds, participants were provided with statistical group feedback about the beliefs of other participants as a means of promoting consensus before the second survey was completed.2 This process is depicted in Figure 18. The scenarios tested in Round 1 were: 1. Each industry develops a work ability strategy to be implemented by individual workplaces.3 2. Each workplace adopts an age management approach to its workforce. 3. Each employee is given the right to request flexible working arrangements.3 4. Easily accessible information and guidance about managing and working in an ageing workforce is available to employers and employees. 5. Effective leadership ensures the ageing workforce is an issue of high priority. 6. A national dialogue regarding what rights, obligations and privileges should be granted to individuals at different ages is undertaken. 7. Employers have a positive duty to achieve employment equality, including for older workers. 8. Employers have a responsibility to consult with employees and/or their ­representatives in relation to equality matters. The experts were asked to evaluate whether the reforms would be desirable, important and practicable in the UK context.4 Questions were assessed on a four-point assessment scale (1 = very low or negative, 4 = very high or positive),5 and partici2  M Turoff and SR Hiltz, ‘Computer-Based Delphi Processes’ in M Adler and E Ziglio (eds), Gazing into the Oracle: The Delphi Method and Its Application to Social Policy and Public Health (Boston MA, Jessica Kingsley, 1996) 58; G Aichholzer, ‘The Delphi Method: Eliciting Experts’ Knowledge in Technology Foresight’ in A Bogner et al (eds), Interviewing Experts, Research Methods Series (Basingstoke/ New York, Palgrave Macmillan, 2009) 252. 3 The results from this scenario can be viewed elsewhere: A Blackham, ‘Emerging Options for Extending Working Lives: Results of a Delphi Study’ in S Manfredi and L Vickers (eds), Challenges of Active Ageing: Equality Law and the Workplace (Basingstoke/New York, Palgrave Macmillan, 2016). 4 Definitions of these terms were provided to ensure respondents had a similar understanding of what the terms meant. These definitions were adapted from the literature: see M Turoff, ‘The Policy Delphi’ in HA Linstone and M Turoff (eds), The Delphi Method: Techniques and Applications (Boston MA, Addison-Wesley, 1977) 91–92; E Ziglio, ‘The Delphi Method and Its Contribution to Decision-Making’ in M Adler and E Ziglio (eds), Gazing into the Oracle: The Delphi Method and Its Application to Social Policy and Public Health (London, Jessica Kingsley, 1996) 32–33. For a full description, see Blackham (n 3 above) 159–60. 5  Derived from Aichholzer (n 2 above) 263.

208 

The Delphi Method

Figure 19:  Relationship between Round 1 and Round 2 Delphi Survey Instruments

pants were asked to rate their confidence in their assessment of each s­ cenario (1 = no confidence, 4 = high confidence).6 Finally, participants were given the opportunity to suggest additional or different policy options during the first survey round. Round 2 included: 1. Re-worded scenarios from the first survey to clarify or develop s­ cenarios in response to participant comments. 2. Repetition of scenarios that led to disagreement in the first round, with the aim of allowing participants to revisit or reconsider their original opinions in light of the feedback provided. 3. New scenarios posed by respondents in the first survey round. These scenarios were: a. Introduction of a statutory right to take paid time off to fulfil caring responsibilities;7 and b. Do nothing and maintain the status quo. 4. New implementation scenarios, posing possible means of implementing ­scenarios that achieved a strong level of consensus in the first round. The process of developing the Round 2 survey is depicted in Figure 19. Following Rayens and Hahn,8 in this study I regard both: —— scenarios having an inter-quartile range (IQR) of more than 1.00;9 and 6 Self-rating is a meaningful way of identifying expertise and can improve the accuracy of responses: HA Linstone and M Turoff, ‘Evaluation—Introduction’ in HA Linstone and M Turoff (eds), The Delphi Method: Techniques and Applications (Boston MA, Addison-Wesley, 1977) 234. 7  The results from this scenario can be viewed elsewhere: Blackham (n 3 above). 8  Rayens and Hahn (n 1 above) 312–13. 9  The IQR is the difference between the upper and lower quartiles (that is, the range within which the middle 50% of responses fall). Rayens and Hahn describe this as the ‘inter-quartile deviation’

The Delphi Survey

 209

Table 18:  Response Rate for Delphi Roundtable by Survey Round Round

Number of participants

Response rate

1

13

77%

May–July 2013

Period of survey

2

8

62%

October 2013–January 2014

—— scenarios having an IQR of 1.00 and receiving positive responses from between 40 and 60 per cent of respondents as having a degree of disagreement worthy of exploration in Round 2. Scenarios that did not satisfy this definition were regarded as achieving consensus.10 Results were adjusted to exclude respondents who self-identified as having ‘no confidence’ in their response to ensure a minimum level of knowledge about each area.11 ­Following this adjustment, the answers of two respondents were excluded for two scenarios in Round 1 (four answers in total); and the answers of one respondent were excluded for five scenarios in Round 2 (five answers in total). The response rate for each round of the Delphi is included in Table 18. Before presenting the results of the survey, it is necessary to flag some preliminary words of warning and caution in relation to these data and findings. In this survey (and, indeed, in the interviews in Chapter four), the experts were participating in their roles as advocates and lobbyists, with strong and partisan views on the issues at hand. Therefore, it was foreseeable that this Delphi survey might not generate consensus. However, it could still clarify and articulate the experts’ views on particular scenarios and issues.12 These results ultimately reflect the experts’ views, which are inherently subjective, far from neutral, and driven by the experts’ own situated context.13 The experts are participating in a system of reproduction of knowledge and expertise, which is more likely to favour conservative approaches to change and renewal.14 Further, these experts are also stakeholders, and have a vested interest in the results of this research, which may distort

(IQD), which is calculated by halving the IQR: ibid 312. However, in the authors’ calculations, it is clear that they are using the IQR, not the IQD. 10  For ease of reporting, results will be presented in the form: S : D(IQR) = Y, D(%P) = Z; I(IQR) =, X I(%P) =; F(IQR) =, F(%P) =). That is, for a given scenario X, the ‘desirable’ (D) IQR is Y, percentage of positive responses is Z, and so on for important (I) and feasible (F). For full reporting of responses, see A Blackham, ‘Extending Working Life for Older Workers: An Empirical Legal Analysis of Age Discrimination Laws in the UK’ (PhD Thesis, University of Cambridge, 2014). 11  Aichholzer (n 2 above) 267. 12  M Turoff, ‘The Design of a Policy Delphi’ (1970) 2 Technological Forecasting and Social Change 149, 153; W Rauch, ‘The Decision Delphi’ (1979) 15 Technological Forecasting and Social Change 159, 163. 13  See L Van Audenhove, ‘Expert Interviews and Interview Techniques for Policy Analysis’ (Brussels, Institute for European Studies Research Colloquium, May 2007). 14  See ibid.

210 

The Delphi Method

their responses.15 Therefore, drawing unreflectively on expert knowledge runs a real risk of ‘the non-validated confirmation and, thus, the legitimisation of social hierarchies’ and norms.16 While the experts’ views should be considered, they are not the only indicator of what change is desirable or possible. With these words of caution, I present the results of the survey in the sections that follow.

II. Results A. Scenario 2: Each Workplace Adopts an Age Management Approach to its Workforce As noted in Chapter six, age management is a key feature of the Finnish approach to older workers. The adoption of age management strategies in Member States is one of the Guiding Principles for Active Ageing and Solidarity between Generations, as jointly agreed by the EU’s Social Protection Committee and the Employment Committee.17 Work on age management has been undertaken in the UK, resulting in various sources of information and guidance for employers,18 though much of this work has originated from non-governmental organisations, such as TAEN and the CIPD. Despite this guidance, age management continues to ‘[strike] British ears rather awkwardly’.19 Therefore, this scenario tested the possibility of adopting age management ideas in the UK context. By focusing on workplaces, the scenario reflects an emphasis on employers as agents of change and the broader role of employers in supporting older workers, consistent with the prevailing approach in Finland. This scenario was given to respondents with a short definition of ‘age management’ to ensure that all respondents had a similar understanding of what the term meant. The results for this scenario demonstrate significant consensus across the experts: the vast majority (92 per cent, IQR = 1) felt the scenario was desirable and

15 A Wroblewski and A Leitner, ‘Between Scientific Standards and Claims to Efficiency: Expert Interviews in Programme Evaluation’ in A Bogner et al (eds), Interviewing Experts, Research Methods Series (Basingstoke/New York, Palgrave Macmillan, 2009) 236. 16 A Bogner et al, ‘Introduction: Expert Interviews—an Introduction to a New Methodological Debate’ in A Bogner et al (eds), Interviewing Experts, Research Methods Series (Basingstoke/New York, Palgrave Macmillan, 2009) 2–3. 17  See Council of the European Union, ‘Council Declaration on the European Year for Active Ageing and Solidarity between Generations: The Way Forward’ (2012) 17468/12 SOC 992 SAN 322, 8. 18  See, eg, C Ball, ‘Defining Age Management: Information and Discussion Paper’ (London, TAEN, 2007); C Ball, ‘Age Management at Work: Adopting a Strategic Approach’ (Employment Relations Comment, March 2013); E Parry and L Harris, ‘The Employment Relations Challenges of an Ageing Workforce’, Future of Workplace Relations Discussion Paper (Acas, December 2011); CIPD, ‘Managing Age: New Edition 2011’ (May 2011). 19  Ball, ‘Defining Age Management’ (n 18 above) 1.

Results

 211

important, and the majority (77 per cent, IQR = 0) felt it was feasible. Respondents’ support for this scenario is evident in the comments, which included:20 Adopting an age management approach would have organisational and economic benefits. (1) It is particularly important given the ageing workforce (1) and skills shortages in younger workers. (1) An age management approach may have helped employers better manage the phase out of the DRA. (1)

However, the respondents’ comments also indicate concern regarding how this scenario might be implemented in practice, particularly in relation to SMEs. Respondent comments included: This may not be easily applicable to SMEs, where it can be difficult to adapt a job to suit an employee’s circumstances. (1) It is unlikely that an age management approach will be adopted in every workplace. However, there are examples in other countries of SMEs engaging with age management. (1) To be successful, this scenario will require strong sectoral leadership (1) and regulatory actors (1) to drive reform. It is unlikely that employers will adopt an age management approach without this. (1) The government’s age positive campaigns and employer guidance materials are already directed to implementing this scenario. (1)

The respondents variously thought that sectoral leadership, leadership from regulatory actors, information and guidance and attitudinal campaigns could help to promote age management approaches at the workplace level. To explore these ideas further, in Round 2 a series of ‘implementation scenarios’ were posed on this topic. The scenarios posited a range of measures, from persuasion and education to incentives and statutory regulation,21 to examine how age management approaches could be achieved at the workplace level. Respondents were also asked who should be responsible for each implementation scenario. Using confidence-adjusted results, there was consensus among respondents that promoting the business case for age management,22 providing additional information and guidance,23 and effective leadership24 were desirable, feasible and important implementation measures for securing organisational

20  The number of respondents making each comment or sub-comment is identified in brackets. The comments have been thematically consolidated and amended for coherence; they are not direct quotes. 21  See DC Walsh and NP Gordon, ‘Legal Approaches to Smoking Deterrence’ (1986) 7 Annual Review of Public Health 127. 22 S 2(1): D(IQR) = 1, D(%P) = 85.7; I(IQR) = 1, I(%P) = 71.4; F(IQR) = .5, F(%P) = 85.7. See n 10 above for an explanation of these results. 23 S 2(2): D(IQR) = 1, D(%P) = 100; I(IQR) = 1, I(%P) = 100; F(IQR) = 1, F(%P) = 87.5. 24 S 2(4): D(IQR) = 1, D(%P) = 100; I(IQR) = 1, I(%P) = 100; F(IQR) = .25, F(%P) = 87.5.

212 

The Delphi Method

age ­management. While training and development opportunities were seen as desirable and important,25 respondents disagreed regarding their feasibility,26 perhaps due to the potential cost of such programmes. When asked who should be responsible for implementing such initiatives, respondents placed emphasis on the role of employer bodies, industry bodies and professional organisations in leading change. While some respondents identified government as playing a role in this area, more reliance was placed on non-governmental bodies. In contrast to the broad support and consensus regarding information and persuasion measures, respondents were far less supportive of regulation and the use of financial incentives in this area: neither financial incentives27 nor regulations to promote age management28 were seen as being feasible or important. Respondents questioned where funds would come from to finance incentives, and doubted whether there would be sufficient political will to introduce regulations. Therefore, respondents appeared far more supportive of ‘persuasion’ measures and nongovernmental activity to promote age management approaches at the workplace level. This is consistent with the traditional ‘hands-off ’ approach to governance in the UK (see Chapters three and four) and the prevailing reliance on ‘soft law’ and exhortation in the area of equality.29 However, while soft law is a useful adjunct to legal regulation, it is ‘a poor substitute’ for legal intervention.30

B. Scenario 4: Easily Accessible Information and Guidance about Managing and Working in an Ageing Workforce is Available to Employers and Employees During the expert interviews, respondents suggested that additional guidance be created to assist employers in managing their obligations towards the ageing workforce. The experts emphasised that this guidance needed to be accessible, available and understandable by employers. Respondents also suggested the creation of an (online) ‘one-stop-shop’ to signpost relevant information, and the provision of opportunities for employers to collaboratively work through issues, including through facilitated opportunities to network, discuss issues around ageing and share best practice. Alternatively, organisations could be given access to tailored advice from advisors about how to adapt legal provisions to their workforce. In this context, Scenario 4 tested the relative desirability and importance of enhancing information provision in this area. The results show a striking 25 S

2(3):

D(IQR) = 1, D(%P) = 100; I(IQR) = 1, I(%P) = 87.5. F(IQR) = 2, F(%P) = 62.5. 27 S 2(5): D(IQR) = 1, D(%P) = 50; I(IQR) = 1.25, I(%P) = 37.5; F(IQR) = 1, F(%P) = 37.5. 28 S 2(6): D(IQR) = 1, D(%P) = 50; I(IQR) = 0, I(%P) = 12.5; F(IQR) = 1.25, F(%P) = 25. 29 L Dickens and M Hall, ‘Fairness—up to a Point. Assessing the Impact of New Labour’s ­Employment Legislation’ (2006) 16 Human Resource Management Journal 338, 348. 30 ibid. 26 S

2(3):

Results

 213

degree of consensus: all respondents believed the scenario was desirable, and 92 per cent felt it was feasible and important (IQR = 1 for all). However, the comments echo the results of the expert interviews: while information is necessary, it is not sufficient, and the information provided must be accessible, readable and relevant. Comments included: Information and guidance are necessary but not sufficient. (1) There also needs to be leadership on the issue and a change in management attitudes. (1) As well as being physically accessible (eg online), information also needs to be readable and relevant. (1) Information and guidance are already available. The issue is about spreading it further (1) and providing employers with tailored guidance when required. (1)

Thus, the real challenge is how information is disseminated: meaningful ­dissemination may require activities other than the production of information, like tailored advice for individual employers. To explore these ideas further, in Round 2 a series of ‘implementation scenarios’ were posed on this topic. The scenarios reflected the respondents’ suggestions and ideas to examine how information and guidance could be made more accessible (see Chapter four). The majority of implementation scenarios generated limited consensus: an online ‘one-stop shop’ was seen as desirable, but there was no consensus regarding whether it was important or feasible;31 an advisory service, while desirable and important, generated no consensus on feasibility,32 perhaps due to the difficulties of securing funding in a climate of austerity;33 the development of further guidance failed to generate consensus regarding its feasibility;34 and the re-drafting of existing guidance was not seen as important and did not achieve consensus regarding its feasibility or desirability.35 The striking exception was the provision of facilitated opportunities for employers to ‘network’, discuss issues around ageing and share best practice. All respondents felt this implementation scenario was important (IQR = .25), feasible (IQR = 1) and desirable (IQR = 1). This is consistent with the promotion of organisational peer review in accordance with an experimentalist architecture (see Chapter two). However, most experts (6) felt that responsibility for this reform should lie with employer or industry bodies and professional bodies, again indicating that government should play a limited role in this area.

31 S

4(1):

D(IQR) = 1, D(%P) = 100; I(IQR) = 1.25, I(%P) = 75; F(IQR) = 1.25, F(%P) = 75. D(IQR) = .25, D(%P) = 100; I(IQR) = .25, I(%P) = 75; F(IQR) = 1, F(%P) = 50. 33  Comments included: This would require statutory funding or employer/industry funding, which may be difficult in an austerity climate (1), particularly given the recent cuts to the advice sector and the EHRC (1). However, another comment was: Employer/industry bodies may see the value of such a service in financial terms and take it on (1). 34 S 4(3): D(IQR) = 1, D(%P) = 100; I(IQR) = 1, I(%P) = 87.5; F(IQR) = 1.25, F(%P) = 75. 35 S 4(4): D(IQR) = 1, D(%P) = 50; I(IQR) = 1, I(%P) = 37.5; F(IQR) = 1.25, F(%P) = 62.5. 32 S

4(2):

214 

The Delphi Method

C. Scenario 5: Effective Leadership Ensures the Ageing Workforce is an Issue of High Priority In the organisational case studies, leadership was seen as a key driver of best practice. Further, the expert interviews identified a need to build employer ­ ­capacity and leadership skills generally. Governmental leadership was also identified by one expert as being important for changing employer attitudes towards older workers and encouraging organisations to adopt best practice. Therefore, leadership at governmental and organisational levels may be a key driver of better age-aware practices. Scenario 5 therefore tested the relative desirability and importance of leadership in the context of an ageing workforce. The question was framed broadly on purpose, to allow respondents to interpret ‘leadership’ as that at a governmental, organisational and/or sectoral level. This scenario generated significant consensus: respondents saw leadership as desirable, feasible and important.36 Further, the comments recognised the importance of leadership in achieving cultural change (a belief supported by the organisational case studies in Chapter seven). However, the comments raised ­concerns that leadership would not be enough to effect meaningful change, and queried the capacity of line managers and organisational heads to lead change in this area. Comments included: While this would make a significant difference, management skills are not yet sufficiently developed to achieve this. (1) It is difficult to engage line managers with long-term planning when many have become managers by circumstance. (1)

This reinforces the need for capacity building in organisations, as identified by the expert respondents in Chapter four.

D. Scenario 6: A National Dialogue Regarding what Rights, Obligations and Privileges Should be Granted to Individuals at Different Ages is Undertaken The expert interviews and broader literature identified the need for a social dialogue regarding what is acceptable age discrimination, what rights and obligations should be provided at different ages and what exceptions should be allowed to age equality provisions. A dialogue of this nature may promote better social ­policy.37

36 S

5: D(IQR) = 1, D(%P) = 92; I(IQR) = 1, I(%P) = 37  See further A Walker, ‘Ageing in Europe: Policies

92; F(IQR) = 1, F(%P) = 77. in Harmony or Discord?’ (2002) 31 International Journal of Epidemiology 758, 760; H Sarfati, ‘Social Dialogue: A Potential “Highroad” to Policies Addressing Ageing in the EU Member States’ (2006) 59 International Social Security Review 49, 53.

Results

 215

Scenario 6 tested this suggestion. The majority of respondents (69 per cent, IQR = 1) rejected the idea of a national dialogue, and there was consensus that the suggestion was not desirable, feasible or important. This was unexpected given that the scenario was derived from interviews with the same expert respondents. However, the comments revealed significant concern that a dialogue could perpetuate discrimination, or be ineffective at achieving concrete change. Comments included: This may perpetuate stereotypes around ageing and could be discriminatory. (3) A national dialogue is unlikely to encourage employers to adopt sensible approaches to managing ageing. (1)

Further, respondents felt the question was unclear, which was reflected in the low percentage of confident responses (62 per cent, IQR = 1). The scenario was re-written for Round 2 to clarify its meaning and integrate the respondents’ comments. The scenario was reframed to focus on discussion and awareness-raising rather than dialogue more broadly.38 The re-worded scenario was received far more positively in Round 2,39 indicating that the issues in Round 1 were at least partly attributable to a lack of clarity. However, there was still a lack of consensus on whether the scenario was feasible, and respondents again expressed concern that the discussion might fail to achieve meaningful outcomes. Therefore, expert views indicate that broad discussion and awareness-raising is not a scenario that should be prioritised.

E. Scenario 7: Employers Have a Positive Duty to Achieve Employment Equality, Including for Older Workers Imposing positive duties on employers to achieve employment equality or fair participation would be a key means of encouraging a more proactive and less compliance-focused organisational response to the equality framework.40 Scenario 7 considered whether positive duties could or should be introduced in the UK for all employers. The respondents generally regarded this scenario positively, and saw it as both desirable and feasible (84 per cent, IQR = 1). However, there was no consensus regarding whether it was important (75 per cent, IQR = 1.25). This positive response was surprising, as imposing positive duties on the private sector 38  Amended Scenario 6: There is broad-ranging discussion and awareness-raising regarding the needs of older workers. 39 S 6(A): D(IQR) = 1, D(%P) = 100; I(IQR) = 1, I(%P) = 87.5; F(IQR) = 1.25, F(%P) = 62.5. 40  L Dickens, ‘Equality and Work-Life Balance: What’s Happening at the Workplace’ (2006) 35 Industrial Law Journal 445, 447; L Dickens, ‘The Road is Long: Thirty Years of Equality Legislation in Britain’ (2007) 45 British Journal of Industrial Relations 463, 474; S Fredman and S Spencer, ‘Delivering Equality: Towards an Outcome-Focused Positive Duty—Submission to the Cabinet Office Equality Review and to the Discrimination Law Review’ (June 2006) 6–7. See Chapter 3 and Chapter 6 of this volume.

216 

The Delphi Method

was predicted to generate significant consternation among employer groups. It was therefore anticipated that the scenario would generate far less consensus. The respondents’ comments suggest that confusion or multiple interpretations may have led to an artificially positive response. Indeed, one respondent noted that the positive duty was already in place under legislation, perhaps confusing it with the PSED. Other respondents felt the question was unclear, and asked for clarification regarding how the duty might be implemented, enforced and defined. The scenario was therefore re-written for Round 2 to clarify its meaning and reflect the respondents’ comments.41 The reframed scenario explicitly noted that the duty would be imposed by statute, and would apply to public and private sector employers. Examples were also given of how the duty might be implemented and enforced, drawing on Hepple and others’ framework.42 Following its re-wording and clarification, the scenario received a markedly less positive reception. While respondents still saw the scenario as desirable (87.5 per cent, IQR = 1), there was no consensus regarding whether it was feasible or important, and most respondents felt it would not be feasible (62.5 per cent, IQR = 1.25). Therefore, a positive duty may not be a priority for governmental reform or, alternatively, other reforms should take precedence. These responses reflect the perceived impact of a deregulatory governmental agenda on the viability of legal reform.

F. Scenario 8: Employers have a Responsibility to Consult with Employees and/or their Representatives in Relation to Equality Matters To enhance the effectiveness of reflexive legislation in the UK, Hepple and others propose a duty on employers to engage with other groups in decision-making.43 This proposal was tested in Scenario 8. While a responsibility to consult was seen as both desirable (83 per cent, IQR = 1) and important (92 per cent, IQR = 1), there was no consensus regarding whether it would be feasible (75 per cent, IQR = 1.25). Some respondents felt this scenario should be feasible as consultation provisions already apply in other contexts. However, other experts felt the scenario was unlikely to eventuate as it would increase the burden on business, and was therefore contrary to the government’s deregulatory agenda. Respondents also voiced concerns regarding how the responsibility to consult would operate in practice, and what benefits or improvements it would offer (if any).

41  Amended Scenario 7: A statutory duty is imposed on all employers (public and private) to achieve employment equality. 42  B Hepple et al, Equality: A New Framework—Report of the Independent Review of the Enforcement of UK Anti-Discrimination Legislation (Oxford, Hart Publishing, 2000) 71. See Chapter 3 of this volume. 43  ibid. See Chapter 3 and Chapter 6 of this volume.

Analysis

 217

Given these concerns, the feasibility of Scenario 8 was reconsidered in Round 2. After viewing the results and comments from Round 1, respondents were asked to reassess whether the scenario would be feasible. Consensus was achieved when respondents reconsidered the feasibility of the scenario (85.7 per cent, IQR = 0): respondents therefore saw the scenario as desirable, feasible and important. However, the comments indicated that a number of other reforms would need to accompany this scenario to make it effective, namely stronger rights for trade union representatives (1), protection of the role of trade union equality representatives in the workplace (1), and stronger references to equality in guidance and rights on collective bargaining (1). The results for this question indicate that the Delphi process is a successful method for achieving consensus. This may reflect a ‘survival bias’: two representatives of employer associations failed to participate in Round 2, potentially distorting the results. However, closer analysis of respondent behaviour over the two rounds shows that survival bias does not account for the consensus achieved. Half of the respondents changed their responses to Scenario 8 over the two rounds, with an average change of 1. This implies that the survey feedback moderated conflicting perspectives. The Delphi method therefore appears to have resolved or reduced the respondents’ disagreement, supporting its efficacy as a consensusbuilding exercise.44

G.  Do Nothing In Round 1 of the survey, respondents were asked to suggest new or different scenarios for testing in Round 2. One respondent suggested that policymakers do nothing, or maintain the status quo. This was tested with the other respondents in Round 2. No respondent saw the scenario as desirable (with the scenario’s original proponent failing to participate in Round 2) (IQR = 1), and the majority of respondents did not regard the scenario as important (37.5 per cent, IQR = 1).Further, there was no consensus regarding whether the scenario was feasible (75 per cent, IQR = 1.25). Therefore, while it may be difficult to achieve change in this area, it is clear that respondents regarded doing nothing as undesirable.

III. Analysis Over the two rounds of the Delphi process, respondents endorsed four scenarios as being desirable, feasible and important: adoption of age management approaches 44  For further analysis of the effectiveness of the Delphi method in this context, see A ­Blackham, ‘Using the Delphi Method to Advance Legal Reform: A New Method for Empirical Labour Law Research?’ in A Ludlow and A Blackham (eds), New Frontiers in Empirical Labour Law Research (Oxford, Hart Publishing, 2015).

218 

The Delphi Method

by workplaces; making information and guidance available and easily accessible to employers and employees; effective leadership; and employers having a responsibility to consult with employees and/or their representatives on equality matters. While consensus was achieved on these scenarios, it is less clear how they should be implemented in practice. In relation to age management, respondents only supported implementation via persuasion measures, rather than incentives or regulation. It is unclear whether persuasion alone will be effective: indeed, previous chapters have shown existing measures (which focus on persuasion and education) to be ineffective for achieving positive change. Respondents also placed emphasis on non-governmental activity in implementing these initiatives, and regarded government as having a more limited role. This is consistent with the experts’ attitudes to law and government intervention, as reported in Chapter four. Further, most implementation scenarios were not seen as being important or desirable to enhance the provision of information and guidance. Consensus was only achieved on the provision of facilitated opportunities for employers to network. Again, responsibility for providing these opportunities was seen as lying with professional organisations and employer or industry bodies, rather than government. Similarly, while respondents supported the leadership scenario, ‘leadership’ was interpreted as organisational leadership, rather than leadership at a governmental level. At the same time, respondents expressed concern that individual managers lacked the capacity and skills to lead change. Overall, the Delphi survey indicates that the expert respondents do not believe legal change is a viable option in the UK context. Indeed, there was a marked preference for information and persuasion mechanisms and non-governmental activity. While these results are informative, they should also be used with a degree of care: as noted above, experts are more likely to favour conservative approaches to change and renewal.45 Thus, it is unsurprising that the experts’ responses embody a hands-off, laissez-faire approach to government intervention, mirroring government documents and the prevailing approach to change. While the experts’ views should be taken into account when developing a programme for change, they should not be seen as determinative of what change is desirable or possible.

IV. Conclusion This Delphi study highlights a range of scenarios that might be pursued in the UK to address the implications of ageing for labour law, and demonstrates that experts support the adoption of these scenarios. Indeed, respondents firmly rejected doing nothing as a viable course of action. Therefore, this study strongly supports the case for targeted change to address demographic ageing.

45 

See Van Audenhove (n 13 above).

Conclusion

 219

However, the results of the Delphi survey also indicate that experts feel g­overnment should play a limited role in this area, and that regulation and ­statutory intervention are not the preferred course of action. Therefore, based on the results of the Delphi study, legal reform appears unlikely to garner support. Instead, employer and industry bodies and professional organisations may need to play a larger role in promoting positive change. The results of the surveys hint at the impact of government austerity m ­ easures and a deregulatory agenda on the perceived viability of legal reform. The former Coalition government (in power at the time of the survey) aimed to eliminate ‘unnecessary burdens on business’,46 to pursue the government’s ‘vision’ of a labour market ‘with minimal intervention by the Government’.47 In this context, the respondents’ concern about reform feasibility was wholly reasonable: it is unlikely that new regulations will be imposed on business in this climate. This will severely limit the UK’s capacity to respond to demographic change. As noted in previous chapters, existing measures that focus on employer education and persuasion have had limited success in promoting positive change. The experts’ support of non-governmental persuasion mechanisms is unlikely to lead to substantial change or improvement. While the consensus-generating scenarios identified in this chapter should be vigorously pursued, as they appear to be easily achievable and command broad support, they are unlikely to be sufficient to achieve meaningful and lasting change for older workers. Thus, while expert views revealed in the Delphi survey provide a clear programme for reform, they should not be seen as reflecting the limits of what change is necessary or desirable in the UK.

46  BIS, ‘Flexible, Effective, Fair: Promoting Economic Growth through a Strong and Efficient Labour Market’ (October 2011) 7. 47  ibid 4.

9 Conclusion In the face of demographic ageing, age discrimination laws are seen as a key legislative tool for increasing participation rates for older workers, reducing financial burdens on social security systems, and upholding the dignity and rights of individuals of all ages. If age discrimination laws are to be used to address the ageing ‘challenge’ in the UK, we must understand how they are operating in practice. This book has drawn on reflexive law theory and an empirical mixed methods research design to consider multiple facets of two questions: —— How are UK age discrimination laws operating in practice? —— How (if at all) could UK age discrimination laws be improved?

I.  How are the Laws Operating in Practice? Using doctrinal, qualitative and quantitative research methods, I have shown that UK age discrimination laws have multiple practical limitations. The content of the legislation is unclear and underdetermined, particularly in relation to EJRAs and the use of positive action. In this space, employers are increasingly expected to define what ‘compliance’ entails, with limited oversight by the EHRC, unions and ETs. Qualitative expert interviews, statistical analysis of the WERS data set and organisational case studies indicate that most UK employers have defined ‘compliance’ narrowly, and few employers are adopting proactive age-aware practices that respond to older workers’ specific needs. ‘Compliance’ in the UK therefore appears to be fairly limited,1 and involves eliminating overtly discriminatory practices, rather than addressing broader limitations or issues associated with age. This is unsurprising, given that UK legislation does not expect or require organisations to adopt age-aware practices: the EqA is designed to provide a ‘floor of rights’ that employers can then build upon. However, there is limited evidence that employers are doing much beyond what they believe the EqA requires.

1 L Dickens and M Hall, ‘Fairness—up to a Point. Assessing the Impact of New Labour’s Employment Legislation’ (2006) 16 Human Resource Management Journal 338, 346.

How are the Laws Operating in Practice?

 221

Indeed, employers in the organisational case studies believed they legally could not make any special provision for their ageing workforce, even if they had wanted to. This may reflect a perception that age-aware measures are legally risky and may give rise to claims by younger workers of age discrimination or unjustified positive action. In sum, then, the UK’s limited approach to age equality—which is beset by legal uncertainty, a relatively weak legislative framework,2 and prevailing cultural norms that favour formal over substantive equality—is unlikely to facilitate older workers’ workforce participation or improve the quality of opportunities available for older workers. UK age discrimination legislation is limited in the extent to which it is encouraging employers to create opportunities for older workers to remain in or return to the workforce. It has had limited success in addressing ageist attitudes in the UK, and ageism remains socially acceptable in many contexts. The legislation has also had limited success at achieving organisational change: best practice employers are already operating beyond legal requirements, and small and private employers demonstrate limited responsiveness to legal change. Therefore, the law is failing to achieve its instrumental or intrinsic objectives. That said, it is clear that the UK legislative framework is influencing employer responses to ageing: consistent with ideas underpinning the EqA, the UK case study organisations all aspired to a model of age neutrality, with a focus on equal treatment and formal equality of opportunity. Therefore, while the EqA is failing to secure proactive employer responses to ageing, it is clearly having some impact on organisational practice: law does matter, which means legal change can influence what goes on at the organisational level. This study also casts light on organisational characteristics that may encourage employers to go beyond a ‘compliance’ orientation. As detailed in Chapter seven, organisational leadership, business factors, organisational culture, law and unions may play some role in encouraging ‘best practice’ organisational approaches to age equality. Drawing on these findings, we can posit some broad linkages between different levels of government intervention and organisational drivers of change, as depicted in Figure 20.3 Government approaches based on encouragement and persuasion are most likely to affect organisational leadership and the culture of an organisation. This may, in turn, encourage more age-aware organisational ­measures.4 Further, promotion of the ‘business case’ for equality may encourage employers to re-evaluate their assessment of business factors. However, it is

2  See L Dickens and M Hall, ‘Legal Regulation and the Changing Workplace’ in W Brown et al (eds), The Evolution of the Modern Workplace (Cambridge, Cambridge University Press, 2009) 349. 3  The categorisation of government actions is adapted from DC Walsh and NP Gordon, ‘Legal Approaches to Smoking Deterrence’ (1986) 7 Annual Review of Public Health 127. 4  See A Ollier-Malaterre et al, ‘Looking up to Regulations, out at Peers or down at the Bottom Line: How Institutional Logics Affect the Prevalence of Age-Related HR Practices’ (2013) 66 Human ­Relations 1373.

222 

Conclusion

Figure 20:  Relationship between Government Action and Organisational Drivers

unlikely that restorative dialogue of this nature will make any dramatic change to the ­position of older workers if we do not also review employers’ legal ­obligations.5 At the same time, legal rules need to be complemented by more autonomous forms of self-regulation, such as inter-organisational benchmarking.6 Thus, when looking to review the EqA, we must consider changes to both soft and hard law measures, across the full height of the ‘enforcement pyramid’.

II.  How might the Laws be Improved? If age discrimination legislation is to achieve its posited instrumental and intrinsic ends, it is clear that the current legislative framework is not fit for purpose. Therefore, we must consider how these ends might better be achieved. First, a reflexive law approach modelled on the enforcement pyramid could be strengthened via a positive duty on employers to achieve equality, or a responsibility to consult regarding equality matters. The Delphi method demonstrated the experts’ enduring lack of consensus regarding the imposition of a positive duty on employers; however, a responsibility to consult was regarded as desirable, feasible and important. An experimentalist approach to reflexive law could be strengthened via: the

5  A Klarsfeld et al, ‘Social Regulation and Diversity Management: A Comparative Study of France, Canada and the UK’ (2012) 18 European Journal of Industrial Relations 309, 322. 6 ibid.

How might the Laws be Improved?

 223

adoption of clear and evolving goals and performance measures for employers; regular reporting and peer scrutiny of employer actions and outcomes, including by extending equality reporting to age as a protected characteristic; the p ­ ooling of information between employers; the use of incentives and disincentives to encourage participation; centralised coordination of the review of employer ­ action; and promotion of employer benchmarking. The Delphi respondents endorsed the promotion of employer benchmarking via facilitated opportunities for employers to network, discuss issues around ageing and share best practice. However, respondents did not support the use of financial incentives to encourage participation or reform. Secondly, drawing on the comparative experience of Finland, UK organisational responses to ageing could be strengthened via the adoption of age management approaches. While the experts endorsed the desirability, feasibility and importance of age management in the UK, it was less clear how this approach should be implemented in practice. In the Delphi survey, respondents emphasised the role of employer bodies, industry bodies and professional organisations in promoting the business case for age management, providing additional information and guidance to employers, and securing effective leadership. Respondents also endorsed the provision of facilitated opportunities for employers to network, ­discuss issues around ageing and share best practice but, again, felt that this responsibility should lie with employer or industry bodies and professional bodies, not government. In contrast to the broad support and consensus regarding information and persuasion measures, respondents did not support the use of regulation or financial incentives to achieve change. Therefore, through this research I have identified five priorities for reform going forward: —— introducing a responsibility for employers to consult on equality matters; —— developing effective leadership regarding the ageing workforce; —— providing facilitated opportunities for employers to network and share best practice; —— providing information to employers on age management; and —— promoting the business case for age management. Given the expert consensus around these reforms, they should be vigorously ­pursued by governments and stakeholders alike. Indeed, the respondents in the Delphi survey explicitly rejected maintaining the status quo: doing nothing to address ageing is not a viable option. While the expert respondents in the D ­ elphi survey favoured action by employer or industry bodies and professional bodies, not government, it is clear that government could play a significant role in prompting and leading change in this area. Indeed, this was recognised by the experts in the qualitative interviews. Governments cannot avoid or devolve their responsibilities in the face of demographic ageing: ultimately, extending working lives is a government priority, and government action is required to compel action by other stakeholders.

224 

Conclusion

Going beyond the reforms that achieved consensus in the Delphi survey, other legal reforms should be considered as part of any programme to address demographic change. More particularly, in Chapter three I put forward three key reforms to the existing legal framework: —— limit or amend existing exceptions to age discrimination legislation (­ including by prohibiting EJRAs); —— clarify and simplify if positive action provisions in the EqA; and —— rethink the EqA’s primary reliance on individual enforcement, including by strengthening the position of the EHRC and through the imposition of positive duties on employers. These reforms need to be debated and discussed, and are likely to provoke both strong support and strong opposition. In this contentious space, the tripartite development of reforms in Finland provides one model for how legal changes of this nature could be developed, with the ultimate goal of ensuring that changes obtain broad acceptance and ‘buy-in’ before they are adopted. At the same time, government reform can be used to lead change, and stakeholder opposition to reform needs to be seen in the context of changing normative standards. Beyond these doctrinal reforms, we also need to revisit the orientation and goals of age discrimination laws in the UK. This research has identified pervasive tensions in the UK regarding what is required (or should be required) to achieve age equality.7 In the UK, age equality is generally equated to ‘age-neutral’ work ­structures,8 where age has ‘little relevance’ in decision-making and the determination of rights and responsibilities.9 In this conception, age discrimination legislation is designed to remove age-based barriers that prevent individuals from participating in employment.10 In the organisational case studies, all UK organisations aspired to create age-neutral policies and work practices. While UK organisations may aspire to be age-neutral, this ignores the ongoing importance and prevalence of age norms as a form of social ordering and control.11 Age norms and standards remain firmly entrenched in the UK, as age continues to be a category we deploy to help make sense of the world.12 Thus, a truly age-neutral society is arguably not feasible in practice.13 Instead, age-­ neutral behaviour may cloak and obscure practices that are indirectly

7 See, eg, G Kirton and A Greene, The Dynamics of Managing Diversity: A Critical Approach, 3rd edn (Oxford, Elsevier, 2010) 212–13. 8  HT Schrank and JM Waring, ‘Older Workers: Ambivalence and Interventions’ (1989) 503 Annals of the American Academy of Political and Social Science 113. 9  C Duncan, ‘The Dangers and Limitations of Equality Agendas as Means for Tackling Old-Age Prejudice’ (2008) 28 Ageing & Society 1133, 1138. 10  P Uhlenberg, ‘Introduction: Why Study Age Integration?’ (2000) 40 The Gerontologist 261, 261. 11  Duncan (n 9 above) 1145. 12  ibid 1146. 13 ibid.

How might the Laws be Improved?

 225

discriminatory, and may perpetuate or legitimise inequality.14 For example, older workers may be excluded from the workforce through age-neutral rationales such as ‘business financial well-being’ and ‘long-term stability’, particularly where older workers are perceived as being more expensive.15 Age neutrality may also deprive older workers of existing age-related benefits (such as those tied to length of service)16 and benefits tailored to workers’ specific needs.17 Therefore, while age-neutral practices are arguably necessary for increasing workforce participation rates for older workers, they are inadequate for actively facilitating workforce participation.18 Alternatively, age equality could be pursued by ‘age-specific’ interventions or programmes that seek to accommodate workers’ individual needs, taking age into consideration. This approach values difference, rather than seeking to dissolve difference (as is the case in age-neutral policies).19 This is evident in the Finnish organisational case studies, where specific measures have been put in place to adapt work to older workers’ needs, and is inherent in the very notions of age management and work ability. Age-specific measures are closely related to positive action and positive duties to achieve equality, and may be one means of securing substantive equality of opportunity and equality of results. It is therefore unsurprising that age-specific measures appear antithetical to the UK context, with its strong focus on formal equality of opportunity and equal treatment. Indeed, in the UK, age-consciousness is often conflated with ageism and age discrimination. The need to reconcile a desire to treat people equally with an acknowledgement of individual needs and differences is a central dilemma in equality policy-making in the UK.20

14  VJ Roscigno et al, ‘Age Discrimination, Social Closure and Employment’ (2007) 86 Social Forces 313, 325. 15 ibid. 16  The fact that age neutrality undermines specific benefits for older workers is not a reason to reject it out of hand. While older workers may be seen as ‘deserving’ of additional payments or benefits, it is essential that any exceptions to the principle of equal treatment be seriously scrutinised and justified, rather than being accepted as automatically positive: M Sargeant, ‘The Employment Equality (Age) Regulations 2006: A  Legitimisation of  Age  Discrimination in  Employment’ (2006) 35 Industrial Law Journal 209, 222. Indeed, Duncan argues that seniority-based payment systems may render age discrimination a rational means of reducing employment costs: C Duncan, ‘Assessing Anti-Ageism Routes to Older Worker Re-Engagement’ (2003) 17 Work, Employment & Society 101, 106. Age-related benefits may both harm and benefit older workers, and should be subject to scrutiny. 17  Duncan (n 16 above) 109; Duncan (n 9 above) 1152. 18 T MacDermott, ‘Older Workers and Extended Workforce Participation: Moving beyond the “Barriers to Work” Approach’ (2014) 14 International Journal of Discrimination and the Law 83, 90. Indeed, age-neutral processes may actually force older workers to retire earlier: see V Beck and G Williams, ‘The (Performance) Management of Retirement and the Limits of Individual Choice’ (2015) 29 Work, Employment & Society 267, 273. 19  Duncan (n 16 above) 110. 20  G Healy et al, ‘Inequalities, Intersectionality and Equality and Diversity Initiatives: The Conundrums and Challenges of Researching Equality, Inequalities and Diversity’ in G Healy et al (eds), Equality, Inequalities and Diversity: Contemporary Challenges and Strategies, Management, Work and Organisations (Basingstoke, Palgrave Macmillan, 2010) 3.

226 

Conclusion

While age-specific measures are well-established in the Finnish context, it appears that this tradition is shifting: both the Finnish council and the Finnish experts felt that a focus on older workers had become ‘old-fashioned’, and was being replaced by a ‘life-course’ perspective that accommodates individuals of all ages. However, this is still distinct from the age-neutral model in the UK: age is still relevant in Finland, though the focus is now on individuals of all ages, rather than older workers specifically. Therefore, the Finnish approach does not deny the impact of age, but seeks to recognise the diverse consequences of age and ageing on all workers, and how age might interact with other protected characteristics. As a result, the Finnish approach engages with ideas of intersectionality and multiple discrimination.21 This is a more sophisticated approach than the UK focus on age neutrality. It is timely for researchers and policy-makers to consider how organisations can adopt policies and processes that acknowledge and accommodate the impact of age, but do not condone age discrimination. The experiences of comparative jurisdictions, including Finland, may prove enlightening for the UK in this area. However, this will also require a broader review of how we conceive of ‘equality’ in the UK. At present, the EqA embodies a narrow legislative conception of ‘equality’, namely equal treatment and formal equality of opportunity. This legislative focus encourages employers to define ‘compliance’ narrowly as adherence to ‘fair’ and formalised procedures,22 rather than as achieving substantive equality of opportunity or outcomes through age-aware practices. As illustrated by the WERS analysis in Chapter five, few employers have monitoring or review practices in place to assess the efficacy of their formal policies. Thus, there is a risk that UK employers will become focused on compliance as ‘an end in its own right’, distorting and derailing the substantive purpose of equality law.23 A legislative focus on formal equality of opportunity and equal treatment can therefore restrict substantive equality in practice. If age discrimination laws are to achieve instrumental and intrinsic ends, we must review the broader goals of equality legislation. This has the potential to challenge the very conception of ‘equality’ adopted in the UK legal framework. At a deeper and broader level, we also need to address the enduring ambivalence towards age and ageing in the UK. As noted in Chapter one, age is generally seen as ‘different’ to other protected characteristics, and age discrimination is often regarded as less harmful or detrimental than other forms of discrimination. This ambivalence is embodied in the terms of the Framework Directive and the EqA: unlike other protected characteristics, direct age discrimination can be objectively justified, and there is a general acceptance in the EU that mandatory retirement (an archetypal form of age discrimination) is an acceptable tool 21 

See further ibid 4–10. M Noon et al, ‘The Equality Effects of the “Hyper-Formalization” of Selection’ (2013) 24 British Journal of Management 333, 333–34. 23  ibid 333–34. 22 

How might the Laws be Improved?

 227

for workforce management. Law’s ambiguity towards age and ageing may have significant ­practical consequences. Law declares and affirms prevailing social values and ideas:24 by treating age discrimination as ‘different’ and less harmful than other forms of discrimination, the law condones ageism and perpetuates ageist values. In this book I have questioned whether age is inherently different to other protected characteristics, and have critiqued the legal justifications for mandatory retirement in EU and UK law. Rather than reflecting any inherent differences between age and other protected characteristics, the legislation reflects a different social value placed on age equality. In distinguishing between legitimate and illegitimate age discrimination, the law has struck a balance that favours the instrumental ends of age discrimination laws over any intrinsic objectives.25 While organisational efficiency would not generally be allowed to ‘trump’ equality for other protected characteristics, age is seen as less worthy of protection. Therefore, rather than any inherent difference in the nature of protected characteristics, age equality is just less socially valuable than other types of equality. This fundamentally undermines the legal rationalisations of age discrimination under EU and UK law, which merely perpetuate and reinforce negative social norms around ageing. Thus, rather ironically, allowing economic ends to trump intrinsic objectives may undermine the instrumental aims of age discrimination law in the long-term, as it encourages ageist views to endure. Before concluding this work, I would flag a word of caution on the use of reflexive law. This book has used reflexive law to critique and model changes to equality law in the UK. Reflexive law is helpful for explaining why age discrimination law has had limited impact on organisational practice, and to shift our focus towards more creative regulatory interventions in this area. However, reflexive law should not be confused with soft or non-law. As noted in Chapter three, the focus of UK governments in relation to equality law has been primarily on ‘restorative dialogue’ and the use of information and persuasion mechanisms.26 While the enforcement pyramid posits dialogue and self-regulation as the default forms of intervention in a reflexive law model,27 the UK government’s focus on education and persuasion mechanisms has not resulted in concrete organisational change. This is understandable, as reflexive law requires more than just restorative dialogue to achieve change: there must also be increasingly interventionist action where dialogue and self-regulation fail. However, this intervention is actually being undermined by government reforms: as individual enforcement becomes increasingly difficult, 24  JR Gusfield, Symbolic Crusade: Status Politics and the American Temperance Movement, 2nd edn (Chicago IL, University of Illinois Press, 1986) 4–5. 25  S Fredman, Discrimination Law, Clarendon Law Series, 2nd edn (Oxford, Oxford University Press, 2011) 106. 26  J Macnicol, ‘The Age Discrimination Debate in Britain: From the 1930s to the Present’ (2005) 4 Social Policy and Society 295, 298. 27  J Braithwaite, Restorative Justice and Responsive Regulation (Oxford, Oxford University Press, 2002) 30.

228 

Conclusion

and the power and resources of the EHRC are progressively undermined, any progress beyond dialogue appears unlikely. This demonstrates the fundamental limitations of reflexive law in overcoming a deregulatory governmental agenda.28 While reflexive law is posited as a key means of avoiding or ‘transcending’ the deregulation debate, it appears that it is actually becoming a means of furthering governments’ retreat from social and employment regulation.29 The UK experts’ focus on information and persuasion mechanisms, at the expense of legal reform, indicates that a reflexive law approach based on the enforcement pyramid may justify and theoretically validate a deregulatory push. Similarly, a focus on peer review and scrutiny of employer action, consistent with the experimentalist architecture, decentres government responsibility and action in achieving change. Indeed, in the Delphi survey, the experts felt that even information and persuasion activities should not be the role of government. Therefore, the use of a reflexive law framework may condone and enhance the push for deregulation and a withering away of governmental responsibility, rather than providing a way out of the deregulation/regulation divide. It may also place additional burdens on third sector organisations to pick up former government roles and responsibilities, even where they lack the resources and capacity to fulfill such a role. Finally, reflexive law is concerned with process, not substance, and has little to say about the normative merits of law. Reflexive measures may not be beneficial from a normative perspective. Thus, reflexive law should be used with a degree of caution and scepticism.

III. Conclusion Labour law is likely to have increasing relevance for older workers in coming years. The current legal approach to demographic ageing centres on age discrimination legislation as a tool for achieving instrumental and intrinsic ends. However, in this book I have demonstrated the substantial limitations of the laws for achieving either of these aims. There is limited evidence that the EqA is creating opportunities for older workers, reducing demand on social security systems, or upholding the dignity of older workers. The enduring pervasiveness of age discrimination, limited prevalence of measures to support older workers, and low levels of ageawareness among UK employers indicate that significantly more needs to be done before UK age discrimination legislation will be effective in practice.

28  cf I Ayres and J Braithwaite, Responsive Regulation: Transcending the Deregulation Debate, Oxford Socio-Legal Studies (Oxford, Oxford University Press, 1992). 29  See C Parker, ‘Twenty Years of Responsive Regulation: An Appreciation and Appraisal’ (2013) 7 Regulation & Governance 2, 5, 7; R Rogowski, Reflexive Labour Law in the World Society (Cheltenham, Edward Elgar, 2013) 38.

Conclusion

 229

In this book, I have identified five key proposals for reform to enhance the efficacy of the EqA. These reforms have achieved consensus across a range of experts as being desirable, feasible and important. Therefore, they should be vigorously pursued and should be readily implementable. However, these reforms will not go far enough to address the implications of ageing for labour law. The current framework for age equality in the UK, which focuses on equal treatment and formal equality of opportunity, minimalistic and weak government intervention, and broad discretion for employers with limited oversight, is unlikely to prove sufficient for achieving the instrumental or intrinsic objectives of age discrimination legislation. Therefore, these findings show substantial cause for pessimism regarding the impact of age discrimination laws. Without broad and fundamental change, the UK approach to age equality is likely to remain, at best, insipid.

Appendix 1—Glossary of Statistical Terms and Tests Test

Description1

Akaike information criterion (AIC)

A measure of the quality of a statistical model, comparing the goodness of model fit with the model complexity.

Bivariate correlation

Exploration of the extent to which two variables are related.

Chi-square test of independence / Pearson chi-square (χ2)

A statistic for categorical data, which determines whether there are statistically significant differences between the observed and expected frequencies of variables.

Conditional change model

A statistical model for assessing change that does not assume that a score at a later time point is independent of the response at the first time point. The original response is included in the model as a predictor variable.

Cramer’s V coefficient (V)

Tool for comparing multiple chi-square statistics across contingency tables of varying sizes. It is a measure of the relative strength of an association between two variables, ranging from 0 to 1 (perfect association). A Cramer’s V of .10 is often regarded as a minimum threshold for suggesting a substantive relationship between two variables.

Degrees of freedom (df )

The number of values that are free to vary in calculating a statistic.

Mann–Whitney U test

Test of the statistical significance of differences between two groups that have been measured on an ordinal scale.

Negative binomial model

Like the Poisson log-linear model, but uses a different probability distribution, based on the number of failures in a sequence of binomial (Bernoulli) trials. Suits over-dispersed count data.

Ordered probit model

Form of regression analysis using an ordinal dependent variable. Probit is short for ‘probability unit’.

Ordinal scale

Where categories can be ranked but the distances between categories are unknown.

Paired sample t-test

Statistical test that compares correlated samples, such as scores from the same subjects on a pre-test and post-test. (continued)

Appendix 1

 231

Test

Description1

Pearson’s correlation coefficient (Pearson’s r)

A statistic showing the degree of linear relationship between two variables that have been measured on interval or ratio scales.

Poisson log-linear model

Model for studying relations between categorical variables, using a Poisson form of probability distribution. Transforms equations by taking their natural logs to make them linear.

Spearman’s Rank Order correlation (Spearman’s rho)

A statistic showing the degree of one-directional relationship between two variables that are arranged in rank order and measured on an ordinal scale.

Note: 1  Adapted from D Cramer, The Sage Dictionary of Statistics: A Practical Resource for Students in the Social Sciences (London, Sage Publications, 2004); WP Vogt, Dictionary of Statistics and Methodology: A Nontechnical Guide for the Social Sciences, 3rd edn (Thousand Oaks CA, Sage Publications, 2005).

232

INDEX

Acas conciliation  72, 97, 98 helpline  100 Age Action Alliance  99–100, 103 Age-aware practice score see also Workplace Employment Relations Study and best practice  122–123 and organisational characteristics  115–127, 130–134 calculation  113 change over time  127–134 spread across workplaces  113–114 Age discrimination legislation see Discrimination law; Equality Act 2010; EU Framework Directive; Finland non-regulatory action  76, 77 older workers’ experiences of  166–169 voluntary code of practice  6–7, 53 Age management  90, 151, 153, 185, 210–212, 218, 223 Age neutrality  202, 203, 221, 224–226 Age Positive campaign  103 Attitudes towards older workers  94–97, 102–103, 166–169 in Finland  154–156 Best practice organisations adoption of best practice policies  172–178 barriers inhibiting best practice  177, 204–205 business needs and market pressures  174–175, 190–191, 197–198 Finnish local council (case study 4)  185–187 Finnish manufacturing company (case study 5)  188–189 government intervention to encourage  221–222 legal change  92–93, 172–173, 192–195, 197 national context  176, 202–204 organisational characteristics  90, 115–127, 130–134, 198–202, 205 organisational culture  174, 192, 198 organisational leadership  175–176, 191, 197, 198 peer benchmarking  79, 176, 223 pension provision  201

trade unions  117–118, 121–122, 124, 125, 127, 132, 134, 176–177, 195–197 types of organisations  177–178 UK city council (case study 3)  183–184 UK large enterprise (case study 2)  181–182 UK small enterprise (case study 1)  179–181 WERS data set  122–123 Case law Commission v Hungary (2012)  51–52 Engel v Transport and Environment Committee of London Councils (2013)  63, 65, 66, 67 Hampton v Lord Chancellor (2008)  63, 64, 65, 66, 67 John v Ministry of Justice (2014)  63, 65 Lindsay v Department for Employment and Learning (2013)  63, 64 Martin v Professional Match Game Officials Ltd (2010)  63, 64, 65, 66, 67 R (Age Concern England) v Secretary of State for Business Enterprise and Regulatory Reform (2009)   56–57 R (Age UK) v Secretary of State for Business Innovation & Skills (2009)  57 Seldon v Clarkson Wright & Jakes (2012)   58–67 White v Ministry of Justice (2014)  63, 64, 65 Willey v England and Wales Cricket Board Ltd (2015)  63, 64, 65, 67 Change fatigue  106 Collective bargaining see Trade unions Command and control regulation  74–75 Comparative perspective on age discrimination see Finland Compliance approaches  106–109, 173, 203, 220 expert perspectives  90–92, 106 moving beyond  108–109 WERS findings  113–115 Compulsory retirement see Mandatory retirement Default retirement age  Finland  142–143 UK  55–58, 68

234 

Index

Delphi method overview  16, 39–41, 206 process  206–210 results of scenarios  210–217 age management  210–212 consultation  216–217 doing nothing  217 information and guidance  212–213 leadership  214 national dialogue regarding rights, obligations and privileges  214–215 positive duty on employers  215–216 Dialogue on age  101–102, 214–215 Dignity  9, 48–49, 60, 63–64, 87–88 Discrimination law aims  6–13 Equality Act 2010 see Equality Act 2010 equality, concept and meaning of  10–12, 159, 224, 226 EU Framework Directive see EU Framework Directive EU regulation see EU Framework Directive Finnish regulation see Finland instrumental ends  7–9 failure to achieve  221 intrinsic ends  9–12 failure to achieve  69, 221 irrational prejudice  7–8 role  6–7 comparison with Finland  159 expert views of  92–93, 106, 109 UK regulation see Default retirement age; Employer-Justified Retirement Age; Equality Act 2010; Mandatory retirement Diversity management   13, 197 Doctrinal research method  29–31 Empirical legal research  24–26 Employer-Justified Retirement Age  58 case law  58–67 legal uncertainty  62, 68, 69, 88 legitimate aims  58–60 dignity  60 intergenerational fairness  59 limitations   68–72, 87–89 proportionality  61 reform  73, 83, 87 Seldon decision, consequences of  61–67 Equality Act 2010 claims and remedies  70–72 direct discrimination, justification of  53, 69, 86–87 enforcement  70–72, 74, 97–98, 194–195 exceptions  53–54, 72–73, 87 expert views of  86–89 implementation by employers, expert views of  89–93, 106 indirect discrimination, justification of  53, 69

legal uncertainty  87, 88, 93–94, 98–100 practical impact of  93–94 reforms to address  98–100 limitations  68–72, 86–89, 221 mandatory retirement  55–58 see also Employer-Justified Retirement Age positive action  53–54, 69–70, 73–74, 93 Public Sector Equality Duty  54–55, 78, 79, 117 impact on best practice  193 reform  80–83, 215–216 reflexive law approach  74–83 reform  72–74, 222–224 Equality and Human Rights Commission enforcement action  70, 74 limitations of efficacy  71, 74, 107 potential role  80 Equality law see Discrimination law; Equality Act 2010; EU Framework Directive; Finland EU Framework Directive  42–52 equal treatment principle  43–44 exceptions  43–44 genuine occupational requirements  43 implementation in Finland  139–142 implementation in the UK  6, 52 limitations  68–72 mandatory retirement  44–52 legitimate aims  45–49 dignity  48–49 fair innings  46–47 workforce planning  49 proportionality  49–52 positive action  44 EU regulation see EU Framework Directive  Everyday Ageism Project  103 Finland and ageing  136 attitudinal change  154–156 comparative analysis  37–38 comparison with the UK differences   157–159 similarities   160 statistical impact  160–169 enforcement of discrimination law  141–142, 146–148 financial incentives  145 impact on best practice   190, 193 implementation of law  145–146 by employers  150–154 involvement of the social partners  148–150, 195 mandatory retirement  142–143 National Programme for Ageing Workers  136, 154–155 non-discrimination law  138–142 positive duties  139–140, 143–144 role of law  157, 159

Index Finnish local council (case study 4)  185–187 Finnish manufacturing company (case study 5)  188–189 Framework Directive see EU Framework Directive Information and guidance  75–77, 98–100, 153–154, 212–213 Intergenerational fairness  59, 63, 201 fair innings  46–47, 50, 51 Labour law implications of ageing for  2–6 Leadership  214 organisational leadership  175–176, 191, 197, 198  Legal doctrinal research see Methods  Mandatory retirement see also Employer-Justified Retirement Age and consent  50–51, 66–67, 68 EU regulation  44–52 Finnish regulation  142–143 UK regulation  55–67 normal retirement age  55–58 Methods comparative research  37–38, 138 Delphi method see also Delphi method  39–41 doctrinal research  29–31 empirical legal research  24–26 expert interviews  36–37 integrating different research methods  33–34 mixed methods  26–29, 35–36 organisational case studies  38–39, 178–179 qualitative research methods  32–33 quantitative research methods  31–32 statistical analysis  37 Mixed methods research  26–29, 35–36 Non-discrimination see Age discrimination; Discrimination law; Equality Act 2010; EU Framework Directive; Finland Organisational case studies see Best practice organisations; Methods Pensions  2, 3–5, 7, 50, 161, 166, 201 Performance management  48, 60, 64, 93, 100–101, 200 Public Sector Equality Duty  54–55, 78, 79, 117 impact on best practice  193 reform  80–83, 215–216

 235

Reflexive law see also Systems theory  17–24, 93 as a basis for reform  80–83, 222–223 consultative duties  83 positive duties  80–82 as a model for age discrimination law  23, 75–80 caution on the use of  227–228 enforcement pyramid  19–21, 75–76, 80, 142 experimentalist architecture  20–23, 75, 79–80 Athena SWAN Charter award scheme  22–23 institutional requirements  77–78 Responsibility for reform  104–105, 218 role of government  89, 104–105, 218, 219, 223 in Finland  159 Retirement ages see Default retirement age; Employer-Justified Retirement Age; EU Framework Directive; Finland; Mandatory retirement Systems theory  18–19, 74–75, 95 Trade unions and age-aware practices  117–118, 121–122, 124, 125, 127, 132, 134 and best practice  176–177, 195–197 duty to consult  83, 216–217 in Finland  148–150, 195 justifying retirement ages  50–51 limited UK presence  107, 169 UK Business Champion for Older Workers  103 UK city council (case study 3)  183–184 UK large enterprise (case study 2)  181–182 UK regulation see Default retirement age; Discrimination law; Employer-Justified Retirement Age; Equality Act 2010; UK small enterprise (case study 1)  179–181 Unions see Trade unions Work ability  151–153, 159, 203 Workplace Employment Relations Study   111 see also Age-aware practice score age-aware practices over time  127–134 age-aware workplaces by organisational characteristics  115–127 compliance approaches in UK workplaces  113–115 means of analysis  110–113 Workforce planning  49, 201

236