Uncertainty and Challenges in Contemporary Economic Behaviour 9781800430969, 9781800430952, 9781800430976

Citation preview

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Uncertainty and Challenges in Contemporary Economic Behaviour

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

This page intentionally left blank

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Emerald Studies in Finance, Insurance, and Risk Management

Uncertainty and Challenges in Contemporary Economic Behaviour

EDITED BY

ERCAN ÖZEN University of Usak, Turkey

SİMON GRIMA

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

University of Malta, Malta

United Kingdom – North America – Japan – India – Malaysia – China

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Emerald Publishing Limited Howard House, Wagon Lane, Bingley BD16 1WA, UK

First edition 2020 Copyright © 2020 Emerald Publishing Limited Reprints and permissions service Contact: [email protected] No part of this book may be reproduced, stored in a retrieval system, transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without either the prior written permission of the publisher or a licence permitting restricted copying issued in the UK by The Copyright Licensing Agency and in the USA by The Copyright Clearance Center. Any opinions expressed in the chapters are those of the authors. Whilst Emerald makes every effort to ensure the quality and accuracy of its content, Emerald makes no representation implied or otherwise, as to the chapters’ suitability and application and disclaims any warranties, express or implied, to their use. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

ISBN: 978-1-80043-096-9 (Print) ISBN: 978-1-80043-095-2 (Online) ISBN: 978-1-80043-097-6 (Epub)

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Contents

List of Contributors

vii

About the Contributors

ix

ESFIRM Editorial Board

xvii

Series Editors’ Introduction

xix

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Chapter 1  Challenges and Perspectives of International Migration in Europe within the Brexit Framework Graţiela Georgiana Noja, Mirela Cristea, Petru Ştefea and Ciprian Panzaru

1

Chapter 2  The Impact of US Economic Policy Uncertainty on Developing Countries Under Different Economic Cycles: A Nonlinear Approach Deniz Erer and Elif Erer

21

Chapter 3  The Investigation of Zombie Firms’ Existence in Borsa İstanbul Manufacturing Industry Emre Kaplanoğlu and Canan Yükçü

37

Chapter 4  Forensic Accounting Theory Peterson K. Ozili

49

Chapter 5  Econometric Analysis for the Period Between 2003 and 2018 as Regards the Functioning of Interest Channel of Monetary Transmission Mechanism in Turkey Berna Kaçar and Huriye Gonca Diler

61

Chapter 6  A Conceptual Framework for Behavioral Accounting Ahmet Coşkun and Mehtap Karakoç

77

Chapter 7  Theories of Financial Inclusion Peterson K. Ozili

89

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

vi   Contents

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Chapter 8  Shareholder Activism: What Does It Refer to? Yeşim Şendur117 Chapter 9  Relationship Between Interest Rates, Exchange Rate and Investor Sentiment in Turkey N. Serap Vurur

127

Chapter 10  Effect of Climate Change on Financial Institutions and the Financial System Peterson K. Ozili

139

Chapter 11  Testing for Asymmetric Causality Between Developed and Emerging Markets Letife Özdemir

145

Chapter 12  An Assessment for Financial Performance of Banks Listed in Borsa Istanbul By Multiple Criteria Decision Making Eser Yeşildağ, Ercan Özen and Ender Baykut

159

Chapter 13  100 Quotes from the Global Financial Crisis: Lessons for the Future Peterson K. Ozili

185

Chapter 14  Labouring Behind Closed Doors: The Working and Living Conditions of Filipino Live-in Care Workers in Malta Mario Thomas Vassallo and Manwel Debono

195

Chapter 15  Too Small to Hedge? The Case for Derivatives Hedging for a Small Island State Konrad Farrugia, Matthew Attard and Peter J. Baldacchino

217

Chapter 16  The Effect of Micro Entrepreneur Arrangements to City Culture in Terms of Urban Esthetics and Street Peace in the Urbanization Process Ahmet Fidan

259

Chapter 17  A Study of Kosovo’s Fiscal Policies and Tax System: Key Issues to Align with European Standards Gani Asllani, Simon Grima and Sharon Seychell

273

Index

285

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

List of Contributors

Gani Asllani

University “Haxhi Zeka”, Kosovo

Matthew Attard

University of Malta, Malta

Peter J. Baldacchino

University of Malta, Malta

Ender Baykut

Afyon Kocatepe University, Turkey

Ahmet Coşkun

University of Uşak, Turkey

Mirela Cristea

University of Craiova, Romania

Manwel Debono

University of Malta, Malta

Huriye Gonca Diler

Afyon Kocatepe University, Turkey

Deniz Erer

Independent Researcher, Turkey

Elif Erer

Independent Researcher, Turkey

Konrad Farrugia

University of Malta, Malta

Ahmet Fidan

Ordu University, Turkey

Simon Grima

University of Malta, Malta

Berna Kaçar

Afyon Kocatepe University, Turkey

Emre Kaplanoğlu

Ege University, Turkey

Mehtap Karakoç

University of Uşak, Turkey

Graţiela Georgiana Noja

West University of Timisoara, Romania

Letife Özdemir

Afyon Kocatepe University, Turkey

Ercan Özen

University of Uşak, Turkey

Peterson K Ozili

Central Bank of Nigeria, Nigeria

Ciprian Panzaru

West University of Timisoara, Romania

Yeşim Şendur

İzmir Katip Çalebi University, Turkey

Sharon Seychell

University of Malta, Malta

Petru Ştefea

West University of Timisoara, Romania

Mario Thomas Vassallo

University of Malta, Malta

N. Serap Vurur

Afyon Kocatepe University, Turkey

Eser Yeşildağ

University of Uşak, Turkey

Canan Yükçü

Ege University, Turkey

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

This page intentionally left blank

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

About the Contributors

Gani Asllani holds MA and PhD in Finance from University of Pristina, Kosovo. He teaches public fnance, taxation and competition rights. His work in the area of market economy, competition, tax policy, economic development appeared in relevant international journals. He is an Associate Professor at University “Haxhi Zeka,” Peja within the Law Faculty. During his 20 years of experience, he has contributed to the creation of the tax system in Kosovo, making tax analysis and expertise, participating in several working groups at international level (IMF, OECD, and Robert Shuman Center for Advance Studies). Matthew Attard is currently reading for a Master of Science in Applied Actuarial Science at the University of Kent. He has previously worked as a Client Account Manager with a leading Maltese insurance manager. His focus was assisting insurers in the implementation of the Solvency II requirements primarily their capital requirements and regulatory reporting processes. Prior to this, he was engaged as audit and assurance senior with one of the Big Four audit frms, specialising in the audits of local insurance undertakings.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

He is a Certifed Public Accountant and a student member of the Institute and Faculty of Actuaries (United Kingdom). He holds a Master’s in Accountancy from the University of Malta and his research interests comprise: corporate fnancial reporting, enterprise risk management and stochastic risk modelling. Peter J. Baldacchino is Advisor to Rector – Financial Affairs and Head, Department of Accountancy, F.E.M.A, University of Malta. He holds both a PhD and an MPhil from Loughborough Business School. He is also a Fellow of the Chartered Association of Certifed Accountants, a Fellow of the Malta Institute of Accountants and a Certifed Public Accountant. He specialises in the corporate governance and fnancial strategy of listed entities and cooperatives, and also in internal/external/public sector auditing with particular reference to the implications thereon of state smallness. He lectures in postgraduate professional and business programs in these areas and has related publications in various international refereed journals. He is also the Director at the University Group of companies and both the Director and Audit Committee Chairperson at the Central Bank of Malta. He has extensive experience in the governance of large Maltese organisations, including listed groups and cooperatives, with past positions including chairmanships and memberships of audit and risk management committees in large organisations.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

x    About the Contributors Ender Baykut received his BSc in Management (in English) (2011), MSc in Finance (2013), PhD in Finance (2017) from Afyon Kocatepe University. Now he is an Assistant Professor of Finance in the Department of Management (in English) at Economics and Administrative Sciences Faculty of Afyon Kocatepe University. His current research interest includes different aspects of fnance especially in Corporate Governance and Financial Econometrics. He has co-authored one book and more than 20 international indexing articles, and more than 20 international conferences participation. Ahmet Coşkun (June 1, 1976) received his Bachelor’s degree in Finance (1997) from Faculty of Economics and Administrative Sciences, MSc in Management (2008) from Afyon Kocatepe University, Social Science Institute and PhD in Management (2013) from Adnan Menderes University, Social Science İnstitute. He is an Assistant Professor in Banking and Finance and working for Uşak University and he is studying Accounting.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Mirela Cristea is a Professor at the University of Craiova (Romania), Faculty of Economics and Business Administration, Department of Finance, Banking and Economic Analysis, Center for Banking and Financial Research. Her main research directions and teaching activities are insurance and private pension funds, banking administration, interdisciplinary research in economics. She obtained the scientifc title of PhD in Economics at the University of Craiova and graduated the Post-Doctoral School in Economics and Applied Sciences in Economics at the Romanian Academy from Bucharest, on the subject of private pension system in Romania. She received two awards from the Faculties on Economics Association from Romania, in 2007 and 2013, for two books on life insurance and pension system subjects. In addition, she is a reviewer and member of the scientifc and organisational committees for numerous journals and conferences. Manwel Debono, PhD (Melit), C.Psychol (UK) is a Chartered Occupational Psychologist and a Senior Lecturer at the University of Malta. He served as the Director of the Centre for Labour Studies for fve years. He has been involved in numerous local and European research projects about industrial relations and working conditions. He has contributed to the European Foundation for the Improvement of Living and Working Conditions since 2003, and to the European Employment Policy Observatory for over a decade. His research interests include industrial relations, working conditions, human resources and career development. Huriye Gonca Diler (March 23, 1976) received a Bachelor’s degree in Economics (2000) from Faculty of Economics and Administrative Sciences, an MSc in Economy (2003) and a PhD in Economy (2011) from Afyon Kocatepe University, Social Science Institute. She is an Assistant Professor in Economy Policy and working for Afyon Kocatepe University. She is studying Econometrics. Her main duties are involved in teaching Economic Growth, Economic History, History of Economy Thought, Mathematical Economics at the Department of Economy.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

About the Contributors    xi Since 2018, she has been working as an Assistant Director in Afyon Kocatepe University’s Lifelong Education, Application and Research Center. Deniz Erer was born in Izmir, Turkey, in 1987. She holds a BA in Econometrics from Dokuz Eylul University, Turkey, in 2009, a Master’s degree in Econometrics from Dokuz Eylul University in 2011 and a PhD in Economics from Ege University in 2018. Her areas of research include international capital movements, exchange rates, monetary policy, fnancial markets and fnancial instability. She participated in the TUBITAK (The Scientifc and Technological Research Council of Turkey) Project and Ege University Scientifc Research Project. She has published many studies in international journals and participated in many international conferences and congress on economics and fnance.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Elif Erer was born in Izmir, Turkey, in 1987. She holds a BA in Econometrics from Dokuz Eylul University, Turkey, in 2009, a Master’s degree in Econometrics from Dokuz Eylul University in 2011 and a PhD in Economics from Ege University in 2018. Her areas of research include international capital movements, exchange rates, monetary policy, fnancial markets, fnancial instability and public fnance. She has participated in Ege University Scientifc Research Project, published many studies in international journals and also participated in many international conferences and congress on economics and fnance. Konrad Farrugia is an Actuary based in the UK, specialising in Solvency II methodology and valuation, internal model calibration and experience analysis. Previously, he was a Senior Manager with the Malta Financial Services Authority, where he played an active role in the implementation of the Solvency II Directive in Malta. He started his career as a Junior Auditor with one of the Big Four audit frms. He is a Fellow of the Institute and Faculty of Actuaries (United Kingdom), a Chartered Enterprise Risk Actuary and a Certifed Public Accountant. He holds a Master of Science in Applied Actuarial Science with Distinction from the University of Kent, United Kingdom, and a Bachelor of Accountancy (Hons.) from the University of Malta. He is a Visiting Lecturer with the Department of Accountancy at the University of Malta and his main research interests include auditing, risk management and corporate valuation. Ahmet Fidan is an Assistant Professor in Dr Ordu University, graduated from I. U. Faculty of Political Sciences in 1993. He did his Master’s degree in I. U. Human Resources Management Department in 1996. He completed his PhD in M. U. Social Sciences Institute Local Administrations and Decentralization Department. He started his life as a Government Offcer in Istanbul Metropolitan Municipality. He worked in IETT General Management between 1997 and 1999. He worked as an Academic in Balikesir University since 2002 to 2011. He was also on duty as Chairman Advisor in Balikesir Municipality with dual assignment. He was transferred to Avrasya University Faulty of Economics and Administrative Sciences Public Management Department as Deputy Dean in 2011. He was

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

xii    About the Contributors t­ransferred to Ordu University Unye FEAS in 2012. He is managing AEF Culture and Research Center for which he is the establisher. He is the Chief Editor of Journal of Urban Academy, Turkey Interactive Columnist Newspaper, and Journal of Environmental and Natural Studies. His Presidentships include Blacksea Poets and Author Association, Blacksea Nature and Environment Association. Has got Turkey Second Prize in its Category by National Productivity Center MPM 2000. Simon Grima, PhD (Melit.), MSc (Lond), MSc (BCU), BCom (Hons) (Melit.), FFA, FAIA (Acad), is the Head of the Department of Insurance which he set up in 2015 and started and coordinates the MA and MSc Insurance and Risk Management degrees together with the Undergrad degree in Insurance. He served as the President of the Malta Association of Risk Management and the President of the Malta Association of Compliance Offcers between 2013 and 2015, and between 2016 and 2018, respectively. Moreover, he is the Chairman of the Scientifc Education Committee of the Public Risk Management Organization. His research focus is on governance, regulations and internal controls, and has over 30 years of experience varied between fnancial services, academia and public entities. He has acted as co-chair and is a member of the scientifc program committee on some international conferences and is a chief editor, editor and review editor of some journals and book series. He has been awarded Outstanding Reviewer for Journal of Financial Regulation and Compliance in the 2017 Emerald Literati Awards.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Berna Kaçar was born in Sultandağı/Afyonkarahisar in 1991. Firstly, she went to İshaklı school and Sultandağı high school. In 2016, she graduated from Afyon Kocatepe University, Faculty of Economics and Administrative Sciences, as best student of the Economy Department. She received MSc in Economy (2019) from Afyon Kocatepe University, Social Science Institute. She has got Computer Management, Pedagogical Formation Training and Computer-Aided Accounting Certifcates. Emre Kaplanoğlu (May 21, 1978) received his BSc in 2001, MSc in BusinessFinance (2005) and PhD in Business Accounting (2012) from Ege University. Now he is an Associate Professor of Accounting and Coordinator in Program of Accounting and Tax Applications, Bergama Vocational School, Ege University, Turkey. His current research interests include different aspects of accounting and fnance. He has (co-)authored 6 book chapters, more than 20 papers, more than 20 conferences participation and taken part in 4 research and 2 regional development projects. Mehtap Karakoç (September 20, 1980) received her Bachelor’s degree in Business Administration (2004) from Faculty of Economics and Administrative Sciences, Dokuz Eylül University, and an MSc in Management (2008) and a PhD in Management (2012) from Afyon Kocatepe University, Social Science Institute. She is currently an Assistant Professor in Accounting Information System at Uşak University. She is also studying Accounting.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

About the Contributors    xiii Graţiela Georgiana Noja is an Associate Professor at the West University of Timisoara (Romania), Faculty of Economics and Business Administration, Department of Marketing and International Economic Relations. She is the Director of East European Center for Research in Economics and Business. She completed her studies at West University of Timisoara, respectively, PhD in Economics, Master’s in Management and European Integration and BA in International Economic Relations. In addition, she was a Visiting Researcher at the University of East Anglia, School of Economics, she developed several national and international projects, attended numerous research seminars and has a wide membership in various educational organisations, research networks, scientifc and review committees of top tier journals and international conferences. Her main research and teaching activities are developed within the framework of economics and international business area, with a focus on world economy, globalisation and European economic integration.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Letife Özdemir received her BSc in Business (2002), MSc in Business (2005) and PhD in Business Finance (2011) from University of Afyon Kocatepe. Now she is an Associate Professor of Finance in Department of Logistics Management, School of Applied Sciences, University of Afyon Kocatepe, Turkey. Her current research interests include different aspects of Finance. She has authored 19 papers and more than 10 conferences participation. She has been involved in two scientifc projects and has awarded the best third prize in the 21st Finance Symposium. She is a member of TEMA (Turkey Combating Soil Erosion, for Reforestation and the Protection of Natural Resources Foundation). Ercan Özen received his BSc in Public Finance (1994), MSc in Business-Accounting (1997), PhD in Business Finance (2008) from University of Afyon Kocatepe. Now he is Associate Professor of Finance in the Department of Banking and Finance, School of Applied Sciences, University of Uşak, Turkey. His current research interests include different aspects of Finance. He has co-authored three book chapters and more than 40 papers, more than 30 conferences participation, is a member in International Program Committee of three conferences and workshops. He is the Chair of International Applied Social Sciences Congress. He is a certifcated Accountant, a member of Aegon Finance Association and a member of TEMA (Turkey Combating Soil Erosion for Reforestation and the Protection of Natural Resources Foundation). Peterson K Ozili is an established author. He has published research in a wide range of outlets such as journal articles and book series in the fnance and economics discipline. His research focuses mainly on fnancial sector issues particularly in fve areas: fnancial stability, fnancial development, fnancial inclusion, banking and forensic accounting. He has previously published in the British Accounting Review, the International Journal of Managerial Finance, Review of Accounting and Finance and International Finance Review, among others. He currently works as an Economist at the Economic Policy Directorate of the Central Bank of Nigeria.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

xiv    About the Contributors Ciprian Panzaru is an Associate Professor at the West University of Timisoara (Romania), Sociology Department. He coordinates disciplines such as Sociology of Migration, Labour Economics, Social Mobility and Economic Sociology. His areas of interest include agent-based modelling, computational sociology, international migration, labour market issues and population forecasting. He attended the postdoctoral program in Economics, fnanced from European Social Fund, studying the effect of demographic changes on social security system. He was a Visiting Researcher at the Department of Applied Economics of the Free University of Brussels. Currently, he is a PhD Supervisor and leads at the West University of Timisoara, a research group on social and economic complexity. Yeşim Şendur, upon graduating, started her career as a Banker in SME’s banking. She is a Professor of the International Trade and Finance Department at the Adana Alpaslan Türkeş Science and Technology University now. She earned her PhD in Finance from Çukurova University. Her felds of study are corporate governance and dividend policy in general.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Sharon Seychell, BA (Melit.), MA (Melit.), is a Visiting Lecturer at the University of Malta, in Research Method Techniques for Risk Management, Sociology of Risk, as well as Governance, Compliance and Risk Management. She has over 15 years’ experience working in fnancial services and other service industry frms, working in the areas of fnancial management, operations, internal audit, compliance and risk management. She is a member of the Malta Association of Compliance Offcers and holds a Bachelor’s and Master’s degree in Sociology specialising in Risk Management and Internal Controls. Her current research interests include sociology of fnancial services, anomie, risk management and business ethics. Petru Ştefea is the Dean of the Faculty of Economics and Business Administration within the West University of Timisoara (Romania). As Professor and PhD Supervisor at the same faculty, he pursues research and teaching activities on topics related to fnancial analysis, economic analysis, fnancial output analysis, structural funding, investment fnance and effciency. During his academic career, he was also the Vice-Rector of the West University of Timisoara (between 2008 and 2016), a member of CECCAR as expert accountant since 2001 and a member of ANEVAR as company assessor since 1997. He has a wide expertise in national and international projects, and comprehensive membership in various educational and professional organisations and research centres. Mario Thomas Vassallo, PhD (Sheffeld), MA(European Studies), BA (Hons) (Melit) is a Senior Lecturer at the Department of Public Policy, University of Malta. He is the Rector’s Delegate to the Institute for Public Services and his research interests include multi-level governance, politics in arts and literature, and evidence-based policy. He is the editor of the book series Public Life in Malta, and a regular contributor on Campus FM, the offcial radio station of the University of Malta. He serves as a board member of the Islands and Small

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

About the Contributors    xv States Institute and the Foundation for Social Welfare Services. For over a decade, he worked as an HR professional in the telecommunication sector. N. Serap Vurur received her BSc in Economy (2002), MSc in Business-Accounting (2004), PhD in Business Finance (2009) from University of Afyon Kocatepe. Now she is an Assistant Professor of Finance in Department of Accounting, School of Applied Sciences, University of Afyon Kocatepe, Turkey. Her current research interests include different aspects of fnance. She has co-authored more than 15 papers, more than 20 conferences participation. She is a member of TEMA (Turkey Combating Soil Erosion, for Reforestation and the Protection of Natural Resources Foundation). Eser Yeşildağ was born in Hatay City (Turkey) in 1983. In 2005, he graduated from Adnan Menderes University, Faculty of Economics and Administrative Sciences, Department of Business Administration. He holds a Master’s degree in 2008 and a PhD degree in Business Administration in 2013. He wrote his thesis on fnance. Between 2006 and 2010, he worked as an investment expert in Evgin Securities, Şeker Investment and Anadolu Investment Brokerage Houses. Between 2010 and 2013, he worked as a lecturer in Çivril Vocational School of Pamukkale University. In 2013 he started to work with the title of Assistant Professor in the School of Applied Sciences of Uşak University. In 2014, he was appointed as the Deputy Director of the School of Applied Sciences. He teaches fnance courses such as Financial Mathematics, Financial Management, Derivative Markets and Financial Literacy.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Canan Yükçü (September 17, 1965) received her BSc in (1986), MSc in BusinessFinance (2006) from Dokuz Eylül University. Now she is a Lecturer and Coordinator for Program of Accounting and Tax Applications, Ege Vocational School, Ege University, Turkey. Her current research interests include different aspects of accounting and fnance. She has co-authored 10 book chapters, more than 10 papers, and participated in over 20 conferences.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

This page intentionally left blank

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

ESFIRM Editorial Board

Marta Borda Wroclaw University of Economics and Business, Poland Christophe Courbage University of Applied Sciences Western Switzerland (HES-SO), Switzerland Tomasz Dorożyński University of Lodz, Poland Joseph Falzon University of Malta, Malta Simon Grima University of Malta, Malta Pierpaolo Marano Catholic University of the Sacred Heart – Milan, Italy

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Ibish Mazreku University Haxhi Zeka, Kosovo; University of Tetova, North Macedonia; University of Malta, Malta Ercan Özen University of Uşak, Turkey Inna Romānova University of Latvia, Latvia Ramona Rupeika-Apoga University of Latvia, Latvia Darko Tipurić University of Zagreb, Croatia Eleftherios I. Thalassinos University of Piraeus, Greece

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

This page intentionally left blank

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Series Editors’ Introduction

The Emerald book series: Emerald Studies in Finance Insurance and Risk Management Volume 1 includes studies on Uncertainty and Challenges in Contemporary Economic Behaviour contributed mainly by authors invited from participants in the 3rd International Applied Social Science Congress (C-IASOS2019) held in Çeşme, İzmir, Turkey, between April 4–6, 2019, and the University of Malta’s Department of Insurance. A study of the challenges and perspectives of international migration in Europe within the Brexit framework is carried out in Chapter 1. The study shows that during 2000–2019, among EU-10, the main destination country for immigrants and asylum seekers in terms of welfare and living standards is Germany (both for economic and humanitarian migration), along with the United Kingdom (in the case of economic migration). This situation tends to remain unchanged for the following years even in the Brexit context, as refected by the 2020–2025 forecast scenarios. The authors of Chapter 2 evaluated the impact of US economic policy uncertainty on developing countries under different economic cycles: a nonlinear approach. According to the results of the study, the macroeconomic variables in Turkey’s economy signifcantly and strongly respond to the changes in the EPU index during the periods in which global risk perception is low; nonetheless, so-called responses weaken due to be adopted policy of “wait and watch” by investors during the periods in which global risk perception is high. In Chapter 3, the authors aim to fnd out the existence of zombie frms in the Borsa Istanbul manufacturing industry. The authors stated that 62 of 109 frms which traded on Borsa İstanbul Manufacturing Industry between 2008 and 2018 were classifed as zombie frms because they had interest coverage ratios below 1, for three or more consecutive years. The author of Chapter 4 clarifes the forensic accounting theory and explains why and how the choice of methods or techniques is used to detect creative accounting or manipulations in fnancial and non-fnancial reporting, and shows that the outcome of using such methods or techniques depends on the accounting and non-accounting decisions taken into consideration by the forensic accountant or investigator. In Chapter 5, the authors aim to determine the effects of monetary policy practitioners in Turkey, CBT (Central Bank of Turkey), on macroeconomic variables via the interest channel of monetary transmission mechanism. The results of the study manifest that the interest channel directly affects fxed capital investment and real gross domestic product.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

xx    Series Editors’ Introduction The authors of Chapter 6 analyze the theoretical bases and extent of behavioral accounting, which focuses on the human behavior factors being observed while creating or using fnancial reports. In Chapter 7, the author presents theories that can be used in fnancial inclusion research and policy debates. The study shows that fnancial inclusion theories are explanations for observed fnancial inclusion practices. It also shows that the ideas and perspectives on fnancial inclusion can be grouped into theories to facilitate meaningful discussions in the literature. The author of Chapter 8 aims to investigate the concept of shareholder activism by carrying out a literature review. The study shows that shareholder activism has a signifcant infuence on corporate governance policy that a frm adopted in recent years. Shareholder activism increases levels of shareholder engagement in frm decisions and fosters a long-term corporate governance culture. In Chapter 9, the author analyzes the relationship between interest rates, the exchange rate and investor sentiment in Turkey. The results of the study show that foreign exchange and commercial credit interest rate variables are carefully monitored by market players and are effective and infuential in the formation of future expectations. The author of Chapter 10 focused on a discussion of the fact that (a) climate change is emerging as an important issue, increasing uncertainty in the business circle, and (b) fnancial institutions through their inaction seem to be unmoved by climate change risk despite the potential for climate change events to affect the fnancial institutions and the fnancial system. In Chapter 11, the author tests the asymmetric causality between developed and emerging markets. Results of the tests show that there is a weak correlation between developed and emerging markets. Moreover, results show that there is a long-term relationship between the MSCI Emerging market index and the DJIA index. In Chapter 12, the authors measure the fnancial performance of commercial banks listed on Borsa Istanbul by the Multiple Criteria Decision-Making and investigate the relationship between fnancial performance and market return. Findings show that there is no generally signifcant correlation detected between fnancial ratios and market returns. The author of Chapter 13, with the use of 100 quotes, presents some thoughts, in the form of quotes, on what caused the fnancial crisis, why it was severe and what can be done to prevent another crisis in the future. In Chapter 14, the authors seek to explore the grounded realities of live-in care workers in Malta. They conclude with brief policy suggestions to trigger improvements in the wellbeing and dignity of migrant carers. In Chapter 15, the authors delve into the determinants and praxis of the derivative hedging instrument usage (DHIs) of Malta, a small island state. Empirical evidence is also provided in relation to the impact of DHI usage and the adoption of a hedge accounting model in entities’ fnancial statements. Chapter 16 highlights the effect of micro entrepreneur arrangements to city culture in terms of urban aesthetics and street peace in the urbanization process.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Series Editors’ Introduction    xxi

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Finally, in Chapter 17, the authors address the key issues of the fscal policies and tax system in Kosovo, in order to align with the contemporary tax principles and requirements of the European Standards. Results of the study confrm the need for tax reforms in order to have an adequate tax system oriented to indirect tax; changing the structure of tax collection for border tax into domestic tax, simplifcation of the legal procedures, improvement of management and audit systems, reducing the informal grey economy and to have a gradual growth-friendly fscal policy.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

This page intentionally left blank

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Chapter 1

Challenges and Perspectives of International Migration in Europe within the Brexit Framework Graţiela Georgiana Noja, Mirela Cristea, Petru Ştefea and Ciprian Panzaru Abstract Introduction: International migration, a complex, dynamic and multifaceted process, grasps important challenges for the European economies, through its advantages and pitfalls.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Aim: This research is conducted to examine the fundamental credentials of immigration in Europe and its perspectives within the Brexit framework, an ongoing process that induced profound implications. Method: The authors have applied the cluster analysis and structural equations as the main research methods on a balanced panel comprising 10 receiving countries (most targeted by migrants), members of the European Union (EU-10), for the 2000–2019 timespan (2019 being a Brexit milestone year).The authors have separately extrapolated a sample for 2020–2025 that was further used to identify some perspectives after the Brexit timeline in terms of migration determinants and effects on EU-10 host economies. Cluster analysis is based on a key scenario related to wellbeing, living standards (income level) and poverty risk at destination. These credentials are essentialfor the migration decision and important elements of migrant labour market integration strategies with keen economic consequences, further assessed through the structural equation models. Findings: Results show that during 2000–2019, among EU-10, the main destination country for immigrants and asylum seekers in terms of welfare and living standards is Germany (both for economic and humanitarian migration), along with the United Kingdom (in the case of economic migration). This situation tends to remain unchanged for the following years even in the Brexit context, as refected by the 2020–2025 forecast scenarios. Uncertainty and Challenges in Contemporary Economic Behaviour, 1–19 Copyright © 2020 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-80043-095-220201002

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

2   Graţiela Georgiana Noja et al. Immigration effects on labour market outcomes and economic welfare are extremely signifcant, being largely discussed within the chapter. Keywords: Immigration; Brexit; economic welfare; forecasting; clusters; structural equation JEL classifcation: F22; F15; C53

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

1. Introduction International migration is a sheer phenomenon with strapping economic, social and demographic effects on both developed and developing countries, regardless of the geographical area examined. Worldwide ampleness of migration fows makes this issue extremely relevant for rigorous cross-national scientifc research. Even though an overview of the migration phenomenon can serve to grasp a complex image on it at a synchronic level, we can attest that the migration process represents a keen issue that can’t and shouldn’t be studied different from a diachronic perspective. Hence, we entail that despite its current status of novelty, the migration phenomenon has an important historical background. Its origin is evocated even from the ffteenth to seventeenth centuries, marked by important discoveries of new territories that have subsequently represented destinations of signifcant migration fows. The American society itself, well known as the ‘melting pot’, has built its groundings on a major number of migrants that have helped to transform this once new territory into one of the world’s greatest powers. Even though until a certain point the migration process was considered a study issue only from an historical perspective, to the extent that it was in the past and also for what it represents nowadays, we highlight that the international migration involves a greater horizon of economic and social implications that transform this phenomenon into a study object for economics, sociology or law. Several aspects speak to the relevance of the research study and questions dealt with in this chapter. International migration represents a major research theme, frstly because of its magnitude, since over 250 million migrants all over the world and, particularly, in Europe, have confgured this research issue into one that is globally signifcant, as long as ignoring it can cause important economic and demographic unbalances (Thalassinos, Cristea, & Noja, 2019). Second, migration is a major research topic through its two-fold interdependence with the development concept. It relates to the developed countries that attract not only the labour force needed in production, but also the know-how from other countries, the skills/capabilities migration being a long-term loss for the origin country, but a tremendous gain for migrant receiving economies. A further argument in favour of studying this research thematic is the current migration situation in Europe, framed by the challenges faced by the European countries in front of the refugees’ crisis and increased labour migration fows,

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Challenges and Perspectives of International Migration    3 while EU countries are also under the sheer implications of the ongoing Brexit process (Rupeika-Apoga, Romanova, Bule, & Thalassinos, 2019). On that discursive note, the importance of a migration integrative study is even greater since we account for a phenomenon that during the years has gained completely different valences. In other words, migration during the nineteenth and twentieth centuries had different amplitude than today, particularly in terms of intentionality. In that vein, the twenty-frst-century migration is one extremely intense deployed based on full awareness decisions, widely shaped by the information and communication technology. Even more, current international migration is one in which migrants maintain their roots in the migrant sending country. Thus, compared to previous centuries when this phenomenon was targeting new colonial territories, a present migration study is compelled to observe the specifc ways in which the process is initiated, takes amplitude and impacts the origin and destination countries. Compared to previous studies, the general objective of this chapter focusses on analysing both economic (immigrants) and humanitarian (asylum applicants) migration indicators, along with other economic and labour market specifc credentials, based on data for 10 EU migrant receiving countries (EU-10), namely Belgium, Denmark, Finland, France, the United Kingdom (UK), Germany, Austria, Sweden, Italy and Spain. The analysed period is 2000–2019, with a sub-sample extrapolated for 2020–2025. We have applied two econometric procedures, namely: (1) cluster analysis, in order to check for dissimilarities among the EU-10 countries, considering welfare, earnings and poverty risk at destination both for the economic and the humanitarian migration fows; and (2) structural equation modelling (SEM), for overall (direct indirect, total) assessment of economic and humanitarian migration fows in terms of labour market insertion of the foreign population, and economic welfare. The rest of the chapter is structured as follows: after developing a literature review grounded on core defnitions and concepts of international migration, by induction it further identifes the diverse strands of thought that have amended different authors’ views on international migration over the last decade. Then, the data and econometric methods are detailed, completed with own empirical fndings. Final section discusses the fndings along with their policy implications and outlines important concluding remarks.

2. Theoretical Framework and Brief Literature Review 2.1. Defnitions and Concept of International Migration Framing a general setting that encompasses the analysis of both demographic changes and connective socio-economic factors, migration becomes a primordial concept in the research endeavour set to examine and explain the migratory phenomenon in Europe within the Brexit context. Defned in a minimal manner through the geographical movement of persons across a certain border with the purpose to get permanent or temporary residence

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

4   Graţiela Georgiana Noja et al. at destination (Haupt & Kane, 2004), migration has a series of implications that are viewed as determination contexts for the migratory process. Another defnition of migration is given by the International Organization for Migration (IOM, 2004) that views migration as the process of population movement across borders or within a country. The same organisation states that migration is a phenomenon that doesn’t depend on structure, size or causes. According to the United Nations (UN, 2016), international migration represents the cross-border movement of individuals with the aim to establish in a different country than the origin one. Moreover, the UN organisation links migration to the 2030 Agenda for Sustainable Development as it recognises for the frst time the contribution of migration to sustainable economic development (‘11 out of the 17 Sustainable Development Goals contain targets and indicators that are relevant to migration or mobility’). A further defnition describes the migration concept as the population movement concomitant with a change in residence (Petit, 2000). Going deeper into the concept, the same author considers international migration not only through the change of residence in another state, but also through a change in the legal status. Rossi (2008), another author which has extensively studied the migration issue, describes this phenomenon as the movement of persons in different countries, in the case of international migration or as a movement inside the borders of a country, for the general migration concept, while a relatively older perspective of Shaw (1975) attests that migration is the permanent movement of persons on a signifcant distance and lapse of time. Concluding, the international migration concept denotes the movement of individuals across borders from various economic or humanitarian reasons.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

2.2. Theoretical Approaches of International Migration Beyond the numerous defnitions set to explain the migration concept (endeavour that in certain situations can result in epistemic approaches rather than concrete measurements of the migration phenomenon that could allow for tailored policy interventions) migration was further studied by several authors whose fndings enrich the scientifc literature. Valtonen (2008) describes the migration process in a dual perspective. By referring to migration in the same already familiar terms of population movements, Valtonen (2008) sets a clear line between internal and international migration. Nevertheless, a more important statement made by the author entails that the expression of ‘moving persons’ refers not only to individuals that are voluntarily moving inside or outside the origin country, but also to those constrained by different factors to adopt the migration decision. The study performed by Barrell, Gerald, and Riley (2007) focussed on the macroeconomic effects of EU enlargement and, implicitly, of migration. Their fndings reveal that as a result of the 2004 integration wave, most of the Old EU Member States have enforced new restrictions placed on the immigrants coming from the New EU Member States. Unlike the existing migration trend at that time, there were, however, certain countries that opted out of these restrictions

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Challenges and Perspectives of International Migration    5 and implemented fexible immigration policies. These countries were Denmark, Ireland, Sweden and even the UK. The consequences were soon to come. Thus, according to Barrell et al. (2007), if up until the 2004 adhesion year, these four countries registered relatively low migration infows, from the moment they adopted stable and fexible migration policies, these countries became primordial destinations for both economic and humanitarian migrants. On the opposite side, the neighbouring countries of the New EU Member States that until the 2004 adhesion were most targeted by migrants have registered a signifcant decrease in the number of immigrants afterwards. Hooper, Desiderio, and Salant (2017, p. 27) have shown that ‘while integration and employment policies remain a broadly national competence’, the EU represents the support ‘on the frontlines of efforts to improve newcomer labour market integration outcomes’. Current migration fows in Europe are widely shaped by the UK’s decision to leave the European Union and the ongoing Brexit process. Headline estimates grasp a patterned link between international migration and economic growth when relate to the long-term impact of Brexit. Tetlow and Stojanovic (2018) entail that ‘one important way in which Brexit may have an impact on economic growth is by precipitating changes to immigration policy’. Hence, new immigration policies tend to centre on attracting certain types of migrants according to their educational background, while enforcing restrictive regulations for EU nationals. Within the Brexit challenges, Suciu, Cristea, and Noja (2018, pp. 124–125) revealed:

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

a strong positive immigration impact upon host countries’ labour markets refected through important increases in employment rates (both overall and mainly for the foreign population) … and the overall impacts of immigration fows, both labour (economic) and humanitarian, tend to have positive expected effects refected on the host countries’ economies. Referring to the UK, Dennison and Geddes (2018, p. 1137) appreciated that the Brexit referendum ‘seems likely to lead to a decline in “Europeanised” migration policy in the UK’, on the medium term. Also, Portes and Forte (2017, p. S31) empirically attested that the Brexit process will induce a downturn in immigration fows to the UK, and: Brexit-induced reductions in migration are likely to have a signifcant negative impact on UK GDP per capita (and total GDP), with marginal positive impacts on wages in the low-skill service sector. Kierzenkowski, Pain, Rusticelli, and Zwart (2016, p. 26) state that ‘immigrants have contributed on average 0.7 percentage points to GDP per year since 2005’, hence the Brexit will induce unfavourable effects upon GDP. Even more, on the Brexit outcome, the same authors (Kierzenkowski et al., 2016, p. 28) also drawn

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

6   Graţiela Georgiana Noja et al. into attention that ‘lower immigration and weaker FDI could reduce the pool of skills’, with adverse impact on economic welfare. On this line, Booth, Howarth, Persson, Ruparel, and Swidlicki (2015) attested that policies and strategies designed to support the migration phenomenon will signifcantly redound to the increase of the total GDP by 2030. After reviewing the literature, we can attest that there are different strands of thoughts regarding the determinants and economic consequences of international migration in Europe, particularly in the Brexit frame of reference, that need to be strengthen with comprehensive studies to further support accurate policy interventions.

3. Data Analysis and Research Methodology

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

3.1. Data – Selected Indicators According to our general objective, frstly, we have compiled a balanced panel (2000–2017 sample) and considered the total immigration fows and asylum seekers, along with a complex set of indicators that we’ve further used in the empirical analysis. The data were afterwards extrapolated until 2019 to better capture the Brexit impact upon labour mobility within the European Union (since 2019 is the timeline advanced for a possible outcome of this ongoing process). Extrapolation is used even though it may be subject to more uncertainty, yet the data are sampled periodically and therefore the linear extrapolation approximates next data points on short periods of time, by expanding the given data to make predictions about what can happen in the Brexit framework. We rely on short-range forecasting so we’ve separately extrapolated a sample for 2020–2025 (EU-10) that was used in both cluster analysis and SEM to identify some perspectives after the Brexit timeline (2019) in terms of migration determinants and effects on EU-10 host economies. These essentials were further confgured in designing the migrant labour market integration general models. We’ve further applied the standardisation procedure (based on the mean and standard deviation of each time series) on indicators, to cover for data benchmarking between countries (OECD, 2005, 2015). Numerous indicators were used as proxies for the variables of developed models, both for the cluster analysis and in SEM, respectively: (i) International migration indicators: economic migration as infows of foreign population (IMIG); humanitarian migration as infows of asylum applicants (ASYL). (ii) Economic activity and labour market specifc indicators: Gross Domestic Product per capita in Euro (GDP_cap); unemployment rate for the foreignborn population (UR_FBP) (%); annual net earnings of a two-earner married couple with two children (NE) (Euro); educational level refected through the tertiary education (levels 5–8) (EDU_TL); at–risk–of–poverty rate (POV_RK) (%).

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Challenges and Perspectives of International Migration    7 The main databases used for collecting the data were European Commission – Eurostat, OECD – International Migration Database, World Bank – World Development Indicators, United Nations High Commissioner for Refugees (UNHCR). As shown in Fig. 1, immigration infows are particularly keen in Germany and the UK among the EU-10 receiving countries considered in our analysis, as well as in France, Austria and Spain (Fig. 1(a)). Asylum seekers, on the other hand, are mostly found in Italy (through the main Libyan humanitarian migration corridor), but also in Germany, Sweden and Finland (Fig. 1(b)). As regards the labour market insertion of migrants in EU-10 countries, offcial data and own extrapolation for 2019 highlight that Germany and the UK have the lowest unemployment rates of the foreign population, while France, Finland and Spain still struggle with high pressures of increased unemployment among the foreign population residing in these countries (Fig. 1(c)). Fig. 2 entails important differences among the EU-10 receiving economies in terms of economic welfare, earnings levels and poverty rate. Hence, France, Italy and Spain have relatively low annual net earnings of a two-earner married couple with two children (below the 59,000 Euro threshold), while in Germany, Austria and Sweden these earnings are signifcantly higher, even more in Denmark and the UK, where they exceed 100,000 Euro (Fig. 2(b)). GDP per capita is extremely high in Sweden and Austria (Fig. 1(a)), these two countries also having the lowest poverty rates (Fig. 2(c)) and the highest levels of tertiary educational attainment (Fig. 3(b)).

3.2. Research Methods

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Grounded on these data and after reviewing the literature, we’ve applied the cluster analysis and SEM techniques as main research methods pursued in our analysis.

[29337,128516] (128516,185406.5] (185406.5,428864] (428864,1845793]

(a)

[12850,69660] (69660,128022.5] (128022.5,179240] (179240,490520]

(b)

[3.60,11.79] (11.79,13.84] (13.84,15.79] (15.79,20.59]

(c)

Fig. 1.  (a) Flows of Immigrants; (b) Asylum Seekers; and (c) Unemployment Rate of Foreign-Born Population, EU-10, 2019. Source: Own process of Eurostat and OECD data.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

8   Graţiela Georgiana Noja et al.

[8735.14,11048.14] (11048.14,23650] (23650,37500] (37500,49800]

[12.40,14.69] (14.69,17.25] (17.25,21.09] (21.09,30.30]

[45722,58766] (58766,64561] (64561,70506] (70506,113749]

(a)

(b)

(c)

Fig. 2.  (a) GDP per Capita; (b) Annual Net Earnings; and (c) At-Risk-ofPoverty Rate, EU-10, 2019. Source: Own process of Eurostat and OECD data.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

[13.40,40.49] (40.49,42.15] (42.150,45.20] (45.20,57.09]

(a)

[13.20,33.10] (33.10,35.00] (35.00,39.79] (39.79,75.89]

(b)

Fig. 3.  Educational Attainment: (a) Secondary and (b) Tertiary, EU-10, 2019. Source: Own process of Eurostat and OECD data. Cluster analysis is built on the Ward method specifc for hierarchical clustering, being processed through the standardised values of the indicators. The Ward method relies on squared Euclidian distance and a minimum variance criterion by fnding the pairs of clusters and minimising the total within-cluster variance (Cornish, 2007). Cluster sampling is performed separately on both timespans 2000–2019 and 2020–2025 and for both dimensions of the migration process, economic and humanitarian. It is a benefcial method that takes advantages of random and stratifed samples, thus allowing us to sort the countries into groupings based on similarity and hence ‘reveal the outliers and dimensionality of previously unnoticed interesting relationships’ (Ferreira & Hitchcock, 2009, p. 1926).

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Challenges and Perspectives of International Migration    9

IMIG

ε1

ε2

ASYL

UR_FBP

GDP_cap

NE

ε3

POV_RK

ε4

EDU_TL

Fig. 4.  General Confguration of the SEM Model. Source: Authors’ research.

The SEM model captures in a new integrative approach the interlinkages (direct, indirect, bidirectional and total) between the economic and humanitarian immigration, labour market insertion and economic welfare credentials. The general confguration of the SEM model is refected in Eq. (1) and Fig. 4:



b y +  + b y + c x +  + c x = ε 1m mt 11 1t 1n nt 1t  11 2t b y +  + b y + c x +  + c x = ε  21 2t 2 m mt 21 1t 2 n nt 2t (1)  ............  y + c x ++ c x = ε b y +  + b mm mt m1 nt mn nt mt  m1 mt

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

where t is the number of observed time periods, bij represents the yij endogenous variable’s parameters, cij are the xij exogenous variable’s parameters, i=1, …, m, j=1, …, n; ε comprise the error term (residuals) (Cristea & Noja, 2019). The SEM model is processed through the maximum likelihood estimation (MLE) method for both time periods, 2000–2019 and 2020–2025. Based on the literature review and own methodological approach, we settled to evaluate the following hypotheses: H1. There are signifcant dissimilarities among the EU-10 countries, considering economic welfare, earnings and poverty risk at destination as key elements for the economic and humanitarian migration fows. H2. There are overall (direct indirect, total) signifcant implications of economic and humanitarian immigration fows upon the labour market insertion of the foreign-born population and economic welfare.

4. Results and Discussion 4.1. EU-10 Immigration Cluster Analysis Results The research conducted within the chapter focussed, in an initial phase, on EU-10 cluster forming and analysis, based on a key scenario related to wellbeing

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

10   Graţiela Georgiana Noja et al. (welfare), living standards (income level) and poverty risk at destination, these credentials being essential for the migration decision and important elements of migrant integration strategies. In order to set the number of clusters we used the Calinski–Harabasz criterion (cluster stop) and the method of graphical representation through dendrograms. Thus, we have correlated the number of immigrants (IMIG) and asylum applicants (ASYL) residing in each host country with the GDP per capita (GDP_cap), annual net earnings (NE) and at–risk–to poverty rate (POV_RK), and further applied the cluster analysis through the Ward method (Table 1). The variables used for cluster analysis were previously standardised in order to cover for data benchmarking among countries, as described in the previous section. The detailed results of the cluster analysis for both time periods are described in Table 1. The correlation matrixes and dendrograms, resulted from the research (Figs. 5 and 6), allowed for a proper identifcation of three main clusters comprising the EU-10 countries, both in terms of economic and humanitarian migration (Tables 2 and 3). The frst cluster (C1) comprises four receiving countries for both time periods considered, with medium attraction for immigrants and refugees (in absolute terms) and reduced immigration impacts, namely Sweden, Italy, Spain, Austria, along with Belgium, for the humanitarian one (Table 3). However, relative to their population, Sweden and Austria have the largest share of refugees compared to the other analysed countries (The UN Refugee Agency, 2017). The forecasting results associated with the 2020–2025 period refect a much more intense attractiveness for fve of the EU-10 countries considered, especially for the humanitarian migration, where a signifcant increase on the asylum seekers infows can be noticed particularly in cluster 3 comprising Germany, but also in cluster 1 comprising France, Belgium, Finland and Italy. Moreover, the Brexit spillovers mainly refect in a reposition of the UK among the EU countries with low levels of asylum applications for the following years. At the same time, the results obtained after applying the cluster analysis (C3) (Tables 2 and 3) point out that during 2000–2019, among EU-10, the main destination country for immigrants and refugees in terms of welfare, living standards and poverty risk, will be Germany (both for economic and humanitarian migration), along with the UK (in the case of economic migration). This situation tends to remain unchanged for the following years even in the Brexit context, as refected by the 2020–2025 forecast scenarios. Therefore, the results confrm OECD (2015) data revealing that the migration fows have registered a shift in patterns over the years, most emigrants from Central and Eastern European countries (especially from Romania and Poland) selecting Germany as the main destination country, along with the UK, in their search for better living standards, whereas Italy and Spain experience a slight decrease in migrants’ preferences compared to previous years. The results obtained after applying the cluster analysis (C1) (Table 3) revealed that the Nordic countries will be less targeted by the immigrants compared to previous years, while Italy and Spain might regain their importance as migrant favourite destinations in light of the Brexit process, while the UK tends to be largely affected by it only in terms of humanitarian migration.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

N

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

4

Source: Own research.

POV_RK

−0.333

IMIG 4 0.349 GDP_cap 4 −1.106 NE 4 −1.120 POV_RK 4 1.321 Immigrants 2020–2025 IMIG 4 −0.376 GDP_cap 4 −4.921 NE 4 1.974 POV_RK 4 −1.191 Asylum applicants 2000–2019 ASYL 4 −0.232 GDP_cap 4 −0.714 NE 4 −0.697 POV_RK 4 0.452 Asylum applicants 2020–2025 ASYL 4 7.142 GDP_cap 4 −3.872 NE 4 0.716

Mean

Cluster 1 (C1)

Immigrants 2000–2019

Indicator

5 5 5 5 5 5 5

0.653 0.537 0.686 1.024

3.103 2.024 0.691 5

4 4 4 4

0.568 1.345 1.599 0.776

1.806

4 4 4 4

N

0.747 0.248 0.450 0.452

SD

−1.037

2.586 −0.972 3.266

0.780 0.385 0.973 −0.423

−0.119 0.077 0.660 −0.344

−0.603 0.287 0.512 −0.654

Mean

Cluster 2 (C2)

1.222

2.638 2.511 3.239

1.597 1.080 1.071 0.881

0.744 1.245 1.000 2.099

0.434 1.049 0.869 0.540

SD

1

1 1 1

1 1 1 1

2 2 2 2

2 2 2 2

N

−1.153

21.935 0.475 2.109

10.722 0.111 1.288 −0.421

7.124 −0.250 5.385 −0.765

2.024 −0.004 1.236 0.119

Mean

SD

0.248

3.814 0.123 0.279

3.223 0.104 0.239 0.210

2.384 0.821 3.579 0.469

1.404 0.629 1.447 1.002

Cluster 3 (C3)

1.8282

108.9617*** 15.6892*** 7.8283*

160.376*** 38.1751*** 83.6621*** 21.3445***

180.0098*** 116.1076*** 22.98059*** 2.088447

164.3301*** 44.7729*** 75.9233*** 164.1004***

F

Table 1.  Cluster Analysis Results (Economic and Humanitarian Migration within EU-10), 2000–2019 and 2020–2025.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

0.0603

0.7927 0.3550 0.2155

0.6195 0.2793 0.4593 0.1781

0.8633 0.8029 0.4464 0.0683

0.6252 0.3125 0.4353 0.6249

R2

Challenges and Perspectives of International Migration    11

12   Graţiela Georgiana Noja et al. –2

0

2

–2

0

2

–10 –5

10 5

IMIG_st

–10

5

L2squared dissimilarity measure

Dendrogram for IMIG_Welfare cluster analysis

10 10 4

3

7

4

2

2

8

5

5

6

6

5

6

0

5

10

0

5

10

Dendrogram for IMIG_Welfare cluster analysis

1500

5

400

600

0

200 0

L2squared dissimilarity measure

–5

0 POV_r_st

1000

–2

2 0 –2 –4

500

POV_r_st

0

0

5

EARN_st

–5

10

10

0

2

5

GDP_capita_st

–5

EARN_st

2 10 5 0

0

GDP_capita_st

0

–4 –2 0

IMIG_st

0

2 0 –2

0

1

3

4

1

4

2

2

7

8

9 10 5

5

6

6

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Fig. 5.  Correlation Matrixes and Dendrograms Associated with the Cluster Analysis for Labour Migration: 2000–2019 (Left) and 2020–2025 (Right). Source: Own process of panel data. The most preferred destination by immigrants and refugees (C3) are again Germany (both for economic and humanitarian migration) and the UK (only in case of economic migration), along with France and Finland. Thus, we can attest that H1 is fulflled, with the mention of almost similar situation for the 2000–2019 period, and the following years in the Brexit context, the 2020–2025 forecast scenario.

4.2. SEM Models and Results To assess the interlinkages between international migration, labour market insertion and economic welfare in the Brexit setting, at the level of EU-10 receiving countries, we’ve processed the confgured SEM model on two lapse of time, 2000–2019 and 2020–2025. The results are graphically represented in Fig. 7 and synthesised in Table A1. The results obtained show that the labour market insertion of economic migrants (IMIG) is improved for the period 2020–2025, than 2000–2019, through a higher reduction of the unemployment rate of the foreign population (UR_FBP) (a coeffcient of –0.311 compared to –0.208, very statistically signifcant at a threshold of 0.1%), under the unfavourable bidirectional infuence of tertiary education (EDU_TL) (a coeffcient of –4.474 for 2020–2025, compared

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Challenges and Perspectives of International Migration    13 2

–2

0

2

–10

15 10 5 0

ASYL_st

2

2 0 –2 –4

POV_r_st

–2

5

10 15

–5

0

5

9

4

5

6

9

6

3

6

3

8

5

5

5

6

6

0 POV_r_st

0

10 20 30

0

5

10

Dendrogram for ASYL_Welfare cluster analysis

4000 0

L2squared dissimilarity measure

1000 800 600 400 200 0

L2squared dissimilarity measure

Dendrogram for ASYL_Welfare cluster analysis

5

3000

0

0

10 EARN_st

–5

30 20 10 0

GDP_capita_st

0

2

2

–10

5 EARN_st

–4 –2 0

0 –5

–2

0

ASYL_st

GDP_capita_st

0

–5

2000

0

1000

–2

4

3

3

9

9

2

2

5

5

7

8

10

6

6

6

Fig. 6.  Correlation Matrixes and Dendrograms Associated with the Cluster Analysis for Humanitarian Migration: 2000–2019 (Left) and 2020–2025 (Right). Source: Own process.

Table 2.  Labour Migration: Cluster Results (Modelled through Welfare, Earnings and Poverty Risk at Destination).

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Clusters (C) 2000–2019

Immigrants (avg. level)

Immigrants (avg. level)

Clusters (C) 2020–2025

Low to medium

Low

Belgium, Denmark, C1 Finland, France

C1

Sweden, Italy, Spain, Austria

C2

Belgium, Denmark, Finland, France

Low

Low to medium

Austria, Sweden, Italy, Spain

C2

C3

UK, Germany

High

Very high (increased earnings)

UK, Germany

C3

Source: Authors’ research.

to –0.246 for 2000–2019, statistically signifcant at a threshold of 5%) and positive bidirectional impacts of humanitarian migration (ASYL) (a coeffcient of 11.57 compared to 1.443, very statistically signifcant at a threshold of 0.1%).

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

14   Graţiela Georgiana Noja et al. Table 3.  Humanitarian International Migration – Cluster Results (Modelled through Welfare, Earnings and Poverty Risk at Destination). Clusters (C) 2000–2019

Asylum Applicants (avg. level)

Asylum Applicants (avg. level)

Clusters (C) 2020–2025

C1

Belgium, Sweden, Italy, Spain

Low

Medium

Belgium, Finland, France, Italy

C1

C2

Denmark, Finland, France, UK, Austria

Medium

Low

Denmark, UK, Austria, Sweden, Spain

C2

C3

Germany

Very high

Very high

Germany

C3

Source: Authors’ research.

IMIG 1.4

ASYL

-.25 -.12

.11 1.6

-.21

.29

EDU_TL

-.014

UR_FBP

12

ASYL

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

-4.5 -13

-.23

.99

GDP_cap

-.09

.011

NE

ε3

1.2

POV_RK

ε4

.9

NE

ε3

6.7

ε4

2.1

.32

-.38

.13

-.083

1.9

IMIG

ε2

.85

.59

.59 4.6

ε1

(a)

1.2 10

-.31

2.5

EDU_TL

16

ε2

.99

5.9 -.026

6.3 40

ε1

.1

UR_FBP

-.84

GDP_cap

-2.2

-.67

2.1

-.0035

POV_RK

.062

-.77

(b)

Fig. 7.  Results of the SEM Model: (a) 2000–2019 and (b) 2020–2025, EU-10. Source: Authors’ research. On the opposite side, the humanitarian migration (ASYL) will generate a lower labour market insertion of the foreign population for the period 2020–2025, by inducing an upward trend in the unemployment rate of foreign-born population (UR_FBP) (a coeffcient of 0.101, very statistically signifcant at a threshold of 0.1%). We can see that the tertiary education will not encourage the humanitarian migration (especially for the period 2020–2015, with a coeffcient of –12.98, very statistically signifcant at a threshold of 0.1%). Also, tertiary education will

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Challenges and Perspectives of International Migration    15 neither inset a better integration of the foreign-born population on the EU-10 host countries labour markets (positive coeffcients for both considered time lapse periods, more emphasised for 2000–2019), under the adverse effects of economic and humanitarian migration fows. Overall, the economic welfare into the EU-10, measured by the GDP per capita (GDP_cap), will be downsized (more visible for the period 2020–2025, having a coeffcient of –0.842, very statistically signifcant at a threshold of 0.1%, compared to 2000–2019, with a coeffcient of –0.230, statistically signifcant at a threshold of 1%), under the direct impacts of the unemployment rate of foreign born population (UR_FBP), and indirect impact of economic and humanitarian migration, along with tertiary education (migrant selection process at destination according to the educational background). These results complement the work of Booth et al. (2015) and are in line with those obtained by Portes and Forte (2017), which proved that diminished immigration fows will induce a negative impact upon GDP per capita (even upon total GDP). Also, negative implications for the period 2020–2025 will reverberate upon earnings (a coeffcient of –0.0260, although statistically insignifcant). As regards poverty risk, positive effects will be generated under the infuence of all considered variables both for 2000–2019 (a negative coeffcient for –0.381, very statistically signifcant at a threshold of 0.1%) and for 2020–2025 (a coeffcient of –0.00354, although statistically insignifcant). These results are similar to those obtained by Suciu et al. (2018, p. 124), which showed that ‘additional labour immigrants and asylum seekers are drawn by improved living standards (refected by reducing POV_R)’, for the 2000–2019 time span period. Based on SEM results, we can attest that H2 is fulflled, being entailed well-marked infuences particularly for the 2020–2025 period than for the 2000–2019.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

5. Concluding Remarks This study examined the intercausalities between immigration, labour market outcomes and economic development in Europe within the Brexit framework. Particularly, the chapter brought new insights on the challenges and opportunities encountered by 10 receiving countries, members of the European Union, most targeted by migrants. To handle on these research questions, we’ve applied two econometric techniques, namely cluster analysis (confgured through the Ward method) and SEM (processed by the MLE method). The research strengthens the existing literature and sheds new light on the immigration effects upon most targeted receiving countries within the EU, along with the challenges and perspectives faced in the Brexit framework. Both public discourse, and policy makers or international organisations render global the keen need for comprehensive investigations of the migration phenomenon, particularly of its socio-economic implications. Results show that during 2000–2019, among EU-10, the main destination country for immigrants and asylum seekers in terms of welfare and living

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

16   Graţiela Georgiana Noja et al. standards is Germany (both for economic and for humanitarian migration), along with the UK (in the case of economic migration). This situation tends to remain unchanged for the following years even in the Brexit context, as refected by the 2020–2025 forecast scenarios. Based on our main fndings, overall, the economic welfare of EU-10 countries, measured by the GDP per capita, net earnings and poverty risk, will be downsized for both periods, 2000–2019 and 2020–2025, being entailed unfavourable high-marked impacts particularly for the 2020–2025 lapse of time. The research is not without limitations ascribed to the low availability of accurate international migration data that astonishes the amplitude of migration fows in Europe. Future research guidelines are confgured to further assess the specifc ways of immigrants’ integration into host economies with positive spillovers on labour market performance and economic development.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

References Barrell, R., Gerald, J. F., & Riley, R. (2007). EU enlargement and migration: Assessing the macroeconomic impacts. ESRI Working Paper No. 203.The Economic and Social Research Institute (ESRI), Dublin. Booth, S., Howarth, C., Persson, M., Ruparel, R., & Swidlicki, P. (2015), What if? The consequences, challenges and opportunities facing Britain outside the EU. London: Open Europe. Cornish, R. (2007). Statistics: cluster analysis. Mathematics Learning Support Centre. Retrieved from http://www.statstutor.ac.uk/resources/uploaded/clusteranalysis.pdf. Accessed on September 21, 2017. Cristea, M., & Noja, G. G. (2019). European agriculture under the immigration effects: New empirical evidence. Agricultural Economics (Zemědělská Ekonomika), 3(65), 112–122. Dennison, J., & Geddes, A. (2018). Brexit and the perils of ‘Europeanised’ migration. Journal of European Public Policy, 25(8), 1137–1153. European Commission. (2019). Eurostat database. Brussels: European Commission. Ferreira, L., & Hitchcock, D. (2009). A comparison of hierarchical methods for clustering functional data. Communications in Statistics – Simulation and Computation, 38(9), 1925–1949. doi:10.1080/03610910903168603 Haupt, A., & Kane, T. (2004). Population handbook. Washington, DC: Population Reference Bureau. Hooper, K., Desiderio, M. V., & Salant, B. (2017). Improving the labour market integration of migrants and refugees – Empowering cities through better use of EU instruments. Brussels: Migration Policy Institute Europe (MPI). International Organization for Migration (IOM). (2004). Glossary on migration. International migration law (pp. 1–78). Geneva: International Organization for Migration. ISSN 1813-2278. Kierzenkowski, R., Pain, N., Rusticelli, E., & Zwart, S. (2016). The economic consequences of Brexit: A taxing decision. OECD Economic Policy Papers, 1(16), 1–36. OECD. (2005). Handbook on constructing composite indicators: Methodology and userguide. Statistics Working Paper. Organization for Economic Cooperation and Development, Paris.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Challenges and Perspectives of International Migration    17

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

OECD. (2015). Is this humanitarian migration crisis different? Migration policy debates. Paris: Organization for Economic Cooperation and Development. Petit, V. (2000). Les migrations internationales. In Y. Charbit (Ed.), La population des pays en développement (pp. 99–128). Paris: Les études de la Documentation Française. Portes, J., & Forte, G. (2017). The economic impact of Brexit-induced reductions in migration. Oxford Review of Economic Policy, 33(suppl_1), S31–S44. Rossi, A. (2008). The impact of migration on children in developing countries. Unpublished manuscript prepared for the Youth Migration Conference, 24–26 April, Bellagio, Italy. Rupeika-Apoga, R., Romanova, I., Bule, L., & Thalassinos, E. Y. (2019). The impact of population ageing and social stratifcation: The case of Latvia. International Journal of Economics & Business Administration, 7(1), 49–63. Shaw, R. P. (1975). Migration theory and fact. Philadelphia, PA: Regional Science Research Institute, University of Pennsylvania. Suciu, M. C., Cristea, M., & Noja, G. G. (2018). Immigration effects within the EU–Brexit framework: An empirical analysis. Economic Computation & Economic Cybernetics Studies & Research, 52(4), 113–130. doi:10.24818/18423264/52.4.18.08 Tetlow, G., & Stojanovic, A. (2018). Understanding the economic impact of Brexit. London: Institute for Government. Thalassinos, I. E., Cristea, M., & Noja, G. G. (2019). Measuring active ageing within the European Union: Implications on economic development. Equilibrium. Quarterly Journal of Economics and Economic Policy, 14(4), 591–609. doi:10.24136/eq.2019.028 The UN Refugee Agency (2017). Figures at a glance. Geneva: United Nations High Commissioner for Refugees. UN (2016). International migration report 2015: Highlights (ST/ESA/SER.A/375). New York, NY: United Nations, Department of Economic and Social Affairs, Population Division. Valtonen, K. (2008). Social work and migration: Immigrant and refugee settlement. Aldershot: Ashgate. World Bank. (2018). World development indicators. Washington, DC: World Bank.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

18   Graţiela Georgiana Noja et al.

APPENDIX Table A1.  SEM Results for EU-10 Countries, 2000–2019 and 2020–2025 Time Periods. Variables

(1)

(2)

2000–2019

2020–2025

IMIG

−0.208*** (0.0618)

−0.311*** (0.0496)

ASYL

−0.0141 (0.0357)

0.101*** (0.0272)

EDU_T

0.133** (0.0482)

0.0625 (0.0375)

_cons

0.0106 (0.0691)

−0.674** (0.251)

UR_F

−0.230** (0.0716)

−0.842*** (0.241)

_cons

−0.0900 (0.0705)

−2.202*** (0.321)

GDP_capita

0.592*** (0.0742)

−0.0260 (0.125)

_cons

0.319*** (0.0763)

2.079*** (0.416)

−0.381*** (0.0657)

−0.00354 (0.0700)

−0.0830 (0.0675)

−0.774*** (0.233)

0.107 (0.0889)

1.227** (0.410)

0.593*** (0.152)

6.344*** (0.815)

0.292** (0.0965)

2.531*** (0.516)

UR_F

GDP_capita

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

EARN

POV_R GDP_capita _cons mean(IMIG) _cons mean(ASYL) _cons mean(EDU_T) _cons

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Challenges and Perspectives of International Migration    19 Table A1.  (Continued ) Variables

(1)

(2)

2000–2019

2020–2025

0.847*** (0.0847)

0.988*** (0.180)

0.995*** (0.0995)

5.950*** (1.086)

1.153*** (0.115)

6.701*** (1.223)

0.903*** (0.0903)

2.105*** (0.384)

1.579*** (0.158)

10.09*** (1.842)

4.644*** (0.464)

39.89*** (7.284)

1.861*** (0.186)

15.99*** (2.919)

1.443*** (0.217)

11.57*** (2.990)

−0.246* (0.122)

−4.474* (1.738)

−0.121 (0.208)

−12.98*** (3.666)

200

60

var(e.UR_F) _cons var(e.GDP_capita) _cons var(e.EARN) _cons var(e.POV_R) _cons var(IMIG) _cons var(ASYL) _cons var(EDU_T) _cons

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

cov(IMIG,ASYL) _cons cov(IMIG,EDU_T) _cons cov(ASYL,EDU_T) _cons N

Source: Authors’ research. Note: Standard errors in parentheses, * p < 0.05, ** p < 0.01, *** p < 0.001.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

This page intentionally left blank

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Chapter 2

The Impact of US Economic Policy Uncertainty on Developing Countries Under Different Economic Cycles: A Nonlinear Approach Deniz Erer and Elif Erer Abstract Introduction: Uncertainty plays an important role on economic stability and macroeconomic variables. Economic agents postpone decisions about investment and consumption in periods in which uncertainty is high. This situation affects economic growth negatively. Recently, uncertainty has focused on policy uncertainty. At this point, economic policy uncertainty (EPU) comes to the forefront. EPU is defned as conception that economic agents do not forecast consequences of economic policies adopted by policy makers and future economic policies. In terms of developing countries, statements presented by policy makers in the United States especially may appear as a source of uncertainty in developing economies.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Aim: Therefore, the aim of this study is to analyze the effects of US EPU on macroeconomic variables for Turkey and Brazil, Russia, India, China and South Africa for periods in which global risk perception is low and high. Method: The authors used monthly data from January 1998 to December 2018. For this purpose, the authors used Threshold VAR. VIX index takes in consideration as global risk perception. The authors used US EPU index proposed by Baker’s vd. (2016) in order to measure EPU in the United States. Besides, the authors used macroeconomic variables such as industrial production index, infation and exchange rate. Findings: As is seen from the results of the analysis, for Turkey’s economy the macroeconomic variables signifcantly and strongly respond to the changes in the EPU index during the periods in which global risk perception

Uncertainty and Challenges in Contemporary Economic Behaviour, 21–35 Copyright © 2020 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-80043-095-220201003

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

22    Deniz Erer and Elif Erer is low; nonetheless, the so-called responses weaken due to the adopted policy of “wait and watch” by investors during the periods in which global risk perception is high. Keywords: Economic policy uncertainty; real effective exchange rate; infation; global risk perception; threshold VAR; developing countries; nonlinear modelling

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

1. Introduction Uncertainty and specially policy uncertainty have increased particularly since 2008 Global Financial Crisis (Baker, Bloom, & Davis, 2016; Bloom, 2014). Uncertainty, which was defned by Knight (1921) as that economic agents cannot forecast probability of events, has important effects on macroeconomic variables and economic stability. Bernanke (1983), Bloom (2009) and Aastveit et al. (2013) expressed that monetary policy is less effective in the periods of high uncertainty. Rising uncertainty causes that economic agents will have postponed investment decisions by the time having more information (Hanias, Curtis, & Thalassinos, 2007). Although the effects of uncertainty on investments and economic growth have been analyzed since 1980 (Bernanke, 1983; Guiso & Parigi, 1999; Rodrick, 1991), researches relating to the effects of monetary policy and policy uncertainty on macroeconomic activity have again become important in recent years (Baker et al., 2016; Bloom, Bond, & van Reenen, 2007; Jens 2017). According to Bloom (2014), the reason is that uncertainty has risen after 2008 Global Financial Crisis. Also, it was expressed in IMF Report (2018) that global economic slowdown stems from policy uncertainty in largest economies, policy instability and slowdown of investment growth in developing economies. It is diffcult to measure uncertainty. However, various variables representing this concept were offered in the literature. These variables are stock price volatility (Bloom, 2009), return volatility of frms (Gilchrist, Sim, & Zakrajsek, 2014), macroeconomic policy uncertainty (EPU) (Jurado, Ludvigson, & Ng, 2015) and EPU index (Baker et al., 2016). In this study, we used EPU. The so-called term is defned that economic agents cannot forecast results of current and future economic policy adopted by policy makers (Sahinöz & Cosar, 2018). Policy uncertainty overwhelmingly results from US economy (IMF, 2013). Baker et al. (2016) developed an index to measure EPU for the United States since 1985. This index states: the frequency of articles in 10 leading US newspapers that contain the following triple; “economic” or “economy”; “uncertain” or “uncertainity”; and one or more of “congress”, “defcit”, “Federal Reserve”; “legislation”, “regulation” or “White House.” (Baker et al., 2016) Policy uncertainty in leading economies in the world such as US economy can spread rest of the world, which can affect negatively the so-called economies.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Impact of US EPU on Developing Countries    23 Thus, literature relating to international spread of uncertainty has increased in recent years (Berger, Grabert, & Kempa, 2017; Carriere-Swallow & Cespedes, 2013; Gabauer & Gupta, 2018; Kamber, Karagedikli, Ryan, & Vehbi, 2016; Trung, 2018). In this study, we investigated the effects of US EPU on Turkey and Brazil, Russia, India, China and South Africa (BRICS) economies for the periods of high and low global risk perception during January 1998–May 2018. For this purpose, we used Threshold VAR (TVAR) method and considered VIX index representing global risk perception as threshold variable.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

2. Literature The effects of uncertainty on investment and economic growth were started to analyze especially after 1980 (Bernanke, 1983; Rodrick, 1991). Earlier studies relating to the effects of uncertainty focused on “wait and see” approach. According to this approach, frms adopt “wait and see” approach rather than bear the cost of uncertainty on output when uncertainty increases, which leads to decrease in investment and output (Bernanke, 1983; Bloom, 2009). Households reduce their consumption and rise their precautionary saving (Kimball, 1990; Leduc & Li, 2016). This effect has contractionary pressure on output in short run while the so-called effect is uncertain in long run. The reason is that higher level of savings encourages investment and increases long-run economic growth (Bloom, 2014). However, Fernandez-Villaverde et al. (2011) stated that uncertainty can negatively affect economic growth due to outfow of a part of savings in a small open economy. Some studies in literature explained the effect of uncertainty on real economic activity through risk Premium channel (Arellano, Bai, & Kehoe, 2012; Christiano et al., 2014; Gete & Melkadze, 2018). These studies expressed that high level of uncertainty increases risk premium and cost of external fnancing and therefore it negatively affects economic growth. Although the studies relating to the effect of uncertainty on economic activity started to analyze after 1980, these studies again came into prominence because uncertainty started to rise again after 2008 Global Financial Crisis. Therefore, the so-called studies focused on the effects of EPU (Bachmann, Elstner, & Sims, 2013; Baker et al., 2016; Bloom et al., 2007; Born & Pfeifer, 2014; Caggiano, Castelnuovo, & Figueres, 2017; Jens, 2017; Kido, 2018; Thalassinos & Thalassinos, 2018). They showed EPU has negative effect on economic activity. Bloom et al. (2007) expressed that high level of uncertainty reduces the effect of demand shock on investment Born and Pfeifer (2014) examined the effects of policy uncertainty on monetary and fscal policies within the frame of New Keynesian in business cycle. From the results of the study, they concluded that policy uncertainty plays an important role on business cycle. Caggiano et al. (2017) analyzed the effect of US EPU on unemployment for periods of expansion and contraction in the economic activities using Smooth Transition VAR model. They showed that response of unemployment to EPU is statistically and economically higher in contraction period. Kido (2018) investigated the effects of US EPU on

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

24    Deniz Erer and Elif Erer global fnancial markets using stock market price, exchange rates and commodity price by FAVAR model. From the results, he concluded that an increase of US EPU affects stock markets negatively; however, its effect on China stock market is lower. Also he showed that USD and Japan yen appreciate against shock of US EPU although other currencies depreciate. According to him, this situation states that USD and Japan yen are safe havens. As examined the studies relating to Turkey, Akkus (2017) analyzed the effects of political instability in developing countries and US EPU on 33 developing economies for the period of 1994–2013. He stated that political instability in developing countries and US EPU affect negatively the so-called economies. Sahinöz and Cosar (2018) calculated EPU for Turkey since 1998 and examined the effects of the so-called uncertainty on real exchange rate, real interest rate, real GDP, real investment and real private consumption using VAR for the period of 2006Q4–2016Q4. From the results of this study, they concluded that policy uncertainty has negative effects on economic growth, consumption and investment, and negative effect of high level of uncertainty on investment is higher than economic growth and consumption. Korkmaz and Güngör (2018) analyzed the effect of global EPU on returns of frms traded in Istanbul Stock Exchange using GARCH models for the period of 1997:1–2018:4. They found that the so-called uncertainty has positive infuence on the volatility of stock prices.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

3. Data and Methodology In this study, we analyzed the effects of US EPU proposed by Baker et al. (2016) on Turkey and BRICS economies for high and low periods of global risk perception. For this purpose, we utilized TVAR model in which parameters vary across regimes. TVAR model was proposed by Tsay (1998). This model is nonlinear estimation method and is used to acquire regime switching, asymmetry and multiple equilibrium. TVAR model distinguishes economy to different regime based on value of threshold variable. Therefore, we can see how economy changes across regimes. Linear VAR model can been shown as follows: p

q

i =1

i =1

Yt = ∑ AY i t−i +∑ Bi X t−i +εt (1)

t

where Y is a vector of endogenous variables in period t; Xt is a vector of exogenous variables in period t and εt a vector of structural shocks. A and B are coeffcient matrix. −1



Yt = [ EPU,IP,CPI,I,REER, VIX ] (2)



X t = [ DUMMY ] (3)

EPU refects US EPU index. This index is proposed by Baker et al. (2016) and states that the frequency of articles in 10 leading US newspapers that contain the

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Impact of US EPU on Developing Countries    25 following triple; “economic” or “economy”; “uncertain” or “uncertainty”; and one or more of “congress”, “defcit”, “Federal Reserve”; “legislation”, “regulation” or “White House”. We obtained the so-called data set from www.policyuncertainty. com web site. Macroeconomic variables in this study are industrial production index (IP), consumption price index (CPI), interbank rate (I) and real effective exchange rate index (REER). These variables were obtained from Federal Reserve Economic Data St. Louis Fed database. VIX index representing global risk perception was considered as threshold variable in TVAR model. This variable is acquired from Yahoo Finance web site. All variables are in logarithmic form. Besides, we removed seasonality using Tramo–Seats method. TVAR model with two regimes can be written as follows (Atasanova, 2003; Balke, 2000): q q  p 1   p  1 + I [ ct−d < γ ]∑ Ai2Yt−i + ∑ Bi2 X t−i  + εt (4) Yt = I [ ct−d ≥ γ ]∑ AY i t−i + ∑ Bi X t−i    i=1  i=1   i =1 i =1 where ct-d is threshold variable, I [ ct−d ≥ γ ] is indicator function which takes value 1 if ct−d ≥ γ , otherwise 0. Ai1, Ai2 , Bi1 and Bi2 are regime parameters. γ is threshold value. Therefore, the economy is in Regime 1, in which global risk perception is high, if threshold variable (VIX index) is below the threshold value (γ), otherwise in Regime 2, in which global risk perception is low.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

4. Empirical Results We examined stability of the variables using Lee–Strazicich unit root test with two structural breaks. This test was developed by Lee and Strazicich (2003). Null hypothesis states that series have unit root although alternative hypothesis expresses that series is trend stability in test. According to Perron (1989), Model A presents crash model allowing shifts in the intercept and Model C presents trend shift model considering both breaks in intercept and trend. Results from two structural break LM unit root test for level and frst difference of the variables are shown in Table 1. As is seen, breaking dates from the results of this test indicated that economic and fnancial crises occurred in considering period lead to structural breaks on macroeconomic variables. EPU, ITURKEY, IBRAZIL and IRUSSIA are trend stability in level while other variables are trend stability in frst difference at the 5% level of signifcance. After unit root analysis, it is necessary to test threshold effect in multivariate framework. For this purpose, we applied C(d) nonlinearity test proposed by Tsay (1998). The results based on recursive estimation of arranged regression using alternative starting points of m0 = 25 ve m0 = 50 and delay parameters d are seen in Table 2. According to the results from the so-called test, delay parameter d is chosen as 1 corresponding maximum value of Chi-Square test statistics. ChiSquare test statistics is 166.13 and corresponding probability value is 0.000 for d = 1. TVAR model is valid. According to minimum AIC, threshold value is equal to 2.80408. Therefore, Regime 1 refects periods in which global risk perception

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

2007:3

2003:8

2002:5

2009:11

−20.4361 0

−2.3826

−3.9829

−2.1355

−20.3004 0

−1.6795

−16.2659 1

−1.3705

−5.0872

−2.3174

−6.8402

−1.7917

−5.0723

−1.0181

IPBRAZIL

∆IPBRAZIL

IPRUSSIA

∆IPRUSSIA

IPINDIA

∆IPINDIA

CIPTURKEY

∆CIPTURKEY

CIPBRAZIL

∆CIPBRAZIL

CIPRUSSIA

∆CIPRUSSIA

CIPINDIA

1

5

6

0

1

4

5

2

1

2

3

2008:3

2001:6

2009:9

2000:5

2000:7

2002:1

2003:10

2009:1

2000:3

2005:5

2001:2

∆IPTURKEY

1

−2.2967

2004:11

IPTURKEY

8

−3.5763

D1t

%5

Critical Values

LM Statistics

−3.5649 −9.9065

−3.5649 −5.1481

−3.5644 −21.2164

−3.5644 −4.7707

−3.5644 −21.5693 −3.5647 −16.5621

−3.5655 −8.4230**

−3.5644 −7.9262

−3.5644 −3.8072

−3.5653 −6.8414

2014:1

−3.5644 −3.7554

2014:11 −3.5655 −6.2516

2015:3

2003:2

2003:7

2004:3

2006:10 −3.5653 −5.3551

2011:4

2011:11 −3.5647 −3.9417

2009:5

2003:11 −3.5644 −4.0729

2010:2

2014:2

2009:3

2005:1

2007:12 −3.5659 −6.1797

D2t

Breaking Time

EPU

LM Lag Statistics

Model A (Crash Model)

Table 1.  Lee–Strazicich Unit Root Test With Two Structural Breaks.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

1

5

6

0

1

4

5

1

2

0

1

2

3

0

1

8

Lag

2006:8

2000:12

2000:7

2003:3

2003:7

2000:12

2001:4

2007:12

2000:10

2003:1

2003:7

2005:6

2007:8

2005:12

2002:1

2001:8

D1t

2006:8

2000:12

2000:7

2003:3

2003:7

2000:12

2001:4

2007:12

2000:10

2003:1

2003:7

2005:6

2007:8

2005:12

2002:1

2001:8

DT1t

2013:6

2014:2

2007:8

2015:1

2014:11

2003:4

2003:5

2010:1

2009:12

2009:1

2008:9

2008:11

2014:5

2008:12

2008:9

2007:12

D2t

Breaking Time DT2t

2013:6

2014:2

2007:8

2015:1

2014:11

2003:4

2003:5

2010:1

2009:12

2009:1

2008:9

2008:11

2014:5

2008:12

2008:9

2007:12

Model C (Trend Shift Model)

−6.1513

−5.8654

−6.1128

−5.8869

−6.1157

−6.0397

−6.0397

−6.2145

−6.1464

−6.1387

−6.1282

−6.2078

−6.1284

−6.2211

−6.1387

−6.1025

%5

Critical Values

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

2002:10

2002:1

2001:3

2002:11

2002.12

2004:12

2

−3.2576

−1.2964

−1.1737

−15.2333 0

−0.6982

−8.7248

−3.0865

−11.3875 1

−1.7717

−10.8238 1

5

−2.4571

−13.1322 1

6

−6.4694

−7.0755** 6

2

−5.5678

−1.7201

−6.3396

−4.7967

−3.6853

∆CIPSOUTH AFRICA

ITURKEY

I BRAZIL

I RUSSIA

I INDIA

∆I INDIA

ICHINA

∆ICHINA

I SOUTH AFRICA

∆I SOUTH AFRICA

REERTURKEY

∆REERTURKEY

REERBRAZIL

∆REERBRAZIL

REERRUSSIA

∆REERRUSSIA

REERINDIA

∆REERINDIA

7

8

2

2

4

5

0

8

2

4

2006:3

2008:7

2000:6

2001:5

2009:2

2007:12

2012:2

2000:3

2000:6

2001:6

2001:5

2002:10

2003:10

2008:3

∆CIPCHINA

5

−1.1643

CIPCHINA

2008:2

−12.1666 0

∆CIPINDIA

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

−3.5655 −6.4213**

−3.5647 −12.2778

−3.5647 −3.6662

−3.5647 −5.7374

−3.5653 −10.9201

−3.5653 −3.5755

−3.5644 −16.9564

−3.5659 −5.7833

−3.5655 −8.0214

−3.5655 −4.9840

−3.5647 −11.3551

−3.5647 −4.3226

2010:10 −3.5659 −6.9995

2010:6

2011:7

2015:7

2013:6

2005:4

2011:10 −3.5647 −11.1917

2004:2

2016:3

2009:2

2015:6

2010:11 −3.5644 −5.3026

2014:2

2012:4

2

4

5

0

7

8

5

6

1

2

1

2

4

5

0

0

1

2

2

8

−3.5655 −12.0499** 6

−3.5649 −6.7392

−3.5653 −7.5536

−3.5653 −3.3875

−3.5644 −12.4169

2014:11 −3.5647 −8.1996

2008:6

2003:7

2006:7

2007:6

2014:1

2013:8

2010:11

2010:10

2000:8

2000:6

2002:12

2003:7

2001:7

2001:2

2002:9

2000:5

2000:2

2000:1

2011:12

2000:12

2000:3

2001:11

2001:6

2003:3

2007:3

2006:8

2000:8

2010:11

2010:10

2000:8

2000:6

2002:12

2003:7

2001:7

2001:2

2002:9

2000:5

2000:2

2000:1

2011:12

2000:12

2000:3

2001:11

2001:6

2003:3

2007:3

2006:8

2000:8

2013:10

2013:9

2016:4

2010:12

2016:2

2011:8

2003:10

2008:3

2007:12

2004:9

2008:2

2004:9

2015:10

2012:2

2005:11

2013:3

2004:11

2008:1

2009:2

2013:2

2009:5

2013:10

2013:9

2016:4

2010:12

2016:2

2011:8

2003.10

2008:3

2007:12

2004:9

2008:2

2004:9

2015:10

2012:2

2005:11

2013:3

2004:11

2008:1

2009:2

2013:2

2009:5

(Continued)

−6.2215

−6.2215

−5.8654

−6.1272

−5.8826

−6.3203

−6.0583

−6.1336

−6.1180

−6.0397

−6.1387

−6.0708

−6.0583

−6.1001

−6.2603

−6.0794

−6.03335

−6.1284

−6.1944

−6.2244

−6.1513

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

2

%5

LM Statistics

2016:3

2009:3

2013:7 −3.5649 −8.7460

−3.5649 −4.8571

−3.5644 −12.1949

2010:11 −3.5644 −5.5427

D2t

Critical Values

2

3

0

1

Lag

Note: Critical values are obtained from Lee and Strazicich (2003) ** 5% level of signifcance. Lag length is determined according to Akaike information criteria (AIC)

2002:11

2003:9

3

−3.3895

REERSOUTH AFRICA

∆REERSOUTH AFRICA −8.4216

2007:12

2008:10

−9.1325

∆REERCHINA

1

0

−1.2261

REERCHINA

D1t

Breaking Time

Model A (Crash Model)

LM Lag Statistics

Table 1.  (Continued)

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

2002:3

2006:4

2002:3

2000:2

D1t

2002:3

2006:4

2002:3

2000:2

DT1t

2008:11

2009:10

2009:2

2004:9

D2t

Breaking Time DT2t

2008:11

2009:10

2009:2

2004:9

Model C (Trend Shift Model)

−6.1284

−6.3883

−6.1387

−6.0646

%5

Critical Values

Impact of US EPU on Developing Countries    29 Table 2.  C(d) Nonlinearity Test.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Threshold Variable: Global Risk Perception (VIX Index) d

m0

C(d)

Prob.

1

25

166.13

0.00000

1

50

146.87

0.00001

2

25

66.33

0.86333

2

50

59.33

0.95964

3

25

75.04

0.88442

3

50

67.28

0.63592

4

25

80.07

0.47688

4

50

72.52

0.71156

5

25

142.39

0.00002

5

50

125.42

0.00089

6

25

124.60

0.00105

6

50

110.57

0.01339

7

25

43.55

0.99971

7

50

38.57

0.99998

γ

2.80408

AIC

−1,477.06556

is high (logarithmic value of VIX index is above threshold value 2.80408, LVIX ≥ 2.80408) while Regime 2 represents periods in which global risk perception is low (LVIX < 2.80408). Fig. 1 shows regime classifcation of VIX index. The impulse–response functions for Turkey and BRICS, which represent the responses of interest rate, infation rate, exchange rates and industrial production index in the so-called countries to the one-deviation shocks in US EPU index for both the periods in which global risk perception is low and the periods in which global risk perception is high, are shown in Fig. 2, Fig. 3, Fig. 4, Fig. 5, Fig. 6 and Fig. 7, respectively. As is seen from Fig. 2, short-term interest rate in Turkey decreases across the shocks in US EPU and the effects of the so-called shocks continue for a long time during the periods in which global risk perception is low. However; the response of infation rate to the shocks in EPU is positive for only one period and the effects in question immediately disappear. The shocks in EPU cause TL/USD exchange rate to decrease and Turkish liras to appreciate approximately for three months. The shocks in question have negative effect on industrial production index and the effects of the shocks last for two months. The effects of the shocks in EPU index for the periods in which global risk perception is high are different from the periods in which global risk perception is low. The so-called shock has an increasing effect on interest rate and infation in the frst month

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

30    Deniz Erer and Elif Erer 4.25 4.00 3.75 3.50 3.25

Global risk perception is high

Global risk Global risk perception is low perception is high

Global risk perception is low

3.00 2.75 2.50 2.25

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Fig. 1.  Regime Classifcation. Note: the fgure displays regime classifcation of vix index. Shading areas in the fgure represents periods in which global risk perception is low. As is seen, global risk perception increased before 2004 and in 2008 Global Financial Crisis.

Fig. 2.  Dynamic Effects of US EPU Index on Macroeconomic Variables in Turkey.

only and on real effective exchange rate in the frst two months only. We cannot fnd a statistically signifcant effect of industrial production index. Consequently, macroeconomic variables in Turkey more signifcantly respond to changes in US EPU index during the period in which global risk perception is low. Economic agents adopt “wait and see” approach and more weakly respond to the so-called variables during the periods in which global risk perception is high. Responses of macroeconomic variables in Brazil to a shock in US EPU are represented in Fig. 3. As is seen, interbank rate negatively responds to one standard deviation shock in the so-called uncertainty. Besides, Brazil real against USD tends to depreciate in frst two months during the period in which global risk perception is low. However, infation and industrial production index don’t signifcantly respond to the so-called shock. During the period in which global risk

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Impact of US EPU on Developing Countries    31

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Fig. 3.  Dynamic Effects of US EPU Index on Macroeconomic Variables in Brazil.

Fig. 4.  Dynamic Effects of US EPU Index on Macroeconomic Variables in Russia.

perception is high, interbank rate positively responds to a shock in EPU index. Infation rate increases from the second month to the fourth month, Brazil real against USD tends to appreciate during the frst two months and industrial production index decreasingly response only in the second month. Therefore, the effects of EPU index on macroeconomic variables are higher in periods in which global risk perception is high. Fig. 4 displays dynamic effects of changes in EPU index on macroeconomic variables in Russia. A shock in EPU index has decreasing effects on interbank rate and infation during the period in which global risk perception is low. Besides,

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

32    Deniz Erer and Elif Erer

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Fig. 5.  Dynamic Effects of US EPU Index on Macroeconomic Variables in India.

Fig. 6.  Dynamic Effects of US EPU Index on Macroeconomic Variables in China. Russian ruble against USD tends to depreciate from the second month to the fourth month; industrial production index decreases only in the third month. Interbank rate decreasingly responds to the so-called shock during the period in which global risk perception is high. However, infation rate rises during the frst two months contrary to period in which global risk perception is low. Russian ruble doesn’t respond signifcantly. Industrial production index decreases only in the second month. Fig. 5 shows impulse–response functions to EPU shock for macroeconomic variables in India. Interbank rate, infation and industrial production index respond decreasingly to one standard deviation shock of EPU index during the period in which global risk perception is low. The so-called shock doesn’t have a signifcant effect on Indian rupee. A surprising change in EPU index increases

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Impact of US EPU on Developing Countries    33

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Fig. 7.  Dynamic Effects of US EPU Index on Macroeconomic Variables in South Africa. interbank rate and infation approximately two months while it decreases industrial production index during three months and real effective exchange rate during four months. Impulse–response functions displaying dynamic effects of a surprising change in EPU index on China economy is shown in Fig. 6. Interbank rate responds positively in the frst month, but negatively from the second month to the fourth month to one standard deviation shock in EPU index during the period in which global risk perception is low. Besides this shock decreases infation and real effective exchange rate during two months. However, we obtained more different results for periods in which global risk perception is high. A surprising change in EPU index has an increasing effect on interbank rate and the so-called effect takes a longer time than periods in which global risk perception is low. Infation and real effective exchange rate statistically and decreasingly respond to the so-called shock only in the frst month. Fig. 7 displays impulse–response functions refecting dynamic effects of EPU index on macroeconomic variables in South Africa. As is seen, interbank rate responds statistically signifcant and decreasingly to one standard deviation shock in EPU only in the frst month during the periods in which global risk perception is low. Infation increases during four months. Real effective exchange rate responds statistically signifcant and positively to the so-called shock in the frst two months. However, interbank rate responds decreasingly to a surprising change in EPU approximately six months during the periods in which global risk perception is high. Infation decreases approximately seven months in contrast to periods in which global risk perception is low. Real effective exchange rate decreases simultaneously against the so-called shock.

5. Conclusion In this study, we investigated the impacts of US EPU index on the Turkey’s economy and BRICS economies in both low and high global risk perception regimes

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

34    Deniz Erer and Elif Erer using TVAR models over the period of January 1998–May 2018. As is seen from the results of the analysis, for Turkey’s economy the macroeconomic variables signifcantly and strongly respond to the changes in EPU index during the periods in which global risk perception is low; nonetheless, the so-called responses weaken due to be adopted policy of “wait and watch” by investors during the periods in which global risk perception is high. Interbank rate, infation and real effective exchange rate in countries Turkey, Russia and China responded more signifcantly to a surprising change in US EPU during the periods in which global risk perception is low while the so-called variables for countries Brazil, India and South Africa responded more signifcantly during the periods in which global risk perception is high. Consequently, changes in US EPU arise as a source of uncertainty in Turkey and BRICS economies. Investors in Turkey, Russia and China adopted wait and see approach during the period in which global risk perception is high, which caused to respond slowly to infation. The fndings from this study are consistent with of the previous fndings of the studies by Akkus (2017), Sahinöz and Cosar (2018), Bloom et al. (2007) and Caggiano et al. (2017).

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

References Aastveit, K. A., Natvik, G. J. Sola, S. (2013). Economic Uncertainty and the Effectiveness of Monetary Poliicy. Workinh paper 2013/17, Norges Bank. Akkus, Ö. (2017). Ekonomik Politika Belirsizliği ve Politik İstikrarsızlığın Büyüme Üzerindeki Etkisi. Anadolu Üniversitesi Sosyal Bilimler Dergisi, 17(3), 27–42. Arellano, C., Bai, Y., & Kehoe, P. J. (2012). Financial frictions and fuctuation volatility. Minneapolis, MN: Federal Reserve Bank of Minneapolis. Atasanova, C. (2003). Credit market ımperfections and business cycle dynamics: A nonlinear approach. Studies in Nonlinear Dynamics and Econometrics, 7(4), 1–22. Bachmann, R., Elstner, S., & Sims, E. R. (2013). Uncertainity and economic activity: Evidence from business survey data. American Economic Journal Macroeconomics, 5(2), 217–249. Baker, S. R., Bloom, N., & Davis, S. J. (2016). Measuring economic policy uncertainity. Quarterly Journal of Economics, 131(4), 1593–1636. Balke, N. (2000). Credit and economic activity: Credit regimes and nonlinear propagation of shocks. Reviews of Economics and Statistics, 82, 344–349. Berger, T., Grabert, S., & Kempa, B. (2017). Global macroeconomic uncertainity. Journal of Macroeconomics, 53, 42–56. Bernanke, B. S. (1983). Irreversibility, uncertainity and cyclical ınvestment. Quarterly Journal of Economics, 97, 85–106. Bloom, N. (2009). The ımpact of uncertainity shocks. Econometrica, 77(3), 623–668. Bloom, N. (2014). Fluctuations in uncertainity. Journal of Economic Perspectives, 28(2), 153–176. Bloom, N., Bond, S., & van Reenen, J. (2007). Uncertainity and ınvestment dynamics. The Review of Economic Studies, 74, 391–415. Born, B., & Pfeifer, O. (2014). Policy risk and business cycle. Journal of Monetary Economics, 68, 68–85. Caggiano, G., Castelnuovo, E., & Figueres, J. (2017). Economic policy uncertainity and unemployment in the United States: A nonlinear approach. Economics Letters, 151(C), 31–34.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Impact of US EPU on Developing Countries    35 Carriere-Swallow, Y., & Cespedes, L. F. (2013). The ımpact of uncertainity shock in emerging economies. Journal of International Economics, 90(2), 316–325. Christiano, L. J., Motto, R., & Rostagno, M. (2014). Risk Shocks. American Economic Review, 104, 27–65. Fernandez-Villaverde, J., Guerron-Quintana, P., Kuester, K., & Rubio-Ramirez, J. (2013). Fiscal Volatility Shocks and Economic Activity. American Economic Review, 105(11), 3352–3384. Gabauer, D., & Gupta, R. (2018). On the transmission mechanism of country-specifc and ınternational economic uncertainity spillovers: Evidence from a TVP-VAR connectedness decomposition approach. Economics Letters, 171, 63–71. Getea, P. & Melkadzeb, G. (2018). Aggragate volatility and international dynamics. The role of credit supply. Journal of International Economics, 111, 143–158. Gilchrist, S., Sim, J. W., & Zakrajsek, E. (2014). Uncertainity, fnancial frictions and investment dynamics. National Bureau of Economic Research Working Paper 20038. Guiso, L., & Parigi, G. (1999). Investment and demand uncertainity. Quarterly Journal of Economics, 114, 185–227. Hanias, P. M., Curtis, G. P., & Thalassinos, E. J. (2007). Non-linear dynamics and chaos: The case of the price indicator at the Athens Stock Exchange. International Research Journal of Finance and Economics, 11(1), 154–163. IMF. (2013). World Economic Outlook: Apr-13. Washington, DC: International Monetary Fund. IMF. (2018). Brighter prospects, optimistic markets, challenges ahead. World Economic Outlook, January 22. Jens, C. E. (2017). Political uncertainity and ınvestment: Causal evidence from U.S. gubernatorial elections. Journal of Finance and Economics, 124, 563–579. Jurado, K., Ludvigson, S. C., & Ng, S. (2015). Measuring uncertainity. American Economic Review, 105(3), 1177–1216. Kamber, G., Karagedikli, O., Ryan, M., & Vehbi, T. (2016). International spillovers of uncertainity shocks: Evidence from a FAVAR. CAMA Working Papers 2016-61. Kimball, M. S. (1990). Precautionary Saving in the Small and in the Large. Econometrica, 58, 53–73. Kido, Y. (2018). The transmission of US economic policy uncertainty shocks to Asian and global fnancial markets. North American Journal of Economics and Finance, 46, 222–231. Knight, F. H. (1921). Risk, Uncertainty and Proft. Boston MA: Hart, Schaffner and Marx; Houghton Miffin. Korkmaz, Ö., & Ve Güngör, S. (2018). Küresel Ekonomi Politika Belirsizliğinin Borsa İstanbul’da İşlem Gören Seçilmiş Endeks Getirileri Üzerindeki Etkisi. Anemon Muş Alparslan Üniversitesi Sosyal Bilimler Dergisi, 6, 211–219. Leduc, S., & Liu,Z. (2016). Uncertainty Shocks are Aggregate Demand Shocks. Journal of Monetary Economics, 82, 20–35. Lee, J., & Strazicich, M. C. (2003). Minimum Lagrange Multiplier Unit Root Test with Two Structural Breaks. Review of Economics and Statistics, 85(4), 1082–1089. Perron, P. (1989). The great crash, the oil price shock and the unit root hypothesis. Econometrica, 57, 1361–401. Rodrick, D. (1991). Policy uncertainity and private investment in developing countries. Journal of Development Economics, 36, 229–242. Sahinoz, S., & Cosar, E. E. (2018). Economic policy uncertainity and economic activity in Turkey. Applied Economics Letters, 25(21), 1517–1520. Thalassinos, I. E., & Thalassinos, E. Y. (2018). Financial crises and e-commerce: How are they related? SSRN. Retrieved from https://ssrn.com/abstract=3330169 Trung, N. B. (2018). The spillover effect of the US uncertainity on emerging economies: A panel VAR approach. Applied Economics Letters, 26(3), 210–216. Tsay, R. S. (1998). Testing and modeling multivariate threshold models. Journal of the American Statistical Association, 93(433), 1188–1202.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

This page intentionally left blank

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Chapter 3

The Investigation of Zombie Firms’ Existence in Borsa İstanbul Manufacturing Industry Emre Kaplanoğlu and Canan Yükçü Abstract Introduction: The word “zombie” is associated with the “living dead” in our minds from horror movies we watch on TV. Recently, this concept has been used frequently to identify the frms that are still standing while they should have been closed long ago. Zombie frms are apparently active, but their cash fows are only used to compensate interest expense of their debts. Banks prefer to re-fnance these frms and restructure their sunken loans rather than following their debts and moving them out of the bank’s balance sheets. However, these ineffcient and unproductive frms also consume the capital that can be transferred to more productive frms. Therefore, these frms live as long as the cash infow, which is considered as fresh blood and meat for the living dead-zombie, and they consume the resources of other living healthy frms.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Purpose: The aim of this study is to investigate the existence of zombie frms which are listed in Borsa İstanbul manufacturing industry. Methodology: The research period is between the years 2008 and 2018, and interest coverage ratio (ICR) is used for Borsa İstanbul manufacturing frms. There are several explanations of zombie frms in the literature, which are commonly constructed on a scope of proftability of a frm. In this research, the OECD’s preferred explanation and classifcation of zombies is chosen which describes a zombie frm as having an ICR (operating earnings to interest expenses) which is less than 1 over three consecutive years. Findings: It has been noted from this research that ICRs differ in the research period of Borsa İstanbul manufacturing industry frms. About

Uncertainty and Challenges in Contemporary Economic Behaviour, 37–48 Copyright © 2020 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-80043-095-220201004

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

38    Emre Kaplanoğlu and Canan Yükçü 62 of 109 frms traded on Borsa İstanbul manufacturing industry between 2008 and 2018 were classifed as zombie frms because they had ICRs below one for three consecutive years or over. Keywords: Zombie frms; Borsa İstanbul; manufacturing industry; interest coverage ratio; EBIT; interest expense

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

1. Introduction In any economy, productivity can be defned as the effcient use of scarce resources and the growth of productivity depends on labor and capital allocation among productive frms. Additionally, effcient and productive frms continue to expand, and ineffcient and unproductive frms sink, cutting operations and leaving in a well-organized market. In a competitive environment, zombie frms (insolvent businesses, fnancially distressed enterprises, etc.) naturally exit from the market. The concept of zombies comes from a spell made in Haiti. A zombie is the name given to corpses played by sorcerers, and in popular culture, it is expressed as consuming everything it fnds without producing any value. We can say that the concept of zombie is to consume without producing the essence in popular culture (Platts, 2013). The concept of a zombie frm has been used since the late 1980s. The occurrence of theory of zombie frms has already been deeply analyzed, mainly in the case of Japan. The theory of zombie frms can also be used in researches related to other countries, especially developing countries, because there are several studies about other developed countries. One reason for zombie frm emergence is related to the decline in fnancial pressure as a result of low interest rates. Zombie frms are seemingly active, but use a huge part of their cash fow only to pay interest on their debts. These ineffcient and unproductive frms also consume capital that can be transferred to more productive frms. Banks prefer to re-fnance these frms and restructure their bad loans instead of reducing the debts of these frms and taking them out of the balance sheet. Therefore, these frms live as long as the cash fows, which is regarded as fresh blood for themselves, while consuming the resources of other living healthy frms. Banks are also stimulated negatively by zombie frms. The existence of zombie frms may be an unwanted effect of the incentives that banking regulation generates on the part of banks. A bank may choose to re-fnance zombie frms rather than restructuring their loans because the latter would reveal substantial losses on existing loans when faced with restructuring or extending a loan. So the bank’s share value can decrease. If the bank’s capital is already close to its regulatory threshold, then a large-scale restructuring of its loans might cause the need for costly recapitalization. Therefore, there is a paradox that zombie frms make zombie banks or zombie banks make zombie frms (Fukuda & Nakamura, 2011; Goto & Wilbur, 2019; Hoshi, 2006; Nakamura, 2016). Also, zombie banks have responsibilities in banking and other related crises (Thalassinos & Stamatopoulos, 2015) The purpose of this research is to investigate the existence of zombie frms among the frms listed in Borsa İstanbul manufacturing industry. The research on

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

The Investigation of Zombie Firms’ Existence    39 Borsa İstanbul manufacturing industry frms was carried out from 2008 to 2018 and the research was done using the interest coverage ratio (ICR) as per OECD’s defnition. It was found that ICRs of Borsa İstanbul manufacturing industry frms varied during the research period. About 62 of 109 frms that traded on Borsa İstanbul manufacturing industry between 2008 and 2018 were classifed as zombie frms because they had an ICR below 1 for three consecutive years or more. This research is organized in six sections. Section 2 reviews previous studies about zombie frms. Sections 3 and 4 discuss the scope and methodology of the research. Section 5 demonstrates the fndings. In Section 6, conclusions are given.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

2. Literature Review for Zombie Firms According to Kobayashi and Osano (2011), the main banks are forced to perform more successfully in the management of distressed loans when non-main banks pull capital out. The main banks can stay away from bad situations concerning zombie frms on condition that the frm’s cash fow fuctuates (decreases or increases) in connection with non-main banks’ fnanced supply. Fukuda and Nakamura (2011) investigated recovery reasons related to Japanese zombie frms in the 2000s with an extended method of previous studies. They categorized frms as zombies among the listed frms and found that reorganization, such as, decreasing employee strength and selling unnecessary tangible assets was adequate for troubled corporations to improve. Increasing losses supported the recovery but increasing profts through the disposition of main entities had a negative effect on zombie frms’ readjustments to health. Zombie frms’ recovery was also affected by external support and interventions such as debt relief and capital reduction. Jaskowski (2015) showed that lending to a zombie frm can be an ideal solution for a bank in some circumstances. The zombie lender bank can increase lending and avoid losses stemming from asset sales at a very low price. Banks prefer to add capital to their banking system, but it would not contribute to the solution of zombie frm lending. On the other hand, distress of zombie frm lending would decrease by straight buying of the collateral from the market. Kwon, Narita, and Narita (2015) investigated the distribution of resources in Japan by calculating the total productivity growth via producers’ data about factory and equipment among other things, between 1981 and 2000. They found that the support to the total productivity growth of resource distribution declined in the 1990s and turned into a negative at the end of the 1990s. They also indicated that zombie frms avoided decreasing factors of production to misguide banks into supporting them. Imai (2016) estimated that most of the zombie frms’ predominance, borrowing and investment activities between 1999 and 2008 in Japanese consisted of small and medium sized frms. About 4–13% of small- and medium-sized frms were identifed as zombies between 1999 and 2008. Borrowing function reveals that small and medium sized zombie frms did not change their loans as a reaction to differences in estate prices, resulting from ever greening. The marginal q as investment’s proftability did not have benefcial consequences for investments of zombies.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

40    Emre Kaplanoğlu and Canan Yükçü Tan, Huang, and Woo (2016) found that investment from the state increased the operations of zombie frms and suppressed the expansion of healthy frms in China between 2005 and 2007. They also found that the accumulation degree of state-owned banks in the market assists the conditions of zombie frms. McGowan, Andrews, and Millot (2017) showed in an OECD paper that there is an opportunity to develop the insolvency rules in order to lessen the barriers to reorganization of weak frms and the personal costs related with business failure. Problems of zombie frm may relatively stem from banks’ tolerance and a more forceful policy to resolve non-performing loans can be effective by complementary reforms to insolvency regimes. De Barros and Pereira (2017) found that between 5.2% and 12.5% of frms in the Portuguese market, in 2008 and in 2013, respectively, were zombie frms pertaining to non-tradable sectors of construction and services. They also supported the theoretical predictions and previous empirical results that confrm that there is a greater zombie existence in construction and services, which in turn has signifcant negative implications for healthy frms operating in the construction and services sector. Jiang, Li, and Song (2017) examined zombie frms which are trading on China’s Shanghai and Shenzhen markets between 2009 and 2016. They found that zombies are capable of staying vigorously alive with government funding in the form of incentives and bank loans. Lam, Schipke, Tan, and Tan (2017) illustrated the principal role of zombies and their strong connections with state-owned frms in funding debt weaknesses and low productivity with using novel frm-level industrial survey data. They showed that weak frms can generate signifcant gains in the long-term growth per year and also operational reorganization through divestment and decreasing redundancies. Shen and Chen (2017) documented the complication of excess capacity and zombie frms between 2011 and 2013 in China. They found that complications about excess capacity are especially penetrating in heavy chemical industries and public industries in China’s northeastern and western areas. Caballero, Hoshi, and Kashyap (2008) examined the effects of zombie frms that how zombies conceal the normal competitive process. They found that congestion by the zombies decreases the profts for non-zombie frms and industries which are infuenced by zombies have weak productivity, more despondent job generation and elimination. Chen, Du, Wu, and Zhou (2017) analyzed the development trend, as well as the macro-economic environment and industries and the share of zombies in China between 2007 and 2015. They used an altered method that takes its abbreviation FN-CHK from studies of Fukuda, Nakamura, Caballero, Hoshi and Kashyap. They detected zombie frms by integrating irreproducible gains and losses. They also suggested that the state has to improve the supply in the market and manage a well-selected industry policy. Storz, Koetter, Setzer, and Westphal (2017) explored the link between zombie frms and bank health and the magnitudes for aggregate productivity in 11 European countries. They found that zombies were conceivably connected to weak

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

The Investigation of Zombie Firms’ Existence    41 banks and partly stem from bank tolerance toward them. The increasing continued existence of zombie frms overburden for markets and limits the growth of healthy frms. Urionabarrenetxea, Merino, San-Jose, and Retolaza (2018) showed frms with negative equity that carry on operating in spite of having no equity at all as extreme types of zombie frm. They took into consideration the riskiest frms by diverse determining factors in EZIndex between 2010 and 2014. They established a method for grading zombies, according to their risks and deviations over time. Even though zombies are not too legally structured, they are still placed in districts with big industries, and they recommended important improvements that need to be explored by the capable authorities. Hallak, Harasztosi, and Schich (2018) identifed the occurrence of zombie frms over European states and the effects of such incidence on the asset growth, employment growth and productivity of non-zombie frms. The results showed that the larger the share of zombie frms in a country, the lower the growth of non-zombies in that country. They found that young non-zombies were most affected by other zombie frms through congestion effects since zombies were spreading in Europe. He, Li, and Zhu (2018) documented that becoming bankrupt and fruitless is a result of political relationships of frms. In their study, they used data of stateowned companies in China. They found that this situation is more clear-cut for frms placed in areas with widespread state involvement. On the other hand, the existence of zombie frms has greater adverse spreading results in the investment and output of non-zombie frms, when matched with zombie frms in the similar sector without governmental linkages. In addition, adverse spreading results are not detected for frms which have governmental linkages. Huang and Zhang (2018) examined the foreign direct investment of frms without productivity advantages using Chinese manufacturing frm-level data between 2002 and 2005. They found that some zombie frms got fnancial maintenance from government to conduct outward foreign direct investment. Banerjee and Hofmann (2018) investigated an increase in the existence of zombies at the end of the 1980s in 14 developed economies. They found that zombie frms’ increase is connected with decreased fnancial stress due to low interest rates. Goto and Wilbur (2019) focused on small and medium sized zombies in Japan. They found that a lot of zombie frms occurred in small and medium sized frms and the share of zombie frms increased as a result of frm size decreased. Du and Li (2019) identifed zombie frms in the energy sector by an altered method. They applied the survival analysis model to show the infuence of environmental ruling on the walk out of zombies from the sector. They found that environmental ruling can increase the walk out process of zombies and give the energy sector additional capacity. Dai, Qiao, and Song (2019) used an altered recognition method to detect zombie frms from a big sample of coal mining frms in China. They analyzed the existence of zombies spreading across diverse areas and proprietorship forms. The causes of the occurrence of zombie frms and the assessment of the

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

42    Emre Kaplanoğlu and Canan Yükçü performance of several reorganizing practices to stump out the zombies’ problem are investigated. They found that state interferences as well as bank funds are vital reasons for the existence of zombies. The uninterrupted fnancial backing from the state or banks does not provide any support to the restructuring of zombies. Andrews and Petroulakis (2019) explored the links between zombie frms and bank health and the consequences on total productivity in 11 European countries. The results of their study showed that zombie frms are linked with weak banks, and the increasing survival of zombie frms blocks markets and pressures the advancement of healthier frms. Rodano and Sette (2019) showed the results of different methods to assess the presence of zombie frms in Italy. They used OECD’s measure (McGowan et al., 2017) for comparison, which identifes zombie frms based on criteria such as frm age and ICR. They found that profts taken before amortization and depreciation or after amortization and depreciation deliver different results. Profts, after amortization and depreciation, have numerous adverse features, such as the overestimation of the capital portion trapped into zombies, which is inadequate for predicting future performance of frms, thus classifying frms as zombies in a specifc year, having invested heavily in previous years and amortized that investment quickly and unsuitably for cross-country evaluations.

3. Scope

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

We investigated the existence of zombie frms and the Borsa İstanbul manufacturing industry frms were used for this research, however continuously listed frms for the related period were also in the scope. As of January 2019, 178 frms were traded in Borsa İstanbul manufacturing industry. One hundred and nine of them were continuously traded from 2008 to 2018, and examined during this period (Table 1).

4. Methodology Several different explanations of zombie frms are used in the literature. These are mostly grounded on the calculation of a company’s proftability. ICR is the OECD’s favored defnition for zombie frms. It identifes zombie frms, according to their ICR (the ratio of earnings before interest and taxes to interest expenses). Zombie frms are those who have an ICR with less than 1 (Earnings before interest and taxes/interest expense < 1) over three successive years in a period of 10 years (McGowan et al., 2017). In other words, a frm must take on additional debt to cover its interest payments. The research on Borsa İstanbul manufacturing industry frms was conducted between 2008 and 2018 and the ICR was used (ICR is equal to 1 and the frm covers only interest expenses with its earnings before interest and taxes). The higher the rate is, the lesser the risk of the entity’s inability to cover interest expenses. If the ratio is greater than 1, the frm can cover all of the interest expenses with earnings before interest and taxes. If the rate is less than 1, then

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

ADEL

AEFES

AFYON 23

AKCNS

AKSA

ALCAR 26

ALKA

ALKIM 28

ANACM 29

ARCLK 30

ARSAN 31

ASLAN

ASUZU 33

ATEKS

AYGAZ

BAGFS

BAKAB 37

BANVT 38

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

36

35

34

32

27

25

24

22

21

ADANA 20

1

48

47

46

45

44

43

42

41

40

39

52

51

DESA

57

DERIM 56

DENCM 55

CMENT 54

CMBTN 53

CIMSA

CEMTS

CELHA 50

CCOLA 49

BURVA

BURCE

BUCIM

BTCIM

BSOKE

BRSAN

BRISA

BOSSA

BOLUC

BFREN

58 60

67

GEREL

GENTS

FROTO

FMIZP

ERSU

EREGL

ERBOS

EMNIS

76

75

74

73

72

71

70

69

EMKEL 68

EGSER

EGGUB 66

EGEEN 65

DYOBY 64

DURDO 63

DOBUR 62

DMSAS 61

DITAS

DGKLB 59

DEVA

77

80

84

83

82

88

87

94 KRTEK 95

KRSTL

KRDMA 93

KORDS 92

KONYA 91

KNFRT 90

KLMSN 89

KERVT

KENT

KARTN 86

KARSN 85

IZOCM

IZMDC

IHEVA

HURGZ 81

HEKTS

GUBRF 79

GOODY 78

GOLTS

TBORG

SODA

SKTAS

SARKY

PNSUT

PINSU

PETUN

PETKM

PENGD

PARSN

OTKAR

OLMIP

NUHCM

MRSHL

MRDIN

MNDRS

MERKO

LUKSK

KUTPO

Table 1.  Manufacturing Industry Firms in Borsa İstanbul from 2008 to 2018 in the Scope of Research.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

109

108

107

106

105

104

103

102

101

100

99

98

97

96

YUNSA

YATAS

VKING

VESTL

VESBE

USAK

UNYEC

ULKER

TUPRS

TUKAS

TTRAK

TRKCM

TOASO

TIRE

The Investigation of Zombie Firms’ Existence    43

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

44    Emre Kaplanoğlu and Canan Yükçü the frm is having problems with interest payments. It can be said that it needs additional cash fow to overcome this problem. The formula of the ICR is given in Eq. (1).

Interest Coverage Ratio =

Earnings Before Interest and Taxes (1) Interest Expense

To calculate the ICR of the manufacturing industry frms listed in Borsa İstanbul from 2008 to 2018, the authors used data from the fnancial statements announced on the public disclosure platform (www.kap.gov.tr). The calculations of the research were performed using Microsoft Excel 2016.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

5. Results The classifcation of the 109 frms is given in Table 2. The ICR of 62 manufacturing industry frms listed in Borsa İstanbul was below 1 over three consecutive years between 2008 and 2018. Therefore, these frms were classifed as zombie frms, according to the OECD’s defnition. The rest of the 47 frms were classifed as non-zombie frms. Forty-seven frms did not meet the defnition of OECD between 2008 and 2018, so they were classifed as non-zombies. However, some of these non-zombie frms’ ICRs were below 1 in the research period but not over three consecutive years. On the other hand, 62 frms out of 109 frms were classifed as zombie frms. In other words, in the 62 manufacturing industry frms’ ICRs remained below 1 over three consecutive years between 2008 and 2018 and these frms could not cover their interest expenses. According to Table 2, the average of the ICR from 2008 to 2018 was calculated as 7.43 for 47 non-zombie frms and 2.29 for 62 frms which were classifed as zombie frms. The average of the ICR between 2008 and 2018 was calculated as 4.49 for the 109 manufacturing industry frms which are listed in Borsa İstanbul. The annual average ICRs of Borsa İstanbul manufacturing industry frms examined for the scope of this research and the annual average ICRs according to the classifcation (zombies and non-zombies) from 2008 to 2018 are given in Fig. 1. The annual average ICRs of frms classifed as zombies decreased between 2008 and 2012. The lowest rate in 2012 was –4.07. The annual average ICRs of the frms classifed as non-zombie frms increased from 3.99 in 2008 to 7.36 in 2012. Table 2.  Number of Zombie and Non-Zombie Firms in Borsa İstanbul Manufacturing Industry. Total Number of Firms Number of Non-zombie Firms Number of Zombie Firms 109

47

62

Average ICRs between 2008 and 2018 4.49

7.43

2.29

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

The Investigation of Zombie Firms’ Existence    45

Fig. 1.  Annual Average ICRs of Borsa İstanbul Manufacturing Industry Firms from 2008 to 2018. Table 3.  Average Number of Years of Zombie Firms According to ICRs.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Period

ICR1

2008–2018

8

3

2008–2012

4

1

2013–2018

4

2

From 2012 to 2014, both zombie frms and non-zombie frms’ annual average ICRs increased. In 2014, the annual average ICR of zombie frms was 11.85 and the annual average ICR of non-zombie frms was 17.32. From 2014 to 2018, the manufacturing industry frms’ annual average ICRs in both classifcations decreased. Table 3 shows the average number of years according to the ICRs (ICR1) of zombie frms between 2008 and 2018. Between 2008 and 2018, the authors found that the average number of years in which the ICR was less than 1 and more than 1 was eight and three years, respectively. The average number of years in which the ICR was less than 1, between 2008 and 2012, and between 2013 and 2018 were four years. The average number of years in which the ICR was greater than 1 between 2008 and 2012 was 1 and between 2013 and 2018 was two years. Considering the defnition of a zombie frm, that is, ICR is less than 1 during three consecutive years in a period of 10 years, the calculation of the number of average consecutive years in which ICR was less than 1 for zombie frms in Borsa İstanbul manufacturing industry in research period is given in Table 4.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

46    Emre Kaplanoğlu and Canan Yükçü Table 4.  Number of Average Consecutive Years in Which ICR is Less Than 1 for Zombie Firms for Different Periods. Period

ICR 1 (indicating that there should be a greater number of frms that survive the outcome of forensic investigations compare to frms that do not survive the outcome of forensic investigation).

6. Conclusion This chapter is about forensic accounting theory, not about how to conduct forensic investigation. It argues that forensic accountants, having been exposed to the methodology and practice of forensic accounting, need to examine the broader implications of the choice of forensic detection methods for the fair and effcient

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Forensic Accounting Theory    59 resolution of unresolved cases. The objective is to give the reader a critical awareness of the current forensic accounting and investigation environment, taking into account the diverse interests of both the forensic investigator and the frms being investigated.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

References Albrecht, S., & Romney, M. (1986). Red-fagging management: A validation. Advances in Accounting, 3, 323–333. Arokiasamy, L., & Cristal-Lee. S. (2009). Forensic accounting: Public acceptance towards occurrence of fraud detection. International Journal of Business and Management, 4(11), 145–160. Bolgna, G., & Linquist, R. (1995). Fraud auditing and forensic accounting: New tools and technique (2nd ed.). New York, NY: John Wiley & Sons. Buckhoff, T. A., & Schrader, R. W. (2000). The teaching of forensic accounting. Journal of Forensic Accounting, 1(1), 135–146. Bunker, R. J. (2011). Criminal (Cartel & Gang) Insurgencies in Mexico and the Americas: What You Need to Know, Not What You Want to Hear: Testimony before the House Foreign Affairs Subcommittee on the Western Hemisphere at the Hearing ‘Has Merida Evolved? Part One: The Evolution of Drug Cartels and the Threat to Mexico’s Governance.’ Small Wars Journal El Centro, 86. Burns, N., & Kedia, B. (2006). The impact of performance-based compensation on misreporting. Journal of Financial Economics, 79(1), 35–67. Calavita, K., Tillman, R., & Pontell, H. N. (1997). The savings and loan debacle, fnancial crime, and the state. Annual review of sociology, 23(1), 19–38. Crumbley, D. L., Heitger, L. E., & Smith, G. S. (2003) Forensic and investigative accounting. Chicago, IL: CCH Incorporated. Davis, C., Farrell, R., & Ogilby, S. (2010). Characteristics and skills of the forensic accountant. White paper. American Institute of Certifed Public Accountants, pp. 11–26. Denis, D. J., Hanouna, P., & Sarin, A. (2006). Is there a dark side to incentive compensation? Journal of Corporate Finance, 12(3), 467–488. Dhar, P., & Sarkar, A. (2010). Forensic accounting: An accountant’s vision. Vidyasagar University Journal of Commerce, 15(3), 93–104. Dorminey, J. W., Fleming, A. S., Kranacher, M. J., & Riley, R. A., Jr. (2010). Beyond the fraud triangle. The CPA Journal, 80(7), 17. Erickson, M., Mayhew, B. W., & Felix, W. L (2000) Why do audits fail? Evidence from Lincoln savings and loan. Journal of Accounting Research, 38, 165–194. Free, C. (2015). Looking through the fraud triangle: A review and call for new directions. Meditari Accountancy Research, 23(2), 175–196. Hernandez, J. R., & Groot, T. (2007) Corporate fraud: Preventive controls which lower fraud risk. Working Paper, Free University of Amsterdam. Howard, S., & Sheetz, M. (2006). Forensic accounting and fraud investigation for nonexperts. Hoboken, NJ: John Wiley & Sons Inc. Huber, W. (2017). Forensic accounting, fraud theory, and the end of the fraud triangle. Journal of Theoretical Accounting Research, 12(2), 28–48. Johnson, S. A., Ryan, H. E., & Tian, Y. S. (2003). Executive compensation and corporate fraud. Working Paper, WLU/Louisiana State University. Liapis, K., & Thalassinos, I. E. (2013). A comparative analysis for the accounting reporting of “employee benefts” between IFRS and other accounting standards: A case

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

60    Peterson K.Ozili study for the biggest listed entities in Greece. International Journal of Economics and Business Administration, 1(1), 91–116. Lie, E. (2005). On the timing of CEO Stock Option Awards. Management Science, 51(5), 802–812. Lokanan, M. E. (2015, September). Challenges to the fraud triangle: Questions on its usefulness. Accounting Forum, 39(3), 201–224. Messmer, M. (2004). Exploring options in forensic accounting. National Public Accountant, 5, 9–20. Ozili, P. K. (2015). Forensic Accounting and Fraud: A Review of Literature and Policy Implications. International Journal of Accounting and Economics Studies, 3(1), 63–68. Özkul, F. U., & Pamukçu, A. (2012). Fraud detection and forensic accounting. In K. Çaliyurt & S. O. Idowu (Eds.), Emerging fraud (pp. 19–41). Berlin: Springer. Peterson, B., & Reider, B. (2001). An examination of forensic accounting courses: Content and learning activities. Journal of Forensic Accounting, 2(1), 25–42. Ramaswamy, V. (2005). Corporate governance and the forensic accountant. CPA Journal, 75, 68–70. Reding, A. (1997). Facing political reality in Mexico. Washington Quarterly, 20(4), 103–116. Rezaee, Z. (2002). Forensic accounting practices, education, and certifcations. Journal of Forensic Accounting, 3(2), 207–223. Rezaee, Z. (2005). Causes, consequences, and deterrence of fnancial statement fraud. Critical Perspectives on Accounting, 16(3), 277–298. Rezaee, Z., & Burton, E. J. (1997). Forensic accounting education: Insights from academicians and certifed fraud examiner practitioners. Managerial Auditing Journal, 12(9), 479–489. Schuchter, A., & Levi, M. (2016). The fraud triangle revisited. Security Journal, 29(2), 107–121. Silverstone, H., Sheetz, M., Pedneault, S., & Rudewicz, F. (2004). Forensic accounting and fraud investigation for non-experts (Vol. 2). Hoboken, NJ: Wiley. Singleton, T. W. (2010). Fraud auditing and forensic accounting (Vol. 11). Hoboken, NJ: John Wiley & Sons. Stanbury, J., & Paley-Menzies, C. (2010). Forensic futurama: Why forensic accounting is evolving. AICPA Store, June 28. Sullivan, J. P. (2011). From drug wars to criminal insurgency: Mexican cartels, criminal enclaves and criminal insurgency in Mexico and Central America. Implications for global security. Working Paper Series. Fondation Maison des sciences de l’homme, Paris. Thalassinos, I. E., Pintea, M., & Raţiu, I. P. (2015). The recent fnancial crisis and its impact on the performance indicators of selected countries during the crisis period: A reply. International Journal of Economics and Business Administration, 3(1), 3–20. Ugurlu, E., Thalassinos, E., & Muratoglu, Y. (2014). Modeling volatility in the stock markets using GARCH models: European emerging economies and Turkey. International Journal of Economics and Business Administration, 2(3), 72–87. Zahra, S. A., Priem, R. L., & Rasheed, A. A. (2005). The antecedents and consequences of top management fraud. Journal of Management, 31(6), 803–828.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Chapter 5

Econometric Analysis for the Period Between 2003 and 2018 as Regards the Functioning of Interest Channel of Monetary Transmission Mechanism in Turkey Berna Kaçar and Huriye Gonca Diler Abstract Introduction: Monetary policy resolutions issued by central banks play effective role in economy when accompanied with interest variable. In Keynesian approach to fnance, interest is treated as the main determinant underlying fnancial policy resolutions. Thus interest is a pivotal factor in monetary transmission mechanism. Tight monetary policy practices, essentially decreasing money supply, eventually lead to a slump in investments, total demand and national income due to the increase in real interest rates.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Objective: The aim of this study is to determine what type of effects do monetary policy practitioner in Turkey have on macroeconomic variables via the interest channel of monetary transmission mechanism. Methodology: Based on this objective, variables that could help in unveiling CBT overnight interest rates, direct fxed capital investment (GSSO), real gross domestic product (RGDP), industry production index (SUE) and domestic producer price index (YUFE) variables and that could explain monetary functions of transmission mechanism’s interest channel were selected. For the variables constituting the research topic, collected data belong the period of 2003Q1–2018Q3. Findings: In the study relation between the variables has been analyzed under two parts via harnessing Toda–Yamamoto casualty test. In the frst part, results of Toda–Yamamoto causality test from RGDP, GSSO and

Uncertainty and Challenges in Contemporary Economic Behaviour, 61–75 Copyright © 2020 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-80043-095-220201006

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

62    Berna Kaçar and Huriye Gonca Diler interest rate (FO) variables have been presented. The results manifest that interest channel directly affects direct fxed capital investment and RGDP. Interest channel was found to be effective on these variables of the analysis. In the second part, Toda–Yamamoto causality test was harnessed for SUE, YUFE and FO variables. Interest channel did not provide a result that affected YUFE and SUE. Keywords: Turkish economy; monetary policy; monetary transmission mechanism; interest channel; Zivot–Andrews unit root test; Toda–Yamamoto causality test JEL classifcation: E42; E52; E58

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

1. Introduction Finance and monetary policy are the two economy policies that complement each other. Finance policy refers to state-led formation of income and expenditure policies in alignment with specifc objectives such as economic growth and development. Monetary policy relates to compilation of resolutions issued by central banks to meet specifc economic goals via utilizing policy tools. Policy resolutions of central banks necessitate harnessing appropriate and effective policy tools in order to steer national economy. Transmission mechanism means monetary transfer from fnancial markets to real market as a result of resolutions on monetary policy. The way functioning of this mechanism can affect macroeconomic variables is vitally important. Therefore, it is mandatory that to reach their objectives authorities of money implement appropriate policy tools. Despite being a powerful tool, monetary policy may occasionally result in unexpected outcomes. Furthermore, to ensure success in the execution of monetary policy, it is a must to correctly analyze timing and effect of this policy on economy (Mishkin, 1995, p. 4). As regards national economies, there may be a divergence from state to state in the functioning of their monetary transmission mechanism. This divergence is bound to economic parameters such as fnancial structure of a state, index of openness and its role in global economy. In order to exhibit potential effects of real policy tools that central banks implement to ensure fnancial stability on a national economy, there has been a rapid rise as of 1980s in the studies that manifested the functioning of monetary transmission mechanism. Monetary transmission mechanism has thus been harnessed to the end of examining how and why monetary changes can direct a national economy, and through which ways it can shape national production, total demand and the key problem-infation. After 2008 global crisis caused by the instability of fnancial markets in global economy, the salient role of monetary transmission mechanism has been accentuated even further (Central Bank of Turkey (CBT), 2012, p. 3). The aim of this study is to investigate the functioning and results interest channel of monetary transmission mechanism, which is one of the effective policy instruments used to ensure economic stability of the CBT in the period 2003–2018.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Econometric Analysis 2003–2018    63 The period 2003–2018 for Turkey’s economy is referred to as transitional period to strong economy. During this period, strong structural changes have been made in the economy. The effectiveness of the central bank in the economy has also increased. It is important to investigate the functioning and results of the interest rate channel, which is one of the most powerful instruments of monetary policy in this period. This is why it is necessary to do the study. Studies investigating the interest rate channel of monetary transmission mechanism in Turkey have generally used monthly data. Since monthly data are more sensitive to seasonal effects and trends, they may mislead the results of the analyses. However, the use of quarterly data in our study reveals a different result for the literature. On the other hand, the variables used in analyzing the functioning interest channel of the monetary transmission mechanism also differ from other studies. These variables and functioning of interest channel have been analyzed in two sections according to the macrofunctions of the variables. In line with the determined purpose of the study in terms of examined period and used variables, its contribution to the literature is important. This also reveals the originality of the study.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

2. Literature Review In the economics literature, there are studies dealing with from various perspectives the functioning of the monetary transmission mechanism in countries. In some studies, the channels of monetary transmission mechanism have been considered separately, and in some studies, some channels have been considered jointly. In terms of Turkey’s economy, studies of the functioning of monetary transmission mechanism have been also made. Taylor (1995) reported that interest channel is the most noteworthy channel in any monetary transmission. Chirinko and Von Kalckreuth (2003) presented that interest channel and credit channel are signifcant attributes for fxed investment in Germany. In their research, they evidenced that (infation expectation not destructing the change) 100 points of fall in nominal interest rates could lead to 7.55% slump in investment expenditures and approximately 1.40% jump in gross domestic product (GDP) in two years. By using vector autoregression (VAR) and error correction models Guerra, Rodriguez, and Sänchez (1998) examined monetary transmission mechanism for the period of 1985–1995 in Venezuela. Butzen, Fuss, and Vermeulen (2001) analyzed effects of monetary policy of Belgium on corporate investment behaviors via applying data from 1985 to 1998 period. Discrimination was made based on the size of corporations. It then became certain that, unlike large corporations, small corporations were more sensitive to monetary policy. In the end, it was detected that the study was in line with the theory on the functioning of interest channel. In a research Peersman and Smets (2001) applied VAR method on the Euro Area for the period between 1980 and 1998 and compared the reactions that macroeconomic variables showed against a monetary policy shock.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

64    Berna Kaçar and Huriye Gonca Diler Wróbel and Pawlowska (2002) used structural vector autoregressive (SVAR) method and studied the functions of interest channel and credit channel in Poland. Chapsa, Tabakis, and Athanasenas (2018) have used panel unit root test to investigate the catching-up hypothesis in the group of PIIGS (Portugal, Italy, Ireland, Greece, and Spain), while Thalassinos, Pintea, and Raţiu (2015) have examined the impact of the recent crisis on the performance indicators for selected countries. Seyrek, Duman, and Sarıkaya (2004) in their study covering years from 1968 to 1996 compared and contrasted monetary and New-Keynesian monetary transmission mechanisms. Their analysis evidenced that money supply in Turkey was not internal and money supply could, to an insignifcant degree, explain the level of credit volume and miscellaneous variables. Thus, it was concluded monetary transmission mechanism indeed functioned. By employing VAR approach Çiçek (2005) shed light on monetary transmission mechanism in Turkey for the period of 1995:1–2003:2. They concluded that traditional interest channel was the best functioning channel in monetary transmission and around 25% of fuctuations in real activities were due to the effects of interest channel. Mehrotra (2006) analyzed the role of foreign exchange and interest channel in Japan, Hong Kong and China via using SVAR method and concluded that effects of interest shocks were substantial in states of Japan and Hong Kong. Interest rate was not a salient monetary policy tool in China and failed to be supportive in issuing new prices against foreign exchange or interest shocks. In his analysis, Kasapoğlu (2007) employed data from 1990:1 to 2006:7 period and the interaction between monetary policy and real economy on the basis of VAR model. It was determined that traditional interest rate channel functioned over monetary transmission channels while foreign exchange rate created not an effective result on real activity. Yet it signifcantly affected generic level of prices. It was also ascertained that not any statistically signifcant fndings surfaced as regards the existence of stock exchange prices channel and credit channel. To examine the functioning of interest channel in Turkey Sarı (2007) analyzed the period from 1986:04 to 2004:12 by employing structural VAR analysis revealing that a spike in money supply lowered interest rate in the frst two months but it jumped in the third month. This conclusion pinpointed that production was directed by the waves in interest rates as stipulated in the theory hence it was manifested that there was a movement from monetary variables to the variables in real sector. As money supply impacts interest rate, interest rate in turn acts upon production. Bhuiyan (2008) in an analysis of monetary transmission mechanism covering the period from 1994 to 2007 in Canada concluded that monetary shocks operated on real economy through interest channel and exchange rate channel. Bilan and Kryshko (2008) in their study on Ukraine claimed that compared to other channels the interest channel with a transmission mechanism was not superior in comparison. Büyükakın, Bozkurt, and Cengiz (2009) examined interest channel of Turkey by employing Granger casualty test and Toda–Yamamoto method. This model employed annual data of the period from 1990:01 to 2007:09. It was manifested

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Econometric Analysis 2003–2018    65 that results of both methods were signifcantly identical; in Toda–Yamamoto method interest rate acted upon the product via investment expenditures. In Granger casualty test it surfaced that any change in interest rate altered investment expenditures, price level and product. In their paper, Erdoğan and Yıldırım (2009) employed VAR analysis to reveal that during the period of 1995–2007 in Turkey interest channel rendered a shortterm infuence on monetary transmission mechanism. Between 1990 and 2006, Örnek (2009) tested the functioning of monetary transmission mechanism by applying quarterly data and in his analysis the researcher employed action and reaction function as well as variance method within the scope of VAR model. The analysis manifested that although traditional interest rate and exchange rate channels properly functioned, stock certifcate prices and bank credit channels failed to offer a statistically signifcant result. Via employing the method in VAR model Karaca (2010) shed light on the period from January 2002 to September 2009 in Turkey. He concluded that traditional interest rate channel and asset prices effectively functioned whereas foreign exchange rate channel and credit channel were not functional. Cambazoğlu and Güneş (2011) in their Turkey and Argentine focused study employed VAR method. They concluded that in Turkey, in the case of price level reactions against credit shock and interest rate shock monetary channel was functional while in Argentina bank credit channel was functional. In order to test the functioning of traditional transmission mechanism Doğan (2012) formed VAR model method that focused on quarterly data of the periods from 2000:1 to 2011:3 and revealed that traditional transmission mechanism functioned in effect. Via structural VAR method Mohanty (2012) examined the functioning of interest channel in India and at the end of this study it was detected that interest rate was quite an effective monetary policy tool. Arabacı and Baştürk (2013) in their interest channel analysis covering the period from 2001:05 to 2008:05 suggested not to dwell on a single sampling period only but rather analyze as the period before and after 2004. They concluded that when problem of post-2004 fnancial domination was removed interest channel functioned more effectively compared to 2004. Mukhtarov and Aliyev (2014) in their research aimed at exploring the potential effect of interest rate channel in Azerbaijan on real economy employed quarterly data VEC model covering the period of 2002–2012. Their analysis revealed that interest rate channel indeed rendered effects on macroeconomic variables in the short and long terms although they were statistically insignifcant effects. It was also monitored that in all cases were interest rate to render a negative effect on the examined variables. Nonetheless, its short-term effect on consumption expenditures was measured as positive. Turhan and Gümüş (2014) in their analysis of the period from 2004:01 to 2013:11 in Turkish economy employed annual data. They examined contributions of variables that could be effective in VAR model and monetary transmission to explain the fuctuations in industry production and infation. It was concluded that in Turkey interest, foreign exchange rate and credit channels functioned while

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

66    Berna Kaçar and Huriye Gonca Diler it was also detected that monetary policy resolutions on exchange rate channel offered relative signifcance in its transmission to real economy. Gitonga (2015) studied quarterly data of 2005–2013 period in Kenya on the basis of VAR method and tested interest channel transmission mechanism. Mishra, Montiel, and Sengupta (2016) explored functioning of India’s monetary transmission mechanism channels for the period from 2001 to 2014 via employing structural VAR method. They evidenced that although reaction of exchange rate channel against policy shocks was in the right direction, it still triggered an insignifcant effect. It was thus agreed that with this fnding it would be infeasible to obtain a strong enough prediction to refect the limits of empirical mythology in monetary transmission. Uğur, Sancar, and Polat (2016) in their Turkey-based study examined the interaction of monetary transmission channels with real economy via employing VAR model focusing on quarterly data from 1998:1 to 2015:1 period. Their study showed that in the short term all channels functioned as estimated.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

3. Data Series Variables that were employed in the econometric analysis utilized data collected in 2003Q1–2018Q3 period. The reason why the onset of time series employed in the model was set as 2003 is that implicit infation targeting of central bank was effectuated in the period of 2002. Besides, earlier data of the variables had high volatility and movements. The variables in this study, industry production index (SUE), real gross domestic product (RGDP), domestic producer price index (YUFE), direct fxed capital investment (GSSO) and interest rates (FO) were represented by weighted-average overnight interest rate executed in interbank money market. Data pertaining to the variables were gathered from data distribution system of CBT, Turkish Statistical Institute and General Directorate of Budget and Finance Control. Eviews 10 package program was employed in computations. GDP variable was realized by GDP defector. Excluding the interest rate, the logarithm of analyzed variables was measured and all of the variables were freed from seasonality by using moving average method. Graphs of SUE, RGDP, YUFE, direct fxed capital investment (GSSO) and interest rates (FO) are shown in Fig. 1.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Econometric Analysis 2003–2018    67 FOSA

50 40 30 20 10 0

03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 LNG S S O S A

19.6 19.2 18.8 18.4 18.0 17.6 17.2 16.8

03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18

LNRGSYISA

18.4

4.8

17.6

4.6

17.2

4.4

16.8 16.4

4.2

16.0

4.0

15.6

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

15.2

LNS UE S A

5.0

18.0

03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18

6.2 6.0 5.8 5.6 5.4 5.2 5.0 4.8 4.6 4.4

3.8

03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18

LNYUFESA

03

04

05

06

07

08

09

10

11

12

13

14

15

16

17

18

Fig. 1.  Graphs of SUE, RGDP, YUFE, GSSO and FO

4. Methodology 4.1. Stability Analysis Time series are directed by trend, seasonal, erratic fuctuations and conjectural components. Time series properties of the data are analyzed under two headings as deterministic and stochastic properties. Deterministic properties of the series relate to owning or lacking stability, trendy and seasonality effects. Stochastic properties of the series relate to the stability of variables (Tarı, 2014, p. 374).

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

68    Berna Kaçar and Huriye Gonca Diler Augmented Dickey Fuller (ADF) and unit root test will be employed to examine the stability level of variables.

4.2. ADF Unit Root Test Considering that error terms could have been distributed as a different variance or correlation, ADF test avoids including lagged values to the model. It means in ADF test, lagged difference terms are employed. Quantity of lagged difference terms is empirically determined. In order to remove the autocorrelation of error terms, lagged difference terms are integrated to the model (Tarı, 2014, p. 390).

4.3. Zivot and Andrews Unit Root Test It is a must that in order to detect the real results of interrelations between variables structural breakup must be accounted. However, ADF test disregards structural breakup. In Zivot and Andrews (1992) test breakup timing is internally determined. Structural breakup style of Zivot and Andrews test was designed as three models in order to test one unit root: Model 1, a model that allowed level single breakup; Model 2, single breakup in the slope of trend function; and Model 3 a model that allowed single breakup in both the slope of trend function and also level single: Model 1: Yt = a + θDU t (λ) + βt + αYt −1 + ∑ i=1Ci∆Yt−i + ut k

Model 2: Yt = a + µDT ∗t (λ) + βt  +α Yt −1 + ∑ i =1Ci∆ Yt−i + ut k

Model 3: Yt = a + θDU t (λ) µDT ∗t (λ) + βt  +αYt −1 + ∑ i=1Ci  ∆Yt−i + ut k

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

4.4. Toda–Yamamoto Causality Test Toda–Yamamoto is one of the methods that can determine the direction of the relation between variables with no need for identical stability levels and cointegration. Wald test is applied in this casualty analysis. This test presents that the lagged quantity equal to the integration level of series is included to VAR model and in that case the series would obtain Chi-square (χ2) distribution. Toda–Yamamoto casualty test establishes standard VAR model in the level values of the variables. Hence it minimizes any risks that may occur due to incorrect detection of cointegration values of the series (Mavrotas & Kelly, 2001). In Toda and Yamamoto (1995) test, frstly appropriate lag length (p) of VAR model is designated. Next, p lag length is summed with integration level of (dmax) the variable with the highest integration degree. Finally for p + dmax least squares method (EKK) model is predicted related to the original values of series.

Yt = ∑ i=1 max a1iYt−i + ∑ i=1 max b1i X t−i + ∑ i=1 max c1i Zt−i + u1t (1) p+d

p+d

p+d

X t = ∑ i=1 max a2 i X t − i + ∑ i=1 max b2 i Yt − i + ∑ i=1 max c2 i Zt − i +u2t (2) p+ d

p+ d

p+ d

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Econometric Analysis 2003–2018    69

Zt = ∑ i=1 max a3i   Zti +∑ i=1 max b3i   X ti +∑ i=1 max c3i  Yti +u3t (3) p+d

p+d

p+d

5. Findings and Discussion In the study, descriptive statistics were examined for the return series of SUE, RGDP, YUFE, direct fxed capital investment (GSSO) and interest rates. Based on the statistical values obtained by these descriptive statistics, it is examined whether the normal distribution of the series is appropriate. When the return series are examined, it can be seen that the skewness values according to Table 1 have a negative value for all return series (excluding interest rate) and the distribution is skewed to the left. When the kurtosis values are examined, it is seen that the SUE return series is less fat than all return series. According to the statistics of skewness and kurtosis, it is observed that the series do not comply to the normal distribution. This is investigated with the Jarque– Bera test statistic. According to the test statistic results, it is observed that the 1% signifcance level does not normally distribute.

Table 1.  Descriptive Statistics.

Mean Median

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Maximum

FO

GSSO

RGDP

SUE

YUFE

12.33657

18.27983

17.40688

4.330220

5.192216

18.21567

17.60455

4.308791

5.183153

19.46278

18.29978

4.797026

6.002412

16.83194

15.31990

3.822657

4.576835

9.604686 45.26568

Minimum

1.537210

Std. Dev.

8.497215

0.676133

0.760622

0.264684

0.352481

Skewness

1.790122

−0.067228

−1.172945

−0.055381

0.078896

Kurtosis

7.094751

2.144545

3.734748

1.978720

2.188641

2.770114

1.793403

Jarque-Bera

77.66099

1.968440

Probability

0.00000

0.373741

0.000359

0.250309

0.407913

63

63

63

63

63

Observations

15.86301

5.1. Findings of ADF Unit Root Test In this study, stability of the variables was tested in ADF unit root test of which fndings are as listed in Table 2. It is seen that null hypothesis about the acceptance of unit root for the entire series in Table 2 was verifed or in a different saying it is proven that variables are not stabile in fxed-trend level value. Null hypothesis claiming that once the frst difference of variables is removed then it bears unit root is rejected in a signifcance level of 1% thus all variables became stabile in I(1) their frst difference.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

70    Berna Kaçar and Huriye Gonca Diler Table 2.  Results for ADF Unit Root Test. Variables

ADF Test

Result

Level (Trend-Intercept) First Difference (Trend-Intercept) FO

−0.725407 (0.9663)

−5.893797 (0.0000)

I(1)

lnRGDP

−2.327045 ( 0.4130)

−5.858716 ( 0.0000)

I(1)

lnSUE

−3.008240 ( 0.1384)

−8.697635 ( 0.0000)

I(1)

lnYUFE

−0.276149 (0.9897)

−4.368518 ( 0.0049)

I(1)

lnGSSO

−3.469037 (0.0521)

−3.670867 (0.0322)

I(1)

Table 3.  Results for Zivot–Andrews Unit Root Test. Variables

T-Statistic

Result

FO

−3.170505 (0.006873)

No structural breaks

lnRGDP

−3.406802 (0.741144)

There is a structural break

lnSUE

−4.792697 (0.011005)

No structural breaks

lnYUFE

−2.534196 (0.007378)

No structural breaks

lnGSSO

−5.896048 (0.000983)

No structural breaks

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

5.2. Findings of Unit Root Test with Zivot–Andrews Structural Breakup Findings of Zivot–Andrews unit root test are illustrated in Table 3. Here RGDP variable has a structural breakup in a fxed-trend value. In the analyses based on structural breakups a number of instable series exhibit stabile properties. It is manifested that structural breakup in RGDP series did not signifcantly affect results of ADF test and the series remained to be I(1). Despite the presence of a structural breakup signals the fact that RGDP series that included unit root stayed the same.

5.3 Result of Toda–Yamamoto Causality Test As the frst stage of Toda–Yamamoto causality test in this research, standard VAR model was designated via applying level values of RGDP, GSSO and FO data. In determining VAR model, Akaike (AIC), Schwarz Bayesian (SBC), Hannan and Quinn (HQC) data criteria were focused in the selection of lag length. Since AIC and HQ information criteria signaled three lags, lag length of standard VAR model was also established as 3. Maximum lag length for RGDP, GSSO and FO variables is as depicted in Table 4.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Econometric Analysis 2003–2018    71 Table 4.  Maximum Lag Length for RGDP, GSSO and FO Variables. Critical Value (m)

AIC

SBC

HQ

0

7.478748

7.587249

7.520814

1

0.113453

0.547457*

0.281716

2

0.016330

0.775836

0.310789

3

−0.261364aa

0.823646

0.159292a

4

−0.171295

1.239218

0.375558

5

7.478748

7.587249

7.520814

a

Minimum critical value providing maximum lag length.

Table 5.  Results of Toda–Yamamoto Causality Test for RGDP, GSSO and FO Variables. Dependent Variable

MWald Test Statistic

Direction of Causality

RGDP

GSSO

FO

RGDP

–

17.73993 (0.0014)

8.343575 ( 0.0798)

GSSO  RGDP FO  RGDP

GSSO

8.493547 ( 0.0751)

–

22.31648 (0.0002)

RGDPGSSO FO GSSO

FO

6.049847 (0.1955)

4.784531 (0.3101)

–

RGDP FO GSSO FO

The statistics given in the table are χ2 values. Values in parentheses are P-probability values.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

a

Since in the second stage of Toda–Yamamoto casualty test the entire time series pertaining to the variables was I(1) integration level was designed as (dmax) 1. Actual lag length of standard VAR model equals to 3 thus by adding integration level (dmax) to the standard VAR model lag length was increased to 4. The new VAR model formed while lag length was changed had been predicted via Seemingly Unrelated Regression (SUR) method. The reason for applying SUR method is that equations related to the causality analysis estimated in the stage of model prediction took into account heteroscedasticity (changing variance) in error terms and correlation (autocorrelation) between error terms. Based on the explanations above, results of Toda–Yamamoto causality test for RGDP, GSSO and FO variables are as displayed in Table 5. According to the fndings in Table 5, interest channel of monetary transmission mechanism directly impacts fxed capital investment and RGDP in Turkey. In this study, a result supportive of this argument has been obtained. In this study, analysis of the second Toda–Yamamoto causality test belongs to SUE, YUFE and FO variables. These variables represent macroeconomic effects of monetary transmission mechanism’s interest channel from another perspective.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

72    Berna Kaçar and Huriye Gonca Diler Table 6.  Maximum Lag Length for SUE, YUFE and FO Variables. Critical Value (m) 0

AIC

SBC

HQ

3.848430

3.956931

3.890495 a

1

−4.127164

−3.693161

−3.958902

2

−4.397834

−3.638328

−4.103375a

3

−4.434860a

−3.349850

−4.014204

4

−4.260444

−2.849931

−3.713591

5

3.848430

3.956931

3.890495

a

Minimum critical value providing maximum lag length.

Table 7.  Results of Toda–Yamamoto Causality Test for SUE, YUFE and FO Variables. Dependent Variable

MWald Test Statistic

Direction of Causality

SUE

YUFE

FO

–

3.739860 (0.4424)

4.960764 (0.2913)

YUFE  SUE FO  SUE

YUFE

4.865773 (0.3013)

–

8.828681 (0.0655)

SUE  YUFE FO  YUFE

FO

8.246777 (0.0829)

10.67016 (0.0305)

–

SUE

a

SUE  FO YUFE FO

2

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

The statistics given in the table are χ values. Values in parentheses are P-probability values.

Standard VAR model was detected by employing level values of SUE, YUFE and FO variables. Once again in detecting VAR model, AIC, SBC and HQC data criteria were taken into account in selecting lag length. Since AIC and HQ data criteria signaled two lags, lag length of standard VAR model was measured as 2. Maximum lag length for SUE, YUFE and FO variables is as shown in Table 6. Similarly in the second stage of Toda–Yamamoto causality test conducted for SUE, YUFE and FO variables the entire time series was I(1); hence integration level (dmax) was set as 1. In the actual lag length of standard VAR model integration level (dmax) is similarly the equivalent of 1. By adding integration level (dmax) to the standard VAR model, lag length of SUE, YUFE and FO variables was risen to 3. Toda–Yamamoto causality test fndings for SUE, YUFE and FO variables are as manifested in Table 7. According to the fndings of Table 7, interest channel of the monetary transmission mechanism in Turkey did not provide a result that could affect YUFE and SUE. It has thus been suggested that the model supportive of the effect of interest rate on YUFE and SUE did not offer a result that backs up the main argument of this study.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Econometric Analysis 2003–2018    73

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

6. Conclusion In this study, the effects of functioning of interest channel of monetary transmission mechanism in Turkey were examined with respect to analyzed variables. The analysis was based on Toda–Yamamoto causality test. This chapter was divided into two parts by classifying the causality analysis as per the economic functions of employed macrovariables in the analysis. In the frst part, fndings of Toda–Yamamoto causality test were demonstrated in relation to RGDP, GSSO and FO variables. Based on these fndings it is argued that interest channel directly acted upon fxed capital investment and real gross domestic. Interest channel was found to be infuential on these variables used for the analysis. In the second part, the fndings of Toda–Yamamoto causality test were investigated in relation to YUFE and FO variables. Interest channel could not act upon YUFE and SUE. In the studies in the international literature, it is seen that functioning of interest channel varies according to the economic structures of the countries. In studies on monetary transmission mechanism in Turkey, has reached the conclusion that the interest rate channel of infuence on macroeconomic variables. In particular, Büyükakın, Bozkurt, and Cengiz (2009) have applied Granger causality and Toda– Yamamoto causality analyses together and have found that the interest channel was effective on GDP. In the study of Bozkurt and Cengiz (2009), the period 1990:1– 2007:9 was analyzed. In our study, the period of 2003Q1–2018Q3 was examined. Period and time series characteristics of the data differ from previous studies. In addition, structural breakage was also taken into account in the stability analysis of variables in our study. The result obtained after the analysis is consistent with the literature and the hypothesis of the study. This also shows that interest channel is the most powerful policy tool of the central bank. As can be understood from the studies, in every period of Turkey’s economy, interest has been a very effective policy tool. It is suggested that other than Toda–Yamamoto causality test used in this study the interaction of interest channel with macrovariables could be investigated by applying different econometric analysis methods. Since effects of money policy on real variables do not take place solely via interest channel, it is for certain that other channels could also potentially act upon macrovariables. At this point harnessing effective policy tools upon considering the factors impinging on the functioning of transmission mechanism channels is of dramatic signifcance in bolstering and sustaining an economy.

References Arabacı, Ö., & Baştürk, M. F. (2013). Evaluation of the ınterest rate channel of events 2001–2008 period in Turkey. Anadolu University Journal of Social Sciences, 13(2), 15–34.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

74    Berna Kaçar and Huriye Gonca Diler Bhuiyan, R. (2008). Monetary transmission mechanism ın a small open economy: A Bayesian structural VAR approach. Queen’s Economics Department Working Paper No. 1183, pp. 1–27. Bilan, O., & Kryshko, M. (2008). Does monetary policy transmission in Ukraine go through the ınterest rates? EERC Working Paper Series, pp. 1–56. Butzen, P., Fuss, C., & Vermeulen, P. (2001). The ınterest rate and credit channels ın Belgium: An ınvestigation with micro-level. ECB Working Paper 107, pp. 1–43. Büyükakın, F., Bozkurt, H., & Cengiz, V. (2009). Granger causality and Toda-Yamamoto methods of analysis of monetary transmission with channel in Turkey’s interest. Erciyes University Journal of Economics and Administrative Sciences, 33, 101–118. Cambazoğlu, B., & Güneş, S. (2011). Monetary transmission mechanism in Turkey and Argentina. International Journal of Economics and Finance Studies, 3(2), 23–33. CBT. (2012). Monetary and exchange rate policy for 2013. December 25, Ankara. Chapsa, X., Tabakis, N., & Athanasenas, L. A. (2018). Investigating the catching-up hypothesis using panel unit root tests: Evidence from the PIIGS. European Research Studies Journal, 21(1), 250–271. Chirinko, R. S., & Von Kalckreuth, U. (2003). On the German monetary transmission mechanism: Interest rate and credit channels for ınvestment spending. CESIFO Working Paper 838, pp. 1–49. Çiçek, M. (2005). Monetary transmission mechanism in Turkey: An analysis approach with VEC. Economics, Business and Finance, 20(233), 85–105. Doğan, B. (2012). Traditional transmission mechanism: The case of Turkey. Dumlupınar University Journal of Social Sciences, 3(33), 211–220. Erdoğan, S., & Yıldırım, D. Ç. (2009). Interest channel with monetary transmission mechanism in Turkey. Eskisehir Osmangazi University Journal of Economics and Administrative Science, 4(2), 57–72. Gitonga, M. V. (2015). Analysis of ınterest rate channel of monetary transmission mechanism in Kenya. International Journal of Business and Commerce, 4(4), 38–67. Guerra, J., Rodriguez, P. C., & Sänchez, G. (1998). The monetary policy transmission mechanism in Venezuela. BIS Policy Papers, 3, 223–246. Karaca, O. (2010). Monetary policy transmission mechanism and application to Turkey after the 2001 crisis. MS Thesis, Istanbul University, İstanbul. Kasapoğlu, Ö. (2007). Monetary transmission mechanisms: Application to Turkey. Expertise Thesis, CBRT General Directorate of Markets, Ankara. Mavrotas, G., & Kelly, R. (2001). Old wine in new bottle: Testing causality between savings and growth. The Manchester School Supplement, 69(1), 97–105. Mehrotra, A. N. (2006). Exchange and ınterest rate channels during a defationary era evidence from Japan, Hong Kong and China. Journal of Comparative Economics, 35(2007), 188–210. Mishkin, F. S. (1995). Symposium on the monetary transmission mechanism. Journal of Economic Perspectives, 9(4), 3–10. Mishra, P., Montiel, P., & Sengupta, R. (2016). Monetary transmission in developing countries: Evidence from India. In P. Montiel, P. Mishra, & R. Sengupta (Eds.), Monetary policy in India (pp. 59–110). New Delhi: Springer. Mohanty, D. (2012). Evidence on interest rate channel of monetary policy transmission in India. In Second international research conference at the Reserve Bank of India (pp. 1–52). Mumbai. Mukhtarov, S., & Aliyev, K. (2014). Parasal Aktarım Mekanizması Olarak Faiz Oranı Kanalının Etkinliği: Azerbaycan Uygulaması. Journal of Qafqaz University – Economics and Administration, 2(2), 111–118. Örnek, İ. (2009). Operation of monetary transmission mechanism channel in Turkey. Journal of Finance, 156(1), 104–125.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Econometric Analysis 2003–2018    75

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Peersman, G., & Smets, F. (2001). The monetary transmission mechanism in the Euro area: More evidence from var analysis. ECB Working Paper 91, pp. 1–37. Sarı, A. (2007). Through ınterest channel functioning of monetary transmission mechanism in Turkey. Academic Fener Balıkesir University Publications, 5(8), 15–21. Seyrek, İ., Duman, M., & Sarıkaya , M. (2004). Monetary transmission mechanism and monetary policy tool: Transmission mechanism in Turkey. Cumhuriyet University Journal of Economics and Administrative Sciences, 5(1), 201–212. Tarı, R. (2014). Econometrics (10th ed.). Kocaeli: Umuttepe Publications. Taylor, J. B. (1995). The monetary transmission mechanism: An empirical framework. Journal of Economic Perspectives, 9(4), 11–26. Thalassinos, I. E., Pintea, M., & Raţiu, I. P. (2015). The recent fnancial crisis and its impact on the performance indicators of selected countries during the crisis period: A reply. International Journal of Economics and Business Administration, 3(1), 3–20. Toda, H. Y., & Yamamoto, T. (1995). Statistical inference in vector autoregressions with possibly integrated processes. Journal of Econometrics, 66, 225–250. Turhan, İ. M., & Gümüş, N. (2014). On the relative importance of monetary transmission channels in Turkey. Retrieved from https://mpra.ub.uni-muenchen.de/69827/. Uğur, A., Sancar, C., & Polat, M. A. (2016). Empirical evidence on the functioning of the monetary transmission mechanism channel in Turkey: (1998–2015). International Journal of Eurasia Social Sciences, 7(22), 34–55. Wróbel, E., & Pawlowska, M. (2002). Monetary transmission in Poland: Some evidence on interest rate and credit channels. NBP Bureau of Macroeconomic Research Working Paper 24, pp. 1–28. Zivot, E., & Andrews, D. W. K. (1992). Further evidence on the great crash, the oil price shock, and the unit-root hypothesis. Journal of Business and Economic Statistics, 10, 251–270.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

This page intentionally left blank

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Chapter 6

A Conceptual Framework for Behavioral Accounting Ahmet Coşkun and Mehtap Karakoç Abstract The kinds of decisions people make or how they react to certain situations could differ according to the society, atmosphere or environment those people come from. Studies about the infuence of human behaviors on economics, business and actions were initiated by analyzing human behaviors and those studies carry on into behavioral fnance and behavioral accounting.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

In previous years, the models used were based on the assumption that people behave rationally while making decisions. These models lost validity recently and behavioral accounting started to search for the infuences affecting human behaviors. They started considering not only the people who prepare accounting data but also the people who take advantage of this data. People’s environment, cultural differences, psychological and sociological factors have entered into the accounting’s feld of interest as factors that have an infuence on behavior. The aim of this study is to try to analyze the theoretical bases and extent of behavioral accounting, which focuses on the human behavior factors being observed while creating or using fnancial reports. The authors also aim to contribute to the literature by including the neuroaccounting dimension into the analysis. Keywords: Accounting; behavioural accounting; behavioural economics; behavioural fnance; neuro accounting; neuro economics; accounting theories

1. Introduction Etatism resulted from the fact that there was scarcely any private sector in Turkey because of the undercapitalization in the early years of the Republic. Under the Uncertainty and Challenges in Contemporary Economic Behaviour, 77–88 Copyright © 2020 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-80043-095-220201007

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

78    Ahmet Coşkun and Mehtap Karakoç guidance of the State Economic Enterprise, modern accounting practices, which were more developed in Western countries in those years, were brought to Turkey. At the present time, at its simplest, accounting fulflls the functions of saving the operations and actions which could be expressed with money, classifying those operations and actions, and reporting them by summarizing and interpreting them. As a result of the diffculties brought about with technology and the developing era, the concept of making a proft, which is the primary purpose of the sectors trying to stand strong in the challenging competitive markets, is getting more diffcult by the day. In an effort to keep themselves strong in the markets, commercial sectors are turning to different and innovative accounting structures. Due to the daily changing conditions, one of the most affected felds is accounting. Even though accounting comprises objective data, because of the mathematical and numeral concepts it contains, it also comprises subjective data because of different points of view and interpretability. Although the members of this profession prove the accuracy of the gained data by the use of formulas, the reports also include subjective behaviors. In this context, the human element is the most indispensable factor in this phase. It is a fact that a person’s behaviors and psychological make-up are very important in the accounting reports which are interpreted and transferred to third parties. In the meantime, this proves that accounting is a sort of social science and its feld of interest includes judgments and subjective behaviors. Behavioral accounting examines the communication process between the accounting information system and its client. Behavioral accounting determines the attitudes and styles that the person in the decision-making position would want to realize. It is not only traditional accounting reports that affect the decisions of third parties who want to have information about the company. Behavioral accounting also examines the external factors and events of accounting by affecting the relevant people and institutions (Ateş, Çivi, & Öz, 2016, p. 26; Baldacchino, Tabone, Micallef, & Grima, 2018). The fact that the outputs obtained by the managers according to their accounting policies are different or that the managers make different decisions when faced with the same information proves that the cognitive characteristics of the person are an effective factor in the decision phase. Behavioral accounting analyzes the communication process between the accounting information system and its receiver. Behavioral accounting makes the attitudes and manners of the individuals who are about to make decisions more explicit. Since it is not only traditional accounting reports that infuence the decisions of third parties, behavioral accounting analyzes factors and actions that infuence people and institutions related to accounting (Ateş et al., 2016, p. 26).

2. Development of Behavioral Accounting Behavioral accounting is a brand new concept in the feld of accounting, which studies accountants and all of their responses. It analyzes the behaviors of accountants and the people who use accounting data. The concept of behavioral accounting appeared for the frst time in the 1960s, as a newly emerging accounting feld which was brought into the literature after several studies. The offcial

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Conceptual Framework for Behavioral Accounting    79 terminological use of behavioral accounting started with the Journal of Accounting Research in 1967 (Kutluk, 2010, p. 7). The behaviors of the people might result from accounting practices and it might infuence the decisions of said people. The common objective in accounting research is to learn how individuals make information effective. Research about a person’s behavior might have the same function (Kutluk, 2010, p. 4). Moreover, behavioral accounting studies the behaviors which an individual displays in any accounting action. Behavioral accounting does not merely comprise of accounting sciences but is also included in both research and concept. Therefore, in the literature, it is presented as a term blended within sociology and psychology (Gökten, 2017, p. 347). People might have different ideas and cultural features according to the environment they have been socialized in. Naturally, this causes differences in human behavior. In this context, we focus on the individual, analyzing what the human behavior will contribute to the behavioral accounting research (Gönen & Rasgen, 2017, p. 190). Behavioral accounting has been developed in order to make up for the defciency of traditional accounting. Since traditional accounting is not effcient enough to refect the ideas and decisions of third parties using the reports, behavioral accounting emerged so as to meet this defcit (Ateş et al., 2016, p. 26). Whereas traditional accounting is closely related to economics, behavioral accounting is engaged in psychology and sociology (Kayışyapar, 2019, p. 23).

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

3. The Factors Infuencing the Behavioral Dimension of Accounting People are obliged to make a decision in order to reach a conclusion in the accounting action they encounter. Following some studies, researchers claimed that some factors infuence those decision mechanisms. The factors infuencing the behavioral dimension of accounting can be classifed into four groups, such as (1) belief, (2) culture, (3) political structure and (4) ethical understanding (Kayışyapar, 2019, p. 27). Below, these groups are examined in more detail: (1)  Belief: In a person’s learning behavior, from childhood to adulthood, an individual takes what he learns and makes it causal (part of everyday life) and saves it. That learnt and saved information generates certain beliefs which are refected in societies. Belief has an important role in an individual’s decisionmaking process. The infuence of beliefs on decision-making also infuences decisions in accounting (Gönen & Rasgen, 2017, p. 192). (2)  Culture: This is the sum of traditions, customs and beliefs which enables the transferring of behavior and problem-solving behavior from one generation to another (Kayışyapar, 2019, p. 27). ⦁⦁ Social culture: The fact that accounting practices interact with the culture

of the society; this in turn forms cultural values in accounting. Those subcultural values are (Kayışyapar, 2019, p. 28):

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

80    Ahmet Coşkun and Mehtap Karakoç – – – –

Status quo against professionalism. Flexibility against monotony. Optimism against conservatism. Transparency against privacy.

Accounting principles and standards specify a reliability level of accounting knowledge. This specifed reliability level might not be the same among every country due to cultural differences. However, in the globalized world, foreign investors consider the relevancy of this reliability level to international standards as important. Thereby, outer factors could require the reorganization to a reliable level. New reorganizations might need to be relevant to the society in which we are (Kayışyapar, 2019, p. 28; Morais Pereira & Candeias Bonito Filipe, 2018).

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

⦁⦁ Organizational culture: Organizational culture is defned as the total of the

values, norms, beliefs and understandings which are shared among members of a society. Decisions are made by the organization and the decision made by each organization could be different from the others. Thus, an accounting system specifc to the organization could develop (Usul, 2007, p. 36). The function of the organization could become distinct, according to the organization. Some organizations consider accounting only as recordkeeping, while some other organizations consider the function of accounting as going beyond (Kayışyapar, 2019, p. 29). ⦁⦁ Spiritual culture: Societies might have various beliefs and spiritual principles. Those two factors which create a spiritual culture can have an effect on the building of the accounting system (Kayışyapar, 2019, p. 29). Islam has defnite judgments for commercial affairs. If most of the society belongs to Islam, then the accounting system gets shaped accordingly because, while people obey the principles of Islam, they also expect that the accounting system should comply with those principles (Kayışyapar, 2019, p. 29). In other Abrahamic religions, there are no judgments as defnite as in Islam. That fact has no infuence in their accounting systems (Kayışyapar, 2019, p. 29). (3)  Political structure: The bureaucratic process of a country could affect the accounting system of the country. Whether the country has a centralization system or not could lead to the development of different accounting systems (Kutluk, 2010, p. 10). While a country is forming its own accounting system, it should consider political and bureaucratic matters, international trade dimension and sensitivities of the social structure in the country. The important points to build an accounting system are (Usul, 2007, p. 47): – – – –

Intentness of Intentness of Intentness of Intentness of

the management. the parliament. the political parties. the society which will use accounting outcomes.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Conceptual Framework for Behavioral Accounting    81 (4)  Ethical understanding: Ethics comprise studies which are based on moral problems and choices related to those problems. There is an ethical issue behind each decision made in the feld of business and social environment. Therefore, accounting education and ethics education could appear blended. From time to time, an accountant could encounter an ethical dilemma. This dilemma could emerge at the point where his duties and responsibilities get into a confict with his own interests. Within this dilemma, ethical development studies of Piaget and Kohlberg give guidance (Şahin & Akın, 2018, p. 398). The role of ethics in accounting could be different, according to the person. The fact that the person considers the ethical and cultural structure of the society while he is fulflling the accounting function, could decrease the opposite side’s risk of being deceived (Kayışyapar, 2019, p. 31).

4. The Theories Used in Behavioral Accounting The theories used in behavioral accounting and the ones in a close relationship with behavioral accounting are analyzed below:

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

4.1. Expectation Theory Expectation Theory generated by Victor Vroom in 1964 was further developed by Porter and Lawler in 1968. This theory explains that the primary factor motivating the individual is the expectations as a result of the individuals’ behaviors (Marşap & Gökten, 2016, p. 350). As a concept, the word “expectation” expresses attaching the result of a behavior to a reward. High expectation of the individual is linear to the reward expectation and could motivate the person for work. The reward might be a high salary. Moreover, showing high performance to reach the result might enhance a person’s concentration for work (Kutluk, 2010, p. 12). The member of the profession, one of the parts in behavioral accounting, fulflls his fundamental duty to transfer data and report in a nominative approach. They also use the positive accounting approach for the reports, showing the summary of the operations of a term by making predictions and changes in the policies in accordance with IFRS, especially at the end of terms. At this point, the decision on whether to make changes in the direction of one’s own interpretation is related to behavioral accounting. The predictions are directly proportionate to experiences. For instance, public volume of a machine which will be acquired in the future is a predictable situation by analyzing public volume of a machine which was acquired in the past. Thus, the acquisition of an asset is related to expectation theory. The ideas of taking maximum advantage from the asset to be acquired and protecting its presence are expectations. To be more precise, a business always expects an income from an asset (Marşap & Gökten, 2016, p. 350).

4.2. Equality Theory Another effective theory about labor productivity, which also affects motivation, is equality theory. There are several determinants in the practice of equality

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

82    Ahmet Coşkun and Mehtap Karakoç theory. Some of them are the skills of the employee, how much time he spends at work and academic development. The proftability ratio of the company and the utility level of the employee must be in harmony. According to equality theory, when an employee cannot fnd the utility he wants, then his motivation will get lower (Gönen & Rasgen, 2017, p. 193). Employees might be comparing their performance with that of other employees rather than focusing on the utility of his own effort. After comparing his effort with that of his colleagues, if he sees that there is a kind of inequality, this could give rise to some negative consequences leading to low motivation. Employees could show a change in effort or in their work and this could lead to them quitting their job. Bullying could even take place, which can reduce the productivity of the other staff (Kutluk, 2010, p. 13).

4.3. Goal Setting Theory Goals could be directly related to people’s behavior. In this situation, there should not be any confict between the goals of the staff and those of the foundation. Otherwise, the behaviors of the shareholders could get shaped accordingly (Gönen & Rasgen, 2017, p. 193). According to the theory developed by Locke in 1968, there is a frm relation between the acquirability level of the created goal and the motivation of the person. As an example, the effort of the staff with an easy goal and the effort of the staff with a challenging goal might not be the same and the one with an easy goal might have lower motivation (Kayışyapar, 2019, p. 33). While the working individual prepares accounting reports and operations in accordance with his own goals, he should also consider the goals of the foundation. This theory aims to see that the goals of the member of the accounting profession match up with the goals of the foundation.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

4.4. Contingency Approach People can show different behaviors according to the conditions they are in. This theory suggests that determined rules should be fexible and changeable and it objects to a structure with strict rules (Marşap & Gökten, 2016, p. 351). At the present time, there is a continuous change. The present status and the conditions in this status can be refreshed constantly. Societies and people should adapt themselves to this rapid change. Within this change and development, the contingency approach claims that set up rules do not have the quality to keep their validity permanently. For accounting systems, the contingency approach specifes that certain strategies cannot be applied to all situations. Contingency approach explains that all operations will be different according to the present conditions. Therefore, it supports that more fexible and changeable policies should be constituted rather than strict and frm ones. According to the contingency approach, organizational structure undergoes a change as technology and environment change (Kutluk, 2010, p. 14).

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Conceptual Framework for Behavioral Accounting    83 4.5. Agency Theory We can classify the environments of the foundations as intrabusiness and nonoperating. Under those two roofs, there are many parts such as managers, employees, investors and lenders. There emerges a confict since the interests and expectations of all those involved are in line with their positions and expectations. Herein, agency theory gets involved. Agency Theory, which has been on the agenda since the 1970s, got its current status after the new studies carried out by Jensen and Meckling in 1976. It is both used by fnancial sciences and behavioral sciences nowadays. According to this theory, different people with different agendas will try to put forward their interests which might confict with those of others in different positions. Agency theory is comprised of two parts, namely, the principal and the agent. Here, the interests of both parts should be considered (Marşap & Gökten, 2016, p. 352). In a situation, a deal could be made based on two determinants: behavior and outcome. The price of a deal made according to behavior is higher than the price of a deal made according to outcome most of the time. If the principal makes a deal with the agent according to behavior, then an increase in the agent’s motivation can be observed, because making a deal according to behavior means focusing on agent’s effort (Kutluk, 2010, p. 15). The behaviors of people are exhibited through theories. As seen, the fnal shape of the fnancial reporting will be determined according to the behaviors of the accountant brought through the social environment (Gönen & Rasgen, 2017, p. 194).

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

5. Behavioral Accounting Research in the World and Turkey The need for accounting operations and their outputs in every feld of fnancial operations makes accounting universal and carries the science of accounting to every corner of the world. On the other hand, behavioral accounting reveals that accounting is not only based on numbers but also based on human factors. The researches that have made important contributions to the behavioral accounting feld in the world and Turkey are reviewed below. Libby and Luft (1993) examined the contribution of individuals’ cognitive processes to decision-making. At the same time, it was revealed that the decisions of the accountants are infuenced by the level of their skills, knowledge and motivation. Zhang et al. (2001) designed a control game in their studies. In this game, the aim was to search for the diffculties in the process of guessing the possible expense of a case. In the study, two trials were made and the results were refected in a report. Trials comprised of six sessions. As anticipated from the research, it was noted that a case expense was predicted at a much higher level under the infuence of cognitive restrictions. Arnold, Bernardi, Neidermeyer, and Schmee (2006) analyzed how cultural differences among European countries refect to accounting. Two hundred and ninety-four people from eight countries were included into the research.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

84    Ahmet Coşkun and Mehtap Karakoç According to the study, it was found that there is cultural harmony among the employees in the frm and its employment. In the study conducted by Chin and Chi (2008) about behavioral differences of men and women toward accounting, it was observed that men take more risks and women obey ethical rules during accounting operations more than men. In the study by Chung, Cohen, and Monroe (2008), the refection of individuals’ different cognitive features to accounting was analyzed. One hundred and two audit employees were included into the study. According to the research, it was concluded that the employee in a low mood could only reach a limited outlook. Dinç, Abdioğlu, and Büyükşalvarcı (2010) studied the effect of performative acts between accountants and their respondents to earning levels. According to the study which included 2,228 people, it resulted that the relation between accountants and their respondents infuence the level of earning to a low level. Ateş et al. (2016) measured the level of emotional intelligence in accountants by taking behavioral accounting as a base. It was argued that the fact that accountants deal with a lot with numbers, this should not necessarily blind their emotional behaviors. It was found that accountants were good at emotional analyses in the study carried out with 235 people. It was emphasized that emotional control and staying calm are very important in the relation between the accountant and their respondents.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

6. Neuroaccounting: A Research Field in the Way of Behavioral Accounting The term of homo economicus, basic assumption of classic and neo-classic economy, addresses the people who try to obtain maximum beneft with scarce resources and whom are totally assumed to show rational behaviors. However, it is not possible to claim that all people, social creatures who are in interaction with the environment, can show rational behaviors. There are many factors which lead human behaviors. It is not possible that each individual can show the same reaction in every occasion. Sociologists and psychologists have contributed to the development of behavioral dimensions of economy, fnance and accounting by criticizing the rational behavior assumption in the relevant felds (Marşap & Gökten, 2016, p. 346). Neuroaccounting is a feld which combines behavioral accounting, accounting and neuroeconomics. This feld is still being researched, however, the starting point of neuroaccounting is neuroeconomics which carried the techniques of neuroscience to its own feld frst. Neuroeconomics uses the measurement techniques of neuroscience in order to understand the effect of neural networks on economical decisions (Zak, 2004, p. 1737). The experimental studies made in the feld of neuroeconomics encouraged accounting researches to make interdisciplinary studies with neuroscience (Usul & Çağlan, 2018, p. 452). The reason why behavioral accounting research has moved toward neuroscience recently is the fact that the situations which used to be considered as behavioral is now associated with the neurological system. Models related to how the human brain works during a behavior or a decision are generated. The brain is

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Conceptual Framework for Behavioral Accounting    85 considered as a black box and studies are conducted to understand how this black box works (Birnberg & Ganguly, 2012, pp. 1–8; Marşap & Gökten, 2016, p. 353) Neuroaccounting emerged as a discipline which provides a better understanding for the human decision-making process. This discipline analyzes the neurological bases of mental condition of people while they are making decisions (Uslu & Çağlan, 2018, p. 452). Neuroaccounting is the sub-unit of accounting which approaches accounting through the role of the brain in the formation of economic practices and traditions (Ahmad, 2010, pp. 1–3). Neuroaccounting is specifcally designed to understand the role of accounting numbers, interpreting accounting reports and the decision-making process through the theoretical models from cognitive and behavioral neuroscience. It focuses on the potential relationship between decision-making processes in accounting and what is happening in the black box (Baldi, 2017, p. 86). Like neuroeconomics, neuroaccounting aims to determine how neurons interact with each other during the accounting operations. The research focuses on how to determine which neurons are active, the repetitions of different or the same stimulus and how activation rates change as lab conditions change (Dickhaut, Basu, McCabe, & Waymire, 2010, p. 224). Neuroscientists use some methods to gather neural activation data. However, there are some distinctions about the classifcations of those methods in the feld of neuroaccounting. Dickhaut et al. (2010) classifed those methods as Functional Magnetic Resonance Imaging (fMRI) and single cell recording. On the other hand, Birnberg and Ganguly (2012) classifed the methods as intervention required and intervention not required (Uslu & Çağlan, 2018, p. 458). The application of these methods, which are named as psychophysiological techniques in neuromarketing studies, is based on measuring the reaction of the person to any stimulus (Şenel & Darıcı, 2018, p. 386). Some psychophysiological techniques include fMRI, positron emission tomography (PET), electro encephalo graphy (EEG), galvanic skin response (GSR), steady state typography (SST), eye-tracking, heart rate (HR) and Facial Coding (FaCs) (Bercea, 2012; Boz, Yılmaz, & Koç, 2016; Carter & Shieh, 2015). Based on the principle that active regions in brain need more blood, fMRI measures the changes in brain blood fow. By registering radiation spreading from positron emissions, coming from applied radioactive substance, physiological images are taken with PET. EEG measures the electrical activities between neurons simultaneously. GSR is a method which measures the sense of arousal in a sensitive way. SST is a method which records responses of brain activities to a stimulus through instant response measurement. With the eye-tracking method, the topics on which the subjects focus and their concentration duration are measured through eye movements. HR is a method where arousal levels are measured through changing pulse rates. FaCs is a method to measure the changes in emotions by scanning micro expressions on people’s faces. These methods which are used in neuroscience and neuroaccounting studies take the photos of the active regions in the brain at certain times (Birnberg & Ganguly, 2012, p. 8). Neuroaccounting is still a new term and there have been few studies about this. This sub-discipline, like the other disciplines which have taken the prefx neuro-,

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

86    Ahmet Coşkun and Mehtap Karakoç will be studied for many years. The leading studies which were carried out in this feld are explained below. Basu and Waymire (2006), in their studies about the role of systematic recording, mentioned the structure of the human brain and also its relation with recording logic in accounting. Dickhaut, Basu, McCabe, and Waymire (2009), on the other hand, studied the structural parallelism between the brain and accounting principles. Dickhaut et al. (2010) analyzed neuroscientifc connections of accounting principles in their studies. Moreover, they discussed the relation between neuroscience and accounting by offering methods from neuroscientifc researches to contribute to behavioral accounting researches. Birnberg and Ganguly (2012) discussed the neuroeconomical topics from which behavioral researchers can beneft. Just like all neurosciences, there are some criticisms about neuroaccounting. A researcher who wants to make a study about neuroaccounting must incur a high fnancial research cost and get support from expert researchers about neurology (Birnberg & Ganguly, 2012, p. 8). Additionally, in neuroaccounting studies, some factors such as ethical concerns about researches, sincerities of the subjects involved, the objectivity of the results’ interpretations, the adequacy of the interpreters, need to be taken into consideration. Furthermore, making a correct prediction about behaviors, about the outcomes of human brain defned as black box, is only possible if the input is known. While neuromarketing enlighten the black box, it ignores many other inputs. It cannot be stated that the dark sides of the brain are enlightened evenly as the number of the neuroscientifc studies are increasing (Morin, 2011, p. 133). The new theoretical studies in the feld of neuroaccounting will be helpful to fnd answers for the criticisms directed at this new feld.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

7. Conclusion A changing world brings with it new research felds and new sub-disciplines. This reality is also valid for accounting. Accounting, which is a science that deals with objective facts, comes up against the human factor in the analysis and interpretation of the data it produces. Considering that not all of the people can show rational behaviors, it is known that cognitive features have an effect on people’s decision-making processes. This concept, named as behavioral accounting, analyzes the behaviors which individuals encounter during any accounting operation. However, there has been an inclination toward neuroscience since it was determined that some situations considered as behavioral are actually cognitive. In neuroscience, when the process initiated with neuroeconomics encountered some other sub-disciplines later on, it resulted in the emergence of the concept of neuroaccounting. This sub-discipline which studies the effective factors during decision-making, shows the “balance” of the human brain by using the techniques of neuroscience. Whereas some argue that this new discipline has the potential to change the future of accounting, some provide serious criticisms about this feld. The physiological changes created by accounting education or the steps which people take in order to reach the profession underlying their own genetic codes are signifcant

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Conceptual Framework for Behavioral Accounting    87 for the profession of accounting. Moreover, increasing knowledge about human brain will have a signifcant potential for the future in spite of the criticisms about neuroaccounting in the literature.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

References Ahmad, Z. A. (2010). Brain in business: The economics of neuroscience. The Malaysian Journal of Medical Sciences, 17(2), 1–3. Arnold, D. F., Bernardi, R. A., Neidermeyer, P. E., & Schmee, J. (2006). The effect of country and culture on perceptions of appropriate ethical actions prescribed by codes of conduct: A western European perspective among accountants. Journal of Business Ethics, 70(4), 327–340. Ateş, B. A., Çivi, G., & Öz, M. (2016). Davranışsal Muhasebe Bağlamında Muhasebe Meslek Mensuplarında Duygusal Zeka. Sosyal Ekonomik Araştırmalar, 16, 24–53. Baldacchino, P. J., Tabone, N., Micallef, G. L., & Grima, S. (2018). Drivers and drawbacks of an external auditing career. International Journal of Economics & Business Administration, 6(3), 21–46. Baldi, R. (2017). Decision making and neuroaccounting perspective: An eye-tracking ınvestigation on accounting ınformation disclosure. Research. Retrieved from https://etd.adm.unipi.it/t/etd-02052017-190730/. Accessed on July 30, 2019. Basu, S., & Waymire, G. (2006). Recordkeeping and human evolution. Accounting Horizons, 20(3), 1–29. Bercea, M. D. (2012). Anatomy of methodologies for measuring consumer behavior in neuromarketing research. Retrieved from http://www.lcbr-online.com/index_fles/ proceedingsemc12/12emc023.pdf. Accessed on August 1, 2019. Birnberg, J. G., & Ganguly, A. R. (2012). Is neuroaccounting waiting in the wings? An essay. Accounting, Organizations and Society, 37, 1–13. Boz, H., Yılmaz, Ö., & Koç, E. (2016). How to apply neuromarketing tools to business researches? International Human and Nature Sciences: Problems and Solutions Seeking Congress, October 7–9, Sarajevo. Carter, M., & Shieh, J. C. (2015). Guide to research techniques in neuroscience. Burlington, MA: Academic Press. Chin, C.-I., & Chi, H.-Y. (2008). Gender differences in audit quality. American Accounting Association Annual Meeting. Chung, J., Cohen, J., & Monroe, G. S. (2008). The effect of moods on auditors’ ınventory valuation decisions. Auditing: A Journal of Practice & Theory, 27(2), 137–159. Dickhaut, J. W., Basu, S., McCabe, K. A., & Waymire, G. B. (2009). Supplement to neuroaccounting: Consilience between the biologically evolved brain and culturally evolved accounting principles. Working Paper, Dickhaut, J. W., Basu, S., McCabe, K. A., & Waymire, G. B. (2010). Neuroaccounting: Consilience between the biologically evolved brain and culturally evolved accounting principles. Accounting Horizons, 24(2), 221–255. Dinç, E., Abdioğlu, H., & Büyükşalvarcı, A. (2010), Muhasebe Meslek Mensuplarının Davranışsal Değişkenlerinin Kazanç Düzeyi Üzerindeki Etkisini Belirlemeye Yönelik Görgül Bir Araştırma. Ege Akademik Bakış, 10, 1154–1164. Gökten, P. O. (2017). Baumeister’ın Ego Tükenmesi Teorisi Çerçevesinde Denetçi Yargısı. İşletme Araştırmaları Dergisi, 9, 901–911. Gönen, S., & Rasgen, M. (2017). Davranışsal Muhasebe Alanını İçeren Araştırmalara Yönelik Literatür Taraması. İnsan ve Toplum Bilimleri Araştırmaları, 6, 189–198.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

88    Ahmet Coşkun and Mehtap Karakoç

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Kayışyapar, İ. (2019). Davranışsal Muhasebe Çerçevesinde Muhasebe Meslek Mensuplarında İş Stresi (Erzurum İlinde Bir Araştırma). Unpublished Master’s thesis, Social Science Instıtute, Erzincan Binali Yıldırım University. Kutluk, F. A. (2010). Davranışsal Muhasebe Çerçevesinde Etik Konusunun Değerlendirilmesi. Unpublished PhD thesis, Social Science Instıtute, Akdeniz University. Libby, R., & Luft, J. (1993). Determınants of judgment performance ın accountıng setı’ıngs: Abılıty, knowledge, motıvatıon, and envıronment. Accounting Organizations and Society, 18, 425–450. Marşap, B., & Gökten, P. O. (2016). Davranışsal Muhasebe: Kuramsal Yaklaşım. İşletme Araştırmaları, 8, 345–359. Morais Pereira, M. V., & Candeias Bonito Filipe, A. J. (2018). Quality of board members’ training and bank fnancial performance: Evidence from Portugal. International Journal of Economics & Business Administration, 6(3), 47–79. Morin, C. (2011). Neuromarketing: The new science of consumer behavior. Society, 48(2), 131–135. Şahin, Ş., & Akın, O. (2018). Muhasebe Meslek Mensuplarının Ve Muhasebe Dersi Alan Öğrencilerin Ahlaki Gelişim Ve Etik İle İlgili Tutumlarının Karşılaştırmalı Olarak İncelenmesi. Sosyal Bilimler Enstitüsü Dergisi, 10, 397–417. Şenel, S. A., & Darıcı, S. (2018). Psikofzyolojik Tekniklerin Adli Muhasebe Alanında ve Suçun Ortaya Çıkarılmasında Kullanılmasına İlişkin Bir Tartışma: Nöromuhasebe Kavramına Farklı Bir Bakış. Akademik Sosyal Araştırmalar Dergisi, 6(67), 381–391. Usul, H. (2007). Davranışsal Muhasebe. Ankara: Asil Yayın Dağıtım. Usul, H. & Çağlan, E. (2018). Nöromuhasebe. Muhasebe Bilim Dünyası Dergisi, 20(2), 450–465. Zak, P. J., (2004). Neuroeconomics. Philosophical Transactions of the Royal Society B: Biological Sciences, 359, 1451, 1737–1748. Zhang, P., Church B. K., & Ackert, L. F. (2001). Uncertain Litigation Cost and Seller Behavior: Evidence From an Auditing Game. Advances in Accounting, 18, 77–104.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Chapter 7

Theories of Financial Inclusion Peterson K. Ozili Abstract Purpose: This purpose of this chapter is to present several theories of fnancial inclusion. Financial inclusion is the ease of access to, and the availability of, basic fnancial services to all members of the population. Financial inclusion means that individuals and businesses have access to useful and affordable fnancial products and services that meet their needs in a responsible and sustainable way. Financial inclusion practices vary from country to country, and there is need to identify the underlying principles or propositions that can explain the observed variation in fnancial inclusion practices. These set of principles or propositions are called theories. Methods: The chapter uses conceptual discussions to formulate alternative theories of fnancial inclusion.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Findings: The study shows that fnancial inclusion theories are explanations for observed fnancial inclusion practices. It also shows that the ideas and perspectives on fnancial inclusion can be grouped into theories to facilitate meaningful discussions in the literature. Originality/value: Currently, there are no observed or elaborate theories of fnancial inclusion in the policy or academic literature. This chapter is the frst attempt to develop theories of fnancial inclusion. The theories are intended to be useful to researchers, academics and practitioners. The resulting contributions to theory development are useful to the problemsolving process in the global fnancial inclusion agenda. Keywords: Financial inclusion; theory; fnancial access; access to fnance; dissatisfaction theory; vulnerable group; systems theory; community echelon; public service; special agent; JEL classifcation: G00; G21; O16

Uncertainty and Challenges in Contemporary Economic Behaviour, 89–115 Copyright © 2020 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-80043-095-220201008

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

90    Peterson K. Ozili

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

1. Introduction I seize this occasion to break the silence among those of us that have observed the recent trend in fnancial inclusion practice particularly the cross country differences in fnancial inclusion practice and are worried that the modern trend in fnancial inclusion practice lacks an underlying theory or a set of guiding principles that can help to improve our understanding of fnancial inclusion as a prodevelopment initiative in the economics and fnance discipline. As new academics emerge and are eager to understand fnancial inclusion in a fast changing world where sophisticated fnancial innovations have increased rapidly in the formal fnancial sector, it is needful to remind ourselves that the fnancial inclusion literature may become porous and disconnected if there are no set of coherent theories of fnancial inclusion to prescribe some general principles that explain fnancial inclusion practice. Today, there are many studies on fnancial inclusion and I am immensely thrilled by the interesting fndings of empirical research in fnancial inclusion especially policy research and academic research (Demirguc-Kunt, Klapper, & Singer, 2017). But the notable lack of synergy between the policy and academic literatures (Prabhakar, 2019), and the neglect of the role of theory in fnancial inclusion debates and research is worrisome. My frst response to this realization was to ignore it, hoping that some good might come of it. But it is diffcult to ignore the lack of theory, because in the fnancial inclusion literature, ideas have become texts, and opinions have become stylized facts in the literature. In fact, proponents of fnancial inclusion do not challenge the current fnancial innovation products and systems used in achieving fnancial inclusion because they do not want be seen as being anti-innovation or anti-development. Currently, there are no observed or elaborate theories of fnancial inclusion in the policy or academic literature. In fact, I have heard colleagues in top policymaking institutions make statements like “we don’t need a theory of fnancial inclusion.” Some feel that “building a theory of fnancial inclusion is a waste of time.” Others think that “theorizing fnancial inclusion would lead to lack of relevance to practitioners and policymakers.” In sum, the general conception is that it is better to focus on the accumulation of evidence on fnancial inclusion than concentrating on building theories of fnancial inclusion, and even when some fnancial inclusion theories emerge, there will be contentions about what is a theory and what is not a theory, whether we should have a single unifying theory of fnancial inclusion or multiple theories of fnancial inclusion, whether we should rely on old or new theories, and there will be arguments about the degree of abstraction embedded in old and new theories. Despite the disagreements on the value of theory in the fnancial inclusion debate, we can at least agree that we need a framework or a set of principles to help us understand what fnancial inclusion is, how it is achieved and who benefts from fnancial inclusion. These principles are called theories. In this chapter, I present several theories of fnancial inclusion that can be useful to academics and practitioners. These theories are divided into theories of fnancial inclusion benefciaries, theories of fnancial inclusion funding, and

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Theories of Financial Inclusion    91 theories of fnancial inclusion delivery.1 Of course, no theory is perfect and I have been careful to reduce the degree of abstraction in the theories in order to increase the relevance of these theories to both academics, practitioners and policymakers. Each of the theories discussed in this chapter can be expanded into large volumes of texts to take into account a wide range of philosophical dimensions and critical perspectives of the theory but, due to space constraints, I have presented an abridged and concise version of each of the theories and I would gladly welcome any requests to expand each of the theories to be published as a single theoretical paper. I hope the reader would fnd the theories useful for intellectual discourse for future research in fnancial inclusion. While the aim of the chapter is not to critique studies that lack theoretical perspectives, I will make reference to such studies for illustration purposes. This chapter contributes to the fnancial inclusion literature. By articulating a general theory or theories of fnancial inclusion, I present a new and comprehensive statement of what the fnancial inclusion enterprise should be about. I argue that the fnancial inclusion literature – whether academic or policy – can use theories to provide believable explanations for fnancial inclusion objectives and outcomes. Policymakers can also use these theories to justify the various strategies they adopt in achieving fnancial inclusion. The remainder of the chapter is organized as follows. Section 2 presents some foundations for fnancial inclusion. Section 3 discusses the theories that explain who benefts from fnancial inclusion. Section 4 discusses the theories that explain the agent responsible to deliver fnancial inclusion to the benefciaries. Section 5 discusses the theories on how fnancial inclusion is funded. Section 6 concludes.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

2. Some Foundations Financial inclusion is the provision of access to fnancial services to all members of population particularly the poor and the other excluded members of the population (Ozili, 2018). Financial inclusion can also be defned as the delivery of banking services at an affordable cost to the vast sections of the disadvantaged and low-income groups (Dev, 2006). Financial inclusion is also defned as the use of, and access to, formal fnancial services (Sahay et al., 2015). These defnitions have one thing in common which is that they emphasize that each member of the population should have access to available fnancial services. Financial inclusion will bring the excluded population into the formal fnancial sector so that they can have access to formal fnancial products and services (Allen, Demirguc-Kunt, Klapper, & Martinez Peria, 2016). Financial inclusion has been a major policy objective for the government of many developing and emerging countries. Many governments are making tremendous efforts to achieve high levels of fnancial inclusion for the beneft of their citizens. There are many success stories on fnancial inclusion around the 1

These theories may be also categorized into demand-side theories and supply-side theories from an economic perspective.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

92    Peterson K. Ozili World such as India (Nimbrayan, Tanwar, & Tripathi, 2018), Rwanda (Lichtenstein, 2018; Otioma, Madureira, & Martinez, 2019), Kenya (Hove & Dubus, 2019; Ndung’u, 2018) and Peru (Cámara & Tuesta, 2015). In India, the PMJDY scheme improved the level of fnancial inclusion for many of its citizens and became a big success in the early years.2 The later years of the PMJDY scheme witnessed some supply-side challenges such as low supply incentives and low subsidy to providers of fnancial services.3 In Rwanda, community savings and credit cooperatives have been a fnancial inclusion success story. In just three years, these cooperatives, known as Umurenge SACCOs, have attracted over 1.6 million customers and 90% of Rwandans now live within a 5 km radius of an Umurenge SACCO.4 In Kenya, the M-Pesa has been the primary instrument to achieve greater fnancial inclusion for the Kenyan people. The introduction of M-Pesa in 2007 profoundly transformed Kenya’s fnancial system. Through the M-Pesa, the level of fnancial inclusion rose from 26.4% in 2006 to 40.5% in 2009 and the level of fnancial exclusion declined from 39.3% to 33%.5 In Peru, an interoperable mobile money platform called “Modelo Peru” was launched that focuses on bringing mobile fnancial services to those who need it, with the aim of promoting fnancial inclusion. There is also the “Bim” service in Peru that enables any Peruvian with a mobile phone to open a bank account and make payments and this can be done without ever having to visit a bank.6 There is no doubt that the fnancial inclusion programs and policies adopted in some countries have been successful. Yet, the two major concerns that often arise is the concern that fnancial inclusion may spread the risks of the fnancial system to the poor and vulnerable customers in the society and increase the number of high-end (or high-income) consumers benefting from fnancial institutions. The second concern is whether fnancial inclusion should be targeted to those who have never been included in the formal fnancial sector or to those who have been relatively distant to using fnancial products more and more frequently (Liapis, Rovolis, Galanos, & Thalassinos, 2013). The policy literature contains many idealistic interpretations on how to achieve fnancial inclusion while the academic literature is mostly focused on the relationship between fnancial inclusion and poverty levels and income inequality as well as the effect of fnancial inclusion on the economy (Cull, Cull, DemirgüçKunt, & Morduch, 2013; Demirgüç-Kunt & Klapper, 2013; Mehrotra & Yetman, 2015; Morgan & Pontines, 2014; Sarma & Pais, 2011). These two literatures are quite interesting even though there is no synergy between the academic and policy

2

https://www.pmjdy.gov.in/fles/stories/Central-Bank-of-India.pdf https://www.cgap.org/blog/pmjdy-improved-fnancial-inclusion-roadblocks-remain 4 https://www.af-global.org/sites/default/fles/publications/af_case_study_rwanda_ fnalweb.pdf 5 https://www.bsg.ox.ac.uk/sites/default/fles/2018-06/2017-07-M-Pesa-PractitionersInsight.pdf 6 https://www.mastercardcenter.org/insights/modelo-peru-collaboration-createsinteroperability 3

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Theories of Financial Inclusion    93 literatures. But, theories are powerful because they can help to bring the two literatures together. Theories can explain why different ideas exist on what fnancial inclusion objectives should be, and how to achieve fnancial inclusion. Theories can explain the current observations in fnancial inclusion practice and can also explain abnormal deviations that exist in practice so that a coherent and comprehensive system of principles for fnancial inclusion can be developed. Therefore, a good fnancial inclusion theory or set of theories is one that provides a system of ideas to explain fnancial inclusion objectives, processes or outcomes (Thalassinos, Liapis, & Thalassinos, 2014). Some fnancial inclusion observations or practices include the opening of bank accounts with minimal documentation requirements, using ultimatums to compel citizens to obtain a bank verifcation number, granting free debit cards, free insurance policies, using mobile technology to access fnance, adopting a direct government-to-person (G2P) payment system, enrolling for mortgage without having to make compulsory equity down payment, the large-scale use of bank correspondents, among others. To understand why certain fnancial inclusion objectives yield certain outcomes such as those described above, we also need to understand the behavior of the agents involved in the fnancial inclusion process and a theory or principles of fnancial inclusion can help explain this as well. A fnancial inclusion theory is therefore an explanation for observed fnancial inclusion practices. Why do we need a theory or set of theories of fnancial inclusion? We need a theory of fnancial inclusion to achieve a high level of synthesis between fnancial inclusion objectives and fnancial inclusion outcomes. A fnancial inclusion theory or set of theories would provide a system of ideas that can explain fnancial inclusion objectives, processes and outcomes. Secondly, fnancial inclusion theories can consolidate the recent idealistic debates in the policy literature on fnancial inclusion. Thirdly, fnancial inclusion theories can provide a set of principles on which the practice of fnancial inclusion is based, and would make it possible to detect abnormal patterns in fnancial inclusion practice which would elicit further research to improve our understanding on why unexpected deviations exist in practice. Generally, in problem-driven social science research, researchers often use one or multiple theories to assess a problem and use one or multiple theories to solve or prevent a problem. This suggests that the theories for describing a fnancial inclusion problem may differ from the theories for its solution. Similarly, the theories for identifying the benefciaries of fnancial inclusion may differ from the theories for the delivery and funding of fnancial inclusion activities.

3. Theories of Financial Inclusion Benefciary Conficting ideas or perspectives exist on who benefts from fnancial inclusion outcomes. Some studies argue that poor people are the ultimate benefciaries of fnancial inclusion (Bhandari, 2018); others think that women are the benefciaries of fnancial inclusion outcomes (Demirguc-Kunt, Klapper, & Singer, 2013; Ghosh & Vinod, 2017; Swamy, 2014) while some think that the economy and the fnancial system are benefciaries of fnancial inclusion (Kim, Yu, & Hassan,

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

94    Peterson K. Ozili 2018; Mehrotra & Yetman, 2015; Ozili, 2018; Swamy, 2014). Apart from women and poor people, there are other potential benefciaries of fnancial inclusion that have been ignored to a large extent in the literature such as young people, elderly people, institutionalized and ill people, disabled people, and individuals who have been previously expelled from the fnancial sector for various reasons such as committing criminal offenses.7 Below are four theories that explain who benefts from fnancial inclusion.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

3.1. Public Good Theory of Financial Inclusion The public good theory of fnancial inclusion argues that the (1) delivery of formal fnancial services to the entire population and (2) ensuring that there is unrestricted access to fnance for everyone, should be treated as a public good for the beneft of all members of the population. As a public good, individuals cannot be excluded from using formal fnancial services and individuals cannot be excluded from gaining access to fnancial services. All individuals will enjoy basic fnancial services without paying for it. Also, access to fnancial services by one individual does not reduce its availability to others which means that all members of the population can be brought into the formal fnancial sector and everyone will be better-off. Under this theory, all members of the population are benefciaries of fnancial inclusion and nobody is left out. Under the public good theory, any individual or small business that open a formal bank account can be offered free debit cards, they can also use the ATM machines to perform transactions without being charged a transaction fee. Also, the suppliers of fnancial services such as fnancial institutions will have to bear the cost as a sunk-cost of doing banking business. Also, the government can grant subsidy to fnancial institutions to help them cope with any resulting cost problems that may arise from offering free fnancial services. A government can even offer a lump-sum cash deposit into the bank account of all citizens and make owning a formal account the only requirement for individuals to access the free deposits. This means that individuals who cannot pay their current debts and who cannot meet their basic needs at the micro-level will stand a chance to be economically empowered when fnancial inclusion is viewed as public good. This theory has two merits. Firstly, the public good theory suggests that everyone will beneft from fnancial inclusion regardless of status or income level. This means that both the rich and the poor, the fnancially included citizens and the fnancially excluded citizens, will enjoy the benefts of fnancial inclusion. Secondly, as a public good, achieving fnancial inclusion would require public funding rather than private equity because investors would require a premium on private equity which is costly when private funds are used to achieve fnancial

7

The World Bank’s perspective on the benefciaries of fnancial inclusion does not include these potential benefciaries. In fact, the World Bank (2018) concludes that poor households and women as the ultimate benefciaries of fnancial inclusion, which is quite restrictive.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Theories of Financial Inclusion    95 inclusion objectives. Thirdly, as a public good, it gives the government an opportunity to take responsibility for promoting fnancial inclusion. Finally, the public good theory of fnancial inclusion does not recognize private-sector agents as promoters of fnancial inclusion. The public good theory has four demerits. Firstly, treating fnancial inclusion as a public good does not address the real cause of fnancial exclusion in the frst place. Secondly, treating fnancial inclusion as a public good which requires public funding can divert public funds away from other important public projects in order to fund fnancial inclusion projects. Thirdly, the public good theory assumes that fnancial inclusion as a “public good” is free of charge and has no cost to the end-users of fnancial services. When fnancial inclusion is treated as a public good, the level of fnancial inclusion may not be sustainable in the long-term even when supported with public funding if it comes at no cost to end users. Four, the public good theory of fnancial inclusion may have little relevance for developing and emerging economies because banks operating in developing countries and emerging economies are mostly funded on private investments rather than public investments, for this reason, it can be diffcult to make fnancial inclusion a public good8 in such countries.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

3.2. Dissatisfaction Theory of Financial Inclusion The dissatisfaction theory of fnancial inclusion argues that fnancial inclusion activities and programs in a country should frst be targeted to all individuals who were previously on-boarded into the formal fnancial sector but left the formal fnancial sector because they were dissatisfed with the rules of engagement in the formal fnancial sector, or had other unfavorable personal experiences from dealing with frms and agents in the formal fnancial sector.9 This theory suggests that it is easier to bring back people who left the formal fnancial sector because they were dissatisfed if the areas of dissatisfaction in the formal fnancial sectors have been completely resolved. It is easier to bring back this group of individuals into the formal fnancial sector through persuasion than to bring in those who have never been in the formal fnancial sector. The implication of this theory is that the members of the population that left the formal fnancial sector should be the frst target of fnancial inclusion before extending fnancial inclusion policies and programs to other members of the population who have never been on-boarded into the formal fnancial sector. Previously on-boarded individuals may become dissatisfed for several reasons such as when they are victims of fnancial fraud,

8

Actually, in Africa the nature of governments and issues of governance has forced countries to privatized most of the public banks. In the era of public banks, there were problems of break-even, sustainability and proftability for public-sector banks in the provision of fnancial services as a public good by public banks. 9 Such unfavorable experiences may include debit/credit card frauds, long waiting periods before depositors are able to withdraw funds, taking too long before payments are cleared, high transaction costs, excessive bank charges, etc.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

96    Peterson K. Ozili debit/credit card fraud, fnancial theft, long waiting hours before depositors are able to withdraw funds, taking too long before payments are cleared, high transaction costs, excessive bank charges, etc. The dissatisfaction theory of fnancial inclusion has some merits. Firstly, the theory provides a deliberate attempt to deal with the voluntary fnancial exclusion problem which other theories do not address. It reduces the level of voluntary fnancial exclusion by using persuasion to bring back those that left the formal fnancial sector due to dissatisfaction. Secondly, under this theory, it is easy to identify the fnancially excluded members of the population. The previously on-boarded members of the population can be easily identifed because their personal data are stored with fnancial institutions, and they can be reached to be persuaded to return to the formal fnancial sector. It is easier to achieve fnancial inclusion by reaching out to previously on-boarded individuals compared to achieving fnancial inclusion for members of the population that have never been to the formal fnancial sector. Thirdly, achieving fnancial inclusion does not require the use of public funding since it relies strongly on interpersonal persuasive skills and abilities. The dissatisfaction theory has some demerits. Firstly, the theory does not prioritize fnancial inclusion for everybody in the population. It excludes people who have never been to the formal fnancial sector. Secondly, it ignores poor people in remote areas where formal fnancial institutions do not exist. Thirdly, the dissatisfaction theory implicitly assumes that fnancial exclusion is caused by customers’ dissatisfaction with the rules of engagement in the formal fnancial sector. This may not be the case under certain circumstances because individuals can voluntarily withdraw from the formal fnancial sector for other reasons such as religious and personal reasons (Ozili, 2018). Finally, individuals who are dissatisfed with the fnancial sector may have no choice but to remain in the formal fnancial sector if the societal culture relies too much on the fnancial sector.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

3.3. Vulnerable Group Theory of Financial Inclusion The vulnerable group theory of fnancial inclusion argues that fnancial inclusion activities or programs in a country should be targeted to the vulnerable members of society such as poor people, young people, women, and elderly people who suffer the most from economic hardship and crises. Vulnerable people are often the most affected by fnancial crises and economic recession; therefore, it makes sense to bring these vulnerable people into the formal fnancial sector. One way to achieve this is through G2P social cash transfers into the formal account of vulnerable people. Making G2P social cash transfer payments into the formal account of poor people, young people, women and elderly people will encourage other poor people, young people, women and elderly people to join the formal fnancial sector to own a formal account to take advantage of the social cash transfer benefts, thereby, increasing the rate of fnancial inclusion for vulnerable groups. Also, when social cash transfer is working and other tools for achieving fnancial inclusion are provided to the vulnerable people in society, it can make vulnerable people feel that they are being compensated for the current income

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Theories of Financial Inclusion    97 inequality, which gives them an opportunity to catch up with the other segments of society. The implication of theory is that it identifes some members of the population to be vulnerable, and suggests that fnancial inclusion efforts should be the targeted to the vulnerable people in society. The vulnerable group theory of fnancial inclusion has some merits. Firstly, the theory reduces the fnancial exclusion problem by targeting vulnerable groups for fnancial inclusion. It attempts to bring the vulnerable population into the formal fnancial sector. Secondly, under this theory, it is easy to identify the fnancially excluded members of the population. The vulnerable members of the population can be identifed by their income level, gender, age and other demographic characteristics. Thirdly, it may be cost-effective to target only the vulnerable members of the population for fnancial inclusion compared to achieving fnancial inclusion for the entire population. The vulnerable group theory has some demerits. Firstly, the theory does not prioritize fnancial inclusion for everybody in the population. Secondly, it ignores non-vulnerable people outside the formal fnancial sector. Non-vulnerable people also need access to the formal fnancial sector too! Thirdly, it assumes that women are a vulnerable group, which implies that men are not a vulnerable group. This idea is critical because in modern societies women and men compete for equal opportunities, therefore labeling women as vulnerable groups to the exclusion of men could have unintended consequences for fnancial and social inclusion. It could lead to societal resentment among the men toward women. Finally, achieving fnancial inclusion by targeting vulnerable people may lead to increasing social inequality when social policies are designed to favor vulnerable people over others, and it may also lead to income inequality if vulnerable people receive better access to fnancial services than others.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

3.4. Systems Theory of Financial Inclusion The systems theory of fnancial inclusion states that fnancial inclusion outcomes are achieved through existing systems (whether economic system, social structures or fnancial systems), and as a result, greater fnancial inclusion will have positive benefts for the systems it relies on. A signifcant change in one part of the system can signifcantly affect the expected fnancial inclusion outcomes. For instance, imposing regulations on economic agents and suppliers of fnancial services (who are a part of the system) can align their interests with that of the users of basic fnancial services, in order to achieve the expected fnancial inclusion outcome. On the other hand, a signifcant change at the system level (such as replacing the existing national fnancial inclusion plan with a completely new plan) does not necessarily lead to a change in the existing sub-systems because a change in the sub-system must be done at the sub-system level. The theory suggests that fnancial inclusion, as a system, will improve the workings of the sub-systems it rely on, while the effciency and effectiveness of the sub-systems will determine the success or failure of a fnancial inclusion agenda. But in the end, the existing (economic, fnancial and social) sub-systems systems in a country are the ultimate benefciaries of fnancial inclusion, under the systems theory perspective.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

98    Peterson K. Ozili The systems theory of fnancial inclusion has some merits. Firstly, the systems theory recognizes the role of existing economic, fnancial and social systems or structures in a country in promoting fnancial inclusion. Secondly, it provides a macro-perspective on fnancial inclusion compared to other theories with a micro-perspective. Thirdly, systems theory of fnancial inclusion considers how fnancial inclusion outcomes are affected by the interrelationship among the components (or sub-systems) of existing systems that fnancial inclusion relies on. The systems theory has some demerits. Firstly, the existing systems are a refection of the environment. In some environments, the existing systems may not function properly, and as a result, the expected fnancial inclusion outcomes may not be achieved. Secondly, systems theory of fnancial inclusion does not recognize the infuence of factors outside the system that could affect fnancial inclusion outcomes, rather it focuses on the effect of the system and its sub-systems on fnancial inclusion outcomes. Thirdly, systems theory of fnancial inclusion assumes that there is a direct relationship between fnancial inclusion outcomes and systems it relies on.

4. Theories of Financial Inclusion Delivery

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

There are several ideas on who should deliver fnancial services to the people. Some think the government should deliver fnancial inclusion to the people (Aggarwal & Klapper, 2013; Chibba, 2009; Staschen & Nelson, 2013). Others argue that private companies such as banks and Fintech businesses can deliver fnancial inclusion more effciently (Gabor & Brooks, 2017; Ozili, 2018). There are also ideas suggesting that fnancial inclusion can be delivered through cooperation by the public and private sectors (Arun & Kamath, 2015; Pearce, 2011). These expectations regarding fnancial inclusion delivery need an underlying thought-process to establish why these agents are necessary in the frst place to deliver fnancial inclusion; hence, there is need for theories of fnancial inclusion delivery. Some theories or perspectives on fnancial inclusion delivery are provided in the following.

4.1. Community Echelon Theory of Financial Inclusion Community echelon theory states that fnancial inclusion should be delivered to the fnancially excluded population through their communal leaders. The community echelon theory argues that community leaders are infuential in their communities and can use their infuence to encourage or persuade community members to participate in the formal fnancial sector. Community plays an important role in shaping the values of its leaders and members. Community members trust their leaders and believe their leaders would make decisions that are benefcial to them while community leaders ensure that the decisions they make refect the values and ethos held by members of the community. Community leaders can be instrumental in bringing their members into the formal fnancial sector because the strong cultural ties between community leaders and members make it possible for community leaders to encourage their members to participate in the formal

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Theories of Financial Inclusion    99 fnancial sector. If community leaders change their beliefs and preferences toward participating in the formal fnancial sector, community leaders can encourage their members to participate in the formal fnancial sector. Since communal outcomes are mostly predicted by the preferences, beliefs and other idiosyncrasies of community leaders, it makes sense to deliver fnancial inclusion to members of the community through their communal leaders. The most important merit of the community echelon theory of fnancial inclusion is that communal leaders can infuence community members into making changes that improve their welfare. The community echelon theory has some demerits. One, the infuence of communal leaders can rebound if communal leaders are self-serving and corrupt. Secondly, communal leaders can make abrupt decisions that are not in the best interest of the people if the leaders are pressured by corporate promoters of fnancial innovation. Since communal leaders are believable, decisions that are not in the best interest of the community members can be reached and can have negative implications on the welfare of community members. Thirdly, community members already participating in the formal fnancial sector may have worries, anxiety and confusion concerning the formal fnancial sector which they may not articulate to their community leaders. The higher the anxiety, the less likely they will remain in the formal fnancial sector for a long time. Also, serious agency problems may arise in form of nepotism, fraud and corruption. Finally, since there are different leadership styles, it is diffcult to determine which leadership style works best to infuence community members to change their beliefs toward participating in the formal fnancial sector.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

4.2. Public Service Theory of Financial Inclusion Public service theory of fnancial inclusion states that fnancial inclusion is a public responsibility which the government owes its citizens, and the citizens expect the government to promote fnancial inclusion for its citizens. This theory argues that fnancial inclusion should be delivered to all citizens including the fnancially excluded population by the government through public institutions. Under this theory, only the government is instrumental in achieving fnancial inclusion that brings all members of the population into the formal fnancial sector so that each member of the population can have access to formal fnancial products and services. The public service theory has some merits. One, the theory suggest that fnancial inclusion can be achieved when the government chooses to be responsible for fnancial inclusion. Secondly, the government has control over the fnancial system, economic and social structures in the country, which the government can use to achieve its fnancial inclusion objectives, for example, by establishing bank branch networks in the most remote areas of the country to reach the excluded members of the population. Thirdly, there is increased public confdence when the government assumes full responsibility for fnancial inclusion. Members of the population are confdent that all fnancial inclusion activities and programs will work for the greater good of everyone when the government takes full responsibility for fnancial inclusion.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

100    Peterson K. Ozili Some demerits of the public service theory include the following. One, it does not consider private-sector participation in promoting fnancial inclusion. Secondly, it assumes fnancial inclusion will be funded with taxpayers’ money. Tax revenue may be insuffcient to fund large fnancial inclusion programs. Thirdly, under a public service theory perspective where the State is responsible for fnancial inclusion through its public institutions, one disadvantage is that the State can use political power as a means of control over society. The state can provide basic fnancial services to obedient citizens and during good times, and can stop providing basic fnancial services when the citizens collectively choose to revolt against the state.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

4.3. Special Agent Theory of Financial Inclusion The special agent theory of fnancial inclusion argues that the delivery of fnancial inclusion to the excluded population can be hampered by complex issues and technicalities relating to the nature of the community, its people or the geography; therefore, there is need for specialized agents to deliver fnancial inclusion to members of the excluded communities. Under this theory, the special agent is expected to: (1) be a highly skilled and specialized agent, (2) understand the peculiarities of the excluded population, (3) understand the existing informal fnancial system in the communities where the excluded members of the population reside, (4) identify areas for improvement through innovation and (5) devise a means of integrating the local fnancial system into the formal fnancial sector, to bring excluded population into the formal fnancial system so that they can have access to formal fnancial products and services. Under this theory, there is a special agent relationship with the principal. The principal is often the government or foreign organizations while the special agent is often a local bank, non-bank company or other special institutions established for the sole purpose of achieving fnancial inclusion. In a much broader sense, the special agent might be banks, fnancial technology (Fintech) companies or other specialized institutions. The special agent theory of fnancial inclusion has some merits. Firstly, it employs the services of specialized agents to promote fnancial inclusion in the country. Employing the services of specialized agents to promote fnancial inclusion in the country will allow the government to focus on other important and pressing national issues. Secondly, there is high degree of confdence in the ability of special agents to deliver fnancial inclusion to the excluded population. The special agents comprise of skilled individuals and experts affliated with, or in collaboration, with other specialized institutions to harness collaborative inputs in order to reach the common goal of greater fnancial inclusion. Thirdly, there is no ambiguity about the fnancial inclusion targets to be achieved, and the compensation to the special agent is pre-agreed. The special agent knows what the target is, they know the expectations and are committed to meet the target, and they know the compensation to be received for their work, thus there is no ambiguity. Finally, the special agent relationship is not affected by the fundamental principal-agent problems in agency theory because the special agent deals with people, not money! The principal-agent problem in agency theory occurs when

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Theories of Financial Inclusion    101 the agent, who is the manager of a frm, is driven by self-interest to appropriate excess fnancial resources (money) to oneself at the expense of shareholders (see Jensen & Meckling, 1976 for agency theory). The special agent theory of fnancial inclusion has some demerits. Firstly, if the principal is the government, the government may choose its own agency as the special agent, making the government both the principal and the agent. This would defeat the purpose of special agency for promoting fnancial inclusion. The government should not be the principal and the special agent because government agencies are inherently plagued with ineffciencies in the public sector. Secondly, a private special agent will abandon the fnancial inclusion project when there is a breach in contractual terms and conditions for service which may arise from insuffcient compensation to the special agent by the government or the failure of the government to provide agreed funds to fnance the fnancial inclusion projects as agreed in the contract.

4.4. Collaborative Intervention Theory of Financial Inclusion Collaborative intervention theory states that fnancial inclusion should be achieved through collaborative intervention from multiple stakeholders. The theory suggests that joint effort from multiple stakeholders is needed to bring the excluded population into the formal fnancial services. This theory has some merits. One, it encourages a multi-stakeholder approach to achieve fnancial inclusion. Secondly, the collaborating stakeholders have a sense of satisfaction for being a signifcant contributor to a public project. The collaborative intervention theory has some demerits. One, it is diffcult to determine the optimal number of collaborators needed to support a fnancial inclusion agenda. Secondly, some collaborators may become inactive leaving the task for few active collaborators to do. Thirdly, having higher number of collaborators does not guarantee higher probability of achieving fnancial inclusion.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

4.5. Financial Literacy Theory of Financial Inclusion Financial literacy theory of fnancial inclusion states that fnancial inclusion should be achieved through education that increases the fnancial literacy of citizens. This theory argues that fnancial inclusion programs and activities directed at improving the fnancial literacy of the members of the population will increase their willingness to participate in the formal fnancial sector. The fnancial literacy theory has some merits. Financial literacy can make people aware of fnancial products and services that are available to them. When they become aware of existing fnancial products and services that can improve their welfare, they will be willing to participate in the formal fnancial sector by owning a bank account. Secondly, through increased fnancial literacy, people can take advantage of other benefts in the formal fnancial sector. Financial literacy can help people become self-suffcient and can help them have some stability in their personal fnance by helping them distinguish between needs and wants, helping them to create and manage a budget, teaching them to save so that they can pay

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

102    Peterson K. Ozili bills when due, and to plan for retirement. Thirdly, governments that have limited public funds or limited tax revenue to fund fnancial inclusion activities will prefer to use fnancial literacy as a fnancial inclusion strategy since it does not require much public fund to educate the public on the use of fnancial services The demerits of the fnancial literacy theory include the following. One, it addresses the “willingness” not “capacity” to participate in the formal fnancial sector. Financial literacy through education can improve the willingness of people to participate in the formal fnancial sector but it does not necessarily improve “capacity” to participate in the formal fnancial sector where capacity is measured as having money which can be used to perform one or more transactions. This means that people who do not have money (i.e., “capacity”) cannot actively participate in the fnancial sector even if they become fnancially literate.

5. Theories of Financial Inclusion Funding The question, who should fund fnancial inclusion expenditure to the people, – is important. Some think public money (tax-payers fund) should fund fnancial inclusion programs and activities (Marshall, 2004). Others feel that the capitalists in the private sector should fund fnancial inclusion because they contributed to widen the income inequality gap between the poor and the rich (Mohiuddin, 2015). There are also ideas suggesting that fnancial inclusion should be jointly funded by the public and private sectors (Cobb, Wry, & Zhao, 2016; Dashi, Lahaye, & Rizvanolli, 2013). The following are some theories of fnancial inclusion funding.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

5.1. Private Money Theory of Financial Inclusion Private money theory of fnancial inclusion states that fnancial inclusion activities should be funded using private money because private funders will require accountability from the users of their funds and will ensure that private funds are utilized effciently and ensure that fnancial products and services are delivered to the intended fnancially excluded members of the population. The merits of private money theory include the following. One, there is shorter approval time to obtain private funding for fnancial inclusion projects compared to the long approvals for public funding. This is because private lenders can reach decisions more quickly since there are fewer processes through which the approval process must go. Secondly, private funders are often directly involved in fnancial inclusion activities either through equity ownership or through other forms of participation. Thirdly, private funders can take ownership of the project and can gain incomes and profts when they manage fnancial inclusion activities themselves. They can also exchange benefts with the local authorities. Fourthly, it is easy to increase charges from users to meet the cost/budget of a contract with a private operator than by voting in local or national government to increase the cost or budget (Spackman, 2002). Five, private fnanciers can offer better project management skills, innovative facility and risk management in achieving fnancial inclusion objectives. Six, private funders can exert greater pressure on private

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Theories of Financial Inclusion    103 contractors to fnish all fnancial inclusion projects in good time while maintaining high quality. Some demerits of private money theory include the following. One, the cost of raising private funds to fund fnancial inclusion projects is often high. Secondly, funding fnancial inclusion objectives using private money can increase private interests in fnancial inclusion outcomes to the detriment of the excluded population. Thirdly, there may be loss of government control over the fnancial inclusion infrastructure created by private investors due to partial or full private ownership.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

5.2. Public Money Theory of Financial Inclusion Public money theory of fnancial inclusion states that fnancial inclusion programs and activities should be funded using public money. This theory argues that fnancial inclusion programs and activities should be funded from government budgets. There is evidence that public funding for fnancial inclusion is growing faster than private funding (see Dashi et al., 2013). Some merits of the public money theory include the following. One, the government can tax the rich to generate funds for fnancial inclusion projects for the beneft of all, and this would lead to the redistribution of wealth and reduce income inequality for the beneft of the poor and excluded population. Secondly, the cost of raising public funds to fund fnancial inclusion projects is low or negligible. Thirdly, funding fnancial inclusion objectives using public money can prevent unscrupulous individuals from hijacking the fnancial inclusion agenda for their own selfsh beneft. Some demerits of public money theory include the following. One, lack of proper planning is a major problem associated with public funding. Lack of planning can lead to overspending on fnancial inclusion projects, which can ultimately lead to ineffciency. Secondly, using public money to fund fnancial inclusion projects can lead to unnecessary delays in reaching the excluded population such as delays in disbursing funds and delays caused by lobbying, among others. Thirdly, governments with insuffcient funds will be pressured to raise loans to fnance fnancially inclusive projects and will increase the debt burden of the government. Finally, improper delegation of authority may arise when the task of achieving fnancial inclusion is delegated to an incompetent authority or contractor. Usually, governments often make one of its competent agencies responsible for achieving fnancial inclusion objectives. But such competent agency maybe overburdened with the task of achieving fnancial inclusion because agency also has its own statutory duties to perform. This type of improper delegation of authority is common in many countries and can lead to mistakes in implementing fnancial inclusion goals.

5.3. Intervention Fund Theory of Financial Inclusion The intervention fund theory of fnancial inclusion argues that fnancial inclusion activities and programs can be funded by special interventions from diverse related and unrelated funders rather than using taxpayers’ money. It argues that many “special funders” exist in the world such as philanthropists, non-governmental

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

104    Peterson K. Ozili organizations and foreign governments, and these special funders tend to support inclusive fnance for the global population. In some economies, cross-border funding has the largest share of fnancial inclusion funding and much of these funding is allocated to micro-fnance institutions (El-Zoghbi, Gähwiler, & Lauer, 2011). Special funders can voluntarily and selectively choose the fnancial inclusion projects they wish to fund to completion and will provide the “intervention fund” required to achieve the desired fnancial inclusion objectives. The goal of intervention funding for fnancial inclusion is to ensure that poor people and the fnancially excluded population have access to formal fnancial services. This theory has some merits. One, it bypasses the usual political bureaucracy associated with allocating public funds for public projects. Secondly, the special funders can mobilize fnancial and human resources, both locally and internationally, to assist them achieve the desired fnancial inclusion objectives. Thirdly, special funders can create new institutions that are pro-development to help them achieve the desired fnancial inclusion objective, and these institutions will remain in the community to promote development even after the fnancial inclusion projects are completed. The intervention fund theory has four demerits. Firstly, under this theory, special funders would need to develop a methodology to determine which segments of the population are excluded from the formal fnancial sector. Secondly, the method used by special funders to determine which population members are fnancially excluded may not accurately identify the intended target for fnancial inclusion. Thirdly, using intervention funds from foreign governments or foreign donors to fund development projects in a country can damage the reputation of the government of that country.

6. Future Direction: Applying the Theories to Data

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

6.1. Empirical Data Data are records from past observations. Every data or observation from fnancial inclusion practice should be understood in terms of a prior, often implicit theory. This idea is consistent with Popper (1976) who argued that theory should precede observation. Data for fnancial inclusion can be obtained by using direct observations, interviews and surveys. Data obtained from fnancial inclusion practice and the analysis of such data should support or refute the above-discussed theories. Better theories of fnancial inclusion should replace poorer ones if they explain existing observations more effectively. Also, the empirical modeling of fnancial inclusion determinants should take into account the nature of the data and the magnitude of the explained and unexplained variation in the model.

6.2. Case Studies Scholars may use case studies on fnancial inclusion to develop interesting theories. Building a fnancial inclusion theory from case studies is a research strategy that involves using one or more cases to create theoretical constructs and propositions

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Theories of Financial Inclusion    105 from case-based or empirical evidence (Eisenhardt, 1989). Case studies are often descriptions of particular instances of a phenomenon that are based on a variety of data sources (Yin, 1984). Cases can be historical accounts of fnancial inclusion success or failures in several countries. Each case study on fnancial inclusion can serve as a distinct experiment that stands on its own as an analytic unit while multiple case studies will serve as discrete experiments that serve as replications, contrasts and extensions to the existing or emerging theories. Building theories from case studies on fnancial inclusion is likely to become a popular and more relevant research strategy for future studies on fnancial inclusion. Although case studies have some problems such as the small sample associated with case studies and the one-sided interviews that increases the informant bias, these issues can be mitigated by careful choice of sample and sample size, and conducting fair interviews that limit the informants bias. Data for fnancial inclusion can be obtained by using direct observations, interviews and surveys.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

7. Conclusion This chapter presented new theories that can be used in fnancial inclusion research and policy debates. The purpose of this chapter has been to argue that, fnancial inclusion needs to be studied from a theoretical perspective. A summary of the theories have been provided in the Appendix. Several theoretical perspectives have been provided. So far, many fnancial inclusion topics have yet to be suffciently studied, and theories have been underutilized in the fnancial inclusion debate. The theories presented in this chapter can serve as a guide for what needs to be done, but researchers who are not experts in fnancial inclusion will inevitably need more resources in addition to this one. This chapter is a practical description and does not directly address how to test the relative effectiveness of various fnancial inclusion theories or constructs for empirical modeling and critical discourse. Nevertheless, the chapter does suggest ways in which using these theories can change how we think about fnancial inclusion benefciaries, delivery agents and funding. Also, using these theories can remove from the literature extreme idealism. Using these theories will also help researchers make sure that the theoretical constructs they use are actually used all the way through the fnancial inclusion theory building process and will improve evidence summaries and thereby advance our understanding of fnancial inclusion. A possible direction for future research is to develop a glossary of fnancial inclusion success stories across countries. Secondly, future studies can test these theories to identify which fnancial inclusion theories have the highest explanatory power in predicting the success of fnancial inclusion programs and policies in different countries. Of course, many additional theories of fnancial inclusion can be developed, and there is no limit to the number of new theories or ideas that can be developed or explored. Thirdly, there is need to test these theories using a rich sample such as a multi-country sample which is common among panel studies, or using time-series data for country-specifc studies. Finally, although this chapter has argued the need of theory to understand fnancial inclusion, some advice for practitioners and policymakers are stated in

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

106    Peterson K. Ozili the following. Digital fnance and fnancial innovations should be used to achieve fnancial inclusion in a way that minimize tail risk to poor and vulnerable customers. Policies should be developed that encourage competition in the delivery of fnancial services. Governments should consider granting subsidy to providers of fnancial services so that they can offer basic fnancial services to the excluded population at a very low cost or free of charge. Governments should establish a communication channel that allows citizens to express their thoughts and concerns as to whether fnancial services are been offered to them fairly and at a cheap fee and without discrimination.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

References Aggarwal, S., & Klapper, L. (2013). Designing government policies to expand fnancial inclusion: Evidence from around the world. The Journal of Finance, 56(3), 1029–1051. Allen, F., Demirguc-Kunt, A., Klapper, L., & Martinez Peria, M. S. (2016). The foundations of fnancial inclusion: Understanding ownership and use of formal accounts. Journal of Financial Intermediation, 27, 1–30. Arun, T., & Kamath, R. (2015). Financial inclusion: Policies and practices. IIMB Management Review, 27(4), 267–287. Bhandari, B. S. (2018). Life insurance-social security & fnancial inclusion. Bimaquest, 18(2). Cámara, N., & Tuesta, D. (2015). Peru model for fnancial inclusion: E-money potential adopters. Madrid: BBVA Research. Chibba, M. (2009). Financial inclusion, poverty reduction and the millennium development goals. The European Journal of Development Research, 21(2), 213–230. Cobb, J. A., Wry, T., & Zhao, E. Y. (2016). Funding fnancial inclusion: Institutional logics and the contextual contingency of funding for microfnance organizations. Academy of Management Journal, 59(6), 2103–2131. Cull, R., Cull, R. J., Demirgüç-Kunt, A., & Morduch, J. (Eds.). (2013). Banking the world: Empirical foundations of fnancial inclusion. Cambridge, MA: MIT Press. Dashi, E., Lahaye, E., & Rizvanolli, R. (2013). Trends in international funding for fnancial inclusion. Washington, DC: Center for Financial Inclusion. Demirgüç-Kunt, A., & Klapper, L. (2013). Measuring fnancial inclusion: Explaining variation in use of fnancial services across and within countries. Brookings Papers on Economic Activity, 2013(1), 279–340. Demirguc-Kunt, A., Klapper, L., & Singer, D. (2013). Financial inclusion and legal discrimination against women: Evidence from developing countries. Washington, DC: The World Bank. Demirguc-Kunt, A., Klapper, L., & Singer, D. (2017). Financial inclusion and inclusive growth: A review of recent empirical evidence. Washington, DC: The World Bank. Dev, S. M. (2006). Financial inclusion: Issues and challenges. Economic and Political Weekly, 41, 4310–4313. Eisenhardt, K. M. (1989). Building theories from case study research. Academy of Management Review, 14, 532–550. El-Zoghbi, M., Gähwiler, B., & Lauer, K. (2011). Cross-border funding of microfnance. CGAP Focus Note 70.CGAP, Washington, DC.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Theories of Financial Inclusion    107 Gabor, D., & Brooks, S. (2017). The digital revolution in fnancial inclusion: International development in the Fintech era. New Political Economy, 22(4), 423–436. Ghosh, S., & Vinod, D. (2017). What constrains fnancial inclusion for women? Evidence from Indian micro data. World Development, 92, 60–81. Hove, L. V., & Dubus, A. (2019). M-PESA and fnancial inclusion in Kenya: Of paying comes saving? Sustainability, 11(3), 568. Jensen, M. C., & Meckling, W. H. (1976). Theory of the frm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3(4), 305–360. Kim, D. W., Yu, J. S., & Hassan, M. K. (2018). Financial inclusion and economic growth in OIC countries. Research in International Business and Finance, 43, 1–14. Liapis, K., Rovolis, A., Galanos, C., & Thalassinos, I. E. (2013). The clusters of economic similarities between EU countries: A view under recent fnancial and debt crisis. European Research Studies Journal, 16(1), 41–66. Lichtenstein, J. (2018). Financial inclusion in Rwanda: Examining policy implementation and impact on community and household lives. Doctoral dissertation, University of Cambridge. Marshall, J. N. (2004). Financial institutions in disadvantaged areas: A comparative analysis of policies encouraging fnancial inclusion in Britain and the United States. Environment and Planning A, 36(2), 241–261. Mehrotra, A. N., & Yetman, J. (2015). Financial inclusion-issues for central banks. BIS Quarterly Review, March, 83–96. Mohiuddin, S. (2015). Private sector leadership in fnancial inclusion. Corporate Citizenship Center, U.S. Chamber of Commerce Foundation. Retrieved from https://www. uschamberfoundation.org/sites/default/fles/Private%20Sector%20Leadership%20 in%20Financial%20Inclusion_Web.pdf Morgan, P., & Pontines, V. (2014). Financial stability and fnancial inclusion. ADBI Working Paper 488. Asian Development Bank Institute, Tokyo. Ndung’u, N. (2018). The M-Pesa technological revolution for fnancial services in Kenya: A platform for fnancial inclusion. In D. Lee Kuo Chuen & R. H. Deng (Eds.), Handbook of blockchain, digital fnance, and inclusion (Vol. 1, pp. 37–56). London: Academic Press. Nimbrayan, P. K., Tanwar, N., & Tripathi, R. K. (2018). Pradhan mantri jan dhan yojana (PMJDY): The biggest fnancial inclusion initiative in the world. Economic Affairs, 63(2), 583–590. Otioma, C., Madureira, A. M., & Martinez, J. (2019). Spatial analysis of urban digital divide in Kigali, Rwanda. GeoJournal, 84(3), 719–741. Ozili, P. K. (2018). Impact of digital fnance on fnancial inclusion and stability. Borsa Istanbul Review, 18(4), 329–340. Pearce, D. (2011). Financial inclusion in the Middle East and North Africa: Analysis and roadmap recommendations. Washington, DC: The World Bank. Popper, K. R. (1976). The myth of the framework. In E. Freeman (Ed.), The abdication of philosophy – Philosophy and the public good: Essays in honor of Paul Arthur Schilpp (pp. 23–48). LaSalle, IL: Open Court. Prabhakar, R. (2019). Financial inclusion: A tale of two literatures. Social Policy and Society, 18(1), 37–50. Sahay, R., Čihák, M., N’Diaye, P. M. B. P., Barajas, A., Mitra, S., Kyobe, A., … & Yousef, S. R. (2015). Financial inclusion: Can it meet multiple macroeconomic goals? (No. 15/17). Washington, DC: International Monetary Fund. Sarma, M., & Pais, J. (2011). Financial inclusion and development. Journal of International Development, 23(5), 613–628. Spackman, M (2002). Public–private partnerships: Lessons from the British approach. Economic Systems, 26(3), 283–301.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

108    Peterson K. Ozili

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Staschen, S., & Nelson, C. (2013) The role of government and industry in fnancial inclusion. In J. Ledgerwood (Ed.), The new microfnance handbook: A fnancial market system perspective (pp. 71–96). Washington, DC: The World Bank. Swamy, V. (2014). Financial inclusion, gender dimension, and economic impact on poor households. World Development, 56, 1–15. Thalassinos, E., Liapis, K., & Thalassinos, J. (2014). The role of the rating companies in the recent fnancial crisis in the Balkan and black sea area. In Economic Crisis in Europe and the Balkans (pp. 79–115). Heidelberg: Springer. Thalassinos, I. E., Liapis, K., & Thalassinos, E. J. (2014). The role of the rating companies in the recent fnancial crisis in the Balkan and black sea area. In A. Karasavvoglou & P. Polychronidou (Eds.), Economic crisis in Europe and the Balkans. Contributions to Economics (pp. 79–115). Springer International Publishing. doi:10.1007/978-3319-00494-5-6 World Bank. (2018). Financial inclusion. Retrieved from https://www.worldbank.org/en/ topic/fnancialinclusion/overview. Accessed on July 27, 2019. Yin, R. (1984). case study research. Beverly Hills.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Theory

Public good theory

Dissatisfaction theory

S/N

1.

2.

Financial inclusion benefciary

Financial inclusion funding

Category

Merit/Advantage

(1) It deals with the voluntary fnancial exclusion problem; (2) it is easy to identify the fnancially excluded members of the population; (3) achieving fnancial inclusion does not require the use of public funds

(1) Everyone will beneft from fnancial inclusion and nobody will be left out; (2) achieved through public funding; (3) the state takes responsibility for fnancial inclusion

Table A1.  Summary of the Theories of Financial İnclusion.

Appendix

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

(Continued)

(1) The theory does not prioritize fnancial inclusion for everybody in the population; (2) it ignores poor people in remote areas where formal fnancial institutions do not exist; (3) fnancial exclusion may not be caused by customers’ dissatisfaction with the formal fnancial sector

(1) It does not address the real cause of fnancial exclusion in the frst place; (2) it diverts public funding away from other important national projects; (3) fnancial inclusion may become unsustainable when treated as a public good; (4) may not be useful in explaining fnancial inclusion in developing and underdeveloped countries

Demerit/Advantage

Theories of Financial Inclusion    109

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Vulnerable group theory

Systems theory of fnancial inclusion

4.

Theory

3.

S/N

Table A1.  (Continued )

Financial inclusion benefciary

Financial inclusion benefciary

Category

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

(1) It recognizes the role of existing economic, fnancial and social systems or structures in a country in promoting fnancial inclusion; (2) it views fnancial inclusion from a macroperspective; (3) it considers how fnancial inclusion outcomes are affected by the interrelationship among the components of existing sub-systems

(1) It reduces the fnancial exclusion problem by targeting vulnerable groups; (2) it is easy to identify the fnancially excluded members of the population; (3) it is cost-effective to target only the vulnerable members of the population

Merit/Advantage

(1) It does not recognize the infuence of factors outside the system that could affect fnancial inclusion outcomes; (2) it assumes that there is a direct relationship between fnancial inclusion outcomes and systems it relies on

(1) It does not prioritize fnancial inclusion for everybody in the population; (2) it ignores nonvulnerable people outside the formal fnancial sector; (3) it assumes that women are a vulnerable group, which suggest that men are not a vulnerable group

Demerit/Advantage

110    Peterson K. Ozili

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Community echelon theory

Public service theory

5.

6.

Financial inclusion delivery

Financial inclusion delivery

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

(1) Financial inclusion can be achieved when the government chooses to be responsible for fnancial inclusion; (2) the government can use the existing economic, social and political systems to achieve its fnancial inclusion objectives; (3) there is increased public confdence when the government assumes full responsibility for fnancial inclusion

(1) Communal leaders can infuence community members into making changes that improve their welfare

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

(Continued)

(1) It does not consider privatesector participation in promoting fnancial inclusion; (2) it assumes fnancial inclusion will be funded with taxpayers’ money but tax revenue may be insuffcient to fund large fnancial inclusion programs

(1) The infuence of communal leaders can rebound if communal leaders are self-serving and corrupt; (2) communal leaders can make abrupt decisions that are not in the best interest of the people; (3) communal leaders can be pressured by corporate promoters of fnancial innovation to make decisions that are not in the best interest of the community members; (4) serious agency problems may arise in form of nepotism, fraud and corruption

Theories of Financial Inclusion    111

7.

S/N

Special agent theory

Theory

Table A1.  (Continued )

Financial inclusion delivery

Category

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

(1) It requires the services of specialized agents to promote fnancial inclusion in the country; (2) it allows the government to focus on other important and pressing national issues; (3) there is high degree of confdence in the ability of special agents to deliver fnancial inclusion to the excluded population; (4) there is no ambiguity about the fnancial inclusion targets to be achieved, and the compensation to the special agent which is predetermined; (5) the special agent relationship is not affected by the fundamental principal-agent problems in agency theory

Merit/Advantage

(1) The government may choose its own agency as the special agent, making the government both the principal and the agent; (2) the private special agent can abandon the fnancial inclusion project

Demerit/Advantage

112    Peterson K. Ozili

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Collaborative theory

Financial literacy

8.

9.

Financial inclusion delivery

Financial inclusion delivery

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

(1) Financial literacy can make people aware of fnancial products and services that are available to them; (2) fnancial literacy can help people become selfsuffcient and can help them have some stability in their personal fnance; (3) governments that have limited public funds or limited tax revenue to fund fnancial inclusion activities may prefer to use fnancial literacy as a fnancial inclusion strategy since it does not require much public fund

(1) It encourages a multistakeholder approach to achieve fnancial inclusion; (2) the collaborating stakeholders have a sense of satisfaction for being a signifcant contributor to a public project

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

(Continued)

(1) The fnancial literacy theory addresses the “willingness” not “capacity” to participate in the formal fnancial sector

(1) It is diffcult to determine the optimal number of collaborators needed to support a fnancial inclusion agenda; (2) some collaborators may become inactive leaving the task for few active collaborators to do; (3) having higher number of collaborators does not guarantee higher probability of achieving fnancial inclusion

Theories of Financial Inclusion    113

Theory

Private money theory

S/N

10.

Table A1.  (Continued)

Financial inclusion funding

Category

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

(1) There is shorter approval time to obtain private funding for fnancial inclusion projects; (2) private funders are often directly involved in fnancial inclusion activities either through equity ownership or through other forms of participation; (3) private funders can take ownership of the project, and can gain incomes and profts when they manage fnancial inclusion activities themselves; (4) private fnanciers can offer better project management skills, innovative facility and risk management in achieving fnancial inclusion objectives; (5) private funders can exert greater pressure on private contractors to fnish all fnancial inclusion projects in good time while maintaining high quality

Merit/Advantage

(1) The cost of raising private funds to fund fnancial inclusion projects is often high; (2) funding fnancial inclusion objectives using private money can increase private interests in fnancial inclusion outcomes; (3) there may be loss of government control over the fnancial inclusion infrastructure created by private investors

Demerit/Advantage

114    Peterson K. Ozili

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Financial inclusion funding

Financial inclusion funding

Public money theory

İntervention fund theory

11.

12.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

(1) It by-passes the usual political bureaucracy associated with allocating public funds for public projects; (2) the special funders can mobilize fnancial and human resources, both locally and internationally, to assist them achieve the desired fnancial inclusion objectives; (3) special funders can create new institutions that are pro-development to help them achieve the desired fnancial inclusion objective

(1) The government can tax the rich to generate funds for fnancial inclusion projects for the beneft of all; (2) the cost of raising public funds to fund fnancial inclusion projects is low or negligible; (3) funding fnancial inclusion objectives using public money can prevent unscrupulous individuals from hijacking the fnancial inclusion agenda for self-beneft

(1) special funders would need to develop a methodology to determine which segments of the population are excluded from the formal fnancial sector; (2) special funders may use an unfair methodology or criteria to determine which population members are fnancially excluded; (3) using intervention funds from foreign governments or foreign donors to fund development projects in a country can damage the reputation of the country

(1) Lack of proper planning is a major problem associated with public funding; (2) using public money to fund fnancial inclusion projects can lead to unnecessary delays in reaching the excluded population; (3) governments with insuffcient funds may be pressured to obtain loans to fnance fnancially inclusive projects, which will increase the national debt level; (4) improper delegation of authority may arise when the task of achieving fnancial inclusion is delegated to an incompetent authority or contractor

Theories of Financial Inclusion    115

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

This page intentionally left blank

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Chapter 8

Shareholder Activism: What Does It Refer to?1 Yeşim Şendur Abstract Introduction: Shareholder activism comprises a range of activities by public companies’ shareholders who desire some change in the corporation and intervene in the management’s decisions. The goals of activists are various. They may seek to change the company’s strategy, fnancial structure, management, or board in general. More specifcally they may seek to change the capital allocation strategy (stock buybacks, dividends, or company’s acquisitions policies), the board composition, the company’s executive compensation plans, or the company’s certain functions (risk management, audit).

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Purpose: The purpose of this literature review research study is to explore the concept of shareholder activism. According to a point of view, these activist actions stimulate better corporate governance practice in the companies and ultimately lead to an increase in the company’s stock price in the short term. The others claim that activism increases the company’s share price volatility in the long term. In the near future, the impact of shareholder activism will continue to rise and the ways how the companies respond to it is gaining importance. This study sheds light on the types of shareholder activism, when they are likely to approach a company and which tactics they most likely use. Methodology: Considering the rapid expansion of shareholder activism concept in the world the author makes a review of literature on shareholder activism. The structure of this chapter is as follows. First, the characteristics of shareholder activism are introduced. Second, the theoretical background of this concept is given in detail. Third, the types of shareholder activism 1

The abstract of this study was presented at the International Applied Social Sciences (IASOS) Congress on September 21–23, 2017. This study is an extended version of the abstract presented and published in the proceedings book. Uncertainty and Challenges in Contemporary Economic Behaviour, 117–126 Copyright © 2020 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-80043-095-220201009

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

118   Yeşim Şendur are discussed. Finally, the conclusion comprises a summary of shareholder activism. Findings:The study fnds out that shareholder activism has started to have a signifcant infuence on corporate governance policy that a frm adopts in recent years. Shareholder activism increases levels of shareholder engagement in frm decisions and fosters a long-term corporate governance culture. As institutional investors get a higher portion from global equity investments, their role in shareholder activism will increase. There are opinions suggesting that investor activism will lead to better corporate governance practices in frms, leading to an increase in frm share prices in the short term. The shareholder activism phenomenon seems to be on the agenda of all companies in the near future. Keywords: Shareholder activism; corporate governance; institutional investor; board of directors; investor relationship management; shareholder proposals, say on pay, Hedge fund activism JEL classifcation: G23; G29; G30

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

1. Introduction Shareholder activism comprises certain activities by public companies’ shareholders who desire some structural transformation in the company and intervene in the management’s decisions. The activists may aim to change the company’s strategy, fnancial structure, management, or board in general. More specifcally, they may aim to change the company’s capital allocation strategy, dividend payout, and mergers or acquisitions policies. Furthermore, they can give rise to stock buybacks or a change in the frm’s executive compensation plans. Sometimes environmental issues, the change of the climate, shortage of sources, and worker rights will be the reason for activism. Shareholder activism, more broadly investors’ tendency to monitor corporate insiders, is not a new phenomenon in the world. In the USA, institutional investors like insurance companies, pension funds, investment trusts, and mutual funds have been frequently intervened in the management decisions of the invested company for nearly 40 years for creating shareholder value (Brav, Jiang, & Kim, 2009, p. 1). There are some advantages to going to the public for companies. For example, public companies can raise money much easier than private ones; they can beneft from stock options for captivating and retain senior executives (Feldman, 2009, pp. 9–15). But at the same time, being a publicly held company causes some obligations for companies like increasing shareholders’ wealth. If minority shareholders perceive a risk of exploitation by the corporate insiders (the blockholders and management), they take some measures to alleviate this. For example, they may take into consideration both the frm-level corporate governance and country-level investor protection while making an investment decision (Mitton, 2004, pp. 409–411). Shareholder activism is another effective way of using power on the corporations to alleviate the exploitation.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Shareholder Activism    119 Agency theory claims that large block outside ownership may hinder managerial opportunism (Daily, Dalton, & Canella, 2003, p. 372). The large, undiversifed shareholders generally have a critical monitoring role. They can limit the behavior of managers about serving only their own interests (Zeckhauser & Pound, 1990, p. 149). Finally, the theory suggests that dividend payments stem from minority shareholders using their power on the corporations. These mentioned behaviors of large block outside shareholders and minority shareholders are the instances of shareholder activism. The law and fnance researches are interested in the different aspects of shareholder activism. The approach of law researches is to investigate shareholder activism according to the “one-share, one-vote” principle and to search if new fnancial instruments violate this principle. On the contrary, fnance researches generally have examined whether activist actions of the shareholders (shareholder proposals, say-on-pay, etc.) improve the frm performance or corporate governance or not (Ma & Liu, 2016, pp. 34–35). The structure of this chapter is as follows: Section 1 introduces the concept of shareholder activism in general. Section 2 gives information about the theoretical background of shareholder activism. The type of shareholder activism is given in detail in Section 3. In Section 4 shareholder activism is discussed according to the differences between Continental Europe and the Anglo-Saxon model of corporate governance. Finally, the conclusion comprises a summary of shareholder activism.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

2. Theoretical Background The number of researches related to shareholder activism phenomenon increases day by day, especially in the developed economies. In parallel with the development of the capital markets, the shareholder activism phenomenon is going to attract the practitioners and academics interest more in emerging economies. Jensen’s (1986) work on agency costs suggests that, if there are no alluring investment options, frms can prevent the conficts of interest between the insiders and the external shareholders by paying excess cash fows to shareholders as dividends. The confict of interest between shareholders and managers over payout policies is severe when the organization generates a big amount of cash. Payouts to shareholders reduce the resources under managers’ control, thereby reducing managers’ power. The problem is how to motivate managers to disgorge the cash rather than investing it at below the cost of capital or wasting it on organization ineffciencies. The agency theory is based on the assumption that conficts of interest arise between corporate insiders and outsiders and hence managers may conduct actions according to their self-interest, which may not always be benefcial for shareholders and such conficts lead to agency cost (AlNajjar, 2009, p. 2). Corporate governance is an internal system which is intended to work for the interests and needs of the shareholders and stakeholders by directing the company actions to the right direction and also controlling the management activities in a way that meets the overall expectations of the company’s shareholders and stakeholders (Mohamed, 2013, p. i). The presence of activist shareholders among the frm’s investors ensures a driving force of “good” corporate governance practices that should give rise to improve effciency and performance of the frm

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

120   Yeşim Şendur

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

(Shleifer & Vishny, 1997). The scandals at Enron, WorldCom, and Parmalat have led to the loss of confdence and erosion of trust in capital markets. Consequently, investors are driving companies to develop effcient corporate governance practices (Filatotchev, Jackson, Gospel, & Allcock, 2007, p. 6). According to Mitton (2004), the positive relationship between corporate governance and dividend payouts is limited primarily to countries with strong investor protection, suggesting that frm-level corporate governance and country-level investor protection are complements rather than substitutes. Shareholder activism is an important means for providing corporate governance and investor protection in a frm. Filatotchev et al. (2007) searched the literature on the “Key drivers of ‘good’ corporate governance” according to seven fundamental areas. One of these areas is shareholder activism. Their research suggests that shareholders are becoming more pro-active in terms of their infuence upon the strategic decisions that drive the frm’s dynamics and there is an important difference in the role and interests of large blockholders and smaller, portfolio-oriented investors. After then Filatotchev and Dotsenko (2013) investigate kinds of activist shareholders, their objections and results of their activities related to investor activism in the UK, and they also search the impact of activism in terms of stock-market response. Findings indicate that abnormal stock-market returns are related to the forms of activism, type of investor, and the nature of investor proposals. The role of institutional investors is growing in shareholder activism. Yücel (2015) mentioned shareholder activism in his study with the perspective of the models of corporate governance: Continental Europe Model and the Anglo-Saxon Model. The companies have a 100% free foat ratio and their biggest investors are the fnancial institutions in Anglo-Saxon countries. Thus, the biggest fear of the companies is the activist shareholders. The pyramidal ownership structure is common in Continental Europe Model and there are no principal rights for the protection of the minority shareholders. Thus, activist shareholders began to increase their effects in Continental Europe.

3. Types of Shareholder Activism The effectiveness of shareholder activism in terms of abnormal stock-market returns majorly depends on the forms of activism and the types of activists (Filatotchev & Dotsenko, 2013, pp. 5–6). Activism could be categorized as hedge fund activism, shareholder proposals, “just-vote-no” campaigns, Say on Pay advisory vote (Ma & Liu, 2016, p. 34).

3.1. Hedge Fund Activism During the past decade, hedge fund activism has emerged as a new form of corporate governance mechanism that brings about operational, fnancial, and governance reforms in the corporations (Brav et al., 2009, p. 1). Hedge fund activism occurs when an investor, usually a hedge fund or other investor aligned with a hedge fund, seeks to effect a signifcant change in the company’s strategy, fnancial structure, management, or board (https://corpgov.law.harvard.edu, 2015).

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Shareholder Activism    121 Activist investors who collaborate with hedge funds may behave according to their short-term interests that are not compatible with the company’s long-term strategies (Ergin & Erturan, 2017, p. 103; Suryanto, Thalassinos, & Thalassinos, 2017; Suryanto & Thalassinos, 2017). Fig. 1 shows the activist hedge funds in terms of geographical distribution. North America is the most favored target for activist funds to invest in, with 41% of funds, whereas only 8% of all activist funds are actively targeting Europe. The proportion of activist funds pursuing investments in the Asia-Pacifc is 15%, while 4% of funds have a specifc preference for other regions. The remaining 32% of activist funds have a more diverse regional focus and these funds take a global stance when investing (Preqin Special Report, 2014, pp. 4–5). Minority shareholders transfer the duty of controlling the company to institutional investors, thus they get rid of the controlling cost (Yücel, 2015, p. 66). Institutional investors comprise hedge funds, pension funds, mutual funds, and insurance companies. The biggest activist hedge funds are shown in Table 1 according to the asset in control. As a sample of hedge fund activism, Elliott Management, the biggest hedge fund activist of the world, went to war in the Republic of Samsung to stop the South Korean conglomerate from going through with an unfair deal in 2015. According to a point of view hedge fund activism improves a company’s stock price (at least in the short term), operational performance, and other measures

8%

4% North America

15%

41%

Global Asia-Pasific Europe Rest of World

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

32%

Fig. 1.  Activist Hedge Funds in Terms of Geographic Focus. Source: Data from Preqin Hedge Fund Analyst (2014). Table 1.  The Biggest Activist Hedge Funds in Terms of Asset under Management, 2017. Fund

Firm Location

Asset under Control

Elliott Management

USA

$56.0B

Carl Icahn

USA

$34.2B

Third Point Partners

USA

$19.4B

Value Act Capital Partners

USA

$17.6B

Source: Activist Insight, as of June 30, 2017.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

122   Yeşim Şendur of share value, but the others claim that hedge fund activism increases the company’s share price volatility in the long term.

3.2. Shareholder Proposals In some cases, institutional investors submit a proposal to ask policy changes or disclosure on some issues. Individual investors may submit lots of shareholder proposals, but they usually lack the backing to drive real change. Public pension funds, labor pension funds, or individual investors are usually sponsoring the shareholder proposals. These investors believe that these changes may promote more effective corporate governance (including more reliable fnancial reporting), and that good governance enhances shareholder value. The goal of these investors is usually to encourage one of four types of change: (1) A change to the board’s governance policies or practices or a change to the board composition (e.g., increase board diversity, name an independent director as chair); (2) a change to the company’s executive compensation plans; (3) a change to the company’s oversight of certain functions (e.g., audit, risk management); and (4) a change to the company’s behavior as a corporate citizen (e.g., political spending or lobbying, environmental practices, climate change, or resource scarcity preparedness, labor practices). Table 2 shows the number of shareholder proposals in detail according to the types. Table 2.  Common Shareholder Proposals and Their Support between January 1 and June 30, 2017.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Type of Shareholder Proposal

Examples

Number of Proposals Submitted Voted on Received Majority Support ⦁⦁ Proxy access Governance/ ⦁⦁ Board 409 200 54 shareholder declassifcation ⦁⦁ Independent ⦁⦁ Majority rights board chair voting standard ⦁⦁ Climate change ⦁⦁ Human Social/ 453 194 6 ⦁⦁ Board diversity political rights ⦁⦁ Discrimination ⦁⦁ Labor issues practices ⦁⦁ Corporate political spending ⦁⦁ Clawbacks ⦁⦁ Required Executive 97 50 0 compensation ⦁⦁ Pay ratio vesting terms disclosure for equity awards Source: PWC (2018); http://www.pwc.com/us/governanceinsightscenter/.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Shareholder Activism    123 3.3. “Just Vote No” Campaigns “Just Vote No” campaigns are low-cost shareholder activist tools whereby activists encourage their fellow shareholders to withhold votes toward a director’s election to express dissatisfaction with management performance or the frm’s corporate governance structure (Del Guercio, Seery, & Woidtke, 2008, p. 8). Generally, it would require support from a majority of outstanding shares – not just a majority of the votes cast at the meeting, which is much lower. A “just vote no” campaign can infuence the candidate to voluntarily withdraw from the election. If the level of “negative” vote was relatively signifcant, a director may be replaced during his/her subsequent term.

3.4. Say-on-Pay Many people believe that the 2008 global fnancial crisis is due to the self-seeker and excessive risk-taking behavior of executives, which are inspired by unhealthy pay practices. As a result, regulations and corporate governance reforms have been done in many countries. One of such reforms is Say-on-pay, which gives shareholders a voice in the determination of executive compensation practices (Adeleye & Akomode, 2014, p. 1). Shareholders can beneft from the right to interfere with the pay-setting process by voting on the compensation proposed by the board of directors (Ferri & Göx, 2018, p. 1). Say on pay legislation has expanded in countries around the world, including the USA, Australia, Belgium, the Netherlands, and Sweden since 2002 (Thomas & Van der Elst, 2015, p. 655).

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

4. Corporate Governance Models and Shareholder Activism Corporations must create value for stakeholders (fnanciers, customers, management, employees, government, and the community) to maintain their existence. Corporate governance has a key role in creating value for shareholders. Corporate governance is the framework of rules and practices by which a board of directors ensures accountability, fairness, and transparency in a company’s relationship with all. Moreover, the corporate governance aims to reduce the agency cost which stems from the conficting interests of stakeholders, protect the rights of minority shareholders, and fnally determine the procedures for audit and control (Miller & Modigliani, 1961, pp. 411–432). In the literature, there are two main models of corporate governance: Continental Europe Model and Anglo-Saxon Model. The mechanism of corporate governance differs both according to the countries corporate governance system and the companies own properties (Kula, 2006, p. 12). The previous research in the corporate governance area is focused on corporations with diffused ownership especially the conventional US/UK model of corporate control (Anglo-Saxon Model), little is known about the behavior of frms with concentrated ownership structure (Continental Europe Model) (Holderness & Sheehan, 1988, p. 319). The companies are large and they also have a distributed ownership structure in the Anglo-Saxon countries (USA, England, and Canada). And also they

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

124   Yeşim Şendur generally have a 100% free foat ratio and their biggest investors are the fnancial institutions. On the other hand, the managers can be a partner in the company via share purchase programs. This system provides a very strong position for managers (Silva, Goergen, & Renneboog, 2007, pp. 3–5). Nevertheless, the agent of the institutional investors fulflls the function of monitoring and auditing strong managers. Ownership and control rights are completely separated. Both the shareholders and the managers eventually try to maximize their interests. The conficts of the interests give rise to the emergence of agency costs. The biggest fear of the companies in these markets is the activist shareholders. In that, the activist investors can change the managers of the badly managed companies by lobbying. They can also exercise control over the companies to pay dividends or to merge with another company. In the model of Continental Europe, the companies include the suppliers and the workers in management. Generally, this is a legal obligation in countries like Germany and Japan. The intervention of the workers to management mitigates the agency cost and increases the transparency of the company. Sustainability of the company becomes a priority in this model of corporate governance. The free foat ratio is low and there is a concentrated ownership structure. The owners of the big companies are generally the holding companies or the banks. Besides the owners of these holding companies or banks are generally families. Holding companies or banks have complicated subsidiary relations. The pyramidal ownership structure is common in this model (Silva et al., 2007, pp. 3–5). Pyramidal ownership is a structure that the control of a corporation provided through a top-down designing chain of ownership relations. Pyramidal ownership structure causes some imbalances in the use of control in favor of the companies that are at the top of the chain. As there are no principal rights for the protection of the minority shareholders, legal sanctions become an obligation (Yılgör & Yücel, 2012). The activist shareholders began to increase their effects in Continental Europe in recent years. However, it is very limited in comparison to the USA (Yücel, 2015, p. 67). Therefore, most of the previous studies have focused on companies and investors in the USA and generally effects in a single form of shareholder activism (e.g., shareholder proposals). The importance of analyzing the issue from the perspective of emerging countries like is rising.

5. Conclusion Today all big publicly held companies are suffering from shareholder activism (Vodafone, HSBC, Samsung, etc.). Executives have to cope with the critics and interventions of activist shareholder more than ever to keep their positions (Ergin & Erturan, 2017, p. 103). The forces driving shareholder activism are several. If shareholders feel the possibility of exploitation by corporate insiders (managers or large blockholders) and a loose in their invested capital, if the company has underperformed its peers, if they feel concerns about the company’s executive compensation plan, the company will be the target of an activist movement. The issues that are should be on the agenda of the activist investors are; the

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Shareholder Activism    125 structure of the board, performance of the company, managers, wages, acquisitions, mergers, capital allocation strategy, etc. There are different ways of shareholder activism. The form of activism is closely related to the type of investor and what they want. Hedge fund activism is a sample of the activist movements of institutional investors. The increasing shares of the institutional investors in the ownership structure of the frm transform the way of shareholder activism. The other methods used by the activist investors are shareholder proposals, “just-vote-no” campaigns, say on pay advisory vote. Shareholder activism can be discussed according to the two main models of corporate governance: Continental Europe Model and Anglo-Saxon Model. The possibility of being a target of an activist movement is higher in the Anglo-Saxon countries than in Continental Europe countries. Because most of the companies in the Anglo-Saxon countries are open to the public and have a diffused ownership structure. In the Continental Europe countries, the free foat ratio is low and the concentrated ownership structure is widespread among the companies. But with the increase in the free foat ratio of the companies and the legal protection of the shareholders, shareholder activism has started to expand in Continental Europe in recent years. To avoiding shareholder activism, it is important to understand the reasons that will cause the movement. When activists knocking on the door, directors should listen to the demands and try to provide a consensus. The government and investor associations promote shareholder activism in general. In the near future, the impact of shareholder activism will continue to rise and the ways how the companies respond to it is gaining importance.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

References Adeleye, I., & Akomode, T. (2014). Executive compensation and say-on-pay: Recent developments and lessons from around the world. Journal of Corporate Governance, 80(3), 1. Al-Najjar, B. (2009). Dividend behaviour and smoothing new evidence from Jordanian panel data. Studies in Economics and Finance, 26(3), 182–197. Brav, A., Jiang, W., & Kim, H. (2009). Hedge fund activism: A review. Foundations and Trends in Finance, 4(3), 1–16. Del Guercio, D., Seery, L., & Woidtke, T. (2008). Do boards pay attention when institutional investor activists “just vote no”? Journal of Financial Economics, 90(1), 84–103. Ergin, E., & Erturan, İ. E. (2017). Shareholder activism: Telecommunication industry in Turkey. Marmara University Faculty of Economics and Administrative Sciences Publication, 39(1), 101–116. Feldman, D. N. (2009). Reverse mergers: And other alternatives to traditional IPOs. New York, NY: Bloomberg Financial Series. Ferri, F., & Göx, R. F. (2018). Executive compensation, corporate governance, and say on pay. Foundations and Trends in Accounting, 12(1), 1–103. Filatotchev, I., & Dotsenko, O. (2013). Shareholder activism in the UK: Types of activists, the form of activism, and their impact on a target’s performance. Journal of Management & Governance, 19(1), 5–24.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

126   Yeşim Şendur Filatotchev, I., Jackson, G., Gospel, H., & Allcock, D. (2007). Key drivers of ‘good’ corporate governance and the appropriateness of UK policy responses (pp. 27–33). London: The Department of Trade and Industry and King’s College London. Holderness, C. G., & Sheehan, D. P. (1988). The role of majority shareholders in publicly held corporations: An exploratory analysis. Journal of Financial Economics, 20(1–2), 317–346. Jensen, M. C. (1986). Agency cost of free cash fow, corporate fnance, and takeovers. The American Economic Review, 2(76), 323–329. Kula, V. (2006). Kurumsal yönetim: hissedarların korunması uygulamaları ve Türkiye örneği [Corporate governance: The case of Turkey practices and the protection of shareholders]. İstanbul: Papatya Yayıncılık. Ma, V. C., & Liu, S. J. (2016). Exploring the research fronts and main paths of literature: A case study of shareholder activism research. Scientometrics, 109, 33–52. Miller, H. M., & Modigliani, F. (1961). Dividend policy, growth, and the valuation of shares. The Journal of Business, 4, 411–432. Mitton, T. (2004). Corporate governance and dividend policy in emerging markets. Elsevier Publishing, 5, 409–426. Mohamed, H. O. S. (2013). The infuence of corporate governance on dividend policy. Yüksek Lisans Tezi, Yıldız Teknik Üniversitesi, İstanbul. Preqin Hedge Fund Analyst. (2014). Preqin Ltd. Retrieved from http://www.preqin.com/ Preqin Special Report. (2014). Hedge Fund activist report. Retrieved from http://docs.preqin.com/reports/Preqin_Special_Report_Activist_Hedge_Funds_June_14.pdf PWC. (2018). How might the changing face of shareholder activism affect your company? Retrieved from http://www.pwc.com/ Shleifer, A., & Vishny, R. W. (1997). A survey of corporate governance. The Journal of Finance, 52(2), 748–749. Silva, L. C., Goergen, M., & Renneboog, L. (2007). Dividend policy and corporate governance. Oxford: Oxford University Press. Suryanto, T., Thalassinos, E. Y., & Thalassinos, I. E. (2017). Board characteristics, audit committee and audit quality: The case of Indonesia. International Journal of Economics & Business Administration, 5(3), 47–57. Suryanto, T., & Thalassinos, I. E. (2017). Cultural ethics and consequences in whistleblowing among professional accountants: An empirical analysis. Journal of Applied Economic Sciences, 6(52), 1725–1731. Thomas, R. S., & Van der Elst, C. (2015). Say on pay around the world. Washington University Law Review, 92(3), 655. Yılgör, A., & Yücel, E. (2012). Examination of the frms’ ownership structure: Deductions for ownership and control divergence. International Journal of Management Economics and Business, 16, 41–57. Yücel, D. C. (2015). Türkiye’de aktivist yatirimci olur mu? [Would it be activist investors in the future of Turkey?] Kurumsal Yönetim Dergisi [Corporate Governance Journal] – TKYD, 27, 64–67. Zeckhauser, R. J., & Pound. J. (1990). Are large shareholders effective monitors? An investigation of share ownership and corporate performance. In R. G. Hubbard (Ed.), Asymmetric ınformation, corporate fnance, and ınvestment (pp. 149–180). London: University of Chicago Press.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Chapter 9

Relationship Between Interest Rates, Exchange Rate and Investor Sentiment in Turkey N. Serap Vurur Abstract

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Purpose: Investor sentiment in fnancial markets has a close relationship with the general mood prevailing in the environment such as economic, social and political life. Future economic expectations are important for both investors and policymakers. Investor sentiment and macroeconomic variables are likely to affect each other. Emerging countries are particularly sensitive to interest and foreign exchange risk. Turkey is an important emerging country. The effects of interest rate and exchange fuctuations are high in this country. The aim of this study is to reveal the relationship between investor sentiment and interest and foreign exchange rates in Turkey. Methodology: This study investigates the relationship between economic confdence index, exchange rates and interest rates in Turkey during the period between January 2012 and November 2019 using monthly data sets. The economic confdence index is used to represent the investor sentiment in the study. Interest rate variables are the deposit interest rates and the commercial credit interest rates. The representative of the US dollar currency variables is included in the analysis. This chapter used the time series vector error correction model approach of stationarity test, cointegration test and Granger causality test. Findings: According to the causality test, there is a two-way relationship between economic confdence index and exchange rate, and there is unidirectional causality from commercial credit interest rate to economic confdence index. The results show that foreign exchange and commercial credit interest rate variables are carefully monitored by market players and are effective and infuential in the formation of future expectations.

Uncertainty and Challenges in Contemporary Economic Behaviour, 127–138 Copyright © 2020 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-80043-095-220201010

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

128    N. Serap Vurur Originality/value: The study shows the direction of the relationship between economic confdence foreign exchange and commercial credit interest rate. Policymakers can shape expectations by taking into account the direction of the relationship. Keywords: Investment sentiment; exchange rates; interest rate; Johansen cointegration test; VECM approach Granger causality analysis; behavioral fnance JEL classifcation: G40; C58

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

1. Introduction All theories developed up to behavioral fnance theory assume that investors behave rationally. Traditional fnance theory remained insuffcient in explaining investor behavior. This situation caused the behavioral fnance theory to come to the foreground. Behavioral fnance theory suggests that investors make decisions based on psychological factors, not rational expectations. The traditional approach assumes that market participants are rational and only take statistical data into account while guiding their investments. In traditional models, investors have no role in forming asset prices. Behavioral fnancier says that noisy investors can affect fnancial markets and infuence asset prices due to the limited arbitrage opportunities (De Long, Shleifer, Summers, & Waldmann, 1990; Shleifer & Vishny, 1997). Noisy investors are also driving prices away from rational expectations (Gallimore & Adelaide, 2002). Fama (1981), Fischer and Merton (1984) and Barro (1990) revealed in their applied studies that the change in stock returns can predict changes in total real investment. Recent studies show that some of the investors may be rational, but the majority of them act on lack of knowledge, inability to evaluate information or make the same choices of someone else. Baker and Wurgler (2006) defne the irrational behavior of investors as “investor sentiment.” Some of the researchers such as Barberısa, Shleiferb, and Vishnya (1998), Brown and Cliff (2004) and Daniel, Hirschleifer, and Subrahmanyam (1998) describe investor sentiment as the tendency of individual investors to noise trade rather than information, while others have used it specifcally to address investor optimism or pessimism. Empirical research does not yet reveal how to accurately defne and measure “investor sentiment.” Investor sentiment in fnancial markets is closely related to the general expectations prevailing in the economic, social and political environment. The fact that fnancial market participants have different expectations, attitudes and behaviors based on sociological and psychological factors in their investment decisions makes it diffcult to predict future fnancial market conditions. At this point, leading economic indicators play an important role in predicting the contractions and growths that may occur in forward-looking economic activities (Pociovalisteanu, Thalassinos, Tirca, & Filho 2010). Confdence indices are also used as an indicator of investor sentiment which are also among the leading indicators of the economy (Ugurlu, Thalassinos, & Muratoglu, 2014).

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Investor Sentiment in Turkey    129 The economic confdence index is calculated since 2012 in Turkey and is one of the important leading indicators of Turkey. It will be used as an indicator of investor sentiment. The economic confdence index is a leading indicator summarizing the expectations and trends of consumers and producers regarding the general economic situation. The index consists of consumer confdence indicator and seasonally adjusted manufacturing (real sector), service, retail trade and construction industry confdence indicators (Tuik, 2019). With this feature, the economic confdence index shows the confdence in the economy not only of one certain segment but also of different segments. The fact that the economic confdence index is greater than 100 indicates the investors’ optimism about the market, and the fact that it is less than 100 indicates the pessimism about the market (Ozen, 2019). As one of the emerging countries, Turkey’s economy has been in the process of liberalization since the early 1980s. Exchange rate and interest rates appear as two very important factors affecting the country’s economy. Hopper (1997) revealed in his study that the exchange rate was affected by investor sentiment more than economic factors. Interest rates are usually determined by market actors. It affects investor behavior by shaping it according to economic expectations (Ozen, Vurur, & Grima 2018; Tufan, 2008). In the literature, Serna and Arribas (2008), Andreou, Osborn, & Sensier (2000) and Fornari and Mele (2006) studies have demonstrated the ability of interest rates to predict real economic activities and being a leading indicator. Empirical studies on the subject have been conducted in developed countries. The structural differences in developing countries may produce different results. The effects of these relations in developing countries remain uncertain. Turkey is among the rapidly growing countries of recent years, so it is thought that it will be useful to reveal these relations in this case. At the same time, this study includes innovation in terms of examining the relationship between both interest and exchange rate variables and investor sentiment at the same time, unlike previous studies. In the following sections of the study, primarily the studies examining investor sentiment are included; then, the data set, variables and methodology used are explained. In the last part of the study, the results regarding the investigation of the relationship between investor sentiment and interest and exchange rate, evaluations about these results and suggestions for future studies are included.

2. Literature Studies in the literature are focused on investor sentiment and stock market index return. The frst study using the confdence index to represent investor sentiment was done by Ottoo (1999). Ottoo (1999) found that changes in stock returns had a 10% impact on trust, but no relationship from trust to return was revealed. Macroeconomic variables have been used as a control variable together with sensitivity measurements (Debata, Dash, & Mahakud, 2018; Fernandes, Gonçalves, & Vıerıa, 2010; Ferrer, Salaber, & Zalewska, 2016; Lemmon & Portnıaguına, 2006; Namourı, Jawadi, Ftıtı, & Hachicha, 2018). Lemmon and Portnıaguına (2006) used confdence as a measure of investor optimism and examined the time series relationship between investor sentiment

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

130    N. Serap Vurur and stock return. No relationship was found between confdence, stock returns and macroeconomic activity in the period before 1977. However, in the last 25 years, it has become a much better barometer of its confdence, economic activities and investor attitudes. Fernandes et al. (2010), as a representative of investor sentiment, investigated whether the economic sensitivity index predicts future total stock market returns and industrial index returns in Portugal between September 1997 and April 2009. Even after including macroeconomic factors, they found the relationship between expected return and sensitivity signifcantly negative. Ferrer et al. (2016) investigated the stock market–confdence relationship for the USA and 12 EU countries through case studies. For each country, they included confdence indices, stock market indices and macroeconomic variables between January 1990 and December 2010. When the fndings are analyzed, it was concluded that the confdence index–stock relationship is not universally positive. Namourı et al. (2018) investigated the relationship between investor sentiment and stock market returns for G7 countries between June 1987 and February 2014. During the analysis period, G7 countries added macroeconomic variables as a control variable to the model it created to measure the effects of investor sentiment on stock return. Among the variables added were interest rates industrial production growth and world stock market returns. They found that investor sentiment affects stock returns nonlinearly, but the effects did not change according to market conditions (Thalassinos, Ugurlu, & Muratoglu, 2015). Debata et al. (2018) examined the impact of local and foreign investor sentiment on the stock market liquidity of 12 developing countries by including macroeconomic variables in the analysis. When the results are analyzed, it is seen that the sensitivity of foreign investors has affected the emerging stock market liquidity signifcantly. While measuring the relationship between investor sentiment and stock return, there are studies that examine interest rate, exchange rate and investor sentiment as leading indicator features, apart from studies that include macroeconomic variables (Andreou, Osborn, & Sensier, 2000; Fornari & Mele, 2006; Serna & Arribas, 2008). Serna and Arribas (2008) investigated the relationship between interest rate volatility and confdence in Germany and England. The relationship between investor sentiment and interest rate volatility is negative. Volatility in interest rates explains 70% of confdence in Germany and the UK. The leading indicator feature of interest rates in Andreou et al. (2000) was also explored. Andreou et al. (2000) determined that the interest variable is the most important leading indicator for the USA, England and Germany. It also tested the predictive power of this variable in production volatility. Fornari and Mele (2006), on the other hand, investigated the predictive power of interest rate for economic activities in the USA. However, very little analysis focused on the predictive power of the interest rate. The investor sentiment relationship of the exchange rate was revealed by Hopper (1997). Hopper (1997) found that the exchange rate was more infuenced by investor sentiment than economic factors. Cairns, Ho, and Mc Cauley (2007) analyzed the exchange rate investor sentiment relationship in Asia Pacifc countries by

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Investor Sentiment in Turkey    131 using the VIX index as a sensitivity indicator. Cairns et al. (2007) concluded that there is a positive relationship between investor sentiment and the exchange rate. Dae-Sun (2012) found a high relationship between exchange rate and sensitivity in his study in the Korean market. Rahman (2018) revealed the relationship between the exchange rate and investor sentiment in his study in Indonesia. Investor sentiment ensures that the exchange rate remains stable. Thalassinos and Politis (2011, 2012) in their studies used cointegration analysis to investigate the international stock market and the evaluation of the USD currency and the oil prices. In the literature, there are studies that examine the relationship between interest rate, exchange rate and investor sentiment separately. A study examining the relationship between interest rates, exchange rate variables and investor sentiment could not be found.

3. The Aim of the Study In this study, the author will try to examine the interaction between investor sentiment exchange rate and interest rates. In the study, the relationship between the economic confdence index representing investor sentiment for the periods January 2012 and November 2019 and the relationship of causality and cointegration between exchange rate and interest rates were investigated. In the literature, a study examining the relationship between investor sentiment and interest and exchange rates together could not be reached. Therefore, it is thought that the study will contribute to the literature.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

4. Hypotheses of the Study In this study, the relationship between investor sentiment and interest rates and the exchange rate was estimated empirically. This study is looking for answer to the question of whether there is a relationship between the exchange rate, the interest rate and investor sentiment in Turkey’s fnancial market? The null hypotheses of the study created for interest rates and exchange rate variables are: NH1. There is no causality relationship between investor sentiment and deposit interest. NH2. There is no causality relationship between investor sentiment and commercial credit interest. NH3.  There is no causality relationship between investor sentiment and exchange rate.

5. Data Methodology and Findings In this study, the relationship between interest rates, exchange rate, and investor sentiment will be investigated. For this purpose, the dollar rate was taken as the

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

132    N. Serap Vurur exchange rate. Deposit rates and commercial loan rates were used as interest rates. In this section, the economic confdence index is used for investors’ sentiment in Turkey (Confdence). The variable defnitions used in the study are as follows: CFD: The economic confdence index. EXC: Exchange rate (Dollar). DIR: Deposit interest rate. CIR: Commercial interest rate. For the purpose of this study, the series consisting of monthly data for the period between January 2012 and November 2019 were used. Confdence index data sets from Turkey Statistical Institute, exchange rate, interest rates data sets from the Central Bank of the Republic of Turkey electronic data distribution service are provided. We use Eviews 9.0 on all the empirical analysis. Descriptive statistics on the variables in the study are shown in Table 1. As can be seen in Table 1, the Kurtosis value of all series except the exchange rate series is above 3, which shows that these series are not fat. Since the skewness coeffcient indicating the skewness value of the series is greater than zero in all series, the series is skewed to the right. Confdence has the highest standard deviation among the series, while the exchange rate series has the lowest standard deviation. Finally, the Jarque–Bera coeffcient was reviewed to see that the series did not disintegrate normally. Since the Jarque–Bera coeffcient is statistically signifcant, the data are not normally distributed.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Table 1.  Descriptive Statistics. CFD

EXC

DIR

CIR

Mean

99.9213

2.865746

10.21152

14.87611

Median

101.2640

2.278867

8.932362

14.33641

Maximum

115.7296

6.207686

22.72745

33.13297

Minimum

76.10980

1.407480

5.273166

8.431895

Std.Dev.

8.462385

1.362736

4.141114

5.231224

Skewness

−0.983482

1.024737

1.798593

1.491842

Kurtosis

3.723981

2.898924

5.369666

5.289818

Jarque–Bera

21.78244

20.887734

92.00219

70.13880

Probability

0.000019

0.0000029

0.000000

0.00000

Sum

11,890.61

341.0238

1,215.171

17,770.258

Sum Sq.Dev.

8,450.212

219.1318

2,052.987

3,229.153

Observations

95

95

95

95

Source: Author’s own estimation based on the Eviews 9 statistical software.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Investor Sentiment in Turkey    133 The most valid method used to determine whether a variable is stationary or not and the degree of stationarity is the unit root test (Gujarati, 2004). The most commonly used unit root tests in practice are Augmented Dickey–Fuller (ADF) and Phillips–Perron (PP) unit root tests. ADF and PP tests were used in the study to test the stability of variables. Table 2 shows the unit root test results for the variables. In ADF and PP tests, the H0 (basic hypothesis) series have a unit root, and it is not stationary. Since the absolute values of t-values obtained for ADF test statistics are less than the absolute values of 1% and 5% signifcance levels, the series have unit roots, that is, they are not stationary in the level values of the series. PP test statistic also supports the ADF test statistic. It was concluded that in the frst differences of the non-stationary series in the level values, there is no unit root, that is, the degree of integration is I (1) (Table 2). In the ADF and PP tests, the Null hypothesis (H0) is set up as the series having a unit root, that is, it is not stationary. After determining that they are stationary in the frst differences of the series, the existence of a long-term equilibrium relationship between the series was investigated according to the cointegration method developed by Johansen (1988) and Johansen and Juselius (1990). In order to perform cointegration test, it is necessary to estimate an unconstrained VAR model with the variables used in the model and determine the number of lags of the model. In determining the number of lags the; Likelihood Ratio (LR), Final Prediction Error (FPE), Akaike Information Criteria (AIC), Schwarz(SC) and Hannan Quinn (HQ) criteria were used.

Table 2.  Unit Roots Test.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

ADF Test Intercept

Trend and Intercept

Intercept

0.422232

Series LOGEXC Level Difference LOGCFD Level Difference LOGDIR LOGCIR

Level

PP Test Trend and Intercept

−3.129603

−0.630254

−2.182706

−7.806157* −7.806157*

−6.7360*

−7.168556*

−1.753495

−1.488954 −3.314894***

−3.325024

−11.45267* −11.42448* −11.4677* −1.90488

−2.902786

−11.46773*

−1.399657

−2.383823

Difference

−5.403694* −5.369658* −5.29772*

−5.262930*

Level

−2.286882

−1.84186

−2.287503

Difference

−6.289353* −6.334126* −6.27175*

−6.187981*

−2.937701

Critical values * 1%

−3.43344

−3.96266

−3.43344

−3.96266

***10%

−2.56748

−3.12794

−2.56748

−3.12794

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

134    N. Serap Vurur In this framework, the appropriate optimal lag length for VAR analysis is four according to FPE and AIC criteria, and two according to SC and HQ criteria. The lag length will be taken as two based on the SC Criterion (Table 3). The linear combinations of non-stationary series at a certain integration level are called cointegration. Cointegration tests question whether two time series of the same degree of stationary move together in the long term. In other words, if the series becomes stationary at the same level, there is a long-term relationship (cointegration) between the series. Since the series is stationary at the I (1) level, Johansen cointegration analysis was performed. A cointegration test is used to investigate the existence of a long-term relationship between the series. According to the maximum likelihood approach of Johansen (1988) and Johansen and Juselius (1990), they use maximum eigenvalue and trace statistics to test whether there is a long-term relationship between variables. In the study, both statistical values were taken into consideration for the test. Johansen cointegration test in Table 4 for two lag

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

lengths is also given. As seen in Table 4, the zero hypothesis was rejected, which indicates that there is no long-term relationship between the confdence, exchange rate and interest rate series because trace statistics and maximum eigenvalue statistics values were greater than the critical values at the 1% and 5% signifcance levels. Therefore, it is concluded that there are at least two long-term relationships between the series. At the same time, the cointegration test seems necessary to make an error correction model (ECM). In the long term, the ECM was applied to correct inconsistencies between variables. This analysis was made with the ECM. This model is used to correct imbalances between variables. There is a long-term equilibrium assumption among the variables, and there are deviations from this long-term equilibrium in the short term. ECM is applied to determine how long these deviations will take to disappear in the long term. After the cointegration relationship is found, causality analysis should be done between the series. The presence of cointegration between variables suggests a long-term relationship among the variables under consideration. Then, the vector error correction model (VECM) can be applied. The long-term relationship between economic confdence, USD, deposit interest, commercial credit interest is one cointegrating Table 3.  The Determination of Vector Autoregressive (VAR) Lag Length. Lag 1

LR

FPE

AIC

SC

HQ

1,031.112

0.020190

7.448684

7.936888

7.646734

*

6.878783*

2

123.8972

0.008004

6.522294

7.401060

3

32.22139

0.007709

6.481792

7.751122

6.996722

4

31.05875

0.007431

a

6.439668

*

8.099561

7.113038

5

17.43990

0.008239

6.534180

8.584635

7.365989

Source: Author’s own estimation based on the Eviews 9 statistical software. * indicates lag order selected by the criterion. Maximum lag interval was chosen as 2.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Investor Sentiment in Turkey    135 Table 4.  Results of Johansen Cointegration Tests. Trace Test

5% Critic Value

Probability**

Cointegration

a

54.07904

0.0005

None*

a

41.92590

35.19725

0.0081

At most 1*

17.61590

20.26184

0.1112

At most 2

8.635565

9.164546

0.0629

At most 3

Max.Eigen Statics

%5 Critic Value

Probability

Cointegration

30.90565

25.5808

0.0248

None*

24.31000

22.2962

0.0259

At most 1*

8.979344

15.89210

0.4361

At most 2

8.636555

9.164546

06.029

At most 3

72.83155

a

Trace test indicates two cointegrating equation(s) at the 0.05 level. Rejection of the hypothesis at the 0.05 level. ** MacKinnon–Haug–Michelis (1999) p-values.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

*

vector for Turkey in the period between January 2012 and December 2019. The correction coeffcient was found as −0.158329 according to the error results in Table 5. Accordingly, approximately 0.1583 of the deviations occurring in a short time in economic confdence disappear monthly. So these deviations are 1/| ECM| = 1/|0.158329| which will reach the balance of the long term in a period of approximately seven months. Finally, we build the VECM based on the VAR model and use the Granger causality test to determine the causal relationship between confdence index and other variables. The results from the Granger causality test applied to the ECM are reported in Table 6. According to the causality test, there is a bidirectional relationship between the economic confdence index and exchange rate, and there is unidirectional causality from commercial credit interest rate to economic confdence index. Accordingly, the NH1 hypothesis was accepted, as there was no causality between deposit interest and investors’ sentiment. NH2 hypothesis, which states that there is no relationship between investors’ sentiment and commercial credit interest rate, has been partially accepted. Because there is a causality relationship from commercial credit rate to investors’ sentiment. NH3 hypothesis, which states Table 5.  Vector Error Correction Model. Cointegration Equation CointEq1

D (Economic Confdence)

D (USD)

D (Deposit Interest)

D (Commercial Credit Interest)

−0.158329

−0.000366

0.015314

0.028207

(0.04034)

(0.00203)

(0.00939)

(0.01533)

[−3.92529]

[−0.18007]

[1.63011]

[1.84039]

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

136    N. Serap Vurur Table 6.  Vector Error Correction Granger Causality Test Result. Null Hypothesis CFD does not granger cause exchange rate

ChiSq. 32.04731

Prob.

Result

0.000

H0: Rejection

EXC does not granger cause confdence

8.395202

0.0150

H0: Rejection

CFD does not granger cause DIR

1.843080

0.3979

H0: Accepted

DIR does not granger cause CFD

0.798713

0.6708

H0: Accepted

CFD does not granger cause CIR

1.843080

0.3970

H0: Accepted

CIR does not granger cause CFD

5.747373

0.004

H0: Rejection

Note: Author’s own estimation based on the Eviews 9 statistical software.

that there is causality relationship between exchange rate and investors’ sentiment, was rejected.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

6. Conclusion In this chapter, the causality relationship between interest rates, exchange rate and investors’ sentiment was investigated in the January 2012–November 2019 period. Average interest rates applied to deposits and commercial credit by banks were taken as interest variables. The economic confdence index was used as an indicator of investors’ sentiment. The study aimed to establish the interest rate and exchange rate relationship with investors’ sentiment in the case of Turkey. For this purpose, the Johansen cointegration test, ECM and Granger causality test were used. Cointegration test results show that there are at least two long-term relationships between variables. As a result of the analysis, no causality relationship was found between deposit interest rate and investors’ sentiment However, the unidirectional causality relationship was determined from commercial credit rates to investors’ sentiment. These results support the work of Serna and Arribas (2008), Andreou et al. (2006) and Mele (2006). It confrms the expectation that periodic changes in interest rates will contain information about the future path of the real economy. A bidirectional causality relationship was found between the exchange rate and investors’ sentiment. The result obtained is consistent with the result of Hopper (1997). The exchange rate and investors’ sentiment in Turkey are seen as mutually infuencing each other. The fragility of the Turkish economy against the exchange rate may be the reason for the interaction of investors’ sentiment with

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Investor Sentiment in Turkey    137 the exchange rate. The obtained results show that commercial credit interest and exchange rates can be used to guide investors’ expectations. However, future research would be useful to measure the direction of the interaction but also its response to positive and negative shocks so that it can be interpreted. It will be useful to consider the relationship between positive and negative shocks and investors’ sentiment in future studies.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

References Andreou, E., Osborn, D. R., & Sensier, M. (2000). A comparison of the statistical properties of fnancial variables in the USA, UK and Germany over the business cycle. The Manchester School, 68(4), 396–418. doi:10.1111/1467-9957.00202 Baker, M., & Wurgler, J. (2006, August). Investor sentiment and the cross-section of stock returns. The Journal of Finance, LXI(4), 1645–1680. doi:10.1111/ j.15406261.2006.00885.x Barberısa, N., Shleiferb, A., & Vishnya, R. (1998, September). A model of investor sentiment. Journal of Financial Economics, 49(3), 307–343. doi:10.1016/S0304405X(98)00027-0 Barro, R. J. (1990). The stock market and investment. Review of Financial Studies, 3(1), 115–131. doi:10.1093/rfs/3.1.115 Brown, G. W., & Cliff, M. T. (2004, January). Investor sentiment and the near-term stock market. Journal of Empirical Finance, 11(1), 1–27. doi:10.1016/j.jempfn. 2002.12.001 Cairns, J., Ho, C., & Mc Cauley, R. (2007). Exchange rate and global volatility: Implications for Asia Pacifc currencies. BIS Quarterly Review, March, 41–52. Daniel, K., Hirschleifer, D., & Subrahmanyam, A., (1998, December). Investor psychology and security market under- and overreactions. The Journal of Finance, 53(6), 1839–1885. doi:10.1111/0022-1082.00077 Dae-Sun, J. (2012), The effect of global fnancial volatility on won/dollar exchange rate, Samsung Economic Research Institute. 86–89. Debata, B., Dash, S. R., & Mahakud, J. (2018). Investor sentiment and emerging stock market liquidity. Finance Research Letters, 26, 15–31. doi:10.1016/J. Frl.2017.11.006 De Long, J. B., Shleifer, A., Summers, L., & Waldmann, R. (1990). Noise trader risk in fnancial markets. Journal of Political Economy, 98, 703–738. Fama, E. (1981). Stock returns, real activity, infation, and money. The American Economic Review, 71(4), 545–565. Retrieved from www.jstor.org/stable/1806180. Accessed on January 1, 2020. Fernandes, C. M. D. A., Gonçalves, P. M. M. G., & Vıerıa, E. F. S. (2010). Does sentiment matter for stock market returns? Evidence from a small European market. Journal of Behavioral Finance, 14(4), 253–267. doi:10.1080/15427560.2013.848867 Ferrer, E., Salaber, J., & Zalewska, A. (2016). Consumer confdence indices and stock markets’ meltdowns. The European Journal of Finance, 22(3), 195–220. doi:10.1080/135 1847X.2014.963634 Fischer, S., & Merton, R. C. (1984). Macroeconomics and fnance: The role of the stock market. NBER Working Paper Series, Working Paper No. 1291, National Bureau of Economic Research. Fornari, F., & Mele, A. (2006). Financial volatility and real economic activity. Working Paper. Retrieved from http://fmg.lse.ac.uk/~antonio/fles/predict.pdf. Accessed on December 10, 2019.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

138    N. Serap Vurur Gallimore, P., & Adelaide, G. (2002). The role of investor sentiment in property investment decisions. Journal of Property Research, 19, 111–120. doi:10.1080/09599910110110671.x Gujarati, D. N. (2004). Basic Econometrics, USA: The MC-Graw, Hill Companies. Hopper, G. P. (1997). What determines the exchange rate: Economic factors or market sentiment? Business Review, September/October, 17–29. Johansen, S. (1988). Statistical Analysis of Cointegrating Vectors. Economic Dynamics and Control, 12, 231–54. Johansen, S., & Juselius, K. (1990). Maximum likelihood estimation and inference on cointegration with application to the demand for money. Oxford Bulletin of Economics and Statistics, 52, 169–210. Lemmon, M., & Portnıaguına, E. (2006). Consumer confdence and asset prices: Some empirical evidence. The Review of Financial Studies, 19(4), 1499–1529. doi:10.1093/ Rfs/Hhj038 Namourı, H., Jawadi, F., Ftıtı, Z., & Hachicha, N. (2018). Threshold effect in the relationship between investor sentiment and stock market returns: A PSTR specifcation. Applied Economics, 50(5), 559–573. doi:10.1080/00036846.2017.1335387 Ottoo, M. W. (1999). Consumer sentiment and stock market. fnance and economics discussion series from board of governors of federal reserve system. Retrieved from https://www.federalreserve.gov /pubs/feds/1999/199960/199960pap.pdf. Accessed on December 25, 2019. Ozen E. (2019, May–August). The concept of trust in socio-economic life. European Journal of Marketing and Economics, 2(2), 69–74. Ozen, E., Vurur, N. S., & Grima, S. (2018, November). Investigation of causality between interest rate and deposit investor’s behaviour. Broad Research in Artifcial Intelligence and Neuroscience, 9(4), 177–185. ISSN 2067-3957 Pociovalisteanu, M. D., Thalassinos, I. E., Tirca, A., & Filho, L. W. (2010). Trends and challenges in the energy sector of Romania in the post-accession to the European Union. International Journal of Environmental Technology and Management, 12(1), 3–15. doi:10.1504/IJETM.2010.029957 Rahman, R. E. (2018). Understanding Indonesia’s exchange rate behavior. Retrieved from www.emeraldinsight.com/1086-7376.htm. Accessed on December 15, 2019. Serna, I. M., & Arribas, E. N. (2008). The predictive power of interest rate volatility on economic sentiment: Evidence for Germany and the U.K. Retrieved from https://pdfs. semanticscholar.org/a440/c1fc60fd37775ebf453be1e7310c2efcbb54.pdf. Accessed on December 25, 2019. Shleifer, A., & Vishny, R. W. (1997, March). The limits of arbitrage. The Journal of Finance, LII(1), 35–55. doi:10.1111/j.1540-6261.1997.tb03807.x Thalassinos, I. E., & Politis, D. E. (2011). International stock markets: A co-integration analysis. European Research Studies Journal, 14(4), 113–129. Thalassinos, I. E., & Politis, D. E. (2012). The evaluation of the USD currency and the oil prices: A VAR analysis. European Research Studies Journal, 15(2), 137–146. Thalassinos, I. E., Ugurlu, E., & Muratoglu, Y. (2015). Comparison of forecasting volatility in the Czech Republic stock market. Applied Economics and Finance, 2(1), 11–18. Tuik. (2019). TUIK source merged with the upper line. Retrieved from www. tuik.gov.tr. Accessed on December 5, 2019 Tufan, E., (2008). Behavioral fnance, investigation of social psychological anomalies directing fnancial decisions in behavioral fnance context and evaluation in terms of ISE. Ankara: Imaj Publisher. Ugurlu, E., Thalassinos, E., & Muratoglu, Y. (2014). Modeling volatility in the stock markets using GARCH models: European emerging economies and Turkey. International Journal of Economics and Business Administration, 2(3), 72–87. Retrieved from www. tuik.gov.tr. Accessed on December 5, 2019.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Chapter 10

Effect of Climate Change on Financial Institutions and the Financial System Peterson K. Ozili Abstract Climate change is emerging as an important issue increasing uncertainty in the business circle, and fnancial institutions through their inaction seem to be unmoved by climate change risk despite the potential for climate change events to affect the fnancial institutions and the fnancial system. In this chapter, the effect of climate change on fnancial institutions and the fnancial system are highlighted and discussed. Keywords: Climate change; fnancial institutions; fnancial system; risk; insurance; banks; payment system; fnancial markets; regulation; policy; capital JEL classifcation: G21; G28; Q54

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

1. Introduction I highlight and discuss the effect of climate change on fnancial institutions and the fnancial system. Nature disrupting the delivery of fnancial services is nothing new in world history. Financial institutions have coped with foods, typhoons, mudslides, earthquakes, droughts and storms since the nineteenth century and will continue to do so in future. But what has changed now? Two things have changed! First, fnancial institutions have grown complex and global. This is true because most fnancial institutions have contractual cross-border fnancial obligations and claims with fnancial institutions in many countries around the world, and many fnancial institutions have at least one branch in countries that are prone to climate change events. Secondly, as the number of fnancial institutions’ branch networks continue to multiply in tropical areas where there are extreme weather conditions, the risk of physical damage to fnancial institutions’

Uncertainty and Challenges in Contemporary Economic Behaviour, 139–144 Copyright © 2020 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-80043-095-220201011

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

140    Peterson K. Ozili fxed assets increases too (Akhmadeev, Bykanova, Philippova, Vashchekina, & Turishcheva, 2018; Osadchy et al., 2018) The risk of severe climate change is rising and posing physical threats to many frms, not only fnancial institutions. It is important for us to start worrying about climate change now. Climate change is both a short-term and long-term risk to businesses, and it affects every industry in some way. It manifests differently in nearly every industry with varying impacts across several industries. In fnancial institutions, climate change can have serious consequences, making it important to bring it to the attention of everyone. Financial institutions are frms actively involved in the fow of money and credit from surplus units to defcit units of the economy. These generally comprises of banks, investment companies and insurance companies. Severe climate change also hurts the fnancial system through the negative effects of climate change on fnancial institutions (Rupeika-Apoga, Zaidi, Thalassinos, & Thalassinos, 2018; Thalassinos & Stamatopoulos, 2015).

2. Effect on the Financial Sector 2.1. Climate Change Can Damage Physical Collateral When banks lend to corporate and individual borrowers whose ability to repay is doubtful, banks would require physical collateral whose value is greater than the loan facility issued to them. Such physical collaterals may include residential estates, commercial properties, feet of automobiles or aircraft, equipment, lands, etc. Extreme weather events such as fooding, landslide and typhoons can signifcantly damage some or all of the physical (and immovable) collateral intended to mitigate the exposure to credit risk. The resulting damage to collateral value caused by climate change would be refected as a loss in the bank’s proft and loss statement which would negatively affect the proftability of the affected bank.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

2.2. Climate Change Can Disrupt the Operations of Financial Institutions Financial institutions rely heavily on communication systems through information technology infrastructure. Any damage to such infrastructure due to severe weather or hurricanes can disrupt the business of fnancial institutions for a considerable period of time before recovery. Climate change events such as hurricanes can damage the infrastructure used by fnancial institutions to communicate within the frm such as the internet and IT systems and mast. Severe climate change can disrupt the ability of banks to communicate within the frm, thereby disrupting their operations.

2.3. Climate Change Can Disrupt the Business of Bank’s Customers Climate change events such as hurricanes, landslide and foods can damage the business of specifc banks’ clients and customers. Bank customers such as manufacturing frms and agricultural frms are very prone to signifcant damages

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Effect of Climate Change on Financial Institutions    141 during hurricanes and food events. Although the impact on banks may be indirect, such events would lead to low proft margin to banks that have high exposure to the agricultural sector since the affected companies will not be able to conduct business for a considerable period of time until recovery.

2.4. Severe Climate Change Will Lead to Increased Insurance Claims and Liabilities Climate change events will increase costs for companies in the insurance industry1. Businesses and individuals that have some comprehensive insurance policy with insurance or reinsurance companies are often the frst to make a claim on their policy whenever climate change events occur that affects the insured object. When such events simultaneously affect multiple businesses and individuals, the insurance payouts will be higher and could lead to fnancial diffculty to insurance companies who may have insuffcient funds to pay to policyholders. Also, since other fnancial institutions are the major lender to insurance companies, fnancial institutions could also be affected indirectly by climate change events.

2.5. Increased Insurance Premiums and Reduction in Coverage Insurance companies can increase insurance premiums and the associated deductibles in anticipation of climate change events which the insurance company believes are likely to occur. In other cases, insurance companies can either decrease the availability of insurance coverage for climate change events or may refuse to give any cover for climate change events due to diffculty to measure the likelihood and severity (or frequency) of climate change events.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

2.6. Harsher Working Conditions The employees in fnancial institutions will become less productive when there are record-breaking temperatures making the day unbearable, making it tough to get the job done. Extreme heat and hydration, especially when combined with hard work and long hours, can lead to mental and physical breakdown of employees in fnancial institutions. Severe climate change means it will be colder or hotter than ever before, and that the cold or heat may last signifcantly longer than it did in the past.

1

Despite the recent awareness of climate change in fnancial institutions, some insurance companies still understate the likelihood that a severe climate change event will occur. Even when insurance companies anticipate such events, the diffculty to accurately measure the likelihood of such climate events is an issue that insurance companies are currently dealing with. Severe climate change events are likely to increase costs for companies in the insurance industry because of increased claims related to extreme weather events, which are unpredictable.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

142    Peterson K. Ozili 2.7. Loss of Offce Branch Networks Financial institutions, such as banks, often prepare for climate change by furnishing the offce headquarters with solar panels and other efforts to lower their carbon footprint, but the same proportional effort is not applied to protect the bank branches all over the country and in foreign locations. Consequently, a bank is likely to lose some branches when climate change events occur such as fooding, hurricanes, etc.

2.8. Higher Disclosure Requirement and Costs Regulations or legislations may emerge that would require fnancial institutions to disclose information on the extent to which fnancial institutions are exposed to climate change risk, and disclosure the amount of risk capital they have set aside to mitigate such risk. Regulations may also require fnancial institutions to disclose additional information on how climate change will affect the company’s strategy and operational performance. However, such disclosures come at cost to fnancial institutions. Such disclosure regulations may be imposed by the capital market regulator so that investors can have a good understanding of how public companies intend to manage climate change risk. In the banking sector, disclosure regulations may be imposed by the bank regulators to serve as a forward-looking micro-prudential approach to deal with climate change risk. Such disclosures are costly to fnancial institutions because they would need to hire specialists to accurately quantify their exposure to climate change risk, and to estimate the risk capital to set aside to mitigate such risk.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

3. Effect on the Financial System Climate change can have profound effects on the fnancial system. The fnancial stem is the system that allows the exchange of funds between lenders, borrowers and investors. Climate change affects the fnancial system through its effects on fnancial institutions and fnancial technology, as discussed in the following.

3.1. Damage to Payment System Infrastructure The fnancial system relies heavily on payment systems to settle fnancial obligations and to receive fnancial claims from counterparties engaged in business transactions. In every country, there is an interbank settlement system which helps to settle fnancial claims and obligations for transactions done in one or more jurisdiction. The interbank settlement infrastructure is usually installed in a physical warehouse at a disclosed or undisclosed location. Any damage to this payment infrastructure caused by severe thunderstorms, severe weather or earthquakes will disrupt the fnancial intermediation process since payments will not be cleared when due, which can lead to severe instability in the entire fnancial system.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Effect of Climate Change on Financial Institutions    143 3.2. Climate Change Can Lead to Capital Flight Climate change can make cities and countries fnancially unstable. This is because holders of large capital will remove their money from cities and countries that are prone to severe climate change events, leaving such cities or countries with little fnancial resources to respond to climate change events when they occur. Also, high net-worth individuals will withdraw their money from banks and investment frms situated in climate change prone areas, leaving such fnancial institutions with little capital to do business, leading to undercapitalization in fnancial institutions and reduced level of fnancial intermediation which can make the fnancial system become unstable.

3.3. Uncertainty that Destabilizes Financial Markets Climate risks have the potential to impact fnancial markets. In cities that prone to climate change events, the uncertainty about which climate change event will occur, when it will occur and the severity when it occurs can fuel volatility in fnancial markets. Investors due to fear and uncertainty can hoard capital or provide capital at a high premium which frms cannot afford. Also, fnancial institutions that lack the ability to mitigate climate change risk may withdraw liquidity from money markets by recalling the fnancial instruments they have issued to the market, which can create a “liquidity crunch” or a “run on liquidity” in fnancial markets, and this will have negative consequences on the stability of the fnancial system.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

3.4. Climate Change Can Confuse Regulators Financial system regulators are often economists – fnancial economists, business economists or academic economists! The inability of macroeconomic models to forecast the probability of a climate change event occurring can leave regulators confused about what to expect from climate change events and confused about what to do when such events occur. Even fnancial institutions that were signifcantly affected by climate change events will seek help from the regulator, and in some cases, the regulator may be clueless on what do. Although most regulators facing a climate change problem would seek help from the national government, there is no guarantee that the government has a special solution that can restore the fnancial system to normalcy.

4. Conclusion In this chapter, I have shown the effect of climate change on fnancial institutions and the fnancial system. Despite the observation that climate change poses new and unprecedented challenges to fnancial institutions, some fnancial institutions still do not see the urgency to allocate risk capital for climate change risk because they feel that climate change events have an indirect, not direct, effect on fnancial services. While there have been serious activities by corporations

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

144    Peterson K. Ozili and governments to deal with climate change, these activities have not been transformational enough to reverse climate change. Even so, many fnancial system regulators have not issued policy statements to assess the readiness of fnancial institutions for climate change events. At least, regulators should give their views regarding climate change events in the fnancial sector so that fnancial institutions can understand the role they have to play in preventing climate change events from occurring, and to understand their role in reducing the effect of climate change when they occur. Future studies can examine the effect of climate change on the non-fnancial sector. Future studies can also examine the effect of climate change on the economy.

Declaration of Interest There is no conficting interest in this chapter.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

References Akhmadeev, R. G., Bykanova, O. A., Philippova, N. V., Vashchekina, I. V., & Turishcheva, T. B. (2018). Macroeconomic indicators and their impact on the foreign debt burden: The case of BRICS countries. International Journal of Economics and Business Administration, 6(2), 68–82. Osadchy, E. A., Akhmetshin, E. M., Amirova, E. F., Bochkareva, T. N., Gazizyanova, Y. Y., & Yumashev, A. V. (2018). Financial statements of a company as an information base for decision-making in a transforming economy. European Research Studies Journal, 21(2), 339–350. Rupeika-Apoga, R., Zaidi, H. S., Thalassinos, E. Y., & Thalassinos, I. E. (2018). Bank stability: The case of Nordic and non-Nordic banks in Latvia. International Journal of Economics and Business Administration, 6(2), 39–55. Thalassinos, I. E., & Stamatopoulos, V. T. (2015). The trilemma and the Eurozone: A preannounced tragedy of the Hellenic debt crisis. International Journal of Economics and Business Administration, 3(3), 27–40.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Chapter 11

Testing for Asymmetric Causality Between Developed and Emerging Markets Letife Özdemir Abstract Purpose: Through globalization, fnancial markets have become more integrated and their tendency to act together has increased. The majority of the literature states that there is a cointegration between developed and emerging markets. How do positive or negative shocks in developed markets affect emerging markets? And how do positive or negative shocks in emerging markets affect developed markets? For this reason, the aim of the study is to investigate the asymmetric causality relationship between developed and emerging markets with Hatemi-J asymmetric causality test.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Design/methodology/approach: In this study, the Dow Jones Industrial Average (DJIA) index was used to represent developed markets and the Morgan Stanley Capital International (MSCI) Emerging Market Index was used to represent emerging markets. The asymmetric causality relationship between the DJIA Index and the MSCI Emerging Market Index was investigated using monthly data between January 2009 and April 2019. In the frst step of the study, the Johansen Cointegration Test was used to determine whether there is a cointegration between the markets. In the next step, the Hatemi-J asymmetric causality test was applied to see the asymmetric causality relationship between the markets. Findings: There is a weak correlation between developed and emerging markets. This result is important for international investors who want to diversify their portfolios. As a result of the Johansen Cointegration Test, it was found that there is a long-term relationship between the MSCI Emerging Market Index and the DJIA Index. Therefore, investors who make long-term investment plans should not forget that these markets act together and take into account the causal relationship between them. According to the asymmetric causality test results, a unidirectional causality relationship from the MSCI Emerging Market Index to the DJIA Index was determined. This causality shows that negative shocks in the MSCI Emerging Market Index have positive effects on the DJIA Index. Uncertainty and Challenges in Contemporary Economic Behaviour, 145–158 Copyright © 2020 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-80043-095-220201012

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

146    Letife Özdemir Originality/value: This study contributes to the literature as it is one of the frst studies to examine the asymmetrical relationship between developed and emerging markets. This study is also useful in predicting the shortand long-term relationship between markets. In addition, this study helps investors, portfolio managers, company managers, policymakers, etc., to understand the integration of fnancial markets. Keywords: Developed and emerging markets; DJIA; MSCI; correlation; Johansen cointegration test; Hatemi-J asymmetric causality test JEL classifcation: G10; G11; C32

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

1. Introduction As a result of technological advancement, the time to reach the markets to obtain information about the markets and to operate in the markets has shortened. Despite the physical distances between markets, the rapid spread of information has brought markets closer together. With these developments and globalization, fnancial markets have started to become more integrated. Thus, capital fow has accelerated between fnancial markets. Financial integration has been seen primarily in developed markets. Then, fnancial integration has been spread from developed countries to emerging countries. Thus, the capital fows increased between developed and emerging markets. In addition, with the increase in capital fows to emerging markets, the expectation has been that growth will increase and volatility in fnancial markets will decrease. Integration of fnancial markets has become important in many ways, such as offering investors new investment opportunities, portfolio diversifcation, hedging opportunities, and new funding opportunities for investment plans (GambaSantamaria, Gomez-Gonzalez, Hurtado-Guarin, & Melo-Velandia, 2017). The risk of a portfolio created by using the index returns of the developed and emerging markets is lower compared to the risk of a portfolio formed with the indices of the emerging markets. At the same time, the return on this portfolio is higher than the return on the portfolio created by the index returns of developed markets (Akdag & Ekinci, 2018). The fact that fnancial markets act together is an intriguing subject of how markets affect each other. There are asymmetric information and different investor behaviors in fnancial markets. Therefore, investors may react differently to positive and negative information about the markets or remain unresponsive. The majority of the studies analyzing the relationship between fnancial markets determined that there is a cointegration relationship between the markets. Researches following these studies investigated the causality relationship between the markets using time series analysis methods. The main difference that distinguishes our study from previous studies in the literature is the use of the asymmetric causality test, taking into account that the effects of positive and negative shocks experienced in fnancial markets may be different.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Testing for Asymmetric Causality    147 The aim of this study is to investigate the asymmetric causality relationship between developed and emerging markets using monthly data between January 2009 and April 2019. In the study, the asymmetric relationship between developed and emerging markets is examined with the Hatemi-J asymmetric causality test. The rest of the chapter is organized as follows: In Section 2, studies examining the relationship between fnancial markets are discussed. In Section 3, the econometric methodology is explained. In Section 4, data and empirical results are presented. The study is completed with Section 5 in which empirical analysis results are evaluated.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

2. Literature Many studies have examined the relationship between stock markets (Abdennadher & Hellara, 2018; Al-Yahyaee, Mensi, Al-Jarrah, & Kumar, 2019; Bouri, Lien, Roubaud, & Shahzad, 2018; Caporale, Gıl-Alana, & Orlandoc, 2016; Chan, Lien, & Weng, 2008; Liu & Su, 2010; Ozen & Tetik, 2019; Patel, 2013; Thalassinos, Stamatopoulos, & Thalassinos, 2015, chapter 20). The majority of these studies state that the interaction of markets with each other has increased due to globalization experienced in fnancial markets. In addition to these studies, there are also studies examining how developed and emerging markets affect each other especially during the recent fnancial crisis (Liapis, Rovolis, Galanos, & Thalassinos, 2013; Rupeika-Apoga, Zaidi, Thalassinos, & Thalassinos, 2018; Thalassinos & Thalassinos, 2018; Ugurlu, Thalassinos, & Muratoglu, 2014). Norasyikin (2011) examined the structures of linkages among the world’s fastest emerging economies (Brazil, Russia, India, and China) and the selected developed countries namely, the USA, the UK, and Japan. The period of study is divided into three sample periods: the pre-crisis period – January 10, 2005–July 22, 2007, during the crisis – July 29, 2007–January 10, 2010, and post-crisis period – January 11, 2010, until recently July 21, 2011. The results show that all the stock markets under study are cointegrated in pre-crisis, during the crisis, and post-crisis periods. In a similar study, Nashier (2015) determined integration between the BRICS exchanges (Brazil, Russia, India, China, and South Africa) with the US and UK exchanges. Cıtak and Gözbası (2007) examined the long-term integration relationship between the ISE with the developed USA, German, British, and Japanese markets and the emerging Indian and Malaysian markets using the data between January 1986 and July 2006. As a result, they concluded that there is integration between ISE with the UK, USA, Germany, and India’s main indices. Ozdemir, Olgun, and Saracoglu (2009) examined the dynamic linkages between the equity markets of a center (the USA) and its periphery (emerging markets) using Granger causality. Empirical results show a signifcant causal relation from S&P500 to all emerging stock markets included in the sample, but not vice versa. Ali, Zaheer Butt, and Rehman (2011) investigated the co-movement of Pakistan’s Equity Market with the markets of India, China, Indonesia, Singapore, Taiwan, Malaysia, Japan, USA, and UK by using a cointegration test. The results revealed that there is no co-movement of Pakistan’s stock market with the markets of UK, USA, Taiwan, Malaysia, and Singapore. Also, it was determined that stock prices of the Pakistan stock market move together with stock prices of India, China, Japan, and Indonesia.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

148    Letife Özdemir Lingaraja, Selvam, and Vasanth (2015) examined the dynamic linkages between emerging stock markets in Asia and a developed stock market (DJIA). It was found that the emerging stock markets of Asia, namely, India, Malaysia, and Philippines, recorded dynamic linkages with Dow Jones Industrial Average (DJIA) (USA) and the other fve Asian emerging markets (i.e., China, Indonesia, Korea, Taiwan, and Thailand) did not develop dynamic linkages with USA (DJIA), a developed country. Ghiffari, Kindangen, and Tumewu (2017) aimed to analyze if the DJIA and NIKKEI 225 could affect JCI from 2011 to 2016. In the analysis, it was found that DJIA had no signifcant effect on JCI movements and NIKKEI 225 had a signifcant affected the JCI. Selvam, Lingaraja, and Mahalingam (2013) found out that there is a unidirectional Granger causality relationship from the US stock markets (DJ Small Cap Index, DJ Midcap Index, and DJIA) to the Indian stock market (CNX Midcap, CNX Small-Cap, and NSE Nifty index). Akash, Iqbal, and Kashif (2018) examined the integration and causality relation between the Pakistan Stock Exchange (PSX100) and the Asian, European, and American stock markets between January 1, 2006 and July 26, 2018. PSX-100 has international fnancial integration with all the markets of Asian, European, and American Equity markets. And as a result of the analysis, PSX-100 with S&P500 and IPC with S&P500 bivariate causality relationships were found. As mentioned previously, results showing that there are cointegration and causality relationship between developed and emerging markets have been obtained. These studies acknowledge that the effects of positive and negative shocks in the markets are the same. The main difference that distinguishes our study from the other studies is to reveal the relationship between developed and emerging markets using asymmetric causality test, considering that the effects of positive and negative shocks may not be the same.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

3. Methodology In order to examine the relationship between the MSCI Emerging Market Index and the DJIA index, the cointegration test was used to examine whether there is a long-term relationship between the indices. If there is a long-term relationship, the causality relationship between the indices and whether the relationship differs according to positive and negative shocks was examined with the Hatemi-J asymmetric causality test.

3.1. Cointegration Test Cointegration means that the variables move together in the long run. If all variables are stationary at the same level, cointegration methods are used to test whether there is a long-term relationship between the series. The existence of a long-term equilibrium relationship between variables was investigated according to the cointegration method developed by Johansen (1988) and Johansen and Juselius (1990). The Johansen cointegration test is based on the vector autoregression (VAR) model analysis. The VAR model is shown as follows (Brooks, 2011):

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Testing for Asymmetric Causality    149

yt = b1 yt−1 + b2 yt−2 +… +bk yt−k + ut(1)



∆yt = Π yt − k + Γ 1∆ yt −1 + Γ 2∆ yt −2 + …+ Γ k −1∆ yt −( k −1) + ut (2)

where Γ and Π represent coeffcient matrices. Coeffcient matrix Π contains information about long-term relationships. In Johansen and Juselius cointegration method, two different test statistics, trace test statistics and maximum eigenvalue test statistics have been developed to reveal the existence of the cointegration relationship and the number of a cointegrated vector. These test statistics are as follows:

g

λtrace (r) = −T ∑ ln(1− λˆi )  (3) i =r+1

and

(

)

λmax (r,  r + 1) = −T  ln 1− λˆr+1 (4)

The cointegration vector number is represented by r. Trace test statistics investigate the cointegration relationship as r and maximum eigenvalue test statistics investigate the cointegration relationship as much as r + 1 (Brooks, 2011).

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

3.2. Asymmetric Causality Tests Various tests have been developed in the literature for causality analysis (Granger, 1988; Sims, 1972; Toda & Yamamoto, 1995). These tests accept the effects of positive shocks and negative shocks. But in fnancial markets, there may be asymmetric information and different investor behavior. Therefore, investors may not respond similarly to positive and negative shocks of the same magnitude. This situation may cause the results obtained from the mentioned tests to be misleading. The different causality effect of negative and positive shocks in the fnancial markets can be measured with the asymmetric causality test. The asymmetric causality test developed by Hatemi-J (2012) is intended to investigate the downward and upward causality relationship between the variables. Two integrated variables y1t   and y2t , of which we want to determine the asymmetric causality relationship, are defned as follows (Hatemi-J, 2012):

t

y1t   =  y1t−1 + ε1t = y10 + ∑ε1i  (5) i =1



t

y2t   = y2t−1 + ε2 t = y20 + ∑ε2i  (6) i =1

where t=1,2,…T, constants y10 and y20 are the initial values, and the variables ε1i   and ε2i   signify white noise disturbance terms. Positive and negative shocks are defned as the following:

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

150    Letife Özdemir

  ε1+i = max( ε1i ,0)   ε2+i = max( ε2i ,0)



ε1−i = min( ε1i ,0)   ε2−i = min( ε2i ,0) Therefore, it can be expressed as ε1i   = ε1+i + ε1−i and ε2 i   = ε2+i + ε2−i . It follows that t

t

i =1

i =1

t

t

i =1

i =1

y1t   = y1t−1 + ε1t = y1,0 + ∑ε1+i + ∑ε1−i (7)



y2t   = y2t−1 + ε2 t = y2,0 + ∑ε2+i + ∑ε2−i .(8)



Positive and negative shocks in each variable are expressed in cumulative form: t

y1+t   = ∑ε1+t



i =1

t

y2+t = ∑ε2+t



i =1

t

y1−t   = ∑ε1−t  i =1

t

y2−t   = ∑ε2−t  i =1

The next stage of the asymmetric causality test is to test the causal relationship between variables as a positive and negative shock. Here, the test method will be defned only for the data vector generated for positive cumulative shocks, but the test method will be the same for negative shocks. Assuming that yt+ = (  y1+t ,   y2+t ), the test for causality can be implemented by using the following vector autoregressive model of order p, VAR (p): yt+ = v + A1 yt+−1 +…+ Ap yt+−1 + ut+ ,(9)

Copyright © 2020. Emerald Publishing Limited. All rights reserved.



where yt+ is the 2 × 1 vector of variables, v is the 2 × 1 vector of intercepts and ut+ is a 2 × 1 vector of error terms. The matrix Ar is a 2 × 2 matrix of parameters for lag order r( r = 1,…,  p ). The following information criterion is used to select the optimal lag order (p):

 n 2InT + 2 n 2 (InT )  ˆ ,       j = 0,…,  p.,(10) HJC = In  Ω  j + j   2T

( )

( )

where Ωˆ j is the highest likelihood estimator of the variance-covariance matrix, n is the number of variables and j is the lag order. The optimal lag order was determined considering model selection criteria suggested by Hatemi-J (2012) in the study. After selecting the optimal lag order, the author tested the null hypothesis that there was no causality relationship between variables. The null hypothesis of non-Granger causality is rejected at the α level of signifcance if the Wald test generated in the fnal step is greater than the bootstrap critical value. The bootstrap critical values are produced for three different signifcant signifcance levels. The boots simulations are implemented by using statistical software components written in GAUSS (Hatemi-J, 2012).

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Testing for Asymmetric Causality    151

4. Data, Analysis, and Findings

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

This study examines the asymmetrical relationship between developed and emerging markets. DJIA index is used to represent developed markets and Morgan Stanley Capital International (MSCI) Emerging Market Index is used to represent emerging markets. The DJIA Index is a price-weighted average of 30 signifcant stocks traded on the New York Stock Exchange and the Nasdaq. The MSCI index consists of 27 countries representing 10% of the world markets. These countries are Argentina, Brazil, Chile, Colombia, Mexico, Peru, Czech Republic, Egypt, Greece, Hungary, Poland, Qatar, Russia, Saudi Arabia, South Africa, Turkey, UAE, China, India, Indonesia, Korea, Malaysia, Pakistan, Philippines, Taiwan, and Thailand. The asymmetric causality relationship between the DJIA Index and the MSCI Emerging Market Index was investigated using monthly data for the period between January 2009 and April 2019 (Polyakova, Loginov, Serebrennikova, & Thalassinos, 2019). All data were obtained from the investing database. The data used in the study were purifed from the seasonal effect by using the cumulative seasonal decomposition method (Moving Average Methods). Descriptive statistics of daily data are given in Table 1. As it can be seen in Table 1, the average of MSCI Emerging Market Index in the relevant period is 162.62 and its standard deviation is 50.57. In addition, the kurtosis coeffcient of the series has been determined as 2.26. This value is below 3 and this appears as a more fattened series than usual. Skewness coeffcient was determined to be 0.39, and since this value is greater than zero, it shows that the series is skewed to the right and the left tail is longer. Finally, the Jarque–Bera coeffcient was considered as the descriptive statistical value of the MSCI Emerging Market Index. This statistical value, which shows whether the series is normally distributed or not, has revealed that the series is not normally distributed since it is statistically signifcant. When the descriptive statistics of the DJIA Index are analyzed, it is seen that the series has an average value of 40.16 and the standard deviation is determined as 4.87. The Skewness coeffcient of the series was calculated as –0.85 and this indicates that the series is crooked to the left and the right tail is longer. The kurtosis Table 1.  Descriptive Statistics. MSCI

DJIA

Mean

162.6292

40.16924

Median

162.3019

40.81847

Maximum

269.5644

51.17415

Minimum

69.90928

Std. Dev.

50.57395

21.17215 4.874549

Skewness

0.399888

−0.850267

Kurtosis

2.266064

4.739288

Jarque-Bera

6.038811

Probability

0.048830

30.32432 0.000000

Source: Author’s calculations.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

152    Letife Özdemir coeffcient was calculated as 4.73. Since this value is higher than 3, the series appears as an upright series. When we look at whether the series is normally distributed or not, the value of 30.32 obtained according to the Jarque–Bera test result shows that the series is not normally distributed. Firstly, a correlation matrix was established in order to test the existence of the relationship between the fnancial markets that were included in the research. The degree of correlation between fnancial markets is indicated in Table 2. When the table is examined, it can be said that there is a weak correlation between the fnancial markets. This result is important for investors who want to diversify their portfolio. However, it is not possible to say which market affects the other while making a general evaluation of the degree of co-operation between markets by looking at the correlation values. For this reason, causality analysis has been conducted in order to identify the affected and affecting factors in the interaction between fnancial markets. In order to make causality analysis based on time series, frstly, the author investigated whether there is a unit root in the series. For the unit root test, augmented Dickey–Fuller (ADF) unit root test developed by Dickey and Fuller (1979) and Dickey–Fuller-Generalized Least Squares (DF-GLS) unit root test developed by Elliott, Rothenberg, and Stock (1996) were used. Unit root test results are presented in Table 3. Table 2.  Correlation Matrix. MSCI MSCI

1

DJIA

0.30104

DJIA 1

Source: Author’s calculations.

Table 3.  Results for Unit Root Tests.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Series

ADF Test Intercept

MSCI

Level

Intercept and Trend

Intercept

−4.224005* −4.053944*

−0.569538

1 st Diff. −11.49154* −11.56262* DJIA

Level

DF-GLS Test

−0.463969

Intercept and Trend −1.511744

−2.773409* −10.11757*

−2.706391

1.947454

−2.636996

1 st Diff. −12.28644* −12.23831*

−10.55744*

−11.76258*

Critical Values 1%

−3.485115

−4.035648

−2.584055

−3.553600

5%

−2.885450

−3.447383

−1.943471

−3.008000

10%

−2.579598

−3.148761

−1.614984

−2.718000

Source: Author’s calculations. 1% of the signifcance level.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Testing for Asymmetric Causality    153

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

According to the ADF unit root test result, the absolute values of t-statistics calculated for the MSCI Emerging Market Index are less than the absolute value of the relevant critical values at 1% signifcance level. Accordingly, the null hypothesis was rejected and it was determined that this variable does not have a unit root. Therefore, the MSCI Emerging Market Index is stationary at the level values. The absolute value of t-statistics calculated for the DJIA Index is greater than the absolute value of the relevant critical values at the 1% signifcance level. Accordingly, the null hypothesis was accepted and this variable has a unit root. Therefore, the DJIA variable is not stationary at level values. At this stage, the frst difference of the series was taken and it was determined that the frst awareness of the series was stationary. According to the DF-GLS test, the null hypothesis defned as unit root is not rejected for series at level values. In other words, the MSCI Emerging Market and DJIA Index series are not stationary as they have unit roots in level values. The study was based on the results of the DF-GLS test because the DF-GLS test gives more effective results than the ADF test due to its asymptotic distribution. In this case, the integration degree of the MSCI Emerging Market and DJIA Index series is determined as I(1). The fact that index series are integrated into the same degree does not mean that they always act together in the long run. The fact that the series is stationary in the frst differences allows the use of the cointegration method developed by Johansen (1988) and Johansen and Juselius (1990) for cointegration analysis. In order to use the Johansen–Juselius cointegration test, the appropriate lag length must be determined. The appropriate lag length for cointegration tests was determined with the help of the classical VAR model. Suitable lag lengths for the VAR model are shown in Table 4. The suitable lag length for the estimated VAR model was determined as 1 according to the FPE, AIC, SC, and HQ information criteria. In order to determine whether the VAR model estimated for 1 lag length includes the unit root, the position of the inverse roots of the AR characteristic polynomial within the Table 4.  Suitable Lag Lengths for the VAR Model. Lag

LogL

0 −931.6014 1 −575.2265

LR

FPE

NA

38591.68

694.1563

AIC

SC

HQ

16.23655

16.28428

16.25592

84.14620*

10.10829*

10.25150*

10.16642*

2 −572.7978

4.646204

86.48481

10.13561

10.37430

10.23250

3 −572.5026

0.554352

92.25835

10.20005

10.53421

10.33568

4 −569.8060

4.971213

94.40555

10.22271

10.65235

10.39710

5 −562.8568

12.56895*

89.73328

10.17142

10.69654

10.38457

Source: Author’s calculations. Note: *Indicates the optimal lag length determined by the criteria. LR: Likelihood Ratio; FPE: Final Prediction error; AIC: Akaike Information criteria; SC: Schwarz Information criteria; and HQ: Hannan-Quinn information criterion The information criterion values of the most suitable lag number are shown in bold characters.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

154    Letife Özdemir Inverse R oots of AR C haracteristic Polynom ial 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Fig. 1.  Inverse Roots of AR Characteristic Polynomial of MSCI–DJIA Model. Source: Author’s calculations. unit circle was examined. It is understood from Fig. 1 that all of the reverse roots of the AR characteristic polynomial are located within the unit circle. The fact that the inverse roots are located within the unit circle revealed that the estimated model has a stationary structure. The Johansen cointegration test (Johansen, 1988; Johansen & Juselius, 1990) was used to determine the existence of a long-term relationship between the MSCI Emerging Market Index and the DJIA Index. Cointegration test results with the determined lag length are given in Table 5. According to the maximum eigenvalue and trace statistics obtained as a result of the Johansen cointegration test, the null hypothesis is rejected, that is, the hypothesis that predicts at least one cointegration vector is accepted at the level of 5% signifcance. These results indicate a long-term equilibrium relationship between MSCI Emerging Market and DJIA indices during the analysis period. Although cointegration analysis shows that there is a long-term relationship between variables, it does not provide any information about the direction of causality. The causality relationship between MSCI emerging market and DJIA indices and whether the relationship differs according to positive and negative shocks were examined with the Hatemi-J asymmetric causality test. The results of the Hatemi-J asymmetric causality test are shown in Table 6. According to the results in Table 6, it is seen that there is no causality relationship between the indices in terms of the same directional shocks. In other words, the null hypothesis that the positive (negative) shocks in the MSCI Emerging Market Index is the cause of the positive (negative) shocks in the DJIA Index could not be rejected and vice versa. However, negative shocks in the MSCI Emerging Market

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Testing for Asymmetric Causality    155 Table 5.  Results of Johansen Cointegration Test. Hypothesis

Trace Statistics

5% Critical Value

Probability**

25.87211

0.0204

5.771457

12.51798

0.4899

Max-Eigenvalue Statistics

5% Critical Value

Probability**

19.38704

0.0136

12.51798

0.4899

28.89312

r = 0* r≤1 Hypothesis

23.12166

r = 0* r≤1

5.771457

Source: Author’s calculations. Notes: Existence of one cointegrating vector at 5% signifcance level; *indicates the rejection of the null hypothesis at 5% signifcance level; **indicates MacKinnon, Haug, and Michelis (1999) p-values.

Table 6.  Results for Hatemi-J Asymmetric Causality Tests. Direction of Causality +

+

+

−

−

−

−

+

MSCI → DJIA

DJIA+ → ∕ MSCI+

MSCI → ∕ DJIA MSCI → ∕ DJIA ∕ DJIA MSCI →

Bootstrap Critical Values 1%

5%

10%

2.942

9.790

6.330

4.813

0.002

8.891

4.805

2.996

0.347

8.146

3.960

2.579

7.309*

7.033

3.995

2.728

0.234

11.359

6.739

5.323

+

−

1.504

9.620

6.402

4.634

−

−

DJIA → ∕ MSCI

0.086

8.842

4.521

3.114

∕ MSCI+ DJIA− →

3.149

9.075

6.101

4.975

∕ MSCI DJIA →

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Test Statistics

Source: Author’s calculations. 1% of signifcance level

Index were found to be the cause of positive shocks in the DJIA index at 1% signifcance level. These results show that the negative changes (decrease in the index) in the MSCI Emerging Market Index caused an increase in the DJIA index. According to this result, we can say that investors transfer their investments to the developed market index in order to increase their proftability when the emerging market index decreases, thereby causing the increase of the developed market index.

5. Conclusion The increase in integration in fnancial markets has directed investors and fnancial experts, who want to achieve high returns, to investigate the relationship between fnancial markets. Determining the relationship between fnancial markets is a very important issue in terms of portfolio diversifcation. In this study,

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

156    Letife Özdemir the asymmetric causality relationship between the developed and emerging markets is evaluated by using monthly data from January 2009 to April 2019. The DJIA Index was used to represent developed markets and the MSCI Emerging Market Index was used to represent emerging markets. First, a correlation matrix was created to test the existence of the relationship between the fnancial markets was included in the study. There is a weak correlation between developed and emerging markets. This result is important for international investors who want to diversify their portfolios. To examine the causality relationship between fnancial markets, it is necessary to test whether the index series are stationary. As a result of ADF and DF-GLS unit root tests, it is determined that MSCI Emerging Market and DJIA indices are stationary at frst difference. That is, the degree of integration of indices is I(1). The fact that index series are integrated to the same degree does not mean that they always act together in the long run. The Johansen cointegration test was used to determine the existence of a long-term relationship between the equally integrated MSCI Emerging Market Index and the DJIA index. As a result of the Johansen cointegration test, it was possible to determine that a long-term equilibrium relationship existed between MSCI Emerging market and DJIA indices during the analysis period. This result is similar to Norasyikin (2011), Nashier (2015), Cıtak and Gözbası (2007), Ali et al. (2011), and Akash et al. (2018)’s study results. The causality relationship between the MSCI Emerging Market Index and the DJIA index and whether the relationship differs according to positive and negative shocks were examined with an asymmetric causality test. As a result of the asymmetric causality test, the null hypothesis that the positive (negative) shocks in the MSCI Emerging Market Index are the cause of the positive (negative) shocks in the DJIA Index could not be rejected. In other words, it is seen that there is no causality relationship between indices in terms of the same directional shocks and vice versa. However, negative shocks in the MSCI Emerging Market Index were found to be the cause of positive shocks in the DJIA Index. These results show that the negative changes (decrease in the index) in the MSCI Emerging Market Index caused an increase in the DJIA Index. From this study, the following conclusions can be drawn: (1) Since there is a low correlation between the MSCI Emerging Market Index and the DJIA Index, investors can increase their proftability by using these markets when creating a portfolio. (2) The fact that there is a long-term relationship between the markets shows that these markets move together. Therefore, investors who make long-term investment plans should not forget that these markets act together and take into account the causal relationship between the markets. (3) Investors can increase their proftability if they create their portfolios, considering that the decrease in MSCI Emerging Market Index causes the DJIA Index to increase. This study provides benefts in the following areas: (1) As it is one of the frst studies to examine the asymmetrical relationship between developed and emerging markets, it contributes to the literature. (2) It is useful in predicting the short and long-term relationship between markets. (3) It helps investors, portfolio managers, company managers, policymakers, etc., to understand the integration of fnancial markets.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Testing for Asymmetric Causality    157 The data period and the number of variables used in the study constitute the limitations of the study. In future studies, the data period can be extended or the data can be divided into sub-periods. In addition, the indices of the countries included in the MSCI Emerging Market Index can be considered separately.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

References Abdennadher, E., & Hellara, S. (2018). Causality and contagion in emerging stock markets. Borsa Istanbul Review, 18(4), 300–311. doi:10.1016/j.bir.2018.07.001 Akash, S. I. R., Iqbal, M., & Kashif, H. (2018). The dynamics of international equity markets in Asia, Europe and America in term of economic globalization. Faculty of Business Economics and Entrepreneurship, International Review, 3–4, 9–16. Akdag, S., & Ekinci, M. A. (2018). Portfolio optimization with fuzzy logic and international diversifcation: Apply with stock index of developed and emerging countries. Academic Review of Economics and Administrative Sciences, 11(3), 26–41. doi:10.25287/ohuiibf.423114 Ali, S., Zaheer Butt, B., & Rehman, K. (2011). Comovement between emerging and developed stock markets: An investigation through cointegration analysis. World Applied Sciences Journal, 12(4), 395–403. Al-Yahyaee, K. H., Mensi, W., Al-Jarrah, I. M. W., & Kumar, A. (2019). Testing for the Granger-causality between returns in the US and GIPSI stock markets. Physica A: Statistical Mechanics and its Applications, 531, Art. No. 120950, 1–13. doi:10.1016/ j.physa.2019.04.186 Bouri, E., Lien, D., Roubaud, D., & Shahzad, S. J. H. (2018). Fear linkages between the US and BRICS stock markets: A frequency-domain causality. International Journal of the Economics of Business, 25(3), 441–454. doi:10.1080/13571516.2018.1505241 Brooks, C. (2011). Introductory econometrics for fnance. Cambridge: Cambridge University Press. Caporale, G. M., Gıl-Alana, L. A., & Orlandoc, J. C. (2016). Linkages between the US and European stock markets: A fractional cointegration approach. International Journal of Finance & Economics, 21, 143–153. doi:10.1002/ijfe.1537 Chan, L., Lien, D., & Weng, W. (2008). Financial interdependence between Hong Kong and the US: A band spectrum approach. Review Economic Finance, 17, 507–516. Cıtak, L., & Gözbası, O. (2007). Analysis of integration between IMKB and some leading developed and developing country exchanges on the basis of the main index and main sector indexes. Dokuz Eylül University Journal of Faculty of Economics and Administrative Sciences, 22(2), 249–271. Dickey, D., & Fuller, W. A. (1979). Distributions of the estimators for autoregressive time series with a unit root. The American Statistical Association, 74, 423–431. Elliott, G., Rothenberg, T. J., & Stock, J. H. (1996). Effcient tests for an autoregressive unit root. Econometrica, 64(4), 813–836. doi:10.2307/2171846 Gamba-Santamaria, S., Gomez-Gonzalez, J. E., Hurtado-Guarin, J. L., & Melo-Velandia, L. F. (2017). Volatility spillovers among global stock markets: Measuring total and directional effects. Banco de la Republica de Colombia No. 983, Colombia. Ghiffari, M. F., Kindangen, P., & Tumewu, F. (2017). The causality relationship of Dow Jones Industrial Average (DJIA) and NIKKEI 225 towards Jakarta Composite Index (JCI) period 2011–2016. Journal EMBA, 5(2), 445–454. Granger, C. W. J. (1988). Some recent developments in a concept of causality. Journal of Econometrics, 39(1/2), 199–211. Hatemi-J, A. (2012). Asymmetric causality tests with an application. Empirical Economics, 43(1), 447–456. doi:10.1007/s00181-011-0484-x

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

158    Letife Özdemir Johansen, J. (1988). Statistical analysis of cointegrating vectors. Economic Dynamics and Control, 12, 231–254. Johansen, S., & Juselius, K. (1990). Maximum likelihood estimation and inference on cointegration with application to the demand for money. Oxford Bulletin of Economics and Statistics, 52, 169–210. Liapis, K., Rovolis, A., Galanos, C., & Thalassinos, I. E. (2013). The clusters of economic similarities between EU countries: A view under recent fnancial and debt crisis. European Research Studies Journal, 16(1), 41–66. Lingaraja, K., Selvam, M., & Vasanth, V. (2015). Long run dynamic linkages between emerging stock markets in Asia and a developed stock market (DJIA). Research Journal of Applied Sciences, 10(5), 203–211. Liu, Y. S., & Su, C. W. (2010). The relationship between the real estate and stock markets of China: Evidence from a nonlinear model. Applied Financial Economics, 20, 1741–1749. doi:10.1080/09603107.2010.524616 MacKinnon, J., Haug, A., & Michelis, L., (1999) Numerical distribution functions of likelihood ratio tests for cointegration. Journal of Applied Econometrics, 14(5), 563–77. Nashier, T. (2015). Financial integration between BRICS and developed stock markets. International Journal of Business and Management Invention, 4(1), 65–71. Norasyikin, A. F. (2011). The structure of linkages and causal relationships between BRIC and developed equity markets. International Conference on Information and Finance IPEDR, 21, 72–77. Ozdemir, Z. A., Olgun, H., & Saracoglu, B. (2009). Dynamic linkages between the center and periphery in international stock markets. Research in International Business and Finance, 23, 46–53. Ozen, E., & Tetik, M. (2019). Did developed and developing stock markets react similarly to Dow Jones during 2008 crisis. Frontiers in Applied Mathematics and Statistics, 5(49). doi: 10.3389/fams.2019.00049 Patel, S. A. (2013). Dynamic linkages of developed equity markets with Indian stock market. Vilakshan, XIMB Journal, 10(1), 21–36. Polyakova, A. G., Loginov, M. P., Serebrennikova, A. I., & Thalassinos, E. I. (2019). Design of a socio-economic processes monitoring system based on network analysis and big data. International Journal of Economics & Business Administration, 7(1), 130–139. Rupeika-Apoga, R., Zaidi, H. S., Thalassinos, E. Y., & Thalassinos, I. E. (2018). Bank stability: The case of Nordic and non-Nordic banks in Latvia. International Journal of Economics and Business Administration, 6(2), 39–55. Selvam, M., Lingaraja, K., & Mahalingam, G. (2013). Stock market integration of India and USA: A study on small and medium enterprises. Electronic copy. Retrieved from http://ssrn.com/abstract=2368110 Sims, C. A. (1972). Money, income, and causality. American Economic Review, 62(4), 540–552. Thalassinos, I. E., Stamatopoulos, D. T., & Thalassinos, E. P. (2015). The European sovereign debt crisis and the role of credit swaps. In W. T. Ziemba & A. G. Malliaris (Eds.), The WSPC handbook of futures markets, in memory of Late Milton Miller (Nobel 1990), World Scientifc Handbook in Financial Economic Series (Vol. 5, pp. 605–639). ISBN: 978-981-4566-91-9. doi:10.1142/9789814566926_0020 Thalassinos, I. E., & Thalassinos, E. Y. (2018). Financial crises and e-commerce: How are they related? SSRN. Retrieved from https://ssrn.com/abstract=3330169 Toda, H. Y., & Yamamoto, T. (1995). Statistical inference in vector autoregressions with possibly integrated processes. Journal of Econometrics, 66(1–2), 225–250. Ugurlu, E., Thalassinos, E., & Muratoglu, Y. (2014). Modeling volatility in the stock markets using GARCH models: European emerging economies and Turkey. International Journal of Economics and Business Administration, 2(3), 72–87.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Chapter 12

An Assessment for Financial Performance of Banks Listed in Borsa Istanbul By Multiple Criteria Decision Making Eser Yeşildağ, Ercan Özen and Ender Baykut Abstract Introduction: Decision making is always based on several factors which may affect the possible outcomes, especially in fnancial markets. Instead of having many criteria which may be required for decision making, “Multiple Criteria Decision Making” (MCDM) models might be used as a tool to reduce all criteria into a single one. Purpose: The aim of this study is to measure the fnancial performance of commercial banks listed on Borsa Istanbul (BIST) by the MCDM.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Method: To this end, data from 15 different fnancial ratios from 11 commercial banks were used between the periods of 2002 and 2018. Both TOPSIS and gray relational analysis (GRA) models were used, which are commonly used in the literature for detecting the fnancial performance of listed banks in BIST based on their consolidated fnancial statements. Results: According to the TOPSIS method, while the best bank is QNB Finansbank, HALKB, a public bank, was determined as the best bank using the GRA method. There is no signifcant correlation between fnancial performance indicators and market returns obtained by either method, with exceptions. There is no generally signifcant correlation detected between fnancial ratios and market returns. Accordingly, it is concluded that the bank stock prices in the study are shaped by the infuence of external factors and expectations. The study results include information that can be used for different purposes among bank managers, academics and fnancial investors. Keywords: Banking; MCDM; multi-criteria decision model; TOPSIS; gray relational analysis; AHP

Uncertainty and Challenges in Contemporary Economic Behaviour, 159–183 Copyright © 2020 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-80043-095-220201013

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

160    Eser Yeşildağ et al.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

1. Introduction The banking system constitutes the most important part of the fnancial and economic system everywhere in the world, due to its functions and roles. Banks have to use their resources effectively and effciently due to both their importance for the country’s economy and the high competition in their sector. In this respect, banks’ performance measurement has a more privileged position in contrast to other business types. Measuring the banks performance also allows them to use measures to increase their success. Financial information is used in the economic sector for decision-making purposes, mainly by different players such as shareholders, managers, employees, government, creditors and investors. When decision-makers are faced with a decision-making process, they are faced with too many criteria that will affect their decisions. In this instance, decision making becomes a rather diffcult process. Nowadays, banks are infuenced by the rapid change in the feld of information and technology. As a consequence, they are required to operate within many aspects at the same time. Hence, it is diffcult to make decisions in these institutions as well. It is possible to use many qualitative and quantitative criteria in the performance evaluations of banks. Taking into consideration many criteria and choosing the most suitable one among the alternatives is called multi-criteria problems. These criteria often contradict each other. Hereby, it is generally not an optimal solution for such problems. Accordingly, any alternative solution is not best solution (Erpolat Taşabat, Cinemre, & Şen, 2015, p. 97). Information about banks is most of the time limited to managers and is not shared with external users. However, investors do not have the opportunity to use the internal information of the bank while making their decisions. Investors can make their decisions based on information disclosed to the public at macro and micro levels. That’s why, it is more diffcult for investors to make decisions in an environment where there are many criteria. Thus, investors face a portfolio management problem (Chen, Hung, & Cheng, 2011, p. 139; Thalassinos & Kiriazidis, 2003). Investors can use different techniques when buying bank stocks. The aim of the investor is to guesstimate the price of stocks in the future and to get income on their investment. There are two main types of information that the investor will use to make predictions. These are quantitative and qualitative information (Chen et al., 2011, p. 140; Thalassinos, Hanias, & Curtis, 2012; Thalassinos, Hanias, Curtis, & Thalassinos, 2013). Quantitative information consists of values, such as, earnings per share, net proft margin, return on asset (ROA) and return of equity (ROE) and they refect the bank’s short-term performance. Qualitative information, on the other hand, consists of subjective evaluations of the bank, such as competitiveness, experience, innovation ability, product and service quality, and marketing capabilities, by experts. Based on another perspective, the market price of banks is shaped through two main factors which can be internal and external (Kanalıcı, 1997, pp. 38–57). Internal factors consist of the bank’s feld of activity, capacity, market share, sales, proftability and so on (Aydemir, Ögel, & Demirtaş, 2012; Hanias, Curtis, & Thalassinos, 2007; Lewellen, 2004; RupeikaApoga, Zaidi, Thalassinos, & Thalassinos, 2018). External factors affecting stock

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

An Assessment for Financial Performance of Banks    161 prices include various factors, such as, infation, interest rates, unemployment rate, economic growth, exchange rates and current account defcit. Many studies (Gungor & Yerdelen Kaygın, 2015; Kaya, Çömlekçi, & Kara, 2013; Sayılgan & Süslü, 2011; Ugurlu, Thalassinos, & Muratoglu, 2014) provide evidence that different macroeconomic events affect stock prices. Along with the internal and external factors, the market actor’s expectations for the future also affect the formation of the market price of the stocks. Changes in expectations reduce the correlation between both internal and external information and stock prices. Bank proftability, asset and resource composition, or any external factor can have an impact on the market price. The decision of an investor based on a single factor raises the problem of making the wrong choice. Therefore, there is a need for a performance measurement, in which a large number of criteria can be assessed together. Instead of making decisions based on one or more criteria, the use of multiple criteria decision making (MCDM) methods may allow more accurate decision making. MCDM will help to calculate a better performance indicator for the bank, taking into account a number of bank performance criteria. The aim of this study is to detect the fnancial performance of the banks listed in Borsa Istanbul (BIST) using MCDM models with the help of determined fnancial ratios and to analyze the relationship between the performance measures obtained and the market price. In this study, Development and Investment Banks and Participation Banks in Turkey are not included in the analysis. Therefore, 15 different ratios were calculated using the annual data of 11 banks listed in BIST for the period between 2002 and 2018. The TOPSIS, which is a widely used tool from the MCDM models, is used in the study, where gray relational analysis (GRA) models are handled, and the analytical hierarchy process (AHP) model was used to determine the weight of the fnancial ratios. In the following sections of the study, the literature review, methodology and fndings are provided, respectively. After that, the conclusion of the study and suggestions for future studies are explained.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

2. Literature Review From the literature, one can see that there are many kinds of MCDM models and many studies used these models. Akçakanat, Eren, Aksoy, and Ömürbek (2017) classifed the banks in Turkey based on bank asset size, having small-, medium- and large-scale performance according to MCDM models and evaluated with ENTROPE WASPAS. Total assets, total loans and receivables, total deposits, total equity, number of branches and number of personnel are variables determined as main criteria. The results of the analysis showed that Ziraat Bank is among the large-scale banks, Finansbank among the medium-scale banks and Anadolu Bank among the small-scale banks. Erpolat Taşabat et al. (2015) evaluated the performances of 21 deposit banks in the Turkish Banking system using MCDM methods such as ELECTRE, TOPSIS, VIKOR, PROMETHEE, ORESTE, MAPPAC and WSA. The researchers also used “equal weight, Saaty scoring” weighting techniques while applying these methods and 16 fnancial ratios were used as criteria. The fndings of this study showed that the best four banks in Turkey are Akbank, Yapı Kredi Bank, Garanti Bank and Citibank.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

162    Eser Yeşildağ et al. Çağıl (2011) analyzed 13 Turkish public and private deposit banks and 11 foreign banks in 2008 using MCDM methods to evaluate the performance of the banks during the global fnancial crisis using ELECTRA. In this study, covering the period between 2006 and 2010, 10 ratios of banks were determined as important criteria. Results showed that the crisis affected banks performances. Accordingly, in the crisis period, the performance of some banks went ranked higher and other banks ranked lower. Çalışkan and Eren (2016), who analyzed 20 public, private and foreign capital deposit banks, evaluated their performance using AHP and PROMETHEE, which are among the MCDM methods. In their study, which covered the period between 2010 and 2014, 10 ratios of banks were determined as important criteria. The fndings showed that the best performing bank was Ziraat Bank. Demireli (2010) also measured fnancial performance of state-owned banks in Turkey using the TOPSIS method for a data set between 2001 and 2007. The researcher determined that there was no signifcant positive improvement in the performance of public banks during the study period and that the performance was negatively affected by local and global crises. Studies investigating the fnancial performance of banks in Turkey continue to get published extensively, using MCDM model. Akgül (2019) measured the fnancial performance of banks in the Turkish banking system between 2010 and 2018 using SAW, MAUT and ARAS methods. The author determined that in 2010 banks’ fnancial performance was at its best while in 2018 it was at its worst. It can be said that this is a result of the excessive exchange rate volatility seen in Turkey in 2018. It is observed that there are a limited number of studies in Turkey measuring the fnancial performance of banks by using different MCDM models. There are studies done in other countries that measure the performance of the banks using the MCDM model (Cuong, Hien, & Long, 2018; Dash, 2016; Duarte Júnior, 2018; Siew, Fai, & Hoe, 2017; Wanke, Azad, Barros, & Hassan, 2016). In Iran, Amile, Sedaghat, and Poorhossein (2013) measured the performance of public, private–public and private banks with the TOPSIS method. The authors included non-fnancial factors, such as, service quality, and determined that fnancial performance alone was not suffcient to explain the total performance of banks. Wanke et al. (2016) determined the fnancial performance of 114 Islamic banks from 24 countries. They applied the TOPSIS method and revealed in their study that the highest performance bank is from Turkey. Cuong et al. (2018) also determined the performances of commercial banks in Vietnam with the MCDM model. The authors stated that the AHP model is a suitable model for bank performance evaluation. The fnancial performance of Brazilian banks was also measured by the Todim Fuzzy method (Duarte Júnior, 2018). Hemmati, Dalghandi, and Nazari (2013) also used data envelopment analysis (DEA) and TOPSIS methods in their studies and determined that 9 out of 16 banks in Iran are operating effectively and private banks are more successful. TOPSIS method is one of the most used methods in performance evaluation. Another study using DEA belongs to Özdemir (2019). Siew et al. (2017) conducted one of the studies using the TOPSIS method in the performance evaluation of Malaysian banks. Kosmidou, Pasiouras, Doumpos, and Zopounidis (2004) determined the performances of domestic and foreign banks in the UK by UTADIS MCDM. As a result of the study, it has been shown that domestic

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

An Assessment for Financial Performance of Banks    163 bank performances are superior to foreign banks in accordance with the home advantage theory. One of the purposes of measuring bank performance is the possibility of investors to invest in bank stocks. Assuming that the increase in the fnancial success of banks should refect on stock prices, a positive correlation is expected between the fnancial performance of banks and market prices. Although there are many studies in the literature that determine bank performances, studies on the performance– market price relationship are scarce (Baldacchino et al., 2020; Bektaş & Tuna, 2013; Demir, 2001; Kabajeh, Al Nuaimat, & Dahmash, 2012; Lewellen, 2004; Mandic et al., 2014; Taşkın, 2011). Their studies investigated the relationship between bank fnancial ratios and market price. Lewellen (2004) in his study covering a date set period between 1946 and 2000, determined that there is a relationship between market-based ratios such as market price/book value (P/BV) ratio and price/earnings (P/E) ratio and stock returns. However, P/E ratio affects short-term stock returns and P/BV affects long-term stock returns. In addition, the dividend return rate explains long-term market returns. Kabajeh et al. (2012) investigated the relationship between ROA rate, ROE, return on investment (ROI) ratio and the market price of insurance companies in Jordan in the 2002–2007 period. The authors found a positive but weak relationship between the ROA and ROI and the market price separately. However, there was no relationship between ROE and market price. Demir (2001) determined that the most important ratios which explained stock returns in the banking sector in 1991 and 2000 are P/BV, P/E ratio and proftability ratios (proft per share, equity proftability). Bektaş and Tuna (2013) investigated the variables explaining the stock returns for 2011 in the banking sector using the canonical correlation method. According to the study, ROA and P/BV are the two variables that best explain stock returns. There is only a limited number of studies in the literature which investigate the relationship between fnancial performances of companies by MCDM models and markets prices. The studies of Sakarya and Aytekin (2013), Özen, Yesildag, and Soba (2015) and Işık (2019) are three studies which are found in the literature. Sakarya and Aytekin (2013), after determining the fnancial performances of the banks traded in BIST in the period between 2007 and 2011, they investigated the relationship between the banks’ market returns and calculated measure of fnancial performance. As a result of the analysis, the authors concluded that foreign banks showed higher performance than domestic banks and there was no statistically signifcant relationship between fnancial performance values and stock returns on either a year or a bank basis. Özen et al. (2015) measured the performances of four companies from the BIST food industry with the TOPSIS method. As a result of the study, no statistically signifcant relationship was found between TOPSIS performance scores and the market prices of companies. However, a statistically signifcant relationship was found between the proftability rates of one of the companies with the highest asset size and the market price among these four companies. Işık (2019), on the other hand, determined the fnancial performance of companies in Borsa İstanbul based on 30 indexes by the TOPSIS method and examined their relations with the market price. It can be said that this result arose by using three years of data.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

164    Eser Yeşildağ et al. When the literature is analyzed, many academic studies are found in which bank performances are analyzed by MCDM. However, there are only a few studies addressing the relationship between fnancial performances determined by the MCDM and the stock market prices. Therefore, it is necessary to investigate the relationship between the banks’ fnancial performance-stock price in order to support investors to make investment decisions.

3. Data Set and Methodology Data Set The data set of this study consists of listed banks in BIST from 2002 to 2018. All the banks operate as commercial banks and their fnancial performance was detected by using 15 selected fnancial ratios. The selected fnancial ratios were divided into fve main groups: capital adequacy, asset quality, liquidity, proftability and income/expense structure as per the following: Capital adequacy ⦁⦁ Capital adequacy standard ratio ⦁⦁ Shareholders’ equity/total assets ⦁⦁ (Shareholders’ equity-permanent assets)/total assets

Asset quality ⦁⦁ Total loans/total assets ⦁⦁ Total loans/total deposits ⦁⦁ Loans under follow-up (gross)/total loans

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Liquidity ⦁⦁ Liquid assets/total assets ⦁⦁ Liquid assets/short-term liabilities ⦁⦁ Liquid assets/(deposits + non-deposit funds)

Proftability ratios ⦁⦁ Average ROA ⦁⦁ Average return on shareholders’ equity ⦁⦁ Net proft (losses)/paid-in capital

Income–expenditure structure ⦁⦁ Net interest income after specifc provisions/total assets ⦁⦁ Net interest income after specifc provisions/total operating income ⦁⦁ Non-interest income (net)/total assets

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

An Assessment for Financial Performance of Banks    165 All the data sets were collected from the consolidated fnancial statement of the banks. The fnancial statements of banks were downloaded from the following website: https://www.tbb.org.tr/. The list of banks and their stock exchange codes are provided in Table 1. In the study, MCDM Technique, which is one of the methods most commonly used in performance evaluation since the 1980s, is discussed. As revealed in the literature review phase, TOPSIS and GRA models, which are among the ÇKKV techniques, are used extensively by researchers. The fact that these models stand out individually, especially in the studies in the feld of banking, reveal the necessity of a study in which these two models are used simultaneously. Therefore, TOPSIS and GRA models, which are among the MCDM models, were used as the basic analysis/ranking models, while the AHP model was used as an auxiliary model in determining the weight of fnancial ratios. In addition, a correlation analysis was performed by comparing the performance values obtained with the MCDM models with the annual returns of 11 banks.

3.1 Gray Relational Analysis GRA consists of the sub-headings of gray modeling developed by Deng (1982). The term “gray” in the model means that information is missing or unknown. Similarities or differences between two elements or two sub-systems within a given system are called “gray relationships.” The method used to measure the developments in the degree of changes in the similarities and differences between the elements is called as GRA. GRA is a method for determining the degree of relationship between each factor in a gray system and the factor (reference series) series is compared. Each factor is defned as an array; it can be a row or column. The degree of effect between factors is named as gray relational degree (Feng & Wang, 2000; Sofyalıoğlu, 2011; Üstünışık, 2007). GRA is usually preferred for

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Table 1.  Listed Banks and Codes. Name of Bank

Stock Exchange Code

Akbank T.A.Ş.

AKBNK

Denizbank A.Ş.

DENIZ

Türkiye Garanti Bankası A.Ş.

GARAN

Türkiye Halk Bankası A.Ş.

HALKB

ICBC Turkey Bank A.Ş.

ICBC

Türkiye İş Bankası A.Ş.

ISCTR

QNB Finansbank A.Ş.

QNBFN

Şekerbank T.A.Ş.

SEKER

Türk Ekonomi Bankası A.Ş.

TEBNK

Türkiye Vakıfar Bankası T.A.O.

VAKBN

Yapı Kredi Bankası A.Ş.

YKBNK

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

166    Eser Yeşildağ et al. grouping variables when the sample size is small or the sample distribution is unknown (Feng & Wang, 2000). GRA has been widely used in different disciplines at various times. Hsu and Wen (2000) used GRA in airline destination planning, Lin and Hsu (2002) applied GRA for estimating the sales volume of companies, Bass, Huang, and Russo (1997) used the GRA in the prediction and valuation of risk and Peker and Baki (2011) and Kula et al. (2016) evaluated the fnancial performance of companies through fnancial ratios using GRA. There are three main steps in GRA. The frst step is data pre-processing. Data pre-processing is usually required when the range or unit in one data sequence is different from others or the sequence scatter range is too large. Data pre-­ processing is a method of transferring the original data sequence to a comparable sequence. Therefore, data must be normalized, scaled and polarized frst into a comparable sequence before proceeding to the other steps. The processing is called generation of gray relation or standard processing (Sallehuddin, Shamsuddin, & Hashim, 2008). The steps of GRA can be defned as follows (Feng & Wang, 2000; Wen, 2004; Wu, 2002). Step 1: Creating the decision matrix (X i ) In the frst step of the GRA, the following matrix is created for the reference series of length n.



 x (1) x (2) 1  1  x (1) x 2 (2) X i =  2      xn (1) xn (2) 

 x1 ( n )    x2 ( n )   (1)      xn ( n )  

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Step 2: Creating the reference serial matrix Reference series can be shown as x0 = ( x0 (1), x0 (2),..., x0 ( j ),..., x0 ( n )) . In this series, the term x0 ( j ), j . shows the greatest value of the criterion within the normalized values. The lowest value is taken as a reference for the series to be a minimum. The reference series matrix is created by stating the reference series on the frst line of the decision matrix. Step 3: Normalizing the data set and creating the normalized matrix In this step, the data set is normalized and three possible situations can be encountered: (a) If the expectancy is the larger-the-better, then it can be expressed by Eq. (2):

xi* =

xi ( j ) − min xi ( j ) j

max xi ( j ) − min xi ( j ) j

(2)

j

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

An Assessment for Financial Performance of Banks    167 (b) If the expectancy is the smaller-the-better, then it can be expressed by Eq. (3): xi* =



max xi ( j ) − xi ( j ) j

max xi ( j ) − min xi ( j )

(3)

j

j

(c)  If the expectancy is nominal-the-best (optimal value), then it can be expressed by Eq. (4):

xi* =



xi ( j ) − x0 b ( j ) (4) max xi ( j ) − x0 b ( j ) j

In Eq. (4), the target value of criteria is represented by x0 b ( j ) j . and ıt takes a value between the following interval max xi ( j ) ≥ x0 b ( j ) ≥ min xi ( j ). After these j j steps, the decision matrix in (1) becomes the following equation:  * *  x1 (1) x1 (2)  *  x (1) x2* (2) X i* =  2      x * (1) x * (2) n  n



  x1* ( n )    x2* ( n )   (5)     *  xn ( n )  

Step 4: Creating the absolute values of series The absolute value between x0* and xi* can be found as (∆oi ( j )) by using the following equation:

Copyright © 2020. Emerald Publishing Limited. All rights reserved.



 ∆ (1) ∆ (2) 01 01    ∆02 (1) ∆02 (2) * * ∆0i ( j ) = x0 ( j ) − xi ( j ) =       ∆0 m (1) ∆0 m (2) 

 ∆01 ( n )    ∆02 ( n )  (6)      ∆0 m ( n )  

Step 5: Creating the gray relational coeffcient matrix γ 0i ( j ) =



∆ min+ ξ∆ max (7) ∆0i ( j ) + ξ∆ max

The GRA Coeffcient Matrix is calculated by using the equation provided above. The coeffcient of ζ in Eq. (7) is called a separator and gets a value between 0 and 1. However, it is recommended to take 0.5 in fnancial performance assessments. According to Wen (2004), the coeffcient of ζ does not change the order which will occur after the gray relational degree, if it takes a value between 0 and 1. In addition, it can be calculated as ∆ max = max max ∆oi ( j ) and i j ∆ min = min min ∆oi ( j ) too. i

j

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

168    Eser Yeşildağ et al. Step 6: Calculating the gray relational coeffcient

1 n ∑ γoi ( j ) (8) n j =1

Γoi =

In Eq. (8), the coeffcient of Γoi i. shows the gray relational coeffcient and assumes that the criteria are of equal importance. If the criteria have different weights, this time the following equation must be used.

n

Γoi = ∑ [Wi ( j )γoi ( j )](9) j =1

3.2 TOPSIS TOPSIS method was originally developed in 1981 by Yoon and Hwang. It is based upon the concept that the chosen alternative should have the shortest distance from the ideal solution and the farthest from the negative-ideal solution. Assuming that each attribute takes the monotonically increasing (or decreasing) utility; then it is easy to locate the “ideal” solution which is composed of all the best attribute values attainable, and the “negative-ideal” solution is composed of all the worst attribute values attainable. One approach is to take an alternative, which has the (weighted) minimum Euclidean distance to the ideal solution in a geometrical sense (Hwang & Yoon, 1981). The TOPSIS method evaluates the following decision matrix that contains m alternatives associated with n attributes (or criteria):

Copyright © 2020. Emerald Publishing Limited. All rights reserved.



      Aij =       

x11

x12

...

x21

x22

...

. .x i 1

x

x

. xm1

xm 2

i2

ij

...

x1n  x2 n   .   .x in   .  xmn  

where Ai is the ith alternative considered and xij is the numerical outcome of the ith alternate with respect to the jth criterion. The larger the attribute outcomes, the greater the preference for the “beneft” criteria and the less the preference for the “cost” criteria. Furthermore, any outcome which is expressed in a non-numerical way should be quantifed through the appropriate scaling technique. Since all criteria cannot be assumed to be of equal importance, the method receives a set of weights from the decision-maker. For the sake of simplicity, the proposed method is presented as a series of successive steps (Hwang & Yoon, 1981):

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

An Assessment for Financial Performance of Banks    169 Step 1. Construct the normalized decision matrix: This process tries to transform the various attribute dimensions into nondimensional attributes, which allows comparison across the attributes. One way is to take the outcome of each criterion divided by the norm of the total outcome vector of the criterion at hand. An element rij of the normalized decision matrix R can be calculated as: rij =

xij m

∑x

2 ij

i =1

Consequently, each attribute has the same unit length of vector. Step 2. Construct the weighted normalized decision matrix: A set of weights (w = w1,  w2 …….., w j ,…….wn), ∑ j =1 w j = 1, from the decisionmaker is accommodated to the decision matrix in this step. This matrix can be calculated by multiplying each column of the matrix R with its associated weight  w j . Therefore, the weighted normalized decision matrix V is equal to: n

 wr  1 11 w2 r12 ... wn r1n  wr  1 21 w2 r22 ... wn r2 n  . V =  . . .   . .   wr w r w ... 2 m2 n rmn  1 m1

           

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Step 3. Determine ideal and negative-ideal solutions: Let the two artifcial alternatives A* and A− be defned as:

{

A* = (max vij j ∈ J ),(min vij j ∈ J ' i

i

}

i = 1,2,3,…m} = {v1* , v2* ,..., vn* }

{

A− = (min vij j ∈ J ),(max vij j ∈ J ' i

i

}

i = 1,2,3,…m} = {v1− , v2− ,..., vn− }

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

170    Eser Yeşildağ et al. where j = { j = 1,2, …,n |  j associated with beneft criteria} and j ı = { j = 1,2, ..., n | j    associated with cost criteria}. Then it is certain that the two created alternatives A* and A− indicate the most preferable alternative (ideal solution) and the least preferable alternative (negativeideal solution), respectively. Step 4. Calculate the separation measure: The separation between each alternative can be measured by the n-dimensional Euclidean distance. The separation of each alternative from the ideal one is then given by Si* =

n

∑ (v

ij

− v *j )2  i = 1,2,3,…, m.

j =1

Similarly, the separation from the negative-ideal one is given by Si− =

n

∑ (v

ij

− v −j )2  i = 1,2,3,…, m.

j =1

Step 5. Calculate the relative closeness to the ideal solution: The relative closeness of Ai with respect to A* is defned as:

ci* =s i − / s i − + s i * and 0 < Ci * < 1, i = 1,2,3,…, m. It is clear that Ci * = 1 if Ai =  A* and Ci * = 0 if Ai =  A−. An alternative Ai is closer to A* as Ci * approaches to 1.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Step 6. Rank the preference order. A set of alternatives can now be preference-ranked according to the descending order of Ci * . In the following part of the study, the analysis and evaluation of the fndings are presented, and the analysis results will be interpreted in tables.

4. Findings At this stage, 11 banks listed in BIST in Turkey were ranked according to TOPSIS and GRA models by using 15 different fnancial ratios. Firstly, TOPSIS analysis was applied and scores for 11 banks were obtained. According to these scores, 11 banks were ranked on an annual basis. Banks’ annual rows are presented in Table 2. The banks in the frst three and the last three levels were generally taken into consideration. According to Table 2, QNBFN, which was in the top three for 13 years, draws attention in terms of high performance. HALKB, which was in the top three for 10 years and DENIZ, which was in the top three for 8 years, followed.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

2

1

2

7

4

6

6

6

5

5

5

6

6

7

5

4

5

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

2018

AKBNK

2002

Year/Bank

7

8

7

6

4

5

1

1

3

2

1

3

3

5

4

3

5

1

2

1

4

5

3

4

4

4

4

5

1

8

9

8

6

10

4

1

2

1

1

1

2

3

2

3

3

4

5

8

5

4

9

DENIZ GARAN HALKB

9

3

6

2

9

9

9

9

9

8

8

7

7

10

9

9

6

ICBCT

Table 2.  Ranking of 11 Banks According to TOPSIS Scores.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

2

5

3

5

3

4

6

6

6

5

4

5

6

6

6

5

8

ISCTR

3

6

4

3

2

2

3

2

1

1

2

2

2

4

3

7

3

10

10

10

10

10

10

10

10

10

10

10

10

10

11

10

10

1

QNBFN SEKER

8

9

9

8

7

7

7

7

7

7

7

8

1

2

7

2

7

TEBNK

6

7

8

9

8

8

8

8

8

9

9

9

9

3

1

8

4

11

11

11

11

11

11

11

11

11

11

11

11

11

1

11

11

11

VAKBN YKBNK

An Assessment for Financial Performance of Banks    171

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

172    Eser Yeşildağ et al. According to Table 2, it is understood that the Banks which do not perform well in terms of fnancial performance are mostly the same banks. These banks are YKBNK, SEKER and ICBCT. The other fve banks were in the 4th and 8th rows and did not show major changes in these rows. In the study, the GRA method was used and scores for 11 banks were obtained. With the help of these scores, the annual performances of 11 banks were determined. Banks’ annual fnancial performance rankings are presented in Table 3. In the interpretation of Table 3, the banks in the frst three and the last three positions were generally taken into consideration. According to Table 3, AKBNK, which was in the top three for 13 years, draws attention in terms of high performance. This was followed by HALKB, which was in the top three for 10 years and then DENIZ, which was in the top three for 8 years. In addition to this, HALKB had the highest fnancial performance both for being in the top three for 12 years and generally ranking frst in these years. Subsequently, this bank was followed by GARAN and DENIZ, which ranked in the top three for seven years in terms of fnancial performance. According to Table 3, it is understood that the Banks which do not perform well in terms of fnancial performance are mostly the same banks. These banks are ICBCT, SEKER and TEBNK. If the results of the analysis using TOPSIS and GRA methods are compared, one can conclude that the banks that perform best in terms of fnancial performance according to both methods are HALKB and DENIZ. However, QNBFN, which showed the best fnancial performance in TOPSIS method, was not among the top three banks according to the GRI method. Likewise, GARAN, which ranked third in the GRI method for fnancial performance, was not among the top three banks in the TOPSIS method. According to TOPSIS and GRA methods, when the banks that performed in the last three ranks among 11 banks were compared, it can be concluded that SEKER and ICBCT banks performed worst among the 11 banks in both methods. However, YKBNK, which showed the worst performance among 11 banks in terms of fnancial performance according to the TOPSIS method, was not included in the last three according to the GRA method. Likewise, TEBNK, which is one of the worst performing banks among 11 banks according to GRA method, was not included in the last three according to TOPSIS method. In this part of the study, the signifcance of the scores obtained by TOPSIS and GRA analysis methods as a fnancial performance indicator was tested. For this, frstly, TOPSIS and GRA analysis scores and Bank Returns were compared. Bank returns were calculated by taking into account the stock prices of the 11 banks between 2002 and 2018. Accordingly, the correlation levels between the TOPSIS and GRA scores of the 11 banks and their returns are analyzed, and the correlation levels are shown in Table 4. When Table 4 and TOPSIS correlations are taken into consideration, one can see that the correlation levels between the 11 Bank returns and TOPSIS scores varied between 0.336 and −0.227. At the same time, it is understood that the returns of six banks and TOPSIS scores are negative and the relations are weak. In addition, correlation analysis results showed that the relationships were not signifcant at the 0.05 and 0.01 levels. This shows that TOPSIS scores do not move

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

3

1

2

2

3

4

2

2

3

4

3

3

3

4

3

3

5

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

2018

7

7

8

7

2

6

2

2

5

3

5

3

2

3

6

5

11

3

2

2

2

4

4

5

3

2

5

4

1

6

8

8

7

10

1

1

1

1

1

1

1

1

1

1

1

2

4

5

5

6

5

AKBNK DENIZ GARAN HALKB

2002

Year/Bank

11

11

11

11

11

11

11

11

11

11

11

11

11

10

10

11

6

ICBCT

Table 3.  Ranking of 11 Banks According to GRA Scores.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

4

5

5

5

5

5

4

5

4

4

3

5

8

4

7

10

7

ISCTR

6

9

9

9

8

8

8

8

8

8

8

7

5

7

9

8

4

10

10

10

10

10

9

9

9

10

9

6

8

7

9

3

3

9

QNBFN SEKER

9

8

7

8

9

10

10

10

9

10

10

10

1

1

4

2

8

2

4

4

3

6

7

6

6

7

6

7

6

9

6

1

4

2

8

6

6

6

7

2

7

7

6

7

9

9

10

11

11

9

1

TEBNK VAKBN YKBNK

An Assessment for Financial Performance of Banks    173

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

174    Eser Yeşildağ et al. Table 4.  Correlation Level between Bank Returns and TOPSIS and GRA Analysis Scores.

AKBNK Pearson correlation Sig. (two-tailed) DENIZ

Pearson correlation Sig. (two-tailed)

GARAN Pearson correlation Sig. (two-tailed) HALKB ICBCT ISCTR QNBFN SEKER TEBNK VAKBN

GRA Correlation

0.336

0.401

0.188

0.111

−0.085

−0.359

0.745

0.157

−0.046

0.004

0.862

0.989

−0.121

0.218

Sig. (two-tailed)

0.709

0.496

Pearson correlation

0.169

0.026

Sig. (two-tailed)

0.516

0.920

−0.151

−0.043

Sig. (two-tailed)

0.562

0.869

Pearson correlation

0.095

−0.061

Sig. (two-tailed)

0.716

0.816

Pearson correlation

Pearson correlation

Pearson correlation

−0.274

0.566*

Sig. (two-tailed)

0.287

0.018

Pearson correlation

0.286

0.396

Sig. (two-tailed)

0.266

0.115

−0.076

0.084

0.797

0.776

0.148

−0.321

0.571

0.209

Pearson correlation Sig. (two-tailed)

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

TOPSIS Correlation

YKBNK Pearson correlation Sig. (two-tailed)

AKBNK DENIZ GARAN HALKB ICBCT ISCTR QNBFN SEKER TEBNK VAKBN YKBNK

*Correlation is signifcant at the 0.05 level (two-tailed).

in the same or opposite direction with the annual returns of banks. In other words, the relationship between returns and TOPSIS scores appears to be random. When Table 4 and GRA correlations are taken into consideration, one can observe that the correlation levels between the 11 Bank returns and GRA scores varied between 0.566 and −0.359. Similarly, it was understood that there was a weak correlation between bank returns and TOPSIS scores in general, and it was also found that there was a negative relationship between four bank returns and TOPSIS score. In addition, SEKER is the only bank with a positive correlation between bank returns and GRA analysis scores (0.566), and this relationship is signifcant at the level of 0.05. This means that if the fnancial performance of

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

An Assessment for Financial Performance of Banks    175 SEKER bank increases, its returns will increase. Despite this situation, the correlations between bank returns and GRA scores, except for SEKER bank, are not signifcant at the level of 0.05 and 0.01. In this case, it can be said that there is no signifcant relationship between GRA scores and annual returns of banks. In other words, the relationship between GRA scores and bank returns is random. Finally, the authors investigated whether the correlation that cannot be found between annual bank returns and TOPSIS and GRA scores is among the fnancial ratios. For this, the fnancial ratios of the banks were taken into consideration one by one and the authors investigated whether there is a correlation with the bank share returns. Correlation analysis results between the returns and fnancial ratios of the 11 banks used in the study can be seen in Table 5. When Table 5 is analyzed in general, as a result of the correlation tests, one can observe that there is a correlation between very few fnancial ratios and bank returns and, it is understood that there are very few signifcant relationships. If Table 5 is examined in detail, it is noteworthy to see that there are signifcant relationships between AKBNK (which had the best fnancial performance among 11 banks, especially according to GRA) stock returns and almost all ratios at the levels of 0.05 and 0.01. In addition, there are linear and signifcant relationships between all rates and returns of banks except asset quality rates. Again, one notes that there are signifcant relationships between SEKER (Bank with low performance according to TOPSIS and GRA analyses) stock returns and fnancial ratios. Another detail emerging from Table 5 is that the fnancial ratios with the highest signifcance are Average ROA and non-interest income (Net)/total assets. Signifcant relationships were identifed between these two fnancial ratios and the returns of the three banks. Apart from these two ratios, signifcant relationships were found between some bank fnancial ratios and two bank returns. These rates are as follows: capital adequacy standard ratio (shareholders’ equity-permanent assets)/total assets ratio, liquid assets/total assets ratio, liquid assets/(deposits + non-deposit funds) ratio, average return on shareholders’ equity ratio and fnally net interest income after specifc provisions/total assets ratio. Finally, another important issue is that there was no signifcant relationship between HALKB stock returns and its fnancial ratios, which was successfully identifed in terms of fnancial ratios in both analysis methods.

5. Results There are many criteria that can infuence fnancial investors to make accurate stock investment decisions. These criteria may cause diffculty when deciding to make acquisitions The performance indicator, which is reduced to one dimension instead of many performance criteria, makes it easier to make fnancial investment decisions. Therefore, it is necessary to investigate whether different multi-dimensional decision models are used and whether they are suitable for use. In addition, it is necessary to determine which MCDM is a more suitable model.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Total loans/ total assets

Loans under follow-up (gross)/total loans

0.769 −0.092

0.411

0.025

0.518* −0.341

0.031 0.302

0.197

−0.329

Pearson −0.524* −0.266 correlation

Sig. (twotailed)

0.507

0.481

0.874

Sig. (twotailed)

0.173

0.184

0.724

0.077

0.542* −0.213

0.181

0.133

0.734

0.031

0.033

0.380

0.522* −0.089

Pearson −0.042 correlation

(Shareholders’ Pearson equitycorrelation permanent Sig. (twoassets)/total tailed) assets

Shareholders’ Pearson equity/total correlation assets Sig. (twotailed)

Capital Pearson adequacy correlation standard ratio Sig. (twotailed)

0.350

−0.296

0.224

0.380

0.294

0.331

0.234

0.372

0.121

0.472

0.492

0.179

0.760

−0.080

0.813

−0.062

0.653

−0.118

0.406

−0.216

0.238

−0.302

0.577

0.146

0.693

0.103

0.093

0.421

0.102

0.410

0.667

−0.113

0.068

−0.452

0.489

−0.180

0.959

0.014

0.388

−0.224

0.614

0.132

0.969

0.010

0.507

−0.173

0.877

0.041

0.047

0.155

−0.360

0.831

0.056

0.287

0.489* −0.274

0.082

0.433

0.010

0.604*

0.534

−0.182

0.402

0.243

0.347

0.272

0.459

0.216

0.867

0.049

0.487

−0.181

0.788

0.071

0.715

−0.095

0.302

0.266

0.521

0.167

AKBNK DENIZ GARAN HALKB ICBCT ISCTR QNBFN SEKER TEBNK VAKBN YKBNK

Table 5.  Correlation Levels between Stock Returns and Financial Ratios of Banks.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

176    Eser Yeşildağ et al.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

0.384 0.210 0.418

0.027

0.568*

0.017

Pearson correlation

Sig. (twotailed)

Average return on shareholders’ equity

0.028

0.533*

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

0.070

0.006

Sig. (twotailed)

0.040

0.450

0.636**−0.502*

0.040

Pearson correlation

Sig. (twotailed)

Average ROA Pearson correlation

Liquid assets/ (deposits + non-deposit funds)

0.087

0.427

0.089

0.226

0.535*

Liquid assets/ Pearson short-term correlation liabilities Sig. (twotailed)

0.032

0.003

0.356

0.020

Sig. (twotailed)

0.521*

0.426

0.239

0.556*

Liquid assets/ Pearson total assets correlation

0.218

−0.315

0.667**−0.501*

0.331

Sig. (twotailed)

0.050

Total loans/ Pearson −0.483* −0.251 total deposits correlation

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

0.058

0.560

0.095

0.504

0.659

0.142

0.818

0.075

0.687

0.130

0.397

−0.270

0.401

−0.218

0.320

−0.257

0.471

0.188

0.583

0.143

0.491

0.179

0.180

0.341

0.931

−0.023

0.473

0.187

0.148

0.367

0.542

0.159

0.231

0.307

0.322

−0.256

0.620

0.130

0.629

0.126

0.841

0.053

0.142

0.372

0.786

0.071

0.579

0.145

0.060

0.464

0.015

0.578*

0.061

0.464

0.486

−0.181

0.083

0.433

0.472

−0.187

0.452

0.195

0.198

0.329

0.155

0.361

0.277

0.280

0.223

0.312

0.200

−0.327

0.346

−0.244

0.273

−0.282

0.491

0.179

0.424

0.208

0.636

0.124

0.556

−0.154

(Continued )

0.092

0.467

0.218

0.351

0.442

0.224

0.808

0.072

0.416

0.237

0.512

−0.192

An Assessment for Financial Performance of Banks    177

0.806

0.308 0.095 0.718

0.040

0.491*

0.046

Sig. (twotailed)

0.490

0.221

0.074

0.533

0.223

0.380

0.776

−0.092

0.767

−0.078

0.606

−0.135

0.667

0.112

0.426

−0.207

0.049

0.485*

0.945

−0.018

0.154

−0.362

0.512

−0.171

0.146

0.368

0.737

0.088

0.485

−0.182

0.962

−0.012

0.363

0.235

0.025

0.539*

0.765

−0.078

0.099

0.413

0.094

0.420

0.250

0.295

0.757

−0.081

0.179

0.342

*Correlation is signifcant at the 0.05 level (two-tailed); **correlation is signifcant at the 0.01 level (two-tailed).

0.035

0.515*

0.064

0.503* −0.263

Pearson correlation

Non-interest Pearson income (net)/ correlation total assets Sig. (twotailed)

Net interest income after specifc provisions/ total assets

0.218

0.389

−0.315

0.892

0.595

0.148

0.514

−0.036

−0.223

−0.367

0.170

0.714

−0.108

0.134

0.421

0.337

0.277

0.822

−0.066

0.674

0.110

0.265

−0.287

0.124

−0.388

0.270

−0.284

AKBNK DENIZ GARAN HALKB ICBCT ISCTR QNBFN SEKER TEBNK VAKBN YKBNK

0.139

Net interest Pearson income correlation after specifc Sig. (twoprovisions/ tailed) total operating income

Net proft Pearson (losses)/paid- correlation in capital Sig. (twotailed)

Table 5.  (Continued )

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

178    Eser Yeşildağ et al.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

An Assessment for Financial Performance of Banks    179 This study aimed to analyze the fnancial performance of commercial banks listed in BIST by MCDM models with the help of the determined fnancial ratios and to determine the relationship between the determined fnancial performance measures and the market price. It also explored which performance measure has higher correlations with market prices. Development and Investment Banks and Participation Banks are not included in the analysis. In this study 11 commercial banks, trading constantly in BIST between 2002 and 2018, were ranked by using TOPSIS and GRA. Fifteen different fnancial ratios were used under fve main headings, the AHP model was used to determine the weight of the ratios. According to the TOPSIS method, the best banks are listed as QNBFN, HALKB and DENIZ. According to the GIA method, HALKB, which is a public bank, ranked the best bank. Garanti bank and DENIZ follow HALKB. Although the ranking varies according to two different methods, HALKB and DENIZ are in the top three in both methods. The findings showed that there is no significant correlation between financial performances measured by both methods and market returns except for SEKER bank. This shows that the market prices are shaped by different factors besides the banks’ overall financial performance. On the other hand, when the correlation between price and banks’ financial ratios is analyzed; there seems to be no substantial link between financial ratios and stock returns. In this case, the authors conclude that external factors or expectations have a greater affected on stock prices than fnancial statements and measured fnancial performances by MCDM. When the limited literature fndings are compared, it can be seen that the fndings are compatible with the literature. No relation was found between the performance indicators obtained in the studies of Sakarya and Aytekin (2013), Özen et al. (2015) and Işık (2019) and market prices. Another important study is that these two methods are not successful to refect the correlation between fnancial performances and market price. According to the obtained results, it turns out that performance scores cannot be effective in making fnancial investment decisions for now. The study results contain information that can be used for different purposes by bank managers, academics and fnancial investors.

Limitations of the Study It can be said that the annual fnancial data used in the study, as well as the annual market price data of banks, resulted in a very low correlation between fnancial performance measures and market price. However, the frequency and market price frequencies of the fnancial statements are different from each other. In order to increase the correlation between performance indicators and market prices, fnancial data should be used quarterly and working period should be extended depending on the possibility of providing data.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

180    Eser Yeşildağ et al.

Further Studies In order to fnd a statistically signifcant relationship between performance indicators and market prices, a larger and longer-term data set consisting of market data from developed countries should be used. Thus, it may be possible to see that the company performance will be refected in the market prices when the studies are carried out with market data from different countries. On the other hand, the study can be repeated in future studies using more MCDM models.

Acknowledgment This study is the extended and revised version of conference paper named as “Türkiye’de Faaliyet Gösteren Bankaların Finansal Performansının Çok Kriterli Karar Verme Modelleri ile Sıralanması” which was presented at 1. International Banking Congress in 2018, Ankara/Turkey.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

References Akçakanat, Ö., Eren, H., Aksoy, E., & Ömürbek, V. (2017). Bankacılık Sektöründe Entropi ve Waspas Yöntemleri ile Performans Değerlendirmesi [Performance evaluation with entropy and waspas methods in banking sector]. Süleyman Demirel University Journal of Faculty of Economics and Administrative Sciences, 22(2), 285–300. Akgül, Y. (2019). Analyzing the performance of Turkish banking system between 2010– 2018 years by multi-criteria decision-making techniques. Research of Financial Economic and Social Studies (RFES), 4(4), 567–582, ISSN: 2602-2486. doi:10.29106/ fesa.655722 Amile, M., Sedaghat, M., & Poorhossein, M. (2013). Performance evaluation of banks using fuzzy AHP and TOPSIS, case study: State-owned banks, partially private and private banks in Iran. Caspian Journal of Applied Sciences Research, 2(3), 128–138. Aydemir, O., Ögel, S., & Demirtaş, G. (2012). Hisse Senetlerinin Fiyatlarının Belirlenmesinde Finansal Oranların Rolü [The role of fnancial ratios in determining the prices of stocks]. Management and Economy, 19(2), 277–288. Baldacchino, P. J., Camilleri, A., Schembri, B., Grima, S., & Thalassinos, E. Y. (2020). Performance evaluation of the board of directors listed companies: A small state perspective. International Journal of Finance, Insurance and Risk Management, X(1), 99–119. Baldacchino, P. J., Camilleri, A., Schembri, B., Grima, S., & Thalassinos, E. Y. (2020). Performance evaluation of the board of directors listed companies: A small state perspective. International Journal of Finance, Insurance and Risk Management, X (1), 99–119. Bass, B., Huang, G., & Russo, J. (1997). Incorporating climate change into risk assessment using grey mathematical programming. Journal of Environmental Management, 49, 107–123. Bektaş, H., & Tuna, K. (2013). The performance measurement of enterprises traded in Borsa Istanbul the emerging companies market with grey relational analysis.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

An Assessment for Financial Performance of Banks    181 Çankırı Karatekin University Journal of the Faculty of Economics and Administrative Sciences, 3(2), 185–198. Çağıl, G. (2011). 2008 Küresel Kriz Sürecinde Türk Bankacılık Sektörünün Finansal Performansının ELECTRE Yöntemi ile Analizi [Analysis of fnancial performance of Turkish banking sector during 2008 global crisis via ELECTRE method]. Journal of Finance Letters, 25(93), 59–86. Çalışkan, E., & Eren, T. (2016). Evaluation of the multi-criteria decision making performance of banks (in Turkish). Ordu University Science Technology Journal, 6(2), 85–107. Chen, C. T., Hung, W. Z., & Cheng, H. L. (2011). Applying linguistic PROMETHEE method in investment portfolio decision-making. International Journal of Electronic Business Management, 9(2), 139. Cuong, D. X., Hien, H. T., & Long, T. (2018). Multi-criteria decision-making model evaluating the performance of Vietnamese commercial banks. International Journal of Financial Research, 9(1), 132–141. Dash, M. (2016). Banking performance measurement using multi-criteria decision models ELECTRE and PROMETHEE: The case of Indian banks. International Journal of Operations and Quantitative Management, 22(1), 29–37. Demir, Y. (2001). Hisse Senedi Fiyatını Etkileyen İşletme Düzeyindeki Faktörler ve Mali Sektör Üzerine İMKB’de Bir Uygulama [An application in ISE on business level factors affecting stock price and fnancial sector]. Süleyman Demirel University Journal of Faculty of Economics and Administrative Sciences, 6(2), 109–130. Demireli, E. (2010). TOPSIS multicriteria decision making method: An examination on state owned commercıal banks in Turkey. Girişimcilik ve Kalkınma Dergisi [Journal of Entrepreneurship and Development], 5(1), 101–112. Deng, J. (1982). Control problems of grey systems. Systems and Control Letters, 5(1), 288– 294. Duarte Júnior, A. M. (2018). Applying the TODIM fuzzy method to the valuation of Brazilian banks. Pesquisa Operacional, 38(1), 153–171. Erpolat Taşabat, S., Cinemre, N., & Şen, S. (2015). Farklı Ağırlıklandırma Tekniklerinin Denendiği Çok Kriterli Karar Verme Yöntemleri ile Türkiye’deki Mevduat Bankalarının Mali Performanslarının Değerlendirilmesi [Evaluation of different weighting the technical testing of multi-criteria decision making in Turkey with deposit money banks fnancial performance]. Social Sciences Research Journal, 4(2), 96–110. Feng, C. M., & Wang, R. T., (2000). Performance evaluation for airlines ıncluding the consideration of fnancial ratios. Journal of Air Transport Management, 6(1), 133–142. Güngör, B., & Yerdelen Kaygın, C. (2015). Determination of factors affecting stock price with dynamic panel data analysis. Kafkas University Journal of the Faculty of Economics and Administrative Sciences (KAUJEASF), 6(9), 149–168. Hanias, P. M., Curtis, G. P., & Thalassinos, E. J. (2007). Non-linear dynamics and chaos: The case of the price indicator at the Athens Stock Exchange. International Research Journal of Finance and Economics, 11(1), 154–163. Hemmati, M., Dalghandi, S., & Nazari, H. (2013). Measuring relative performance of banking industry using a DEA and TOPSIS. Management Science Letters, 3(2), 499–504. Hsu, C. I., & Wen, Y. H. (2000). Application of grey theory and multiobjective programming towards airline network design. European Journal of Operational Research, 127(1), 44–68. Hwang, C. L., & Yoon, K. (1981). Multiple attribute decision making: Methods and applications. New York, NY: Springer-Verlag.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

182    Eser Yeşildağ et al. Işık, Ö. (2019). The ınvestigation of association between fnancial performance and stock returns by entropy and TOPSIS methods. Journal of Urban Academy, 12(1), 200–213. Kabajeh, M. A. M., Al Nuaimat, S. M. A., & Dahmash, F. N. (2012). The relationship between the ROA, ROE and ROI ratios with Jordanian insurance public companies market share prices. International Journal of Humanities and Social Science, 2(11), 115–120. Kanalıcı, H. (1997). Hisse Senedi Fiyatlarının Tesbiti ve Tesir Eden Faktörler [Determination of stock prices and affecting factors on the stock prices]. Turkish Capital Markets Board, No: 77, Ankara. Kaya, V., Çömlekçi, İ., & Kara, O. (2013). Hisse Senedi Getirilerini Etkileyen Makroekonomik Değişkenler 2002–2012 Türkiye Örneği [Macroeconomic factors affecting on return on equity: The case of Turkey 2002–2012]. Dumlupınar University Journal of Social Sciences, 35, 167–176. Kosmidou, K., Pasiouras, F., Doumpos, M., & Zopounidis, C. (2004). Foreign versus domestic banks’ performance in the UK: A multicriteria approach. Computational Management Science, 1(3–4), 329–343. Kula, V., Kandemir, T., & Baykut, E. (2016). An investigation of fnancial performances of ınsurance companies and pension fund trading on Borsa Istanbul with grey relational analyze. Afyon Kocatepe University Journal of Faculty of Economics and Administration Sciences, 18(1), 37–53. Lewellen, J. (2004). Predicting returns with fnancial ratios. Journal of Financial Economics, 74, 209–235. Lin, C. T., & Hsu, P. F. (2002). Forecast of non-alcoholic beverage sales in Taiwan using the grey theory. Asia-Pacifc Journal of Marketing and Logistics, 14(4), 3–12. Mandic, K., Delibasic, B., Knezevic, S., & Benkovic, S. (2014). Analysis of the fnancial parameters of Serbian banks through the application of the fuzzy AHP and TOPSIS methods. Economic Modelling, 43, 30–37. Özdemir, L. (2019). Effects on domestic effciency in BIST sustainability ındex: DEA Malmquist total factor productivity analysis. University of Uşak Journal of Social Sciences Sosyal Bilimler Dergisi, 12(1), 33–45. Özen, E., Yesildag, E., & Soba, M. (2015). TOPSIS performance evaluation measures and relation between fnancial ratios and stock returns. Journal of Economics Finance and Accounting, 2(4), 482–500. Peker, İ., & Baki, B. (2011). Performance evaluation in Turkısh ınsurance sector with grey relationship analysis. International Journal of Economic and Administrative Studies, 4(7), 1–17. Rupeika-Apoga, R., Zaidi, H. S., Thalassinos, E. Y., & Thalassinos, I. E. (2018). Bank stability: The case of Nordic and non-Nordic banks in Latvia. International Journal of Economics and Business Administration, 6(2), 39–55. Sakarya, Ş., & Aytekin. S. (2013). Measurement of the relationship between deposit banks performance with stock returns in ISE: An application with PROMETHEE multi-criteria decision making method. International Journal of Alanya Faculty of Business, 5(2), 99–109. Sallehuddin, R., Shamsuddin, S. M. H., & Hashim, S. Z. M. (2008, November). Application of grey relational analysis for multivariate time series. In Ajith Abraham (Ed.), 2008 IEEE eighth international conference on intelligent systems design and applications (Vol. 2, pp. 432–437). Kaohsiung, Taiwan. Sayılgan, G., & Süslü, C. (2011). The effect of macroeconomic factors on stock returns: A study of Turkey and emerging markets (in Turkish). Journal of BRSA Banking and Financial Markets, 5(1), 73–96. Siew, L. W., Fai, L. K., & Hoe, L. W. (2017). Evaluation on the fnancial performance of the Malaysian banks with TOPSIS model. American Journal of Service Science and Management, 4(2), 11.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

An Assessment for Financial Performance of Banks    183

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Sofyalıoğlu, Ç. (2011). Grey evaluation model of process failure mode effect analysis. Ege Academic Review, 11(1), 155–164. Taşkın, F. D. (2011). The factors affecting the performance of the Turkish commercial bank. Ege Academic Review, 11(2), 289–298. Thalassinos, I. E., Hanias, P. M., & Curtis, G. P. (2012). Time series prediction with neural networks for the Athens Stock Exchange indicator. European Research Studies Journal, 15(2), 23–31. Thalassinos, I. E., & Kiriazidis, T. (2003). Degrees of ıntegration in ınternational portfolio diversifcation: Effective systemic risk. European Research Studies Journal, 6(1–2), 119–130. Ugurlu, E., Thalassinos, E., & Muratoglu, Y. (2014). Modeling volatility in the stock markets using GARCH models: European emerging economies and Turkey. International Journal of Economics and Business Administration, 2(3), 72–87. Üstünışık, N. Z. (2007). Türkiye’deki İller ve Bölgeler Bazında Sosyo-Ekonomik Gelişmişlik Sıralaması Araştırması: Gri İlişkisel Analiz Yöntemi ve Uygulaması [Investigation of socio-economic development ranking study of provinces and regions in Turkey: An gray relational analysis methods and applications). Unpublished Master Thesis, Gazi University, Ankara. Wanke, P., Azad, M. A. K., Barros, C. P., & Hassan, M. K. (2016). Predicting effciency in Islamic banks: An integrated multicriteria decision making (MCDM) approach. Journal of International Financial Markets, Institutions and Money, 45, 126–141. Wen, K. L., (2004). Grey systems: Modeling and prediction (pp. 49–144). Tucson, USA: Yang Sky Scientifc Press. Wu, H. H. (2002). A comparative study of using grey relational analysis in multiple attribute decision making problems. Quality Engineering, 159(2), 209–217.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

This page intentionally left blank

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Chapter 13

100 Quotes from the Global Financial Crisis: Lessons for the Future Peterson K. Ozili Abstract The fnance literature has not documented the feeling, the shock and the pain that ordinary people had to go through during the 2008 global fnancial crisis especially in the United States where it all began. In an effort to shed new light on the global fnancial crisis, it has become important to present a view of the fnancial crisis from the lens of those who were affected by the crisis, those who were responsible for the crisis, those who could have prevented the crisis, as well as the views of other observers. The views or quotes in this chapter are concise, useful and thought provoking. They create an opportunity to help reconsider the events of 2008 from a fresh perspective, so that a lot more can be done by everyone, including banks, governments and citizens, to prevent a repeat of those events in the future of fnance. Finally, most of the views or quotes reported in this chapter have within them some important lessons and wisdom to guide us on what to do before another future crisis comes.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Keywords: Systemic risk; fnancial crisis; bank failure; too-big-to-fail; government bailouts; central banks; banks; recessions JEL classifcation: G21; G28

1. Long Before the Financial Crisis 1. “Don’t think money does everything or you are going to end up doing everything for money” – Voltaire, French Philosopher. 2. “In times of crisis human beings don’t have it in them to be rational” – Larry McMurtry, American Novelist.

Uncertainty and Challenges in Contemporary Economic Behaviour, 185–194 Copyright © 2020 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-80043-095-220201014

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

186    Peterson K. Ozili   3. “A banker is a fellow who lends you his umbrella when the sun is shining, but wants it back the minute it begins to rain” – Mark Twain, American writer.   4. “Gold and silver are money, everything else is credit” – J.P. Morgan, American Banker.

2. Shortly Before – and During – the Global Financial Crisis   5. “Everybody wanted to own a piece of real estate to get into the game” – Angelo Mozilo, Former Chairman and CEO Countrywide Financial.   6. “What we know about the global fnancial crisis is that we don’t know very much” – Paul Samuelson, American Economist.   7. “Blaming speculators as a response to fnancial crisis goes back at least to the Greeks. It’s almost always the wrong response” – Lawrence (Larry) Summers, American Economist.   8. “A fnancial crisis is a great time for professional investors and a horrible time for average ones” – Robert Kiyosaki, American Investor.   9. “I don’t think that any economist disputes that we’re in the worst economic crisis since the great depression. The good news is that we’re getting a consensus around what needs to be done” – Barack Obama, Former US President. 10. “The fnancial crisis should not become an excuse to raise taxes, which would only undermine the economic growth required to regain our strength” – George W. Bush, Former US President. 11. “When you’re in a crisis of, you know, tremendous proportions, it’s beyond any human capability to control, you just make the best decisions you can, and you just hope that your intuition is correct” – Rudy Giuliani, Former Mayor of New York City.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

3. On the Cause of the Crisis1 12. “Let us not forget, the fnancial crisis had its roots in the decision by Congress to embark on a course of social justice to get everyone that wanted a home into one, regardless of whether or not they could afford it” – Ed Royce, Former US Representative. 13. “The mortgage market meltdown occurred for a number of reasons, but new and poorly underwritten mortgage products were a signifcant contributor” – Jamie Dimon, Chairman and CEO, JPMorgan Chase & Co. In his testimony to the Financial Crisis Inquiry Commission (FCIC) Washington, DC, January 13, 2010. 14. “The big demand was not so much on the part of the borrowers as it was on the part of the suppliers who were giving loans which really most people couldn’t afford” – Alan Greenspan, Former Chairman of the Federal Reserve board. In Jon Meacham and Daniel Gross’s, “The Oracle Reveals All,” Newsweek interview.

1

https://www.responsiblelending.org/mortgage-lending/tools-resources/QuotesWhat-Caused-Crisis.pdf

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Quotes from the Global Financial Crisis    187 15. “Although the high rate of [mortgage] delinquency has a number of causes, it seems clear that unfair or deceptive acts and practices by lenders resulted in the extension of many loans, particularly high-cost loans, that were inappropriate for or misled the borrower” – Ben Bernanke. Written statement by FED Chairman Ben Bernanke, July 14, 2008. 16. “Even after mortgage loans started going bad en masse, the confusing mix of federal and state agencies that made up the nation’s regulatory structure had diffculty responding. After regulators fnally began to speak up about subprime and the other types of mortgage loans that had spun out of control, such lending was already on its way to extinction. What regulators had to say was all but irrelevant” – Mark Zandi, Chief Economist, Moody’s Analytics. 17. “That conclusion suggests that the best response to the housing bubble would have been regulatory, not monetary. Stronger regulation and supervision aimed at problems with underwriting practices and lenders’ risk management would have been a more effective and surgical approach to constraining the housing bubble than a general increase in interest rates” – Ben Benarke, Chairman of the Federal Reserve Bank. In “Monetary Policy and the Housing Bubble,” a speech given at the annual meeting of the American Economic Association in Atlanta, Georgia, January 3, 2010. 18. “Over the course of this crisis, we as an industry caused a lot of damage. Never has it been clearer how mistakes made by fnancial companies can affect Main Street, and we need to learn the lessons of the past few years” – Brian T. Moynihan, CEO and President, Bank of America. Testimony to Financial Crisis Inquiry Commission (FCIC) Washington, DC, January 13, 2010. 19. “The truth is that many of us in the industry were deeply distressed by the growing practice of pushing high risk loans on borrowers who had no reasonable expectation of being able to repay the mortgage. Disclosures were often less than adequate, and faced with a bewildering array of loan terms, borrowers tended to trust their mortgage banker or broker. The broken trust that resulted has damaged borrower confdence in the mortgage industry” – Scott Stern, CEO of Lenders One. Testimony before the Senate Banking Committee, Washington, DC, April 10, 2008. 20. “When they [banks] start losing money, [they say] hey we gotta get back in the game, we gotta get the skites rolling again, hey let’s create another bubble. You think the dot-com bubble was too big? We’ve got a bigger one for you – we’ll call it the real estate and the credit bubble” – Gerald Selente, Economist and Trend Analyst. Excerpt from Overdose: The Next Financial Crisis, the movie. 21. “Millions of Americans were duped by the federal government and the Federal Reserve into buying homes they could not afford and failed to count the cost. When the fnancial crisis of 2008 hit, they could not keep up the monthly mortgage payments and defaulted” – Mark Skousen, American Economist. 22. “My impression is that future generations will look back on this moment and say ‘this is where they completely lost their minds’” – John Hussman, President of Hussman Investment Trust.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

188    Peterson K. Ozili

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

4. Refections After the Financial Crisis 23. “People aren’t as impressed by homes anymore after they saw how they collapsed in price with the fnancial crisis” – Robert J. Shiller, American Economist. 24. “I think people saw this [the crisis] as ‘… that’s happening over there, that’s not happening here.’ The sense of interconnectedness was not realised until the very last moment” – Andrew Ross Sorkin, New York Times. 25. “If a fnancial institution is too big to fail, it is too big to exist” – Bernie Sanders, US Senator. 26. “September and October of 2008 was the worst fnancial crisis in global history, including the Great Depression” – Ben Bernanke, US FED Chairman during the crisis. 27. “One of the biggest mistakes our government made after the fnancial crisis was not prosecuting the people responsible for the greed, recklessness and illegal behaviour that crashed our economy and ruined the lives of millions of Americans” – Bernie Sanders, US Senator. 28. “Financial crises, banking crises or economic crises are simply the painful outcome when fnancial markets begin to reset” – Peterson K. Ozili, Nigerian Economist. 29. “The banking collapse was caused, more than anything, by bad government policy and the total failure of bad regulation, rather than by greed” – Nigel Farage, British Member of European Parliament. 30. “In the fnancial crisis of 2008, it was fraud right down at the heart of that crisis, and yet not one major bank executive was even charged, much less prosecuted and taken to trial – not one” – Elizabeth Warren. US Senator. 31. “The fnancial crisis revealed important weaknesses in many areas of our fnancial system” – Jerome Powell, Current US Chairman of Federal Reserve. 32. “If they are too big to fail, make them smaller” – George P. Shultz, American Economist and Former US Secretary of Labor. 33. “The fnancial crisis happened because no-one could actually say out loud how bad things were” – Mark Ravenhill, British Playwright. 34. “Here you can shoot the bad guys,’ a mercenary says in Baghdad. In America we give them corporate bonuses” – Michael Robotham, The Wreckage. 35. “I believe that the root cause of every fnancial crisis, the root cause, is fawed government policies” – Henry Paulson, Former US Secretary of the Treasury. 36. “The fnancial crisis is a stark reminder that transparency and disclosure are essential in today’s marketplace” – Jack Reed, US Senator. 37. “There are downsides to implicitly trusting banks, as the 2008 fnancial crisis showed” – Balaji Srinivasan, American Businessman. 38. “The recession won’t be over until we raise a generation that know how to live on what they have today not on what they expect to have tomorrow” – Unknown. 39. “Crisis and deadlocks when they occur have at least this advantage, they force us to think” – Jawaharlal Nehru, Former Prime Minister of India. 40. “I certainly don’t think that fnance should lie at the heart of an economy – it’s a subsidiary function not a primary driving function” – Satyajith Das, Except from the Inside Job movie.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Quotes from the Global Financial Crisis    189 41. “The fnancial crisis and the Great Recession posed the most signifcant macroeconomic challenges for the United States in a half-century, leaving behind high unemployment and below-target infation and calling for highly accommodative monetary policies” – Jerome Powell, Current US FED Chairman. 42. “The fnancial crisis is a stark reminder that transparency and disclosure are essential in today’s marketplace” – Jack Reed, US Senator. 43. “A lot of the evil in the world is actually not intentional. A lot of people in the fnancial system did a lot of damage without intending to” – George Soros, Hungarian-American Investor and Philanthropist. 44. “Managing and navigating through a fnancial crisis is no fun at all” – Howard Schultz, American Businessman. 45. “Forced to confront a reptile or an international fnancial crisis, I’ll take the reptile every time” – Alexandra Petri, American Columnist. 46. “The lesson of history is that you do not get a sustained economic recovery as long as the fnancial system is in crisis” – Ben Bernanke, Former US Chairman of Federal Reserve. 47. “If there had been a Financial Product Safety Commission in place 10 years ago, the current fnancial crisis would have been averted” – Elizabeth Warren, US Senator. 48. “The beauty of a fnancial institution is that there are a lot of ways to go to hell in a bucket. You can push credit too far, do a dumb acquisition, leverage yourself excessively – it’s not just derivatives [that can bring your downfall]” – Charlie Munger, American Investor. 49. “The mortgage crisis is a clear instance of consumers who needed protection. There was predatory lending to people who didn’t know what they were doing” – Nassim Nicholas Taleb, Lebanese-American Essayist.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

5. Rethinking the Economics of Financial Regulation 50. “Many of us like to think of fnancial economics as a science, but complex events like the fnancial crisis suggest that this conceit may be more wishful thinking than reality” – Andrew Lo, American Economist. 51. “I believe that the fnancial crisis of 2008/9 exposed more a lack of ethics and morality – especially by the fnancial sector – rather than a problem of regulation or criminality. There were, of course, regulatory lessons to be learned, but at heart, there was a collective loss of our moral compass” – Paul Polman, Dutch Businessman. 52. “The fnancial crisis was linked to the fact that banks had excessive leverage and too many risky assets. The solution is not to try to dictate to banks what they can do or not do, but to require them to strengthen their capital to absorb potential losses and hold less risky assets” – John Paulson, American Investor. 53. “It is a sobering fact that the prominence of central banks in this century has coincided with a general tendency towards more infation, not less. If the overriding objective is price stability, we did better with the nineteenthcentury gold standard and passive central banks, with currency boards, or

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

190    Peterson K. Ozili even with ‘free banking.’ The truly unique power of a central bank, after all, is the power to create money, and ultimately the power to create is the power to destroy” – Paul Volcker, Former US Chairman of Federal Reserve. 54. “Throughout the 19th century, when there was a laissez-faire mentality and insuffcient regulation, you had one crisis after another. Each crisis brought about some reform. That is how central banking developed” – George Soros, Hungarian-American Investor and Philanthropist. 55. “Financial crises require governments” – Timothy Geithner, Former US Secretary of the Treasury. 56. “The global [fnancial] crisis is caused by pathologies inherent in the global fnancial system itself ” – George Soros, Hungarian-American Investor and Philanthropist. 57. “In a fnancial crisis, only the Fed, as the lender of last resort, might stand between our economy and fnancial catastrophe. We must leave the Fed with the fexibility to provide liquidity in order to stop a fnancial panic” – Stephen A. Schwarzman, American Businessman. 58. “The reality is that fnancial markets are self-destabilizing; occasionally they tend toward disequilibrium, not equilibrium” – George Soros, HungarianAmerican Investor and Philanthropist. 59. “Wall street can never be allowed to threaten main street again. No bank can be too big to fail, no executive too powerful to jail” – Hillary Clinton, US Senator. 60. “Every government intervention [in the marketplace] creates unintended consequences, which leads to calls for further government interventions” – Ludwig Von Mises, Economist. 61. “The trouble with government regulation of the market is that it prohibits capitalist acts between consenting adults” – Robert Nozick, American Philosopher. 62. “I don’t want to be partisan here. But please, tell me how you get out of a business recession by raising business taxes and regulations?” – Lawrence Kudlow, Director of the US National Economic Council. 63. “We cannot solve problems by using the same kind of thinking we used when we created them” – Albert Einstein, Theoretical Physicist.

6. Questioning the Economics Profession 64. “Debates go on to this day about what caused the Great Depression. Economics is not very good at explaining swings in economic activity” – Eugune Fama, American Economist. 65. “The purpose of studying economics is not to acquire a set of ready-made answers to economic questions, but to learn how to avoid being deceived by economists” – Joan Robinson, British Economist. 66. “Since the global fnancial crisis and recession of 2007–2009, criticism of the economics profession has intensifed. The failure of all but a few professional economists to forecast the episode – the after-effects of which still linger – has led many to question whether the economics profession contributes anything signifcant to society” – Robert J. Shiller, American Economist.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Quotes from the Global Financial Crisis    191 67. “... If you look at mainstream economics there are three things you will not fnd in a mainstream economic model – Banks, Debt, and Money. How anybody can think they can analyze capital while leaving out Banks, Debt, and Money is a bit to me like an ornithologist trying to work out how a bird fies whilst ignoring that the bird has wings …” – Steve Keen, Australian Economist. 68. “Many of the problems troubling the world are not caused by economic globalisation” – Xi Jinping, China’s President, Speech from 2017 World Economic Forum, Davos. 69. “Engineers do engineering, i.e. they build bridges. So engineering needs engineers. The economy does NOT need economists. Economists do not make economy, but they try it and that is why we have so much problems with some fnancial models” – Steve Keen, Australian Economist. 70. “GDP is not a good measure of economic performance, it’s not a good measure of well-being” – Joseph Stiglitz, American Economist. Quotes from the 2016 World Economic Forum, Davos. 71. “The position I now favor is that economics is a pre-science, rather like astronomy before Copernicus, Brahe and Galileo. I still hold out hope of better behavior in the future, but given the travesties of logic and anti-empiricism that have been committed in its name, it would be an insult to the other sciences to give economics even a tentative membership of that feld.” ― Steve Keen, Debunking Economics – Revised, Expanded and Integrated Edition: The Naked Emperor Dethroned? (2001)

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

7. Why Liquidity is Important During Financial Crises 72. “Liquidity is the currency of bear markets” – Laurent Bernut, An Ex-Fidelity short seller, ex-Hedge Funds analyst, ex-CPA, algo trader. 73. “It is easier to invest for cash fow during a fnancial crisis by hiding your head in the sand. The longer the crisis lasts, the richer some people will be” – Robert Kiyosaki, American Investor. 74. “Actually, I have been very supportive of a very robust stimulus package from day one. I think this economy has to have a major stimulus initiative because the only group with liquidity is the federal government” – Judd Gregg, Former US Senator. 75. “When I hear complaints about less liquidity, remember there is such a thing as too much liquidity” – Paul Volcker, Former US FED Chairman. During interview at CNBC. 76. “If you can’t get out of whatever you are invested in, you don’t own it, it owns you. You don’t trust me? Buy a Boat, a second home, a pink Ferrari and then try and sell it. May the Gods be with You. So, always go for liquidity. Disinvest now from anything with a lock-up” – Laurent Bernut, An Ex-Fidelity short seller, ex-Hedge Funds analyst, ex-CPA, algo trader. 77. “To be sure, the provision of liquidity alone can by no means solve the problems of credit risk and credit losses; but it can reduce liquidity premiums, help restore the confdence of investors, and thus promote stability” – Ben Bernanke, Former US Chairman of the Federal Reserve.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

192    Peterson K. Ozili

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

8. Preparing for the Next Global Financial Crisis 78. “If we lose the confdence of our foreign lenders and we must not allow that to happen but if that were to happen then there will be a dramatic decline in the dollar, a dramatic increase in interest rates, a signifcant fuelling of infation, a very deep recession and possibly depression that would be felt around the world. We must not allow that to happen or we lose the confdence of our foreign lenders” – David M. Walker, Former US Comptroller General. In the movie Overdose: The Next Financial Crisis. 79. “Regulation is necessary, particularly in a sector, like the banking sector, which exposes countries and people to a risk” – Christine Lagarde, President of the International Monetary Fund. 80. “Central banks don’t have divine wisdom. They try to do the best analysis they can and must be prepared to stand or fall by the quality of that analysis” – Mary Kay Ash, American Businesswoman. 81. “The Central Bank should have a permanent window for discounting high quality securities where banks could go and discount these. It gives peace of mind to the banks. In the absence of this facility, what banks tend to do is to keep a liquidity cushion for emergency requirements. This is a very expensive way of managing liquidity” – Abdul Aziz Al Ghurair, Chairman of Mashreq Bank and UAE Politician. 82. “If we have a fnancial crisis of one form or another, do we have the confdence that this person has the experience to handle it well?” – Joel Naroff, Author and President and Founder of Naroff Economic Advisors. 83. “In fnancial services, if you want to be the best in the industry, you frst have to be the best in risk management and credit quality. It’s the foundation for every other measure of success. There’s almost no room for error” – John G. Stumpf, American Executive. 84. “If you don’t invest in risk management, it doesn’t matter what business you’re in, it’s a risky business” – Gary Cohn, President of Goldman Sachs. 85. “In the fnancial system we have today, with less risk concentrated in banks, the probability of systemic fnancial crises may be lower than in traditional bank-centered fnancial systems” – Timothy Geithner, Former US Secretary of the Treasury. 86. “Diversifying suffciently among uncorrelated risks can reduce portfolio risk toward zero. But fnancial engineers should know that’s not true of a portfolio of correlated risks” – Harry Markowitz, American Economist. 87. “Do not trust fnancial market risk models. Despite the predilection of some analysts to model the fnancial markers using sophisticated mathematics, the markets are governed by behavioural science, not physical science” – Seth Klarman, American Investor 88. “For market discipline to constrain risk effectively, fnancial institutions must be allowed to fail. Under optimal fnancial regulatory and fnancial system infrastructures, such a failure would not threaten the overall system” – Henry Paulson, Former US Secretary of the Treasury. 89. “There is a simple way of avoiding excess risk-taking by the managers of our fnancial institutions. It is to make it a crime. Had a crime for reckless

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Quotes from the Global Financial Crisis    193 management of a fnancial institution been on the books, Northern Rock and Royal Bank of Scotland would not have blown up” – Paul Collier, British Economist and Author. 90. “I’m afraid sometimes certain individual cases of defaults are unavoidable. What we should do is to step up monitoring, properly handle relevant matters and ensure there is no regional and systemic fnancial risk” – Li Keqiang, Economist and Chinese Politician. 91. “Many fnancial disasters can be traced to people who thought they were hedging” – Aaron C. Brown, American Finance Practitioner. 92. “Look at any fnancial institution, at any bank. They’re all photocopies of each other. There’s no diversity of institutions and even less diversity of currency. Therefore, just as you say it’s very logical that an ecosystem like this will collapse, it’s very predictable a monetary system like this collapse, too. And it hasn’t fnished collapsing, by the way” – Bernard Lietaer. Electrical Engineer, Economist, Author and Professor. 93. “Most of the time, your risk management works. With a systemic event such as the recent shocks following the collapse of Lehman brothers, obviously the risk management system of any bank appears, after the fact, to be incomplete. We ended up where banks couldn’t liquidate their risk, and the system tended to freeze up” – Myron Scholes, American-Canadian Economist. 94. “Any bailout of a private company is a bad decision by our federal government. Private companies have a right to succeed, but they also should have the right to fail” – Matt Salmon, Former US Representative. 95. “The government cannot become a bailout organization for companies on the verge of bankruptcy” – Martin Winterkorn, Former Chairman of Volkswagen AG. 96. “Sound regulation and enforcement are needed to ensure that our capital markets remain accessible, competitive and free of bad actors” – Thomas (Tom) Donohue, CEO of the US Chamber of Commerce. 97. “We need to think deeply about whether we can sustain banks that are not only too big to fail, but potentially too big to bail” – George Osborne, Former Chancellor of the Exchequer.

9. Sober Refection   98. “Even if the crises that are looming up are overcome and a new run of prosperity lies ahead, deeper problems will still remain” – Joan Robinson, British Economist.   99. “My daughter asked me when she came home from school, “What’s the fnancial crisis?” and I said, it’s something that happens every fve to seven years” – Jamie Dimon, CEO of JPMorgan Chase. 100. “Why should a fnancial engineer be paid four, four times more than a real engineer. A real engineer build bridges, a fnancial engineer builds dreams. And you know, when those dreams turn out to be nightmares, other people pay for it” – Andrew Sheng, China Banking Regulatory Commission, from the Inside Job 2010 Documentary.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

194    Peterson K. Ozili

10. Conclusion This chapter presented some thoughts, in the form of quotes, on what caused the fnancial crisis, why it was severe and what can be done to prevent another crisis in the future. The quotes were carefully selected for information purposes only, and all politically sensitive quotes were excluded. For decades, people have warned that a recession or fnancial crisis was coming but no one listened. If anybody listened, it was a small group of people and that number was too small to force a change in behavior prior to the crisis. Also, the credit derivatives bubble and the massive profts made by banks prior to 2007 have shown that it can be diffcult to get banks to care about what’s happening outside the house when everyone is enjoying the party inside the house. What can we do to prevent future fnancial crises? Should we increase regulation through activity restriction or through stronger capital regulation? Or, should we regulate executive compensation in fnancial institutions since there is evidence that big bonuses to bank executives is the major source of their excessive risk-taking?

Quote Sources

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

AZ Quotes. Retrieved from https://www.azquotes.com/ Brainy Quotes. Retrieved from https://www.brainyquote.com/ Devell, H. (Producer), & Borgs, M. (Director) (2010). Overdose: The next fnancial crisis [Motion Picture]. Journeyman Pictures. Ferguson, C. (Director). (2010). Inside job [Motion Picture].: Sony Pictures Classics. Goodreads. Retrieved from https://www.goodreads.com/ IZ Quotes. Retrieved from https://izquotes.com/ Quote HD. Retrieved from http://www.quotehd.com/ Wiki Quotes. Retrieved from https://en.wikiquote.org/wiki/Main_Page

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Chapter 14

Labouring Behind Closed Doors: The Working and Living Conditions of Filipino Live-in Care Workers in Malta Mario Thomas Vassallo and Manwel Debono

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Abstract This qualitative study seeks to explore the grounded realities of live-in care workers in Malta. The growing economic affuence in Malta, coupled by an ageing population and the lowest fertility rate in the European Union, is resulting in a greater demand for live-in care givers, particularly from the Philippines. Reinforced through public policy wherein families who employ a qualifed live-in carer are benefting from government subsidy to ease burden on the state’s residential homes, Malta appears to be moving from a passive to a more active international recruitment of domestic migrant workers. This inquiry provides an evidence-based contribution to the appeal of the European Economic and Social Committee of the EU calling for more research about the rights of live-in care workers in Europe which has long remained almost invisible to EU and Member State policymakers. The majority of the fndings refect some of the concerns that have already been identifed in international literature, like higher levels of precariousness, contractual agreements not being honoured, psychological obligations, fraudulent agents and the lack of separation between work and personal life. Other fndings have endogenous characteristics that are closely linked to the island state of Malta, namely its safe environment, Catholic culture, bilingual coexistence of Maltese and English and the competitive nature of Filipino community groups that may discourage further social engagement. The chapter concludes with brief policy suggestions to trigger improvements in the wellbeing and dignity of migrant carers. Keywords: Filipino migrants; live-in care workers; living conditions; Malta; precariousness; social wellbeing; working conditions Uncertainty and Challenges in Contemporary Economic Behaviour, 195–215 Copyright © 2020 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-80043-095-220201015

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

196    Mario Thomas Vassallo and Manwel Debono

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

1. Introduction In today’s globalised world, characterised by international competitiveness, labour outsourcing (Bales et al., 2014; ELCJM, 2007) and the weakening of trade unionism (Blanchfower & Bryson, 2008; Kocer, 2018), matters related to the precariousness of work have become more prevalent (European Parliament, 2014; Pillinger, 2016). Both public administrations and private enterprises are shunning their legal obligations to workers by replacing permanent jobs with subcontracting and temporary work (Alfonso, 2012; Rupeika-Apoga, Romanova, Bule, & Thalassinos, 2019; Thalassinos, Stamatopoulos, & Arvanitis, 2011; Wills, 2009). Poor conditions of work might be even more extreme for overseas domestic workers who perform their duties in private households (Ambrosini & Barone, 2007; ILO, 2013a, b ). They often face very low wages, excessively long hours, have no guaranteed weekly day of rest and, at times, are vulnerable to physical, mental and sexual abuse or restrictions on freedom of movement. The International Labour Organisation (ILO, 2017) estimates that there are at least 67 million domestic workers worldwide, and this number is increasing steadily in developed and developing countries. Female live-in carers who assist the disabled, the elderly or sick family members on 24/7 basis, constitute a signifcant proportion of this category of workers (ILO, 2017). This qualitative study explores the grounded realities of live-in care workers in Malta, the EU’s smallest member state. The growing economic affuence in Malta, coupled by an ageing population, is resulting in a greater demand for livein care workers, particularly from the Philippines. International literature maintains that with its explicit nurses export policy, the Philippines is a major supplier of nurses and health workers globally (Encinas-Franco, 2016). In line with other Western countries, Malta appears to be moving from a passive to a more active international recruitment of care workers. This transition is also being reinforced through public policy wherein Maltese families who employ a qualifed live-in carer to look after their elderly are having half the carer’s wages subsidised by the state in a bid to tackle waiting lists in government homes. Adopting an inductive research strategy, this study identifes and examines the living conditions of Filipino carers in Maltese households. While Malta, as other European countries, prides itself with having strong employment legislation, questions arise as to whether minorities such as domestic workers beneft from such safeguards. In a context wherein the demarcation line between work and personal life is non-existent, are the living and working conditions of imported workers regulated by effective legal frameworks or are such workers at the mercy of their employers? To this effect, this study aims to create more awareness about this specifc group of migrant workers who may be at a greater risk of precariousness when compared to others. The rest of the Chapter is segmented into fve parts. The contextual scenario at the European and Maltese level is, frst, put into perspective. Then, evidencebased fndings from previous international studies are rolled out. Following a brief methodological note, the narratives of eight women from the Philippines employed as live-in carers in Malta, are presented and discussed next. The study concludes with a number of brief policy measures.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Labouring Behind Closed Doors    197

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

2. Contextual Setting The 2018 Ageing Report of the European Commission (2018a) confrms that Europe’s population continues to age signifcantly. Increased life expectancy due to better medical care and low fertility rates across all Member States are the two fundamental elements that are contributing to this demographic phenomenon, having long lasting impacts on the bloc’s labour market and potential economic growth. According to Eurostat, even though French and Swedish women have the highest fertility rates of 1.92 and 1.88 children over their lifetimes respectively, their rates are still below the 2.1 generally accepted standard which is deemed to be the minimum rate for a population to replace itself without inward migration boosting numbers (Financial Times, 2018). Although immigration has led to political upheaval in Europe, policy planners are still committed to encourage it in great numbers in order to sustain the Western standard of economic and social wellbeing. Forming part of the southern European countries, Malta has the lowest fertility rate (just 1.26 children per woman) in the EU (Eurostat, 2019). In parallel to shrinking family size, the National Strategic Policy for Active Ageing 2014-2020 in Malta (National Commission for Active Ageing, 2013) forecasts that the older population segment will continue to grow substantially in the foreseeable future. Malta’s ageing population, coupled with a ‘vigorous’ economic prosperity (European Commission, 2018) and rapid social transformations, is resulting in a greater demand for live-in care givers to assist vulnerable family members. Acute domestic labour shortage in this area is being mitigated by migrant workers, particularly female migrants from the Philippines. Filipinos are the largest group of third country nationals in Malta. Statistics of the Central Bank of Malta (CBM, 2016) demonstrate that most Filipino workers in Malta work in elementary occupations (43.3%) often as cleaners, or in services and sales (39.3%), especially as care workers (Jobsplus, 2018). National statistics show that 23.1% of the Filipinos in Malta work in human health and social work activities, and activities within households (11.0%). The number of registered Filipino workers in Malta grew sixfold between 2010 and 2018 (from 459 to 2,882), and is set to keep on growing. Apart from the particular connection to low skilled occupations and the health sector, the Filipino migrant community has other noticeable characteristics, being in large part made of women (one man for every three women) and young workers (mostly below 39 years old) (Jobsplus, 2018). A number of recruitment agencies have opened for business, offering support to Maltese families who wish to hire Filipino nationals to provide elderly care. A 2018 advert of a private recruitment agency in Malta states that: We are currently focused on recruiting workforce from the Philippines because we believe in the magic factor why Filipinos suit work and fnd it very easy to settle in Malta. They are generally considered as hardworking, loyal, go the extra mile, dedicated, hospitable, good English speakers and quickly adapt to different cultures. [Corporate name withheld]

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

198    Mario Thomas Vassallo and Manwel Debono While echoing the set of traits that are generally attributed to Filipinos (Andres, 1996; Church & Katigbak, 2000; Saito, 2013), the advert’s choice of wording may indicate that the Maltese are taking advantage of the Filipinos’ inner attributes by exploiting their personalities, thus leading to unfair treatment, if not outright exploitation. They are deemed by the locals as a good source of ‘cheap labour’ and ‘their religion and gender also make them less threatening’ (Billiard, 2018, p. 26). Consequently, while on the one hand the input of foreign carers towards the welfare of the vulnerable is fundamental, new questions arise on the impact of such a phenomenon on the wellbeing of migrant workers. Notwithstanding the need to provide answers to this developing reality at national and European levels, scientifc research initiatives remain scarce. The author of Exiled Homes, a notable attempt to expose the realities of Filipino live-in care workers in Malta, states that: While their needs and narratives differ greatly, all of [the Filipinas working as live-in carers] share a similar situation: they are all taken care of by a foreigner with whom they share their most intimate space – their home. In this way, the ‘exiled home’ of the older person becomes a home of inclusion, in which two different cultures meet and integrate, and where the public/private dichotomy characterising working and private life disappears. (Orsini, 2018, p. 34)

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

3. International Literature The conditions of employment of domestic foreign workers, in particular live-in carers, has been extensively researched in Asian countries, as well as in the Middle East, Australia and Canada (Encinas-Franco, 2016; Gahagan, Loppie, Rehman, Maclellan, & Side, 2007; Tejero & Fowler, 2012; Yi-Chung & Yu-Cheng, 2012), but it is loosely studied in Europe. This lacuna led to the appeal of the European Economic and Social Committee of the EU calling for more research about the rights of live-in care workers ‘which has long remained almost invisible to EU and Member State policymakers’ (EESC, 2018). Studies from beyond Europe produced evidence-based claims that such workers suffer from structural violence and discrimination (Bhuyan, Valmadrid, Panlaqui, Pendon, & Pearlita, 2018; Duncan & Loretto, 2004; Robillard, McLaughlin, Cole, Vasilevska, & Gendron, 2018) and low levels of life and job satisfaction (Chowdhury & Gutman, 2012; Iecovich, 2011). Their status of poor wellbeing is signifcantly increased by abusive and fraudulent recruitment practices (Simon-Rusinowitz, Mahoney, & Benjamin, 1998; United Nations, 2015) and the reluctance of domestic authorities to detect wrongdoings (O’Neill Richard, 2000). The Business Mirror (2018) reports about the need to ensure that ‘proper procedures for the recruitment, deployment and arrival of Filipino domestic workers in Jordan’ are to be formulated and implemented by ‘licensed Jordanian recruitment agencies in partnership with licensed Philippine recruitment agencies’. When appraising migrant personnel in Cyprus, Trimikliniotis and Demetriou (2011) raise the issue of defcient immigration and

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Labouring Behind Closed Doors    199 labour regulatory regimes which are, supposedly, intended to safeguard the most vulnerable groups of workers. The lack of legislative enforcement and ‘institutional precariousness’ are caught inside the contradiction between severe and repressive immigration control on the one hand, and ineffective labour market regulation on the other. To this effect, Bhuyan et al. (2018) appeal to social services authorities, labour regimes and immigration control units to ‘increase outreach’ through systematic and regular monitoring mechanisms in order to detect risk factors, violation of contractual agreements and exploitation. International literature also suggests a wide prevalence of live-in care workers who go beyond their contractual obligations because emotional commitments take over that which is stipulated in employment contracts (Dyer, McDowell, & Batnitzky, 2008; Ho, Chiang, Leung, & Cheung, 2019). In her ‘Psychological Contract’ theory, Rousseau (1989) ponders upon the mutual beliefs, perceptions and informal obligations between an employer and an employee. In contrast to the formal written contract of employment which, for the most part, only identifes mutual duties and responsibilities in a generalised form, the psychological contract sets the dynamics for the relationship and defnes the detailed practicality of the work to be done. Billiard (2018, p. 24) differentiates between ‘labour code’ and ‘family code’. The former is always fxed, written and endorsed; the latter is always in a state of fux and, most often, negotiated verbally. This explains why migrant live-in carers are subject to ‘unsocial working hours’ and are likely to cede for their vacation leave entitlements in order not to ‘abandon’ their patients (Moriarty, 2010; Robillard et al., 2018). Researching the case of Taiwan, YiChung and Yu-Cheng (2012) reveal that paid caregivers lack free time and feel lonely and tired; but are nevertheless able to experience leisure during their work, obligatory and discretionary time. They sustain that the line between work and leisure is blurred, and their respondents’ lives expose an intricate portrayal of leisure and caregiving. Gahagan et al. (2007) maintain that migrant caregivers in Canada identify a wide variety of activities as ‘leisurable’ because, for them, leisure is defned as ‘simply getting outside of the house and physical removal from the caregiving location’. Thus, the term ‘leisure’ has different connotations to locals and migrants. ‘Social isolation’ (Robillard et al., 2018), peppered by ‘cultural shock and homesickness’ (Chowdhury & Gutman, 2012) are additional key factors that continue to intensify the vulnerability and marginalisation among migrant domestic workers. Given this scenario suggested by international literature, this study will now turn its attention to the situation in Malta to identify analogous or conficting trends and patterns.

4. Methodology Data are derived from eight semi-structured interviews that were conducted in 2018; each with a duration of approximately ninety minutes. The method of selection involved snowball sampling which was initiated by primary contacts provided by two Filipino community groups that are active in Malta. Throughout the

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

200    Mario Thomas Vassallo and Manwel Debono Table 1.  Interviewees’ Profle.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Interviewee Age Code F1

26

F2 F3

Number of Years in Malta

Partner/ Spouse in the Philippines

Children in the Philippines

College Qualifcation

3

None

None

None

48

2

Yes

1

None

39

11

Yes

1

None

F4

40

8

Yes

1

Bachelor of Science (Education)

F5

48

18

Separated – 2 and 5 not in another grand relationship children

None

F6

51

5

Yes

Honours degree in Communications

F7

39

3

Separated but None in another relationship with a Maltese partner

None

F8

38

5

Yes

None

3

1

process, more than 30 carers were contacted but only 8 accepted to be interviewed on condition that that their identity remains anonymous. Various attempts to recruit more interviewees proved fruitless. In itself, this is an indication that the grounded reality involves a high degree of vulnerability and precariousness because workers in the feld are uncomfortable to talk about their experiences. All of the interviewees are female and their profle details are provided in Table 1. Interviews were conducted in English at a venue identifed by the research subjects. The question guide included open questions which were supplemented by a series of probing questions in a conversational style. Participants were briefed a priori on the research scope by their community associations and, just before the interview, they were asked to sign the consent form maintaining strict ethical standards in terms of confdentiality and anonymity. Interviews were kick-started by questioning them about their decision to come to Malta and the environment they found themselves in. Eventually, interviewees were asked about their day-today duties, relations with the patient and relatives and, eventually, the discussion revolved around their integration within Maltese society. Using Braun and Clarke’s (2006) six step approach to thematic analysis, conversations were recorded, transcribed, coded, themed, reviewed and redefned. As exhibited in Fig. 1, this inductive methodology yielded fve inter-related themes,

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Labouring Behind Closed Doors    201

Fig. 1.  Thematic Coding.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

namely: (1) choosing Malta, (2) working conditions, (3) motivation and relations with employer, (4) social integration, and (5) life beyond work. Each theme is treated distinctly in the coming sections wherein verbatim quotes are woven into the narrative. By their very nature, direct quotes are considered as evidence and, at the same time, as illustrations of emerging themes; thus deepening the readers’ understanding of the phenomena under investigation (Corden & Sainsbury, 2005). Each theme encompasses comparable, sometimes contrasting, narratives. Primary fndings are presented and discussed in the context of international literature to map out patterns and divergences.

5. Choosing Malta The starting point to trace the origin of the respondents’ narratives is by inquiring why they relocated to Malta as migrant workers. In all cases, economic reasons were the primary motivator. Salaries in the Philippines are very low and, when they learnt from friends and family members that conditions of work are better in Malta, they left their homeland in search of a better life that pays more: As a customer offcer in a US company, I had a good job in the Philippines but wages are low …. It is for the salary that I opted to work abroad. F1 Job opportunities exist in cities, but in rural areas and on remote islands there aren’t. We have too many graduates but less jobs. My wage as a carer in Malta is triple that I had in the Philippines where I worked as graduate government offcer. F6

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

202    Mario Thomas Vassallo and Manwel Debono The salary of a carer in Malta is better than that of a teacher in the Philippines. F4 All interviewees maintained that some of their closest friends or family members who have already migrated to Malta introduced them to the island state which is deemed to be a ‘nice and safe place’ to live in: They told me that, in Malta, I will be safe. In the Philippines, we have a lot of earthquakes and typhoons but in Malta the weather is mild and there are no natural disasters. F8 If you want to earn more money, then come to Malta. That was the message I got from a friend of mine. F3 Besides the promise of higher salaries and benign environment, another key trigger that enticed Filipinas to come to Malta is the contractual assurance that, as live-in care workers, they would only be responsible for one patient: I was originally a housekeeper in Hong Kong where I was responsible for a whole family. Then a friend of mine told me to come to Malta because, here, you take care of one person. F2 In the Philippines I worked as a cook with a family of three children. They also had a nanny and a cleaner whom I had to partially supervise as well. As a cook, I was paid the best but, still, it wasn’t enough. F5

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

There is another element attracting Filipinas to come to Malta and work as live-in carers: the Maltese. Interviewees’ own experiences confrm that the Maltese prefer Filipinas rather than local guardians to take care of their loved ones: The Maltese prefer Filipinas because we are hard-working. When you say to me, Stay a bit longer, I will stay… The Maltese will leave when time elapses. F4 We have a lot of patience. Maltese shout and talk like quarrelling all the time. But we don’t react if you shout at us. F6 The Maltese prefer female Filipinas to do this work; maybe because the way we care. We have soft voices. F2 We are affectionate to our employers. This is why the Maltese prefer us. F5 Nobody would reject a Filipina carer because the Maltese don’t want to do this job. F7

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Labouring Behind Closed Doors    203 This preference for middle-aged Filipinas to do care work confrms Anderson and Ruhs’ (2010, p. 27) hypothesis that ‘employers draw on socially meaningful stereotypes or their own experiences generalized’ to explain the suitability of particular nationalities for specifc jobs. Duncan and Loretto (2004) sustain that, sometimes, workers’ appropriateness is determined categorically, based on gender, age, race and/or nationality of the job candidates rather than individual merit. Despite their high hopes to migrate to Malta, almost all of the interviewees lamented over the limited orientation provided by the authorities in the Philippines before leaving homeland: I didn’t know about Malta; not even where it was. No help was provided in the Philippines. F8 When I arrived at the airport, my friend picked me up. I had no idea where I landed. F1 In the Philippines they don’t prepare you for this job. But now things start to change. We have agencies that provide initial training. F3 When I decided to leave the Philippines, I was invited for a one-off meeting, together with hundreds of others, to orientate us about what it will imply to live in a foreign country. F2

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

The topic of ‘recruitment agencies’ was introduced by interviewee F3. Such agencies have become an essential part of the puzzle to understand the true picture of domestic migrant workers. Filipinas constantly sustained that such agencies are ‘nothing more and nothing less’ than exploiters: Two agents brought fve Filipinas each and, when they arrived in Malta, they had to wait for two months before [being] assigned any work. Their life was miserable until they found work. Even their basic needs weren’t assured. Every Filippo pays €5,000 to the agent to come to Malta. They sell their properties in the Philippines to fnance their trip to Malta. And, then, they come here and fnd no [immediate] work. F4 There are Filipinos who lure migrants to come to Malta. They are not accredited by the government and the offcial recruitment agencies. One cannot present himself as an agent, unless he is legally connected to an authorized agency. They will tell you, Give me money for a working permit and health insurance, but, in actual fact, these are already paid. There must be a hidden hand somewhere to let these things happen. F7 Other interviewees refused to comment on the role played by agents, while others talked about them as if they were a ‘necessary evil’ in the system. Because of gross

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

204    Mario Thomas Vassallo and Manwel Debono misinformation or the lack of transparent systems and procedures, migrant workers may end up exploited even before reaching their country of destination. However, the revelation by some that there is the ‘hidden hand’ or the ‘closed eyes’ of the establishment in order for such a criminal racket to thrive, deserves deeper investigation.

6. Working Conditions Working conditions are a key indicator of the dignity and wellbeing of ‘these women who [are] living most of their time with their aged and physically challenged employer’ (Billiard, 2018, p. 22). There are various elements related to such conditions, including contractual obligations, living at the employer’s own house, working on a 24/7 basis and the monitoring capacity of local labour authorities. The Interviewees’ narratives capture some of these realities through both positive and negative experiences. Given that such a job can be classifed as ‘emotional labour’ (Russell, 1983), there is no demarcation line between work and personal life. The house they live in is their workplace. The client they nurse is their employer: It’s a 24/7 job. When my employer was admitted to hospital, he wanted me by his bedside. When I asked why he wouldn’t call the nurse, he said that he only trusted me. One night, I started feeling like going crazy. I went out of hospital and spent three hours night driving to feel what is like to be out. Free! F1

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

I was emotionally involved. I suffered seeing her slipping away slowly. I pity her but at the same I was feeling very anxious because she became aggressive due to dementia. She became uncontrollable. She was not aware anymore of night and date times. F7 F1’s employer is very rich and he could afford to buy a car to his carer to do the errands because his house is situated in a remote, rural area which is not well served by public transport. Nevertheless, material wellbeing is no substitute for losing one’s own freedom. This imbalance becomes starker on the weekly day-off that carers are entitled to. Although stipulated in their contract, almost all of the interviews stated that it is not always easy and automatic to avail themselves of this right due to various reasons: If I am requested to stay indoors on the day-off, I am not paid extra. There are employers who pay, but mine no. F1 What annoys me is that, sometimes, the daughter tells me that she cannot come today, as agreed. If I had a plan to go out on that day, I have to stay inside at the last minute. F3 I only leave on Sundays from 9am and come back by 10pm. It is not 24 hours and I am not paid extra. F6

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Labouring Behind Closed Doors    205 [The employer] didn’t want me to go out [on my day-off] and she paid me extra. There was nobody else to assist. And I stayed because I felt responsible for her. F7 Before 4pm, I have to leave from wherever I am, to go home and prepare dinner. I know that other Filipinas need to fnd a replacement, but my employer doesn’t want other people to stay with her when I am out. F8 Although elucidating different feelings, resulting experiences expose an underlying feature that cut across other factors characterising the working and living conditions of live-in care workers: the asymmetrical conditions that Filipinas are obliged to adhere to. Some can take their day-off; others are requested to fnd a replacement in advance. Some are paid extra when they are asked to refrain from their day-off, others are not. Some are compelled to take less than the 24 hours they are allowed to, while others are forced to break down their off-duty day on two or three time brackets. These lopsided arrangements are also evident with regards to the contractual entitlement to a month paid leave annually during which many opt to visit their homeland. Again, some Filipinas encounter no problem to visit the Philippines, while others are either denied this right by their employer or sanction themselves out of their own sense of obligation: I am entitled for a trip to the Philippines once every year. But I only went once. The second time round, I couldn’t go because his health condition became more fragile. He told me to go but I decided not to because this would have created many problems to him. F1

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

To go to the Philippines, I have to fnd a reliever. Otherwise I cannot go ... F3 I can’t have it annually, but only once every two years with this employer. I spoke to my employer about my rights. He said, If you insist on your rights, you better fnd another employer! F6 The interviewees also distinguished between what Billiard (2018) termed as `labour code’ and `family code’: I have two contracts of employment. The offcial one involves the authorities stipulating that I only do medical care. Then I have another contract with the family. There, the conditions are different because it states that I have to perform both caring and housework. F2 It all depends on the people you work with. It is not the law that determines your conditions of employment. It’s people’s hearts. F4

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

206    Mario Thomas Vassallo and Manwel Debono The situation appears even more acute when all interviewees, except for one, confrmed that they have never been consulted or visited by relevant authorities to check on their wellbeing and whether the contractual labour obligations are being honoured or not: It is not easy to approach [specifc authority named]. If we suffer any abuse, we don’t know whom to ask for help. F2 No authority has ever visited us. If a problem arises, you must frst face the family and show them the contract. If they take no heed, then you forward the complaint to the Community Group. F3 During my eight years of employment here, I never experienced a man of authority knocking on my employer’s door to check that all is right. I don’t report to anyone. We don’t go to the authorities. I don’t want hassles. F4 The experience of another interviewee shows that there are sporadic initiatives by the Maltese state authorities, like the National Statistics Offce (NSO) and Jobsplus, that are trying to assess the working conditions of overseas domestic workers, but their efforts are not deemed ‘adequate’:

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Once I got a surprise visit from NSO. He came but I was attending caring training at school. My employer told him to call me later. Then he phoned and asked me about working conditions. Also, Jobsplus sends a questionnaire online annually. But many don’t answer it because they fear the employer. We are not being adequately monitored to detect exploitation. F6 Notwithstanding this rather bleak scenario, there is ample evidence that this kind of job does have its own unique benefts and some of those interviewed stated that they would retain their current job if given an alternative choice to work in public hospitals or residential homes: As a live-in carer there are advantages. I don’t pay any bills. Everything is free. F8 I prefer my present job. When you work in hospital, you have to rent a fat, buy your own food, pay electricity and water bills. While we, as personal carers, are free from all this. F5 Others sounded a different opinion. The loss of one’s own liberty is a heavy price to pay and, thus, they are actively seeking alternative jobs: I will seek another job. My [lack of] freedom is my greatest frustration. F6

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Labouring Behind Closed Doors    207 So my free time has to work around [patient] time. It’s like paid slavery. F7 It can be concluded that although similar patterns do emerge, interviewees’ narratives are characterised by diverging experiences. The reality of the interviewed overseas domestic workers toiling behind closed doors of Maltese households is not determined by legislation but by family demands and personalities.

7. Motivation and Relations with Employer Financial reasons cannot be the only incentive to survive this kind of job. Selfmotivation plays a central part in providing constant ‘gentle affection and care’. This section explores various aspects related to intrinsic motivation, as well as the kind of relationships with patients and their relatives. ‘The work of the live-in carer is more than just a job’ (Orsini, 2018); it is actually a work of love wherein the meaning of work becomes connection, engagement and commitment. The experiences of the interviewees corroborate this claim: Since I was young, I wanted to work with vulnerable people. It’s a vocation. If you are happy, every problem will be small. F3 When I was in the Philippines, I was already taking care of my elder family members. I know the job. It’s part of who I am. F8 I have emotional attachment. I treat him [my patient] like my father. In time I started to call him daddy. F1

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

When I was being interviewed by the son of my employer, he asked how I will take care of his mother. I said, I will take care of your mother in the same way I would have taken care of my grandmother. F6 The health condition of the patient makes a lot of difference in the life of a caregiver. A person who is still fairly independent, mobile and mentally sane renders the daily toil much more manageable. On the contrary, life can be extremely chaotic for a carer whose patients have a diffcult personality, cannot recognise night from day, cannot eat on their own or are in need of constant supervision: I am taking care of nanna of 94 years. She is still independent but suffering from dementia. She needs constant watch, even during the night. So I sleep when she sleeps. F5 I feel lucky and happy. I take care of a 75-year-old man who has renal problems. Kind-hearted and fully independent. F2 My frst employer was terrible. She was amputated. Her character was not good. The previous carers had to leave her because they couldn’t live with her. F4

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

208    Mario Thomas Vassallo and Manwel Debono He is 65 with severe health complications. Alcoholic, chain smoker and suffers from chronic depression. F1 The quality of life of a live-in carer is not only dependent on the health condition of her patient but is also deeply affected by interactions with the latter’s relatives. Some interviewees had words of praise for the relatives: Family members encouraged me to get formal training in caring. They advised me that it is better to get certifcation so as I can sustain my job in the future. F2 The children appreciate me and give me something extra during Christmas time and on my birthday. F3 Others recounted different experiences: I was asked by the family to sign an agreement that I will not claim anything on the basis of inheritance when the patient dies. I felt so miserable and angry. F8 One of her [employer’s] daughters is angry at me, simply for seeing me sitting down. She wants me to work all the time. I told her, I am not a cleaner; my duty is a carer. But she wants me to clean the windows and the foor. F5

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

The narratives of migrant live-in carers in Malta are in tandem with the research undertaken by Iecovich (2011) who concludes that the quality of relationship, in particular intergenerational relationships, was the most signifcant variable in explaining both caregiving burden and caregiving satisfaction. Vahabi, Pui-Hing Wong, and Lofters (2018) adds that dysfunctional relationships with patients and their relatives place live-in carers at greater risk of compromised mental health.

8. Social Integration A thorough exploration of the working and living conditions of live-in carers must also include the dimension of social integration. To what extent do Filipinas feel accepted in the Maltese community? Do they have any opportunities for community engagement? All interviewees agreed that Malta’s Catholic environment and the wide use of English language help them to integrate very rapidly: We are Catholic. I come from a very strict Catholic family. It makes it easier to integrate. F8 I found God in Malta. I was a Christian by name but not by deed in the Philippines. But, here, my relationship with God developed deeply. F6

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Labouring Behind Closed Doors    209 It becomes easier because we share Christianity as our religion. Maltese people are approachable. If you need help, everyone tries to help you here. They speak English too. F2 Although the Maltese language is reckoned to be diffcult, many of the interviewees hinted that they have already, or are planning to learn it. This would enhance their integration process: Next year I will start learning Maltese. That is my goal. F2 I try to study Maltese but it is diffcult. If you don’t practise it, you will not remember it. F6 I speak English and Italian; and these are widely spoken in Malta. I don’t know Maltese. F7 Unlike the situation in continental Europe wherein Filipinos fnd it harder to integrate, Malta does provide a convivial environment which eases the stress of the frst year. Like Israel, where 50% of the migrant care workers are from the Philippines (Liebelt, 2019), Malta offers opportunities for social inclusion and engagement. The fve Filipino community groups that are active in Malta are crucial to help Filipinas integrate within society, and provide them with incentives and activities that call for social engagement. Volunteering is an excellent way to engage in dialogue, while making new friends and gaining confdence in working towards something meaningful with others: We organise weekly gatherings to share experiences; [hence] relieving pressure and tension. Sometimes meeting someone else who shares the same life experience is a relief. F3

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

I am member in one of the Groups. I go for their social gatherings when they are held on Sundays. F5 However, fve out of eight interviewees stated that they are not members in any of the Groups, although they attend some of their activities like the Christmas Party and the Philippines national day event. A common complaint is that there is too much cutthroat competition among them. This does not entice Filipinas to get involved: Community groups’ relationship is sad. They are very fragmented. If we don’t see eye to eye, we tend to split from the group and form another. F6 Filipino community groups in Malta are a disaster. F7 They do a lot fundraising but I see no projects organised. We don’t know where the money is going. F2

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

210    Mario Thomas Vassallo and Manwel Debono They fght on who is the best to represent us; on who has the greatest number of members. F8 After years of fagrant rivalry, things seem to be changing for the better. In 2018, all of the fve Groups came together and organised one Christmas Party. This was unprecedented and involved a lot of goodwill from all those involved. Furthermore, sports can be another unifying factor as F7 suggests: Sports is uniting us because all organisations are contributing. We are starting the Sports Festival together. We are working on that! F7 The Maltese situation differs considerably from that postulated by international scholars on the grounds of social inclusion and engagement. Whereas evidence-based reports speak about ‘cultural shock and homesickness’ (Chowdhury & Gutman, 2012) and ‘signifcant stress’ especially on their frst year living abroad (Tejero & Fowler, 2012), the Maltese environment is deemed to be much less stressful.

9. Life Beyond Work The last set of narratives encompasses work–life balance. Being a domestic caregiver leaves very little time for anything else. Actually there is no balance at all because, essentially there is no distinction between work and personal life. In this respect, three sub-themes emerged during conversations, namely: (i) family ties, (ii) hobbies and (iii) the ultimate dream of leaving Malta for good. Family connections are fundamental to Filipinas. Establishing almost daily contact through social media with their loved ones back in the Philippines helps them to regenerate and gain energy to move on with the struggles of their work:

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

I have a son studying at the Seminary back home. We keep regular online contact. F4 I have two children and fve grandchildren. I don’t feel far away because we talk on Facebook and call them on Skype. F5 I have my daughter in the Philippines. Social media helps a lot. F8 My youngest was seventeen when I left and my eldest was 21. I have another one of 19 years. They can manage on their own without my presence. F6 Besides maintaining regular contact with family and friends, there is almost no opportunity left for other things. Caring for the elderly, particularly if immobility and dementia are entangled, leaves no time for hobbies and side-line interests. Only two interviewees said that they have time for such activities:

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Labouring Behind Closed Doors    211 On my free day, usually Sundays, I go to church and do bible study. If other social activities interfere with my church life, I go to church frst. F6 At frst I felt lonely but my nanna has a sewing machine and I like to do so many things to make me busy. Sometimes you feel the need to go out; to feel liberated. It is a feeling that comes and goes. F3 F3’s narratives expose the third factor underlying this theme: the ultimate dream of leaving Malta for good. Seven out of the eight interviewed claimed that they do not intend to settle in Malta for the rest of their lives. They are always wondering about the day when they will return to their homeland and reunite forever with their families: I will stay here for fve more years. When my elder son fnishes his studies at the seminary, I want to be there again. F4 I will not stay in Malta forever. One day I will go to Philippines for good. But timing depends on my savings! F5 The fndings in this section are in line with the conclusions of international scholars regarding the diffusion of leisure time in the life of overseas care workers.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

10. Conclusion The majority of the fndings presented in this chapter refect some of the concerns that have already been identifed in international literature, like higher levels of risk, precariousness, unfulflled contractual agreements, psychological obligations, fraudulent agents and the lack of separation between work and personal life. Other fndings have endogenous characteristics that are closely linked to the island state of Malta, namely its safe environment, Catholic culture, bilingual coexistence of Maltese and English, and the competitive nature of Filipino community groups that may discourage further social engagement. The resulting patterns and divergences shed exploratory light on what is really happening behind closed doors where the working and living conditions of care workers are shaped by the temperament, moods and exigencies of the family rather than by the clauses laid out in employment contracts. This study confrms that there are no overall conclusions that are applicable to all cases; each case presents different stories involving positive and negative aspects of humanity. Some live-in care workers enjoy decent wellbeing because the client that they are responsible for is still relatively strong and independent, while their relationship with other family members is built on mutual respect and understanding. Others experience outright exploitation by selfsh employers. This may explain why three out of the eight interviewees confrmed that they will recommend this job to their peers, while the rest expressed caution. F2 and F6 both referred to the element of ‘risk’. There is a lot at stake when

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

212    Mario Thomas Vassallo and Manwel Debono

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

a Filipina leaves her homeland to start a new life in an alien house. Lack of prior information may, eventually, result in abuse. F1 more blatantly said, ‘I will not recommend this job to a friend because I won’t serve as a traitor’. The code of silence among Filipinas who ‘are scared because they fear that they may get fred for speaking out’ (F4) renders the policy landscape even more problematic. This feeling is also reinforced by the low number of subjects who consented to participate in this research despite the various reach outs that have been made. On the basis of this research, the authors recommend six policy actions, namely: (1) Given the potential of illegal recruitment agencies, competent authorities should take a closer look at the situation and avail themselves of information and resources to identify and take action against fraudulent recruiters. The effective monitoring of recruitment practices and the labour market in general, as well as the domestic workers sector in particular, needs the commitment and concerted effort of all the authorities involved. (2) Given the apparent insuffcient monitoring of foreign domestic workers, authorities should be more vigilant by systematically paying surprise visits to those homes requiring 24/7 carers. These visits would serve as an institutional ‘outreach’ to uncover illegal and oppressive practices experienced by immigrant carers. (3) Given the nature and consequences of ‘emotional labour’, live-in care workers should be specifcally trained on how to detach themselves from their patients/clients in order to lessen mental stress levels. Learning how to draw emotional boundaries should be an integral module within the training programme leading to a certifcation in caring work. (4) Given the skills required to do the job effectively and the numerous narratives recounting stories of abusive employer–employee relationships, the authorities may consider a process to professionalise the work of live-in care workers, wherein practitioners are issued with a warrant after the necessary level of competence is ascertained through accredited training and on-the-job experience. In parallel, a code of ethics needs to be compiled and endorsed by both employers and carers. (5) Given the crucial need of having a supporting Filipino community, the various Filipino Community Groups in Malta should strive to develop more collaborative relationships by organising more joint activities, sharing resources and solidifying their voice in domestic and European civil society. (6) Given this type of work that renders limited scope for social integration, the migrant integration policy should include special measures specifcally designed for live-in carers and other domestic carers in order to facilitate their wider engagement. These recommendations may lead to signifcant improvements in the wellbeing and dignity of migrant carers in Malta and elsewhere. Further research is required to evaluate the ramifcations of these brief policy suggestions and lead to a more detailed road map.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Labouring Behind Closed Doors    213

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

References Alfonso, A. (2012) Employer strategies, cross-class coalitions and the free movement of labour in the enlarged European Union. Socio-Economic Review, 10(4), 705–730. Ambrosini, M., & Barone, C. (2007). Employment and working conditions of migrant workers. Dublin: European Foundation for the Improvement of Living and Working Conditions. Anderson, B., & Ruhs, M. (2010). Who needs them? A framework for the analysis of staff shortages, immigration and public policy. In B. Anderson & M. Ruhs (Eds.), Who needs migrant workers? (pp. 15–52).Oxford: Oxford University Press. Andres, T. Q. (1996). Effective discipline through Filipino values. Manila: Rex Bookshop Inc. Bales, R., Audyanto, S. B., Guven, S., Jockel, M., Juniarto, C. B., Junlu, J., & Langmead, D. (2014). A comparative analysis of labor outsourcing. Arizona Journal of International & Comparative Law, 31(3), 579–622. Bhuyan, R., Valmadrid, L., Panlaqui, E., Pendon, E., & Pearlita, J. (2018). Responding to the structural violence of migrant domestic work: Insights from participatory action research with migrant caregivers in Canada. Dordrecht: Springer Science and Business Media. Billiard, E. (2018). The encounter. In G. Orsini (Ed.), Exiled homes (pp. 11–16). Malta: Valletta 2018. Blanchfower, D. G., & Bryson, A. (2008). Union decline in Britain. Institute for the Study of Labor. Retrieved from http://repec.iza.org/dp3436.pdf Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3, 77–101. Business Mirror. (2018). Better deal for Pinoy maids, caregivers eyed in Israel, Makati City, August 31. Central Bank of Malta (CBM). (2016). Assessing the economic impact of foreign workers in Malta. Quarterly Review 2016, 1, 39–44. Retrieved from https://www. centralbankmalta.org/quarterly-review Chowdhury, R., & Gutman, G. (2012). Migrant live-in caregivers providing care to Canadian older adults: An exploratory study of workers’ life and job satisfaction. Dordrecht: Springer Science and Business Media. Church, T. A. & Katigbak, M. S. (2000). Trait psychology in the Philippines. American Behavioral Scientist, 44(1), 73–9. Retrieved from https://www.researchgate.net/ publication/258122412_Trait_Psychology_in_the_Philippines Corden, A., & Sainsbury, R. (2005). Using verbatim quotations in reporting qualitative social research: Researchers’ views. York: Social Policy Research Unit, University of York. Duncan, C., & Loretto, W. (2004). Never the right age? Gender and age-based discrimination in employment. Gender, Work and Organization, 11(1), 95–115. Dyer, S., McDowell, L., & Batnitzky, A. (2008). Emotional labour/body work: The caring labours of migrants in the UK’s national health service. Geoforum, 39(6), 2030–2038. Encinas-Franco, J. (2016). Filpino women migrant workers and overseas employment policy: An analysis from women’s rights perspective. Asian Politics & Policy, 8(3), 494–501. European Commission. (2018a). The 2018 ageing report: Economic and budgetary projections for the 28 member states (2016–2070). Brussels: European Commission. European Commission. (2018b). European Economic Forecast: Summer 2018 (Interim). Brussels: European Commission. Retrieved from https://ec.europa.eu/info/sites/ info/fles/economy-fnance/ip084_en.pdf European Economic and Social Committee (EESC). (2018). Europe needs a proactive policy for the long-term care sector. Ref. No. 13/2018. Retrieved from https://www. eesc.europa.eu/en/news-media/press-releases/europe-needs-proactive-policy-longterm-care-sector

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

214    Mario Thomas Vassallo and Manwel Debono European Labour Court Judges Meeting (ELCJM). (2007). National reports on outsourcing, Bundesarbeitsgericht Erfurt, Germany. Retrieved from https://www.ilo.org/ wcmsp5/groups/public/@ed_dialogue/@dialogue/documents/meetingdocument/ wcms_159885.pdf European Parliament (EP). (2014). Discrimination of migrant workers at the workplace. Luxembourg: Publications Offce of the European Union. Eurostat. (2019). Fertility statistics in the European Union. Brussels: Eurostat Publications. Financial Times. (2018). Highest fertility rates in Europe still below replenishment level, March 28. Retrieved from https://www.ft.com/content/d54e4fe8-3269-11e8-b5bf23cb17fd1498 Gahagan, J., Loppie, C., Rehman, L., Maclellan, M., & Side, K. (2007). Far as I get is the clothesline’: The impact of leisure on women’s health and unpaid caregiving experiences in Nova Scotia, Canada. Health Care for Women International, 28, 4768. Ho, K. H., Chiang, V. C. L., Leung, D., & Cheung, D. S. K. (2019). A feminist phenomenology on the emotional labor and morality of live-in migrant care workers caring for older people in the community. BMC Geriatrics, 19, Article 314. Iecovich, E. (2011). Quality of relationships between care recipients and their primary caregivers and its effect on caregivers’ burden and satisfaction in Israel. Journal of Gerontological Social Work, 54(6), 570–591. International Labour Organisation (ILO). (2013a). Domestic workers across the world: Global and regional statistics and the extent of legal protection. Geneva: ILO. Retrieved from https://www.ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/--publ/documents/publication/wcms_173363.pdf International Labour Organisation (ILO) (2013b), Care work and care jobs for the future of decent work. Geneva: ILO. Retrieved from https://www.ilo.org/wcmsp5/groups/ public/---dgreports/---dcomm/---publ/documents/publication/wcms_633166.pdf International Labour Organisation (ILO). (2017). Who are domestic workers? Retrieved from https://www.ilo.org/global/topics/domestic-workers/WCMS_209773/lang--en/ index.htm Jobsplus. (2018). Filipinos working in Malta 2010–2018. Unpublished data. Kocer, R. G. (2018). Measuring the strength of trade unions and identifying the privileged groups: A two-dimensional approach and its implementation. The Journal of Mathematical Sociology, 42(3), 152–182. Retrieved from https://www.tandfonline. com/doi/full/10.1080/0022250X.2018.1449110 Liebelt, C. (2019). Caring for the ‘Holy Land’: Filipina domestic workers in Israel (Chapter 5). New York, NY: Berghahn Books. Moriarty, J. (2010). Competing with myths: migrant labour in social care. In B. Anderson & M. Ruhs (Eds.), Who needs migrant workers?. Oxford: Oxford University Press. National Commission for Active Ageing. (2013). National strategic policy for active ageing 2014–2020. Malta: National Commission for Active Ageing. O’Neill Richard, A. (2000). International traffcking in women to the United States: A contemporary manifestation of slavery and organized crime. Washington, DC: Centre for the Study of Intelligence. Orsini, G. (2018). Exiled homes. Malta: Valletta 2018. Pillinger, J. (2016). Psychosocial risks and violence in the world of work: A trade union perspective. International Journal of Labour Research, 8(1–2), 35–61. Robillard, C., McLaughlin, J., Cole, D. C., Vasilevska, B., & Gendron, R. (2018). Caught in the same web-service providers: Insights on gender-based and structural violence among female temporary foreign workers in Canada. Dordrecht: Springer Science and Business Media, part of Springer Nature. Rousseau, D. M. (1989). Psychological and implied contracts in organizations. Employee Responsibilities and Rights Journal, 2, 121–139.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Labouring Behind Closed Doors    215

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Rupeika-Apoga, R., Romanova, I., Bule, L., & Thalassinos, E. Y. (2019). The impact of population ageing and social stratifcation: The case of Latvia. International Journal of Economics and Business Administration, 7(1), 49–63. Russell, H. A. (1983). The managed heart: Commercialization of human feeling. Berkeley, CA: University of California Press. Saito, I. (2013). Pakikisama: A Filipino Trait. Semantic Scholar paper, 45–53. Retrieved from https://www.semanticscholar.org/paper/PAKIKISAMA-%3A-A-FilipinoTrait-Isamu-Saito/e07ef91b6d66b14e664958db0bd51f9d80552ae7 Simon-Rusinowitz, L., Mahoney, K. J., & Benjamin, A. E. (1998). Payments to families who provide care: An option that should be available. Generations, 22(3), 69–75. Retrieved from http://www.advancingstates.org/sites/nasuad/fles/hcbs/fles/46/2254/ paymentstofamiliesgenerations.pdf Tejero, L. M., & Fowler, C. (2012). Migration of women from the Philippines: Implications for healthcare delivery. Collegian, 19(1), 59–63. Thalassinos, I. E., Stamatopoulos, V. T., & Arvanitis, E. S. (2011). Gender wage gap: Evidence from the Hellenic maritime sector 1995–2002. European Research Studies Journal, 14(1), 91–101. Trimikliniotis, N., & Demetriou, C. (2011). Labour integration of migrant workers in Cyprus: A critical appraisal. In M. Pajnik (Ed.), Precarious migrant labour across Europe (pp. 73–96). Ljubljana: The Peace Institute – Institute for Contemporary Social and Political Studies. United Nations. (2015). The role of recruitment fees and abusive and fraudulent recruitment practices of recruitment agencies in traffcking of persons. Vienna: United Nations Offce on Drugs and Crime. Vahabi, M., Pui-Hing Wong, J., & Lofters, A. (2018). Migrant live-in caregivers mental health in Canada. Community Mental Health Journal, 54, 590–599. Wills, J. (2009). Subcontracted employment and its challenge to labor. Labor Studies Journal, 34(4), 441–460. Yi-Chung, H., & Yu-Cheng, H. (2012). Work and leisure among Indonesian professional family caregivers in Taiwan: Reality versus dreams. World Leisure Journal, 54(4), 337–349.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

This page intentionally left blank

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Chapter 15

Too Small to Hedge? The Case for Derivatives Hedging for a Small Island State Konrad Farrugia, Matthew Attard and Peter J. Baldacchino Abstract This study delves into the determinants and praxis of derivative hedging instruments (DHIs) usage of Malta, a small island state. Empirical evidence is also provided in relation to the impact of DHI usage and the adoption of a hedge accounting (HA) model in entities’ fnancial statements. A mixed methodology design is deployed involving: (1) a series of statistical models and tests and (2) seven semi-structured interviews with senior professionals.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

The data collected comprise proxy variable values collected from the fnancial statements of 568 frm-years from 107 Maltese entities between the years 2009 and 2014. Greater likelihood of fnancial distress, decreasing investment effciency and increased levels of gearing, are identifed as being signifcant determinants for the use of DHIs. Although DHI usage is low in comparison to larger states, it has been increasing over the period under study. HA is evidenced to be less popular in Malta, but the study evidences correlation between certain DHIs and HA usage. The quantitative statistical model results in evidence with no signifcant earnings volatility (EV) or cash fow volatility (CFV) reduction effects through the application of HA. Albeit, the study fnds a signifcant CFV reduction effect emanating from DHI usage, but no corresponding EV reduction effect. Better education and dissemination of the HA treatment by auditors and regulatory bodies could help propagate the HA treatment, potentially enhancing the EV reduction effectiveness of DHI use. This research provides empirical evidence to substantiate the rationale behind utilising DHIs in smaller island states, especially when coupled with a sound risk management culture. Keywords: Hedging derivatives; hedge accounting; hedge accounting models; volatility; risk management; volatility modelling; small island states; risk management Uncertainty and Challenges in Contemporary Economic Behaviour, 217–257 Copyright © 2020 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-80043-095-220201016

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

218    Konrad Farrugia et al.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

1. Introduction An absolute certainty to the construct of uncertainty and risk is its need to be identifed and managed. Increasing transaction and operational complexity, fnancial and political instability together with more stringent regulatory compliance requirements, have accentuated the need for the implementation of an effective risk management system (Thalassinos & Thalassinos, 2018). Uncertainty around any activity be it on impact or likelihood of occurrence are the two defning elements of risk. Financial risk comprises, but is not limited to: market price risk, interest rate risk (IRR), and foreign exchange and credit risks. Risk management is the eclectic discipline dealing with the measurement, mitigation and treatment of risks (Pociovalisteanu, Thalassinos, Tirca, & Filho, 2010; Thalassinos & Pociovalisteanu, 2007). In the realm of corporate risk management, two signifcant measures of risk are earnings (EV) and cash fow volatility (CFV). EV may be described as the degree of uncertainty around an entity’s ability to generate distributable profts. On the other hand, CFV captures the degree of uncertainty about the entity’s ability to generate cash from its operations and sustain working capital and liquidity requirements. Ergo, if an entity can effectively manage the uncertainty of these two metrics, this should result in increased stability of earnings and cash fows, leading to balance sheet value stability, and a reduction in the probability of bankruptcy or ruin. Hedging can be defned as a risk management process, which is aimed at, restricting, or offsetting particular risk exposures through techniques, strategies, contracts and market positions. Non-contractual or ‘natural’ hedging techniques can often be deployed to mitigate particular risk exposures, such as rate-sensitive asset–liability management or portfolio immunisation in the case of IRR (Redington, 1952) or currency exposure matching. Nevertheless, derivative fnancial instruments or derivative hedging instruments (DHIs) are often entered into to allay some of the uncertainty around critical transactions’ pay-offs. A derivative fnancial instrument is defned as; ‘a fnancial instrument whose value depends on the value of other, more basic underlying variables’ (Hull, 2015, p. 1). Derivatives should be used with caution and require technical knowledge and expertise, given their complex structures deriving value from changes in the underlying variable. In the IFRS fnancial reporting area, derivatives can be designated as hedging instruments, following the accounting treatment outlined in IFRS 9. IFRS 9’s hedge accounting (HA) models1 may be adopted with the aim of matching and recognising any gains or losses in the underlying instrument with opposite gains or losses the designated hedging instrument within the same accounting period. In theory this would then possibly abate artifcial proft volatility inherent to ­recognising the value differentials in different periods.

1

Simply put a hedge accounting model captures an accounting treatment that can be ascribed to a designated hedging relationship consisting of a hedged item and a DHI, with the objective of recognising the risk reduction effect of the DHI in the same period of recognising the movements in the hedged item, thereby reducing reporting volatility.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

The Case for Derivatives Hedging for a Small Island State    219 The objectives of this chapter are: (1) To characterise and identify the attributes of users of DHIs in Malta as a Small Island State, through the use of discriminant analysis within the local sampling frame. (2) To present the empirical evidence, derived from the sampling frame for the possible case of DHIs usage by entities in a Small Island State, particularly in respect of reduction of EV and CFV and the implications of adopting a HA model to account for these DHIs. This chapter is segmented in four sections: The frst two lay out the relevant literature and methodology deployed respectively. The third presents the empirical results of the methodology and their respective implications. The conclusions and the inherent study limitations discussed in the fnal section.

2. Literature Review

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

2.1. The Characteristics of DHI Users: The Rationale for DHI Usage The central argument underlying the rationale behind DHI use is predominantly the protection of an entity’s cash fows from risk exposures (Saint-Leger, 2015). Existing literature seems to show, that the rationale for DHI utilisation is dependent on the perspective adopted: At an individual investor level the motivation for hedging is minimal, as presented in the seminal work by Smith and Stulz (1985), Ugurlu, Thalassinos, and Muratoglu (2014) and Rupeika-Apoga, Zaidi, Thalassinos, and Thalassinos (2018). Therefore, in this study, derivatives hedging is treated as part of an entity specifc fnancial strategy, which in turn is consistent with the irrelevancy theory postulated by Modigliani and Miller (1958), under the assumption of perfect capital markets (Brennan, 1971). Nevertheless, individual entities may still be exposed to systemic risk, especially if they lack the scale and resources to diversify. Here empirical research has been quite broad: identifying a key number of ex-post proxy measures2 evidenced to exert some form of infuence in when deciding to utilise DHIs. When it comes to derivatives’ usage in smaller entities, literature is rather limited. A frequently adopted defnition of a small- to medium-sized entity is that given by the European Commission.3 Risk exposure and size are of notable infuence and ‘the SMEs decision-making unit, the manager’s risk attitude, the manager’s risk perception and the manager’s education level play an important role in explaining smaller entity derivatives usage’ (Pennings & Garcia, 2004). The 2

In respect of entities that utilise DHIS, the variables are identifed as ex-post measures, as they are deemed to be captured in the year/time period DHIs are evidenced to be in use by the entity. 3 Thresholds for an entity classifying as an SME are as follows: an entity with an average number of employees of less than or equal to 250, and either one of the following: annual turnover or Balance Sheet total of less than or equal to: €50 million or €43 million, respectively (European Commission, 2012).

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

220    Konrad Farrugia et al. hedging tools and practices used by SMEs are far less complex (exotic) products in comparison to larger entities. The majority of hedging products sold to SMEs are plain vanilla interest rate or commodity swaps and currency derivatives (Thalassinos, Stamatopoulos, & Thalassinos, 2015, chapter 20; Thind, 2004). Foreign currency hedging instruments are popular as SMEs are not keen to take on currency risk (Baldacchino, Gauci, & Grima, 2019; Greenhalgh, 2012). IRR is another risk prominently hedged by SMEs. In reporting on the sale of interest rate hedging risk products, the United Kingdom’s then Financial Services Authority (FSA, 2013) notes that the more commonly used derivatives by SMES are: swaps, caps and collars.

2.2. Hypothesis Formulation for the Factors Affecting Maltese Entities’ DHI Usage In formulating hypothesis to determine the factors infuencing the choice of DHI usage, sets of classifcation factors were selected. The objective of the classifcation factors is to determine a set of variables whose values differ between DHI users and non-DHI users, based on some underlying economic or fnancial rationale. Given that some of the actual variables identifed in previous studies are not directly measurable and observable, proxy variables considered to be representative of the underlying variable have been used. The classifcation through the use of proxy variables has been carried out through statistical hypothesis testing, as outlined in Sections 2.3–2.7.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

2.3. Effective Reduction of the Entity’s Corporate Tax Rate and Potential Costs of Financial Distress Hedging can be said to increase the present value of a frm’s after-tax cash fows, by reducing the degree of uncertainty around its pre-tax value when it’s effective Income tax rate (ITR) increases proportionately to its pre-tax value (Smith & Stulz, 1985). Hedging reduces an entity’s expected tax liability by smoothing ­variations in income. Judge (2006) states that hedging helps stabilise an entity’s cash fows resulting in a lower present value of fnancial distress costs. Costs of fnancial distress (CFD) include restrictive debt fnancing covenants and higher fnancing costs. Ailing Tangible asset and other short-term liquidity ratios (Nance, Smith, & Smithson, 1993), unsustainable levels of dividend pay-outs, higher degrees of fnancial leverage or gearing and higher fnancing costs, could all be symptomatic of fnancial distress costs. Hedging derivatives such as interest rate swaps and currency forwards may help reduce the variability in an entity’ costs of fnance by stabilising cash fows such as borrowings denominated in a foreign currency or variable interest payments. Entities with higher dividend pay-outs and yields tend to be more prone to incur fnancial distress costs, given an unsustainably high level of dividend payment may result in future inability to sustain such dividend pay-out ratios – which

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

The Case for Derivatives Hedging for a Small Island State    221 may be interpreted as a negative signal on the frm’s value by shareholders (Bartram, Brown, & Fehler, 2009). The following hypotheses are formulated from the ensuing proxy variables: H1. For the frm’s imputed tax rate (ITR) calculated as: ITRi ,t = 1 −

Profit   After  Tax  {PAT }i ,t Profit   Before  Tax  {PBT }i ,t

Expected a priori relationship: ITRDHI users < ITRnon−DHI users H 10  :  MEAN   ITRDHI users = MEAN   ITRNon−DHI users H 1a  :  MEAN   ITRDHI users < MEAN   ITRNon−DHI users For all proxy variables utilised: ‘t’ denotes the sequence of fnancial years in which the proxy variable t = 1, 2, 3, 4, 5, 6 pertaining to fnancial years 2009–2014. ‘i’ denotes the respective entity, which can be dichotomously classifed as utilising DHIs and hence a ‘DHI user’ or otherwise , a ‘Non-DHI user’ via the indicator variable ‘HEDGEi ,t , such that:  1 | entity ‘i ’ was a DHI user in year ‘t’4 4 HEDGEi ,t =   0 | entity ‘I ’ was not a DHI user in year ‘t’ H2. As proxies for the likelihood of not incurring CFD; (a) for the frm’s: current ratio (CuR) calculated as:

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

CuRi ,t =

  Book  Value   of  Current   Assetsi ,t Book  Value   of  Current   Liabilitiesi ,t

Expected a priori relationship: CuRDHI users > CuR  non−DHI users H 2 a0  :  MEAN  CuRDHI users = MEAN  CuRnon−DHI users H 2 aa  :  MEAN  CuRDHI users > MEAN  CuRnon−DHI users

4

An entity was classed as a DHI user in a particular fnancial year if explicit mention was made in the fnancial statements whether in elements of the fnancial statements or disclosed within the notes, irrespective of the amount/extent of derivative contracts/ hedging relationships disclosed.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

222    Konrad Farrugia et al. (b) for the frm’s: tangible asset ratio (TAR) calculated as: TARi ,t =

Balance   sheet   value   for   Property   Plant   and   Equipmenti ,t Balance  Sheet  Value   for  Total   Assetsi ,t

Expected a priori relationship: TARnon−DHI users > TARDHI users H 2b0  :  MEAN  TARDHI users = MEAN  TARnon−DHI users H 2ba  :  MEAN  TARDHI users > MEAN  TARnon−DHI users (c) for the frm’s degree of leverage for the gearing ratio (debt- to- equity ratio) (GR) calculated as: GRi ,t = 

Book  Value   of   Long  Term   Liabilitiesi ,t Book  Value   of  Total   Equityi ,t

Expected a priori relationship: GRDHI users > GRnon-DHI users. H 2c0  :  MEAN  GRDHI users = MEAN  GRnon−DHI users H 2ca  :  MEAN  GRDHI users > MEAN  GRnon−DHI users (d) for the frm’s interest cover ratio calculated as: ICi ,t =

Net  Operating   Profiti ,t

( Finance  Costsi ,t − Finance   Incomei ,t )

Expected a priori relationship: ICDHI users < ICnon-DHI users.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

H 2 d 0  :  MEAN   ICDHI users = MEAN   IC non−DHI users H 2 d a  :  MEAN   ICHedgers < MEAN   IC Non−Hedgers

2.4. Mitigating Ineffcient Investment Underinvestment by entities occurs due to the sub-optimal usage of funds raised (Froot, Scharfstein, & Stein, 1993). This is often associated with high levels of gearing, and tends to be pronounced in entities that have higher growth opportunities (Bartram et al., 2009): Proxy measures of investment effciency include; CAPEX (Capital Expenditure) to Sales Ratio and PPE (Property Plant and Equipment) to Sales. When underinvestment is present, proftable projects, may be refused on the basis that they would result in negligible benefts to the frm’s owners. This occurs when the costs incurred to fund this investment exceed the expected return from this investment (Bessembinder, 1991; Myers, 1977). Hedging has been claimed to address this issue by reducing the investment and frm value’s volatility, to accrue more benefts to shareholders. The following hypotheses are derived:

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

The Case for Derivatives Hedging for a Small Island State    223 H3. As proxies for indicators of possible underinvestment; (a) for the frm’s: PPE to sales (PPES) Ratio calculated as PPESi ,t =

Balance   sheet   value   for   Property   Plant   and   Equipmenti ,t Salesi ,t

Expected a priori relationship: PPESDHI users > PPESnon−DHI users H 3a0  :  MEAN   PPESDHI users = MEAN   PPESnon−DHI users H 3aa  :  MEAN   PPESDHI users > MEAN   PPESnon−DHI users (b) for the frm’s: capital expenditure to sales (CAPEXS) ratio calculated as: CAPEXSi ,t =

Additions  to   PPEi ,t Salesi ,t

Expected a priori relationship: CAPEXSDHI users > CAPEXSnon−DHI users H 3b0  :  MEAN  CAPEXSDHI users = MEAN  CAPEXSnon−DHI users H 3ba  :  MEAN  CAPEXSDHI users > MEAN  CAPEXSnon−DHI users

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

2.5. Reduction in EV and CFV As the theoretical assumption of perfect capital markets does not occur in fnancial markets, entities will be incentivised to protect against any loss in value resulting from the negative perceptions inherent to high values for reported Earnings and Cash Flow volatilities. It is implied from the works of Smith and Stulz and others such as Dolde (1995), Berkman, Bradbury, and Magan (1997), that the higher the volatility of a frm’s earnings and operating cash fows, the greater the incentive to hedge. Dong and Woo Gon (2015) fnd signifcant differences amongst the EV and CFV of DHI users in a sample of hospitality industry operands. Zhang (2008) also empirically identifes reduced risk profles for both effective hedgers and speculators, post the implementation of a derivatives hedging programme. Nevertheless, econometric time series tends to exhibit the phenomenon of ‘volatility clustering’ (Gujarati, 2009). This could have had major distortion effects on the model and statistical tests’ robustness as also identifed by Fagerland and Sandvick (2009). To safeguard the complications that could be brought about by a heteroscedastic5 variance of the time series data, a transformation of the variables into fve categories depending on the quantile distribution of the 5

Heteroscedasticity refers to the property that time series data may have unequal variances through time, and hence violating the fundamental assumption of linear regression, of a constant variance.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

224    Konrad Farrugia et al. data has been constructed. A cumulative logit link function was used to achieve the conversion of the metric volatility capturing the EV and CFV defned below in two respective new variables adopting fve categories each: ‘EVCAT’ for the categorical multinomial transform of variable ‘EV’ and ‘CFVCAT’ for the CFV counterpart. This was achieved by segmenting the data values into fve quantiles corresponding to 20%, 40%, 60%, 80% and 100%6 from the observed values. Probabilities could therefore be estimated, for an entity’s given volatility value to fall within a particular category. The above is captured in the below hypotheses formulated accordingly. H4. As captures of EV and CFVs, respectively; (a) for the ith entity’s earnings volatility (EV) measure in year ‘t’ (EVi , t ) the corresponding multinomial categorical capture7 (EVCATi , j , t ) (j = 1,2,3,4); EVCATi , t , j = f ( EVi , t ) where

 EVCAT 1 if 0 < EVi , t ≤ X 1   EVCAT 2 if X 1 < EVi , t ≤ X 2  f ( EVi , t ) =  EVCAT 3 if X 2 < EVi , t ≤ X 3   EVCAT 4 if X 3 < EVi , t ≤ X 4   EVCAT 5 if X 4 EVCATnon−DHI   users H 4a0  :  MEAN   EVCATDHI   users = MEAN   EVCATnon−DHI   users H 4aa  :  MEAN   EVCATDHI   users > MEAN   EVCATnon−DHI   users (b) or the ith entity’s Operating CFV measure in year t (CFVi , t ) the corresponding multinomial categorical capture (CFVCATi , j , t ) (j = 1,2,3,4); CFVCATi , t , j = g (CFVi , t ) where: CFVCAT 1 if 0 < CFVi , t ≤ Y1  CFVCAT 2 if Y < CFV ≤ Y 2 1 i,t   g (CFVi , t ) = CFVCAT 3 if Y2 < CFVi , t ≤ Y3  CFVCAT 4 if Y3 < CFVi , t ≤ Y4  CFVCAT 5 if Y4 CFVCATnon−DHI   users H 4b0  :  MEAN  CFVCATDHI   users = MEAN  CFVCATnon−DHI   users H 4ba  :  MEAN  CFVCATDHI   users > MEAN  CFVCATnon−DHI   users

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

226    Konrad Farrugia et al. 2.6. The Mitigation of the Agency Problem and Critical Mass of Operations Agency risk occurs where one party appointed to act on behalf of another acts on self-interest. Costs are therefore incurred to ascertain that management’s actions are aligned with those of the entity’s owners. Jensen and Meckling (1976) consider costs such as monitoring costs and the costs of losses from inappropriate actions to be inherent. A re-alignment of management’s interests is possible by means of hedging activity. When management has a signifcant share of voting rights, as can be in a multi-class share capital structure, Bartram et al. (2009) identify an increased likelihood to hedge given increased discretion. Moreover larger entities and listed entities with more access to funds are more likely to use DHIs. Counter arguably, Judge (2006) identifes an alternative rationale to hedge; as smaller entities tend to ‘have a greater incentive to hedge due to the inverse relation between frm size and direct bankruptcy costs’8. This is also attested to by Nance et al. (1993). This could be symptomatic of information asymmetry in the external fnance market. Information about the frm’s actual fnancial standing captured by banks and other external fnancers may be inaccurate or incomplete, leading to higher fnancing costs. Therefore, for smaller entities hedging may be deemed to be a cheaper source of alternative fnance. Smaller states’ corporate entities tend to be smaller and thus for the sample of Maltese entities, it is expected that DHI users tend to have lower total Asset Values, than their larger multinational counterparts, for the reasons explained above. The ffth and fnal sets of hypotheses are therefore derived as follows: H5. For captures of information asymmetry caused by the agency problem and operation size respectively;

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

(a) As a capture of the presence of multi-class share capital structure (possibly indicative of the presence of the agency problem), the indicator variable denoting if a company a corporate entity has multiple share classes ‘ MSCi’,t  1 | entity ‘i ’ multiple share capital classes in year ‘t’ MSCi ,t =   0 | otherwise Expected a priori relationship: MSCDHI users > MSC non−DHI users H 5a0  :  MEAN   MSCDHI users = MEAN   MSC non−DHI users H 5aa  :  MEAN   MSCDHI users > MEAN   MSC non−DHI users

8

This rationale is attested to by Warner (1977) that identifes that smaller frms are more likely to incur CFD, as such costs have a fxed element attributable to them and are therefore a greater burden when a frm lacks size. This is due to the fact that they do not increase proportionately to frm size.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

The Case for Derivatives Hedging for a Small Island State    227 (b) As a capture of the scale of an entity’s operations; For an entity that has its equity listed on the Malta Stock Exchange (MSE) denoted by the indicator variable ‘ MSEi’,t  1 | entity ‘i’ has its equity listed on the MSE in year ‘t’ MSEi ,t =   0 | otherwise Expected a priori relationship: MSEDHI users > MSEnon−DHI users H 5b0  :  MEAN   MSEDHI users = MEAN   MSEnon−DHI users H 5ba  :  MEAN   MSEDHI users > MEAN   MSEnon−DHI users (c) As a capture of the scale of an entity’s operations; For the entity’s size measured by the natural logarithm of a frm’s total balance sheet asset value: SIZEi ,t = ln (Total   Assetsi ,t ) Expected a priori relationship: SizeDHI   users < Sizenon−DHI   users H 5c0  :  MEAN  SIZEDHI   users = MEAN  SIZEnon−DHI   users H 5ca  :  MEAN  SIZEDHI   users < MEAN  SIZEnon−DHI   users

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

2.7. Reporting Volatility Reduction and the Adoption of a HA Regime When adopting a HA model,9 DHI users could at face value, expect a reduction in reported EV and CFV captures. This is primarily inherent to the correction of the reporting mismatch that could occur when recognising changes in the hedged item and hedging derivative in separate reporting periods. Conficting evidence emanates from the literature as to the effectiveness of the adoption of a HA model, under differing accounting frameworks In analysing two mutually exclusive sets of hedging Firms; Effective hedgers and Ineffective Hedgers/Speculators,10 using a multiple regression analysis, Zhang (2008) reports that CFV amongst ineffective hedgers and speculators, is signifcantly reduced, in comparison to deemed effective hedgers, post the adoption 9

Accounting standards such as IFRS 9 and SFAS 113 may allow users to adopt a hedge accounting model which allows the adoptee to recognise movements in the hedged item and hedging instrument (usually a derivative), in the same accounting period to remove any mismatches. Such models include fair value and cash fow hedges. 10 Zhang (2008) defnes ‘effective hedgers’ (EH) as entities whom have statistically signifcant reductions in their volatility measures, post the introduction of a derivatives programme, whilst those that are not successful at allaying volatility, are classed together with those that use derivatives for proft making purposes as ‘ineffective hedgers and speculators) (IS).

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

228    Konrad Farrugia et al. of SFAS 133’s HA model (FASB,1998). On the other hand, no signifcant reduction exhibited in the case of EV. Similarly, Singh (2004) establishes no signifcant change in EV, with the application of HA. Van Dinh and Gong (2013) document that ‘hedge accounting brings great proft to the entities and investors. Simultaneously, it also minimises the risks’. This is contrary to Pirchegger (2006) who highlights using HA as a means to ascertain the shareholder’s best interests. Accordingly the following hypotheses are formed: H6. In the context of the variables defned as ‘EVCAT’ and ‘CFVCAT’ above, only for derivatives usage: (a) As a capture of the ‘ith’ entity in year ‘t’ adopting a HA model, denoted by indicator variable: ‘HAi,t’, in respect of the corresponding value for EVCATi,t,j as the capture for reported EV, where:

1 | entity ‘i ’ has adopted a HA model in year ‘t’ & HEDGEi,t = 1 HAi ,t =  0 | otherwise Expected a priori relationship: EVCATi ,t , j     HAi ,t = 1 < EVCATi ,t , j   HAi ,t = 0 H 6a0  :  MEAN  CFVCATDHI  &  HA   users = MEAN  CFVCATother   cases H 6aa  :  MEAN  CFVCATDHI  &  HA   users < MEAN  CFVCATother   cases

(b) As a capture of the ‘ith’ entity in year ‘t’ adopting a HA model, denoted by indicator variable: ‘HAi , t’, in respect of the corresponding value for CFVCATi , t , j as the capture for reported operational CFV, where:

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Expected a priori relationship: CFVCATi ,t , j    HAi ,t = 1 < CFVCATi ,t , j    HAi ,t = 0 H 6a0  :  MEAN   EVCATDHI  &  HAusers = MEAN   EVCATother   cases H 6aa  :  MEAN   EVCATDHI  &  HA   users < MEAN   EVCATother   cases

3. Methodology 3.1. Research Methodology Outline The main objective of this study is to inter-relate, identifable and signifcant variables which explain: Maltese DHI users’ behaviour, the effect of DHI usage and the adoption of HA models on volatility captures (EVCAT and CFVCAT). A mixed method research design was chosen, to avail of the corroborative synergies from the statistical modelling, with the views of treasury and risk management offcials. Table 1 outlines the methodology applied to answer the research questions. The main research areas of this study are focussed on exploring the differentiating factors of DHI (and HA) users from non-DHI (non-HA) users: with particular focus on their respective earnings and cash fow volatilities. The motivation

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

The Case for Derivatives Hedging for a Small Island State    229 Table 1.  Methodology Overview. Research Approach

Research Question Research Objective I To characterise and identify the attributes of users of DHIs in Malta as a Small Island State

Research Objective II Assessing the impact of derivatives hedging and HA on EV and CFV

Quantitative • A survey or inspection of published fnancial statements • Mann–Whitney-U hypothesis tests • A logistic regression model

• Two proportional odds models for EV and CFV respectively • Mann–Whitney-U hypothesis tests

Qualitative

• Semi-structured interviews

• Semi-structured interviews

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

behind the research tools chosen was primarily the ability of each research tool to contribute towards attaining the research objective. The following research tools were set-up for the quantitative part of the study: (1) To characterise the attributes motivating DHI usage, One-way Mann–WhitneyU hypothesis tests and a logistic regression (logit) model were deployed. The Mann–Whitney-U test is a non-parametric11 test that compares the medians of ordinal independent variables to identify whether they come from the same population. In this case, this test was used to identify whether the hypothesised discriminant variables presented in Section 2 were signifcantly different for DHI users contrasting to non-DHI users and also for HA model adoptees versus non-adoptees. (2) The logit model was utilised to discern the effects of the variables of interest on the log-likelihood of a particular entity to utilise DHIs in a given year. This research tool was particularly well suited, as the model attempts to quantify the infuence of the various hypothesised infuential covariates, on the dichotomous response variable; (non)-DHI utilisation. The logarithmic odds ratio expressed as the dependent variable in the model, is a very powerful metric that can be utilised to express the odds of (non)-DHI usage for a given set of covariates for a particular frm-year. (3) In modelling the effects of DHI usage and HA model adoption on the volatility captures CFV and EV an ordered logit model (proportional odds model (POM)) was utilised to relate the effects on the signifcant variables of interest to the odds of observing higher classes of volatility, thereby garnering an augmented log-likelihood for being classifed within higher volatility categories for a given set of values for the explanatory variables attributable to a particular observation (entity-year). 11

Non-parametric statistical hypothesis tests are used when the variable in question is identifed as being non-conformant to a Normal distribution.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

230    Konrad Farrugia et al. To corroborate the fndings arising from the modelling efforts, seven semistructured interviews with senior professionals in fnancial reporting teams, treasury and risk management positions with a number of the entities under study were conducted. In selecting interviewees, consideration was given to the type of industry, the extent of derivatives usage and history of derivatives usage to enable sample representativeness to a comfortable degree.

3.2. Data and Sampling Frame

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Information about derivatives users in an EU jurisdiction can be found in the respective trade repositories. This disclosure obligation stems from European Markets Infrastructure Regulation (EMIR) issued by the European Securities and Markets Authority (ESMA, 2015). Consequently, all entities engaging in derivative transactions need to have a Legal Entity Identifer (LEI). There are various publicly available databases that keep track of entities that have reporting obligations under EMIR. Similarly, researchers like Gay, Lin, and Smith (2011) have used derivatives user databases. For the purpose of this study, derivative users were randomly selected from such database. Entities whose sole purpose is the active trading and management of investments (such as hedge funds and Collective Investment Schemes) were deemed to be out of scope, given the possible use of derivatives for speculative purposes. A group of entities was selected from the Maltese Registrar of Companies (RoC) database at random as control entities, in order not to bias the sample in favour of DHI users or a particular group. The resulting population comprises both public, private companies, listed and non-listed, together with fnancial12 and non-fnancial entities. The fnal identifed population comprised 756 LEI code users. After allowing for deductions for LEIs attributable to unidentifable entities and funds, the fnal number of 426 LEIs was used. A panel data set13 was then formed by means of a survey of publicly available data disclosed within the entities’ fnancial statements as extracted from the RoC. The time period for which data were collected comprised 2009–2014 (six years).

3.3. Model Specifcations 3.3.1. The Logit Model Explaining Maltese DHI Usage.   P (HEDGEi ,t = 1)    = α + β ITR − β CuR − β TAR + β GR ln  1 2 3 4 i ,t i ,t i ,t i ,t  P (HEDGEi ,t = 0) −β5 ICi ,t + β6 PPESi ,t + β7CAPEXSi ,t + β8RDSi ,t +β9 EVCATi ,t , j + β10CFVCATi ,t , j + β11MSCi ,t



(3.1)

+β12SIZEi ,t + β13MSELi ,t + ui ,t Eq. (3.1) is the basic logistic regression model explaining Maltese DHI usage. 12

A ‘fnancial entity’ is one such undertaking whose main operations is the provision of regulated fnancial services such as; banking, the business of insurance and the management of investment funds. 13 Comprising both longitudinal time-series and industry-sector cross-sectional data.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

The Case for Derivatives Hedging for a Small Island State    231 where: ⦁⦁ α is the estimated static regression intercept, denoting the base-case odds to

hedge given all explanatory terms adopting a value of 0.

⦁⦁ βk : k = {1,…,13} is the logistic regression coeffcient for the kth parameter. ⦁⦁ ui , t is the residual stochastic error term. ⦁⦁ All model estimators were estimated using the method of iteratively re-

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

weighted least squares.

In analysing a dichotomous response variable as the variable ‘HEDGE’ in this case, the logistic regression model (logit model) is highly appropriate, given that the logit transformation enables this transformation of the dependent variable to remove restrictions from the explanatory terms (Rodríguez, 2007). The logarithm of the odds to hedge is the dependent variable. The odds ratio is defned as the ratio of the likelihood to hedge using derivatives to the likelihood of not. The main chosen capture of the overall explanatory power of the model was the ‘Nagelkerke Pseudo-R2’. This test allows for the explanatory power of a model with a given combination of variables over the baseline model, scaled down to take values from 0 (no explanatory power) to 1 (total explanatory power). Given the signifcant amount of variables and respective proxies considered, an iterative building block approach was adopted to derive the parsimonious model. Upon testing for Normality,14 variables were then tested for individual discriminatory signifcance between DHI users and non-DHI users by subjecting them to the respective one-way Mann–Whitney-U tests. The outcome of this nonparametric test, gave an idea of the explanatory power of the respective variable without taking into consideration interactive effects. The parsimonious logit model was then derived iteratively by considering different combinations of proxy variables from each of the 8 proxy groups, maximising the Pseudo R2 Value and the number of signifcant variables. Further detail to this derivation is contained in Appendix 1. 3.3.2. The Proportional Odds (Ordered Logit) Model for Assessing the Impact of DHI usage and HA Designation for Maltese entities:  The POMs in use here attempt to predict the class of volatility of a particular frm from very volatile (class 4) to least volatile (class 1), depending on a number of explanatory variables. In Modelling the respective volatilities, Zhang (2008) identifes numerous variables that are known to impact these respective dependent variables, as evidenced in Eqs. (3.2) and (3.3), respectively. These were used as control variables to isolate the effect of DHI use and HA on these respective variables. This is done by testing for the signifcance in the explanatory power of the HEDGE and HA categorical variables. A detailed explanation of the POM is given in Appendix 3. Thus, the two formulations of the POM for EV and CFV, respectively, are as follows. 3.3.3. POM/Ordered Logit for Maltese EV.  The odds of the observed EV Capture EVi ,t for the ith observed entity in year ‘t’, falling into Categories j or lower relative to falling into higher categories are formulated as follows: 14

As evidenced by both the Shapiro–Wilk and Kolmogorov Smirnoff tests at the 0.01 and 0.05 levels of signifcance appended in Table A2 of Appendix 2.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

232    Konrad Farrugia et al.  p( EV ≤ X )  i ,t j   = a − b HEDGE − b HA − b GR ln  j 1 i ,t 2 i ,t 3 i ,t 1 − p( EVi ,t ≤ X j ) +b4SGRi ,t − b5 NPMi ,t − b6SIZEi ,t + b7 PPEi ,t (3.2) +b8RoAi ,t + b9CFVCATi ,t ,1 + b10CFVCATi ,t ,2 +b11CFVCATi ,t ,3 + b12CFVCATi ,t ,4 + ui ,t All model estimators were estimated using the method of iteratively reweighted least squares. Eq. (3.2) is the POM explaining Maltese entities’ EV. The respective variables are defned in Appendix 4. 3.3.4. POM/Ordered Logit for Maltese Operational CFV.  The odds of the observed Operational CFV Capture CFVi ,t for the ith observed entity in year ‘t’, falling into Categories j or lower relative to falling into higher categories are formulated as follows:  p(CFV ≤ X )  i ,t j   = a − b HEDGE − b HA − b GR + b TRV ln  j 1 i ,t 2 i ,t 3 i ,t 4 i ,t 1 − p(CFVi ,t ≤ X j )   (3.3) +b5TPVi ,t − b6SIZEi ,t + b7 RoAi ,t + b8 PPEi ,t +b9CAPEXSi ,t + b10 EVCATi ,t ,1 + b11EVCATi ,t ,2 +b12 EVCATi ,t ,3 + b13 EVCATi ,t ,4 + ui ,t Once more all model estimators were estimated using the method of iteratively re-weighted least squares. Eq. (3.3) is the POM explaining Maltese entities’ Operational CFV. The respective variables are defned in Appendix 5.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

4. Research Findings and Deliberation The empirical fndings in this section draw from the statistical hypothesis testing, parameter estimation and model ftting and from the data set of 568 entity-years from 107 different Maltese corporate entities. These are further substantiated by the insights obtained from seven semi-structured interviews with local leading treasury and risk management professionals.

4.1. Prevalence of Maltese DHI Usage From the 568 sets of fnancial statements analysed, 261 instances of DHI usage, from 43 different entities (39.25% of the 107 total entities sampled) were identifed. When compared to results from other DHI-usage surveys, such as those by Bartram et al. (2009), the Maltese DHI utilisation rate trails amongst the lower end of the distribution, as evident from Table 2. Interviewees working with non-fnancial entities labelled DHIs as ‘speculative’ and ‘complex’. Nevertheless sound risk management praxis and cultures were some of the resonating themes that stemmed from some of the interview fndings. This concurs with the local fndings (Bezzina, Grima, & Mamo, 2014), from fnancial entities.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

The Case for Derivatives Hedging for a Small Island State    233 Table 2.  Tabulation of Intra-Country DHI Usage Statistics, Except for the Maltese Statistics Derived from the Results of this Study.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Country

Percentage Most Most Popular of DHI Popular FCR IRR DHI in Users   DHI in Use Use (%) (%)

Most Popular CPR DHI in Use (%)

Maltaa

39.3

Forwards (52.91)

Interest rate swap (39.15)

Commodity swap (9.52)

Australia

65.4

Forwards (48.5)

Interest rate swap (38.9)

Commodity swap (18.2)

Belgium

46.2

Forwards (13.8)

Interest rate swap (20)

Commodity swap (3.1)

Czech Republic

30.4

Forwards (8.7)

Interest rate swap (13)

Commodity swap (0)

China

16.7

Forwards (5.6)

Interest rate swap (5.6)

Commodity future (5.6)

Finland

63.8

Forwards (40)

Interest rate swap (30.5)

Commodity futures and commodity options (2.9)

France

64.8

Forwards (31.5)

Interest rate swap (37)

Commodity futures and commodity swaps (1.2)

Germany

44.9

Forwards (21.2)

Interest rate swap (17.8)

Commodity future (1.7)

Italy

58.6

Forwards (30.3)

Interest rate swap (24.2)

Commodity swap (2.0)

Japan

80.9

Forwards (70.7)

Interest rate swap (59.1)

Commodity future (3.9)

New Zealand

95.6

Forwards (75.6)

Interest rate Commodity swap (13.3) swap (75.6%)

Spain

58.6

Forwards (17.2)

Interest rate swap (34.5)

Commodity swaps, commodity futures & commodity options (6.9)

United Kingdom

64.9

Forwards (48.5)

Interest rate swap (31.6)

Commodity forwards and commodity futures (5.1)

64

Forwards (30.2)

Interest rate swap (35.8)

Commodity futures (6.2)

United States

Source: Bartram et al. (2009, p. 38). a The percentage of hedgers using a particular DHI is calculated as the weighted average of the number of instrument identifcations per annum weighted down by the percentage of hedgers identifed in that particular year from the global number of hedging entities identifed over the six-year time frame (189).

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

234    Konrad Farrugia et al. Similarly Sturm & Sauter (2010) identify risk aversion, minimal investment in complex products and a robust regulatory environment as predominant traits within the Mediterranean region. Maltese entities seem to be averse to using derivatives to manage their risks. This was a recurrent theme in 60% of the interviews, most prominently in respect of the lack of forex hedging. This is more pronounced in light of geopolitical risks and events such as Brexit. The risks hedged by Maltese entities are the three most commonly hedged fnancial price risks: foreign currency risk (FCR) (52% of all identifed DHIs), IRR (31.8% of all identifed DHIs) and commodity price risk (CPR) (15.71% of all identifed DHIs). Although Malta has a comparably low rate of DHI usage, over half of the Maltese entities use FCRs to hedge their currency risks. The relative prevalence of local DHI utilisation is somewhat comparable to that of US non-fnancial entities as established in studies by Bodnar, Hayt, and Marston (1996) and Comiskey and Mulford (2009). Among the 261 instances of disclosure of DHI usage in fnancial statements, over 12 different types of DHIs were disclosed as presented in Fig. 1. It should be noted that there is an incremental trend in DHI usage across the six-year study period. In respect of the type of contracts used, Currency Forwards and Futures were the modal DHI identifed comprising 38% of all DHI identifcations, closely followed interest rate swaps (28%) and commodity swaps (7%). DHI choices seem to be refective of local conservatism, as seen in the fact that Maltese DHI users seem to prefer simpler derivatives, namely forwards and swaps. The popularity of the aforementioned instruments evidenced in this study at 72.9%15 of FCR DHIs and 94.87% of IRR DHIs, respectively, is consistent with both the studies of Judge (2006) for UK non-fnancial entities and that by HSBC (2012) the latter of which identifes forwards (56.4% of all FCR DHIs) and swaps (58.9% of all IRR DHIs) as the preferred instruments for hedging FCR and IRR, respectively. This conservatism seems to hinder the aptitude for more complex derivatives (e.g. swaptions) which are better risk mitigation devices and help reduce basis risks. Toggling focus on the users of DHI’s, fnancial entities constitute the vast majority of DHI users. Credit and fnancial institutions can be identifed as using the largest amount of DHIs to hedge their fnancial risks. From all 261 instruments identifed over six-years, credit/fnancial institutions’ DHI usage amounts to 64 DHIs making up (25%) of all DHI usage. This fnding gains heightened signifcance in the light of the fact that banking and credit taking Institutions, comprised approximately 13% of the entire sampling frame. Within this segment the most complex of all DHI arrangements such as interest rate swaptions and portfolio-wide Macro-hedges of IRR were observed. It was interesting to note that (re)insurance companies made up 13.8% of the sample of DHI users, despite the potential capital and economic balance sheet stability that could be attained when applying DHIs. Perhaps a low utilisation 15

This differs slightly from the 38% quoted in Fig. 1 as forwards and futures from all DHI identifcations, as the 72.89% considers only FCR hedging derivatives. The same can be said for the percentage for IRSs quoted for all IRR hedging derivatives.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

The Case for Derivatives Hedging for a Small Island State    235

DHI Identifed

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Currency forwards and futures

2009 2010 2011 2012 2013 2014 Total Percentage of Total 12

14

18

16

20

20

100

38

Currency swaps

2

2

2

2

2

2

12

5

Currency options

2

2

2

2

2

3

13

5

Cross-currency interest rate swaps

1

1

2

2

3

3

12

5

Interest rate forwards and futures

0

0

1

0

0

0

1

0

Interest rate swaps

10

10

13

11

15

15

74

28

Interest rate caps

0

0

1

0

0

0

1

0

Interest rate swaptions

0

0

1

0

0

0

1

0

Commodity swaps

2

2

2

3

4

5

18

7

Commodity options

2

2

2

1

1

1

9

3

Commodity collars

1

1

1

1

2

3

9

3

Commodity forwards

0

0

0

1

1

3

5

2

Non-classifable but disclosed

1

0

0

1

1

3

6

2

33

34

45

40

51

58

261

100

Total

Fig. 1.  Number of DHI Identifcations across the Six-Year Period of Study by DHI Typology.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

236    Konrad Farrugia et al.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Fig. 2.  Number of DHI Identifcations by DHI Typology and Economic Sector. rate is inherent due to the capital requirement borne by an insurer on derivative fnancial instruments such as the subjection to a credit risk and/or counterparty default risk charges. Only 16.7% of (re)insurance entity DHIs were used to hedge IRR, comprising solely of interest rate swaps, with the remainder of identifed DHIs comprised Forwards used to hedge forex risk. This contrasts signifcantly with the banking industry, where the converse was observed, primarily due to the greater prevalence of interest rate-sensitive assets and liabilities. The extent of usage and type of DHIs in use according to economic sectors are set out in Fig. 2 and Table 3.

4.2. The Determinants of Maltese DHI Usage In establishing the possible factors that drive Maltese DHI usage, eight classes of latent variables and their respective proxies were utilised. Table 4 tabulates the descriptive statistics of the respective variables and proxies.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

5

64

25

13

0

0

0

0

0

0

0

0

0

0

0

0

0

1

Currency options

Cross-currency interest rate swaps

Interest rate forwards and futures

Interest rate swaps

Interest rate caps

Interest rate swaptions

Commodity swaps

Commodity options

Commodity collars

Commodity forwards

Unclassifable but disclosed

Total

Percentage of DHI identifcations

Sampling pool percentage

0

0

0

0

1

1

17

1

0

6

12

0

Currency swaps

21

0

Agriculture Banking and Aquaculture

Currency forwards and futures

DHI Identifed

10

5

14

0

0

0

0

0

0

0

8

0

6

0

0

0

Finance and Investment

7

7

18

0

0

0

0

0

0

0

18

0

0

0

0

0

19

14

36

0

0

0

0

0

0

0

6

0

0

0

0

30

7

4

10

0

0

0

0

0

0

0

4

0

6

0

0

0

4

2

4

0

0

0

0

0

0

0

4

0

0

0

0

0

8

10

25

0

0

0

0

0

0

0

4

0

0

2

0

19

9

5

14

0

3

0

0

9

0

0

0

0

0

0

0

2

3

2

4

0

2

0

0

0

0

0

0

0

0

0

0

2

Hospital- Insurance Investment IT and ManuOil, Ore, ity & and Rein- Property and Elec- facturing PetroSteel Gaming surance Property tronics leum and and Development Related Metal Services Trading

Table 3.  Number of DHI Identifcations by DHI Typology and Economic Sector.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

8

8

20

0

0

0

0

0

0

0

0

0

0

0

0

20

Retail Commercial

1

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

Retail Wholesale

2

3

7

1

0

0

0

0

0

0

6

0

0

0

0

0

7

12

32

0

0

6

6

6

0

0

6

0

0

5

0

3

1

5

13

0

0

3

3

3

0

0

1

0

0

0

0

3

100

100

261

6

5

9

9

18

1

1

74

1

12

13

12

100

TelTransporta- Utilities Total ecom- tion, Postage munica- and Related tions Services

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

TAR

Covariate

2b

3

4

5

6

7

8

–

–

Degree of fnancial leverage

Extent of possible underinvestment

EV

Operational CFV

Extent of agency problem

Listing status

Size of operations

Intermediary variable

Intermediary variable

Covariate

Covariate

Covariate

Covariate

Covariate

Covariate

CFV

EV

SIZE

MSEL

MSC

CFVCAT

EVCAT

CAPEXS

Covariate Covariate

PPES

GR

Covariate Covariate

IC

Covariate

499

499

499

499

499

499

497

499

499

499

499

499

499

0

0

5

0

0

1

1

0

0

−27

0

0

0

0

1

499a 499

Minimum

N

49,566

133,615

23

1

1

5

5

90

47

269

112,076

1

1,282

209

2

Maximum

641.42

766.11

17.89

0.25

0.36

3.00

2.99

0.41

1.15

2.38

339.09

0.19

8.13

0.91

1.66

Mean

3,197.39

6,896.36

2.14

0.43

0.48

1.41

1.41

4.88

3.49

12.53

5,046.70

0.25

70.77

9.51

0.48

Std. Deviation

In all instances non-negative values are reported save for the GR (Gearing Ratio), due to allowing for negative equity values. The fnal number of variable readings translates to 499 for 568 frm-years to allow for negation of outliers and extreme values that would distort the data, based on statistics such as Cook’s Distance and Leverage. b Although originally classed as one proxy variable group these were later split into two groups due to the respective individual signifcant discriminatory powers in discerning between DHI and non-DHI users as evident from Table 5.

a

CuR

Covariate

2ab

Financial distress costs

ITR

Covariate

1

HEDGE

Rate of corporate taxation

Dependent

Proxy Variable

–

Proxy Variable Variable Role Group

DHI usage

Latent Variable

Table 4.  Descriptive Statistics Pertaining to Variables in Logit Model 3.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

238    Konrad Farrugia et al.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Tangible asset ratio

TAR

CFVCAT

MSC

MSEL

SIZE

EV

CFV

Operational CFV

Extent of agency problem

Listing status

Size of operations

Intermediary variable

Intermediary variable

CFV (%Δ in cash fow from operations)

EV (%Δ in proft before interest and tax)

Natural logarithm of total assets

Equity listing on MSE

Multi-class shares

Multinomial categorical transform of ‘CFV’ on the 5-point quantile scale

Multinomial categorical transform of ‘EV’ on the 5-point quantile scale

Capital expenditure to sales

CAPEXS

EVCAT

Property plant and equipment to sales

Gearing ratio

GR

PPES

Interest cover ratio

IC

EV

Extent of possible underinvestment

Degree of fnancial leverage

Current ratio

CuR

Financial distress costs

Income tax rate

ITR

Rate of corporate taxation

Variable Description

Proxy Variable

Latent Variable (Determinant)

0.005

0.073

0.000

0.002

0.471

0.003

0.084

0.001

0.002

0.000

0.002

0.007

0.148

0.147

One-Tail Test Asymptotic Signifcance (p-Value)

Signifcantly lower for DHI users

Insignifcant

Signifcantly lower for hedgers

Signifcantly greater for DHI users

Insignifcant

Signifcantly lower for DHI users

Insignifcant

Signifcantly lower for DHI users

Signifcantly lower for DHI users

Signifcantly greater for DHI users

Signifcantly Lower for DHI Users

Signifcantly greater for DHI users

Insignifcant

Insignifcant

Outcome at 0.05 Level of Signifcance

Table 5.  Mann–Whitney-U Test Outcomes Pertaining to Variables in Logit Model 3.1.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Signifcantly lower for DHI users

Insignifcant

Signifcantly lower for hedgers

Signifcantly greater for DHI users

Insignifcant

Signifcantly lower for DHI users

Insignifcant

Signifcantly lower for DHI users

Signifcantly lower for DHI users

Signifcantly greater for DHI users

Signifcantly Lower for DHI Users

Signifcantly greater for DHI users

Insignifcant

Insignifcant

Outcome at 0.01 Level of Signifcance

240    Konrad Farrugia et al.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

4.2.1. Signifcant Determinants at the Individual Proxy Variable Level.  To discern whether a variable has any discriminatory power in distinguishing between the two mutually exclusive categories of DHI users and non-DHI users, at a single variable level, the Mann–Whitney-U Test was employed. In view of each variable’s nonconformity to a normal distribution, this test was used to compare the respective independent means (Fagerland & Sandvick, 2009). The outcome of the tests is tabulated in Table 5. Signifcant differences among DHI users and non-users are evidenced for all gearing and underinvestment proxies. Further discriminatory potential is also noted for both captures of CFV, the TAR proxy for CFD, frm size and listing status. This, together with the fact that the vast majority of DHIs are FCR and IRR hedging devices, further validate the hypothesised rationale that DHIs can be used to safeguard the sustainability of an entity’s fnancing and help prevent, greater fnancing costs and ineffcient funding ventures. The fact that signifcant differences in CFV captures are also noted is also consistent with observing DHIs geared at managing day-to-day operating costs and cash fows such as CPR DHIs, given that a number of entities importing raw materials and commodities from overseas, protect themselves from severe price fuctuations. Finally the fact that statistically signifcant differences in both the size and listing status of an entity are observed when assessing the attributes of DHI users, leads to an interesting fnding: On average, comparably smaller entities tend to have a greater tendency to utilise DHIs, possibly due to lack of the critical mass to devise natural hedges and possibly to help allay borrowing and fnancing costs. Listed entities may also be using DHIs to protect variability around their share price, in an attempt to mitigate volatility around their reported earnings. 4.2.2. Signifcant Determinants resultant from the Logit Model.  The parsimonious logit model defning DHI usage determinants for Maltese entities is defned by Eq. (4.1) below (refer to the parsimonious model construction in Appendix 1):



 P (HEDGEi ,t = 1)    = 16.785 − 2.026TAR + 0.077GR ln  i ,t i ,t  P (HEDGEi ,t = 0) +0.259PPESi ,t + 0.211CFVCATi ,t , j (4.1) −0.915SIZEi ,t + ui ,t

Eq. (4.1) is the parsimonious ftted logit model for Maltese DHI usage determinants. The fve variables; the tangible assets ratio (TAR), a proxy for likelihood of incurring bankruptcy costs, the GR, the proxy for the degree of the entity’ leverage, the property plant and equipment-to-sales (PPES) ratio, approximating Ineffcient investment, the frm’s size captured by the natural log of total assets (SIZE) and the categorical transformation of the entity’s annual percentage change in operating cash fows (CFVCAT) all individually signifcantly contribute towards determining the probability of an entity to hedge. The individual parameter signifcance of the constituents of the parsimonious logit model in Eq. (4.1) are detailed in Table 6. Overall, 43.6% of all variations amongst DHI users and non-DHI users are explained by the model as evidenced by the Nagelkerke Pseudo R2 Statistic. The signifcance of the Model’s predictors in Table 7 over the constant-only model,

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

The Case for Derivatives Hedging for a Small Island State    241 Table 6.  Test of Overall Signifcance for the Fitted Parsimonious Logit Model (Eq. (4.1)). Omnibus Tests of Model Coeffcients for Eq. (4.1) Step 1

Chi-square

df

Sig.

Step

189.067

5

0.000

Block

189.067

5

0.000

Model

189.067

5

0.000

Pseudocoeffcients of Determination for Eq. (4.1) Step

−2 Log-likelihood

1

Cox and Snell R2

452.449

0.315

Nagelkerke R2 0.436

Table 7.  Parameter Estimates for the Parsimonious Logit (Eq. (4.1)). Hedging Determinant Logistic Regression Output for Eq. (4.1) Variable

Coeffcient Value (β)

Std. Error

Wald

df

Sig.

−2.026

0.573

12.515

1

0.000

0.132

Negative

GR

0.077

0.037

4.319

1

0.038

1.080

Positive

PPES

0.259

0.091

8.167

1

0.004

1.296

Positive

SIZE

−0.915

103.702

1

0.000

0.401

Negative

6.468

1

0.011

1.235

Positive

1.665 101.579

1

0.000

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

TAR

CFVCAT

0.211

Constant

16.785

0.09 0.083

Odds of Us- Variable Reing DHIs as lationship to Opposed to propensity for non-Usage DHI Usage

19,481,970 n/a

are evidenced by the signifcance of the model’s log-likelihood ratio Chi-Squared Statistic of 189.7 on fve degrees of freedom. This implies that collectively the identifed variables’ contribution to the model is signifcant at the 0.05 level of signifcance. The motive for the avoidance of the CFD, through taking precautionary steps by utilising DHIs seems to be consistent with the fndings of Bartram et al. (2009) and Judge (2006) which identify the explanatory signifcance of all CFD proxies. This study reports a signifcant decrease in the odds to use DHIs by 87%16 for every unit increment in the TAR proxy. For local entities, this implies that a higher TAR leads to less restrictive borrowing covenants and hence more operative and decision fexibility as fnancial positions stabilise through DHI usage. The percentage probability of hedging as opposed to not hedging is derived as: (ebi – 1) for each respective covariate. 16

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

242    Konrad Farrugia et al. The fact that for a unit increment in frm size DHI usage odds decrease by 60% further reinforces the motive for the avoidance of CFD. This directly corroborates with the theories of Nance et al. (1993) and Warner (1977). They identify that CFD are ‘less than proportional to frm size’ (Warner, 1977). Hence fuelling the need to hedge by smaller entities may lack critical mass to have natural hedges. The statistical signifcance of the GR validates the signifcance of fnancial leverage as infuencing DHI usage. Similar signifcance of leverage is clearly identifed by Géczy, Minton, and Schrand (1997), evidencing signifcant association with greater likelihood of FX derivative use to ease the fnancial burdens of servicing higher levels of debt. This risk becomes more pervasive as fnance costs are bound to increase at higher levels of gearing, with the odds of DHI use increasing by 7%17 for every unit increase in gearing. Two interviewees working in the banking industry expect that this trend is bound to increase as ‘banks are en route of moving large corporate client borrowings into EURIBOR linked, variable-rate borrowings’. The signifcance of the PPES proxy increasing the odds to hedge by 29%18 per unit increment, helps validate that: ‘hedging adds value to the extent that it helps ensure that a corporation has suffcient internal funds available to take advantage of attractive investment opportunities’ (Froot et al., 1993). The fact that the proxy variables indicate possible ineffcient investment and unsustainable fnancing costs (PPES, TAR and GR) concurs with the Smith and Stulz (1985), insofar as treating hedging part of the entity’s fnancing package. This study evidences an increment of 24%19 in the odds to hedge when CFV increases to the threshold levels of the next higher class. Three interviewees from fnancial entities cited the importance of DHI use to protect from variability in the cash fows when realising the FV of fnancial assets. Interviewees working in entities using IRSs justifed their use as a protection mechanism against cash fow variations from borrowing cost increments. FX derivatives however are shown to be rather effective at reducing CFV in studies such as those of Géczy et al. (1997) and Nathan Lael (2000) for non-fnancial entities to help ease the impact of realised exchange fuctuations on operating cash fows. This is explained by the prominent usage of forwards and futures contract at 52.91% usage amongst FCR DHI users. 4.2.2. Accounting Treatment Applied for DHIs and Designated Hedging Relationships by Maltese Entities.  The survey fndings evidence a slight preference towards the non-election to apply a HA model. For all identifed DHIs over the six years, 46.74% were designated as part of a hedging relationship with a hedged item, with this proportion progressively declining across the six-year period under study, as per Fig. 3. An affnity between the type of derivative and the accounting treatment opted for was notable. For instance only 22% of Currency Forwards and Futures were accounted for through a HA model. On the other hand, instruments such as interest

17

Derived as (e0.077–1) ≈ 8%. Derived as (e0.259–1) ≈ 29.5%. 19 Derived as (e0.211–1) ≈ 23.59%. 18

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

The Case for Derivatives Hedging for a Small Island State    243

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Fig. 3.  Number of DHI Identifcations (Non-)Designated for HA. rate swaps (79.7% Hedge Accounted (HA) for) and currency options (92.3% HA) were mostly hedge accounted for. In respect of Commodity DHIs, 88.9% of Commodity Options were hedge accounted for. Both the quantitative and qualitative elements of the study seem to evidence that HA designation is more popular with larger entities that use multiple derivatives as described by Glaum and Klocker (2011). Overall the election of a HA model to account for DHIs and hedging relationships is less used in Malta. A resonating theme stemming from the qualitative interviews in respect of availing of HA models is that auditors have been evidenced to have a signifcant bearing on this decision. Two thirds of interviewees have attested to the auditor’s infuence in the choice of the accounting treatment option. 4.2.3. The Impact on DHI Usage and HA on Local Earnings and Operational CFV Captures.  The POM is used here to distil the effects of DHI usage and HA20 on the transformed EVCAT and CFVAT volatility captures. Although other predictor variables were utilised to isolate, to the greatest extent possible, the effects of the HA and HEDGE variables, interpretation of the results from these excipient variables is out of the scope of this study. The estimated coeffcient output as estimated by the statistical package21 should be multiplied by –1 to derive the direction as per the POM.22 The signifcance of the parameters is tabulated in the ‘Tests of Model Effects’ for EVCAT and CFVCAT, detailed in Tables 8 and 9, respectively. 4.2.4. The Modelled Impact on Maltese EV.  The estimated PoM for Maltese EV (refer to Section 3.3.3 for its derivation) is detailed in Eq. (4.2) below:

20

As captured by the variables HEDGE and HA, respectively. Statistical Package Utilised: IBM SPSS v.20 22 For the Generalised Linear Models in Eqs. (4.2) and (4.3), respectively, the Cumulative Logit Function was used as the link function, to attempt to transform the variance of the stochastic error (u) term as a function of its linear predictors. 21

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

244    Konrad Farrugia et al.  P ( EV ≤ X )  i ,t j   =  a + 0.076HEDGE − 0.292 HA − 0.001GR ln  j i ,t i ,t i ,t 1 − P ( EVi ,t ≤ X j )     (4.2) −0.001SGRi ,t − 0.001NPMi ,t − 0.087SIZEi ,t +1.063PPEi ,t − 0.354RoAi ,t + 0.417CFVCATi ,t ,1 +0.307CFVCATi ,t ,2 + 0.262CFVCATi ,t ,3 +0.126CFVCATi ,t ,4 +  ui ,t

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Eq. (4.2) is ftted POM for Maltese EV for frm-years across 2009–2014. Table 8 overleaf shows the ftted coeffcient estimate for each included variable or model effect. The Wald-Chi Square Test Statistic details the individual signifcance of these coeffcient estimates. Table 9 evidences that in respect of EV; neither DHI usage nor HA have a signifcant impact on EV mitigation, as shown by their p-values at both the 5% and 1% levels of signifcance. Statistically this purely means that neither DHI usage nor the application of HA can infuence the odds of falling into lower categories of EV as opposed to higher categories. At face value the economic interpretation could be that the factual fnance cost reduction and operational cash fow stabilising effects (whose signifcant differences were observed in the results of the logit model and hypotheses tests) are not trickling down to the bottom-line distributable proft level. Alternatively, it may also be the case that this effect is being swamped by other signifcant factors such as greater overhead fuctuations and the effect of non-cash, accruals-based costs such as amortisation and depreciation. 4.2.5. The Modelled Impact on Maltese Operational CFV.  Analogous to the case of EV, the estimated PoM for Maltese Operational CFV (refer to Section 3.3.4 for its derivation) is detailed in Eq. (4.3) below:  P (CFV ≤ X )  i ,t j   =  a + 0.541HEDGE + 0.295HA + 0.001GR  ln  j i ,t i ,t i ,t 1 − P (CFVi ,t ≤ X j ) −0.000TRVi ,t − 0.001TPVi ,t − 0.124SIZEi ,t +0.149RoAi ,t + 1.118PPEi ,t + 0.077CAPEXSi ,t (4.3) +0.558 EVCATi ,t ,1 + 0.426 EVCATi ,t ,2 +0.745 EVCATi ,t ,3 + 0.318 EVCATi ,t ,4 +  ui ,t Eq. (4.3) is the ftted POM for Maltese operational CFV for frm-years across 2009–2014. The parameter estimates and the statistical signifcance of the respective component variables are set out in tables 10 and 11 below. Conversely to case with EV; being signifcant at the 0.05 level of signifcance the DHI usage indicator (HEDGE) is evidenced to have a signifcant contributory effect in explaining variations in categorical percentage changes in cash fows from operations. Such that, for an entity that utilises DHIs, for any given level of CFV, the cumulative probability of falling within that particular category or

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

–3.362 {a1} –2.331 {a2} –1.477 (a3) –0.414 {a4} 0.290 0.000 –0.076 0.000 –0.087 0.001 0.35 –1.06 –0.42 –0.31 –0.26 –0.13 0.000 0.001 –0.001

[EVCAT=1] (a1)

[EVCAT=2] (a2) [EVCAT=3] (a3) [EVCAT=4] (a4)

Estimated Coeffcient (β)

0.895 0.891 0.888 0.283 – 0.222 – 0.049 0.006 0.255 0.349 0.27 0.269 0.264 0.263 – 0 0.006

0.901

Std. Error

–4.086 –3.223 –2.155 –0.262 – –0.511 – –0.182 –0.012 –0.147 –1.746 –0.947 –0.833 –0.779 –0.641 – 0.000 –0.01

–5.128

Lower –0.576 0.268 1.326 0.845 – 0.359 – 0.009 0.01 0.854 –0.38 0.113 0.22 0.256 0.389 – 0.001 0.012

–1.596

Upper

95% Wald Confdence Interval

6.778 2.751 0.218 1.067 – 0.118 – 3.163 0.021 1.918 9.302 2.382 1.305 0.983 0.23 – 3.916 0.035

13.919

Wald Chi-Square

1 1 1 1 – 1 –. 1 1 1 1 1 1 1 1 – 1 1

1

df

Hypothesis Test

0.009 0.097 0.64 0.302 – 0.731 – 0.075 0.884 0.166 0.002 0.123 0.253 0.322 0.632 – 0.048 0.851

0.000

Sig.

Note: One is to observe the change in signs of the estimator values from Table 7 to Eq. (4.2), reason being Eq. (4.2) presents as subject the odds of falling into lower classes of volatility whereas Table 7 presents the estimator values as related to the odds of observing values in higher volatility classes. ** denotes signifcance at the 0.5% Level of signifcance; * denotes signifcance at the 5% Level of signifcance.

[HA=1] [HA=2] [HEDGE=1] (H = 1) [HEDGE=2] (H = 2) SIZE GR ROA PPE** [CFVCAT=1] [CFVCAT=2] [CFVCAT=3] [CFVCAT=4] [CFVCAT=5] SGR* NPM

Dependent Variable (EVCAT) Threshold (aj)

Parameter

EV POM (3.2) Parameter Estimates

Table 8.  Parameter Estimates for the EV POM (Eq. (4.2)).

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

The Case for Derivatives Hedging for a Small Island State    245

246    Konrad Farrugia et al. Table 9.  EV POM (3.2) Tests of Model Effects. Source {Model Effect}

Type Ia

Type III

Wald Chi-Square

df

Sig.

Wald Chi-Square

df

Sig. (p-value)

HA

1.132

1

0.287

1.067

1

0.302

HEDGE

1.248

1

0.264

0.118

1

0.731

SIZE

1.695

1

0.193

3.163

1

0.075

GR

0.005

1

0.942

0.021

1

0.884

ROA

1.575

1

0.209

1.918

1

0.166

PPE

14.573

1

0.000

9.302

1

0.002

CFVCAT

5.054

4

0.282

2.893

4

0.576

SGR

3.916

1

0.048

3.916

1

0.048

NPM

0.035

1

0.851

0.035

1

0.851

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Notes: Dependent Variable: EVCAT; Model: (Threshold), HA, HEDGE (H), SIZE, GR, ROA, PPE, CFVCAT, SGR, NPM. a When analysing variances in the model predicted output; utilising a Type I sum of squares implies each factor is considered in the model in the way it was ordered incrementally, whereas under Type III the least squares estimates are replaced by a weighted sum of square means. The results however do not differ signifcantly.

lower volatility classes is 1.71823 times the cumulative probability of an entity that does not utilise DHIs; thus effectively shifting probability mass to lower volatility classes. The negative magnitude of the estimated coeffcient (–0.54) perfectly captures the negative relationship between DHI usage and CFV. This empirically evidences the CFV reduction effectiveness of DHI usage, amongst the sampling frame of Maltese entities. Gauged by the relative magnitude of the p-value of the hedging variable, this is comparably signifcant, when compared to the p-values of other signifcant variables such as trade payables volatility (TPV). 4.2.6. Comparison of Results with Previous Studies and Further Deliberation.  When comparing the results of the POMs for EV and CFV with previous empirical studies; this evidence is partly consistent with the fndings of Dong and Woo Gon (2015) who evidence both EV and CFV differences amongst DHI users in the Hospitality industry. Zhang’s (2008) linear model also notes no signifcant difference in EV post the implementation of a derivatives hedging programme. The former also identifes that ‘cash fow volatility refects “real actions” in entities’ risk-management activities’, whereas EV also refects accounting infuences, such as accounting policy choices or the HA treatment in our case. This result also empirically proves the theoretical groundings of Smith and Stulz (1985) in respect of the CFV mitigation effect of DHI usage.

23 –b1 = 0.54128

e

≈ 1.718.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

–2.331 {a2} –1.477 (a3) –0.415 {a4}

[CFVCAT=2] (a2)

[CFVCAT=3] (a3)

[CFVCAT=4] (a4) –0.29 0.000 –0.541 0.000 0.124 0.001 –0.149 –1.118 –0.558 –0.426 –0.745 –0.318 0.000 0.077 0.000 0.001

–3.362 {a1}

[CFVCAT=1] (a1)

(β)

Estimated Coeffcient

0.292 – 0.226 – 0.045 0.005 0.18 0.353 0.268 0.261 0.259 0.261 – 0.086 0 0

0.888

0.891

0.895

0.901

–0.868 – –0.984 – 0.035 –0.011 –0.502 –1.811 –1.083 –0.938 –1.253 –0.83 – –0.091 –0.000 0.000

–2.155

–3.223

–4.086

–5.128

0.279 – –0.099 – 0.213 0.009 0.204 –0.426 –0.032 0.087 –0.238 0.193 – 0.246 0 0.001

1.326

0.268

–0.576

–1.596

Std. 95% Wald Confdence Interval Error Lower Upper

1.015 – 5.745 – 7.511 0.037 0.686 10.03 4.33 2.652 8.293 1.488 – 0.805 1.703 4.57

0.218

2.751

6.778

13.919

Wald Chi-Square

1 – 1 – 1 1 1 1 1 1 1 1 – 1 1 1

1

1

1

1

df

Hypothesis Test Sig.

0.314 – 0.017 – 0.006 0.847 0.407 0.002 0.037 0.103 0.004 0.223 – 0.369 0.192 0.033

0.64

0.097

0.009

0

Notes: Dependent Variable: CFVCAT; Model: (Threshold), EVCAT, HEDGE, HA, TRV, TPV, ROA, PPE, SIZE, GR, CAPEXS; No Type III estimators have been calculated here as it was not deemed sensible to do so.

[HA=1] [HA=2] [HEDGE=1] (H = 1) [HEDGE=2] (H = 2) SIZE GR ROA PPE [EVCAT=1] [EVCAT=2] [EVCAT=3] [EVCAT=4] [EVCAT=5] CAPEXS TRV TPV

Threshold (aj)

Dependent Variable (CFVCAT)

Parameter

CFV POM (3.3) Parameter Estimates

Table 10.  Parameter Estimates for the Operational CFV POM (Eq. (4.3)).

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

248    Konrad Farrugia et al. Table 11.  Tests of Model Effects for the Operating CFV POM (Eq. (4.3)). Operational CFV POM (3.3) Tests of Model Effects Source {Model Effect}

Wald Chi-Square

df

Sig.

EVCAT

9.123

4

0.058

HEDGE (H)*

5.745

1

0.017

HA

1.015

1

0.314

TRV

1.703

1

0.192

TPV*

4.57

1

0.033

ROA

0.686

1

0.407

1

0.002

PPE**

10.03

SIZE**

7.511

1

0.006

GR

0.037

1

0.847

CAPEXS

0.805

1

0.369

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Notes: Dependent Variable: CFVCAT; Model: (Threshold), EVCAT, HEDGE, HA, TRV, TPV, ROA, PPE, SIZE, GR, CAPEXS; No Type III estimators have been calculated here as it was not deemed sensible to do so. ** denotes signifcance at the 0.5% Level of signifcance; * denotes signifcance at the 5% Level of signifcance.

In partial consistency with Zhang’s (2008) study and as identifed by Singh (2004) no signifcant EV reduction effect is observed for DHI users availing of Cash Flow Hedging models. Nevertheless the aforementioned fnds a degree of signifcance in CFV post opting for a HA model. A recurrent theme stemming from the semi-structured interviews was the underutilisation of DHIs, brought about by lack of proper education. All seven interviewees attested to hedging as being a key exponent in providing ‘Stability and Protection’. Three fnancial entity interviewees attested to the volatility mitigation benefts of DHI activity to mitigate ‘unsustainable risk’ that may impact the entity’s solvency and proftability. This holds especially true for IRR and FCR in the short run, given that such entities are bound by strict risk control and capital adequacy regulations such as CRD IV and Solvency II. A signifcant need exists for increased awareness and education in relation to both DHI and HA use. Better awareness and education from banks and auditors could help Maltese entities appreciate the use of DHI and application of HA. They can assist in identifying the entities’ risk management needs and adequately matching the use and corresponding accounting treatment of appropriate hedging devices. On the other hand, interviewees from non-fnancial entities stated the benefts of hedging on EV and CFV are more related to operational stability, such as cost and sales price variance mitigation. Again this is crucial in the case of entities whereby commodities may be highly price elastic or benchmarked priced. Zhang (2008) identifes that ‘cash fow volatility refects “real actions” in entities’

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

The Case for Derivatives Hedging for a Small Island State    249 risk-management activities’, whereas EV also refects accounting infuences, such as accounting policy choices or the HA treatment. The above underutilisation of DHIs and HA, as elucidated upon by both the empirical and interview fndings portrays the following situation: It seems to be the case that for Maltese entities, the economic volatility reduction benefts from DHI use are not being effectively translated into the fnancial statements. This is most plausible given the evidenced signifcance of the volatility-reducing effects of DHI usage in the case of CFV, but not in the case of EV. A cause may be the non-designation of effective hedges for the HA treatment, thereby giving way to accounting mismatches in recognising changes in hedged item values and DHI(s) in separate accounting periods. Although DHIs can stabilise factual CFV, this stability might not be enjoyed at a distributable proft level without the use of HA. This claim is supported by the observed decrease in designation of DHIs for the HA treatment in the more recent years.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

5. Summary and Concluding Remarks This study aimed to give an insight into both the determinants of DHI usage and also the impact therefrom on the earnings and operational CFV captures for a sample of Maltese entities throughout the period 2009–2014. The impact on the respective volatility captures in accounting for DHI usage by opting for a HA model was also assessed. Being a small island state, and effectively a microstate (Armstrong & Read, 1995) the Maltese perspective, helps provide an expectation for assessing DHI use effectiveness and praxis for states with similar geographical and demographic characteristics. This study also helps contribute to extant literature pertaining to DHI surveys and effectiveness assessment in adopting the POM to model DHI effectiveness on volatility reduction. It applies the POM to investigate the effects of DHI usage and HA on the transformed EVCAT and CFVAT volatility measures. The identifed determinants for local DHI usage resonate with an overall conservative risk management culture. The signifcance of the proxies for fnancial distress costs and ineffcient investment, symptomatic of higher gearing levels, all corroborate with the interviewees’ motives to attain stability. Cash fow volatility (CFVCAT) signifcance substantiates the desire to have suffcient funds to safeguard frm liquidity. These fndings strengthen the argument that DHIs are essential when it comes to protecting against variability around input/output prices and market-related infuences such as exchange rate or interest rate variability. This study sheds light on the underutilisation of DHIs and brings to the forefront an apparent paradox: the aversion to use DHIs due to risks inherent when utilising complex fnancial products. This explains the use of simpler derivatives such as forwards and swaps. Although these are plain vanilla derivatives, in times of signifcant volatility, they tend to be less effective to the more complex DHIs such as options or swaptions. The above-mentioned signifcant CFV-reductive effect clearly evidences the benefts, even for smaller-scale entities as seen locally can be reaped from DHI usage. Although DHIs are less popular with Maltese entities compared to other

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

250    Konrad Farrugia et al. countries, yet different types of DHIs are used across several industries. This is a clear indicator of the use of DHIs as a risk management tool. Interview fndings showed that DHI use and HA are perceived as effective in curbing EV. The quantitative fndings however do not provide empirical evidence to substantiate these perceptions. However, the CFV POM results show that DHI use is used to decrease operational CFV. Nevertheless, the underutilisation of the HA treatment and the short-term volatility inducing effect of the resultant accounting mismatches24 may be identifed as the leading cause of this ineffectiveness of EV reduction. Perhaps better education of the benefts of IFRS 9 by infuential parties such as auditors, regulators and fnancial stability boards could help encourage the proliferation of the HA treatment. Undoubtedly, the need for further education about effective DHI usage as risk management devices particularly their respective potential benefts and risks, especially in a small island state like Malta is clear. Inherent to the study are a number of limitations. Primarily classing all DHI users into 1 category irrespective of the strategy, the number and type of DHI used is one limitation inherent to the mathematical modelling. Furthermore, the study of volatility variables across a relatively short period of time might also impair generalisability of this study to larger time periods. Future research can be aimed at further improvements on the statistical modelling provided, availing of longer time series and perhaps applying more mathematically robust volatility models such as ARCH and GARCH25 models.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

References Armstrong, H. W., & Read, R. (1995). Western European micro-states and EU autonomous regions: The advantages of size and sovereignty. World Development, 23(7), 1229–1245. Baldacchino, J. P., Gauci, A., & Grima, S. (2019). Family infuence in Maltese listed companies: The implications on corporate governance. International Journal of Economics & Business Administration, 7(1), 85–112. Bartram, S., Brown, W., & Fehler, F. (2009). International evidence on fnancial derivatives usage. Financial Management, 38, 185–206. Bodnar, G., Hayt, G., & Marston, R. (1996). The Wharton survey of derivatives usage by U.S. non-fnancial frms. Financial Management, Winter, 113–133. Berkman, H., Bradbury, M. E., & Magan, S. (1997). An international comparison of derivatives use. Financial Management, Winter, 69–73. Bessembinder, H. (1991). Forward contracts and frm value: Investment incentive and contracting effects. Journal of Financial and Quantitative Analysis, 26, 519–532. doi:10.2307/2331409 Bezzina, F., Grima, S., & Mamo, J. (2014). Risk management practices adopted by fnancial frms in Malta. Managerial Finance, 40(6), 587–612. Brennan, M. J. (1971). Capital market equilibrium with divergent borrowing and lending rates. Journal of Financial and Quantitative Analysis, 6, 1197–1205. doi:10.2307/2329856

24

Inherent to recognising movements in DHI and underlying hedged item in two separate accounting periods. 25 (Generalised) Autoregressive Conditional Heteroscedasticity.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

The Case for Derivatives Hedging for a Small Island State    251 Bruin, J. (2011, February). Newtest: Command to compute new test @ONLINE. Retrieved from http://www.ats.ucla.edu/stat/stata/ado/analysis/. Accessed on January 21, 2016. Comiskey, E. E., & Mulford, C. W. (2009). The non-designation of derivatives as hedges for accounting purposes. Journal of Applied Research in Accounting and Finance (JARAF), 3(2), 3. SSRN. Retrieved from http://ssrn.com/abstract=1346112 Dolde, W. (1995, June). Hedging, leverage, and primitive risk. Journal of Financial Engineering, 4(2), 187–216. Retrieved from http://ssrn.com/abstract=6660 Dong, J. K., & Woo Gon, K. (2015). The relationship between the use of hospitality frms’ fnancial derivatives and cash fow/earnings volatility. Tourism Economics 2008, 14(3), 469–482. European Commission. (2012). Evaluation of the SME defnition: Final report. Kent: Centre for Strategy and Evaluation Services. European Securities and Markets Authority (ESMA). (2015). Post-trading. Retrieved from https://www.esma.europa.eu/regulation/post-trading. Accessed on December 29, 2015. Fagerland, M. W., & Sandvick, L. (2009). The Wilcoxon–Mann–Whitney test under scrutiny. Statistics in Medicine, 28, 1487–1497. doi:10.1002/sim.3561 FASB. (1998). SFAS 133: Accounting for derivatives and hedging activities. Statement 133 Implementation Issue No. G7 (Accounting Standard edn.). Norwalk, CT: FASB. Froot, K. A., Scharfstein, D. S., & Stein, J. C. (1993). Risk management: Coordinating corporate investment and fnancing policies. The Journal of Finance, 48(5), 1629–1658. Gay, J. D., Lin, C. M., & Smith, S. D. (2011). Corporate derivatives use and the cost of equity. Journal of Banking and Finance, 35, 1491–1506. Géczy, C., Minton, B. A., & Schrand, C. (1997). Why frms use currency derivatives. Journal of Finance, 52(4), 1323–1354. Glaum, M., & Klocker, A. (2011). Hedge accounting and its infuence on fnancial hedging: When the tail wags the dog. Accounting and Business Research, 5, 459–489. Greenhalgh, H. (2012). SMEs get smart on currency hedging. Financial Times, February 17. Retrieved from http://www.ft.com/cms/s/0/75724460-58b0-11e1-9f28-00144feabdc0. html#axzz46CrQHeja. Accessed on September 19, 2019. Gujarati, D. N. (2009). Basic econometrics (5th ed.). Singapore: McGraw Hill. HSBC. (2012). Corporate hedge accounting. Financing Solutions Group Report, October 2015. Hull, J. C. (2015). Options, futures and other derivatives (9th ed., p. 1). Upper Saddle River, NJ: Pearson Education Inc. Jensen, M. C., & Meckling, W. H. (1976). Theory of the frm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics (JFE), 3(4). Retrieved from http://ssrn.com/abstract=94043; http://dx.doi.org/10.2139/ssrn.94043 Judge, A. (2006). Why and how UK frms hedge. European Financial Management, 12, 407–441. doi:10.1111/j.1354-7798.2006.00326.x McCullagh, P. (1980). Regression models for ordinal data. Journal of the Royal Statistical Society. Series B (Methodological), 42(2), 109–142. Retrieved from http://www. jstor.org.ejournals.um.edu.mt/stable/2984952 Modigliani, F., & Miller, H. M. (1958). The cost of capital, corporation fnance and the theory of investment. The American Economic Review, 48, 261–296. Myers Stewart, C. (1977). Determinants of corporate borrowing. Journal of Financial Economics, 5(2), 147–175. ISSN 0304-405X. http://dx.doi.org/10.1016/0304405X(77)90015-0. Retrieved from http://www.sciencedirect.com/science/article/ pii/0304405X77900150 Nance, D. R., Smith, C. W., & Smithson, C. W. (1993). On the determinants of corporate borrowing. Journal of Financial Economics, 5, 147–175. Nathan Lael, J. (2000). The choice of hedging techniques and the characteristics of UK industrial frms. Journal of Multinational Financial Management, 10(2), 161–184. Pennings, J. M. E., & Garcia, P. (2004). Hedging behaviour in small and medium-sized enterprises: The role of unobserved heterogeneity. Journal of Banking & Finance, 28, 951–978.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

252    Konrad Farrugia et al. Pirchegger, B. (2006). Hedge accounting incentives for cash fow hedges of forecasted transactions. European Accounting Review, 15(1), 115–135. Pociovalisteanu, M. D., Thalassinos, I. E., Tirca, A., & Filho, L. W. (2010). Trends and challenges in the energy sector of Romania in the post-accession to the European Union. International Journal of Environmental Technology and Management, 12(1), 3–15. doi:10.1504/IJETM.2010.029957 Redington, F. M. (1952). Review of the principles of life offce valuations. Journal of the Institute of Actuaries, 78, 286–340. Rodríguez, G. (2007). Lecture notes on generalized linear models. Retrieved from http:// data.princeton.edu/wws509/notes/. Accessed on January 28, 2016. Rupeika-Apoga, R., Zaidi, H. S., Thalassinos, E. Y., & Thalassinos, I. E. (2018). Bank stability: The case of Nordic and non-Nordic banks in Latvia. International Journal of Economics and Business Administration, 6(2), 39–55. Saint-Leger, R. (2015). Why would a company hedge against the fuctuation of its cash fow? Retrieved from http://smallbusiness.chron.com/would-company-hedgeagainst-fuctuation-its-cash-fow-18321.html. Accessed on October 10, 2015. Singh, A. (2004). The effects of SFAS 133 on the corporate use of derivatives, volatility, and earnings management. Pennsylvania State University dissertation. Smith, C. W., & Stulz, R. M. (1985). The determinants of frms’ hedging policies. Journal of Financial and Quantitative Analysis, 20(4), 391–405. Sturm, M., & Sauter, N. (2010). The impact of the global fnancial turmoil and recession on Mediterranean countries’ economies. Occasional Paper Series No 118. ECB, Frankfurt am Main, p. 36. Thalassinos, I. E., & Pociovalisteanu, M. D. (2007). A time series model for the Romanian stock market. European Research Studies Journal, 10(3–4), 57–72. Thalassinos, I. E., Stamatopoulos, D. T., & Thalassinos, E. P. (2015). The European sovereign debt crisis and the role of credit swaps. In W. T. Ziemba & A. G. Malliaris (Eds.), in memory of Late Milton Miller (Nobel 1990), The WSPC handbook of futures markets. World Scientifc Handbook in Financial Economic Series (Vol. 5, pp. 605–639). Singapore: World Scientifc. doi:10.1142/9789814566926_0020 Thalassinos, I. E., & Thalassinos, E. Y. (2018). Financial crises and e-commerce: How are they related? Retrieved from https://ssrn.com/abstract=3330169 The Financial Services Authority (FSA). (2013). Interest rate hedging products pilot fndings (p. 10). London: FSA. (Now assimilated into the FCA (fnancial conduct authority) and PRA (prudential regulatory authority) at the time of writing.) Thind, S. (2004). Hedging for SMEs. Risk Magazine, April 1. Retrieved from https://www. risk.net/derivatives/structured-products/1506322/hedging-smes Ugurlu, E., Thalassinos, E., & Muratoglu, Y. (2014). Modeling volatility in the stock markets using GARCH models: European emerging economies and Turkey. International Journal of Economics and Business Administration, 2(3), 72–87. Van Dinh, D., & Gong, G. (2013, October). How are derivative accounting applied for hedging activities?. International Journal of Academic Research in Accounting, Finance and Management Sciences, 3(4), 72–90. Warner, J. (1977). Bankruptcy costs: Some evidence. Journal of Finance, 32, 337–348. Zhang, H. (2008). Effect of derivative accounting rules on corporate risk-management behavior. Columbus, OH: Fisher College of Business, Ohio State University.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

The Case for Derivatives Hedging for a Small Island State    253

Appendix 1: Parsimonious Model Selection Overall eight model combinations were identifed choosing 1 proxy variable out of n proxies for the latent variables identifed namely: The CFD (two proxies given: CUR TAR), the indicative extent of ineffcient investment (two proxies given: PPES, CAPEXS), and the extent of debt fnancing/leverage (two proxies given: GR and IC). Therefore, the number of possible logit combinations N was derived  2  2  2     = 8. The remainder fve proxies for their respective latent as: N =     1  1  1  variables only had one proxy each. Table A1 shows that model T3 was chosen as the parsimonious model. It has the second higher Pseudo R2 value (explaining 43.9% of all variations between hedgers and non-hedgers, attributing the rest to randomised (stochastic) variations not ftting within the model) and also having the largest number of individually signifcant predictors. The next best model forgone model T6 had an insignifcantly higher Pseudo R2 value of 44.1% a 0.2%differential but had four individually signifcant variables. Table A1.  Logistic Regression Results for the Identifed Eight Model Combinations.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Model Enumeration

Proxy for Proxy for Costs of Degree of Bankruptcy Leverage

Proxy for Ineffcient Investment

Nagelkerke Pseudo R2 Score (%)

Number of Signifcant Variables (N sig. var.)

T1

CuR

GR

PPES

41.70

3

T2

TAR

IC

CAPEXS

41.70

4

T3

TAR

GR

PPES

43.90

5

T4

CuR

IC

CAPEXS

41.42

2

T5

TAR

GR

CAPEXS

41.6

2

T6

TAR

IC

PPES

44.10

4

T7

CuR

IC

PPES

41.9

2

T8

CUR

GR

CAPEXS

41.1

2

Note: T3 is the chosen model.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

254    Konrad Farrugia et al.

Appendix 2: Non-Normality of Data Collected for the Variables Under Study Table A2.  Evidenced Non-Normality of Data Through the Conduct of the Kolmogorov–Smirnov and Shapiro–Wilk Tests of Normality. Tests of Normality

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Proxy Variable

Kolmogorov-Smirnov Statistic

df

Sig.

ITR

0.462

499

CuR

0.454

TAR

0.227

Shapiro-Wilk

Statistic

df

Sig.

0

0.046

499

0

499

0

0.066

499

0

499

0

0.788

499

0

GR

0.412

499

0

0.134

499

0

IC

0.486

499

0

0.034

499

0

PPES

0.371

499

0

0.311

499

0

CAPEXS

0.467

499

0

0.049

499

0

RDS

0.524

499

0

0.095

499

0

EV

0.456

499

0

0.079

499

0

CFV

0.421

499

0

0.17

499

0

SGR

0.427

499

0

0.145

499

0

TRV

0.454

499

0

0.073

499

0

TPV

0.478

499

0

0.029

499

0

NPM

0.454

499

0

0.084

499

0

ROA

0.462

499

0

0.04

499

0

PPE

0.227

499

0

0.788

499

0

SIZE

0.094

499

0

0.905

499

0

Note: The respective null hypotheses states that the variables conform to a normal distribution are rejected at all levels of signifcance.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

The Case for Derivatives Hedging for a Small Island State    255

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Appendix 3: Explanatory Note on the Use of the POMs for EV and CFV The pivotal problem underlying the assessment of the impact of DHI usage and HA adoption on variability in the selected observed accounting measures EV and CFV, is similar in nature to that of most volatility modelling. Volatility in time series is likely to be non-constant across a given observational time period and exhibits clustering behaviour. Although the consideration of panel data partially counteracts this problem, the ordinal categorical transformations given in Section 2.5’s hypotheses (H4a. and H4b) were deemed necessary to allay any possible complications from volatility modelling, such as heteroscedastic or chaotic behaviour. Thus the problem is essentially decomposed to identifying which factors signifcantly infuence the relative ordered magnitude (stochastic ordering) but not necessarily the strict magnitude of the volatility capture. Thus by means of the cumulative logit link function transformation, the EVCAT and CFVCAT variable measures can be effectively treated as ordinal categorical variables measured on a continuous scale, circumventing the need to assign arbitrary categories (McCullagh, 1980). Given the above-mentioned achieved ordinality, simple and easily explicable inferences can therefore be made most notably in the case where there exists a latent variable, such as the earnings and cash fow volatilities in this case (McCullagh, 1980). Thus, the model aims at predicting the cumulative probability that the transformed categorical dependent variable, for a particular observation falls into a class or classes that are of a lower ranking to it, given the respective combination of model effect values. The core underlying assumption here is that the relationship between each set of odds (of the volatility being within the defned range) and predictive variables estimated for every defned category of the response variable (the volatility) is the same for each category (Bruin, 2011). Simply put, the multinomial ordinal logit is a multi-class adaptation of the dichotomous variable logit model. The mechanics of the POM, enable the expression of the odds of a case being classifed into one class as opposed to all other possible classifcations. Under the core assumption of multinomial sampling across cases, the POM attempts to quantify the effect of an array on covariates, in this case DHI usage and HA model adoption, on the odds of falling into a particular volatility categories. The volatility categories are derived in the manner set out in Section 2.5. This is done by expressing n–1 different logistic equations for n categories, including the alias category. The effects or coeffcients of the estimated from the data, in our case using the method of iteratively weighted least squares.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

256    Konrad Farrugia et al.

Appendix 4: Detailed Variable Descriptions to EQ. (3.2) Recall, Eq. (3.2): The POM explaining Maltese entities’ EV:  P ( EV ≤ X )  i ,t j   = a − b HEDGE − b HA − b GR + b SGR ln  j i ,t i ,t i ,t 1 2 3 4 i ,t 1 − P ( EVi ,t ≤ X j ) −b5 NPMi ,t − b6SIZEi ,t + b7 PPEi ,t + b8RoAi ,t +b9CFVCATi ,t ,1 + b10CFVCATi ,t ,2 + b11CFVCATi ,t ,3

(3.2)

+b12CFVCATi ,t ,4 + ui ,t where: ⦁⦁ j = {1,2,3,4} denote the respective categories for the frst four equal categories

⦁⦁ ⦁⦁

⦁⦁

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

⦁⦁ ⦁⦁

of 80% of the cumulative probability mass observed in the empirical cumulative density function of EVi,t.26 Xj is the respective observed jth quantile from the empirical cumulative density function of Vi,t. αj is the static constant term as estimated for j categories by the statistical package. In the case of this study, denoting the basic odds of the observed EV capture falling within the category ‘j’ or lower, when all other explanatory terms have a value of 0 and taking a value of 0 when j is the alias (5th ) category. bk  : k = {1,…,12} is the ordinal logit regression coeffcient for the kth linear parameter. ui,t is the residual stochastic error term. To control for the effect of changes in operating cash fows, explicitly impacting the EV in a given year, the CFV transformed variable CFVCAT was also included as a model effect, whereby observation ‘n’ for entity ‘’ within year ‘t’ can only have one value from the four categories of CFVCAT or the alias category value of 0.

In the above model, the following control variables27 were identifed and incorporated as identifed by Zhang (2008), the: ⦁⦁ (SGRi,t): sales growth rate. Defned as: the percentage change in sales for entity

‘i’ during year ‘t’ from year t–1.

⦁⦁ (NPMi,t): net proft margin. Defned as: the quotient of net proft before tax to

the sales for entity ‘i’ during year ‘t’.

⦁⦁ (RoAi,t): return on assets. Defned as: the quotient of the annual net proft to

the book value of total assets for entity ‘i’ during year ‘t’.

⦁⦁ (PPEi,t): property plant and equipment ratio. Defned as: the quotient of a frm’s

book value of property plant and equipment to the total assets for entity ‘i’ during year ‘t’.

26

For instance when j = 4 P(EVi,t ≤ X4) = 80% and when j = 3 4 P(EVi,t ≤ X3) = 60% and so on. 27 All other variables such as HEDGE and GR, are as previously defned in Section 2.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

The Case for Derivatives Hedging for a Small Island State    257

Appendix 5: Detailed Variable Descriptions to Eq. (3.3) Recall, Eq. (3.3): The POM explaining Maltese entities’ operational CFV:  P (CFV ≤ X )  i ,t j   =  a − b HEDGE − b HA − b GR + b TRV  ln  j i ,t i ,t i ,t 1 2 3 4 i ,t 1 − P (CFVi ,t ≤ X j ) +b5TPVi ,t − b6SIZEi ,t + b7 RoAi ,t + b8 PPEi ,t +b9CAPEXSi ,t + b10 EVCATi ,t ,1 + b11EVCATi ,t ,2

(3.3)

+b12 EVCATi ,t ,3 + b13 EVCATi ,t ,4 + ui ,t where similarly to Eq. (3.2) in Appendix 4: ⦁⦁ j = {1,2,3,4} denote the respective categories for the frst four equal categories

⦁⦁ ⦁⦁

⦁⦁ ⦁⦁ ⦁⦁

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

⦁⦁

of 80% of the cumulative probability mass observed in the empirical cumulative density function of CFVi,t. Xj is the respective observed jth quantile from the empirical cumulative density function of CFVi,t. αj is the static constant term as estimated for j categories by the statistical package. In the case of this study, denoting the basic odds of the observed operational CFV capture falling within the category ‘j’ or lower, when all other explanatory terms have a value of 0 and taking a value of 0 when j is the alias (5th) category. bk  : k = {1,…,13} is the ordinal logit regression coeffcient for the kth linear parameter. ui,t is the residual stochastic error term. Identically to the EV ordered logit model; to control for the effect of volatility in earnings infuencing the observed CFV, the transformed variable EVCAT was also included as a model effect, with an observation limited to 1 mutually exclusive value of EVCAT or assuming the alias category with a value of 0. The new control variables introduced in this POM, to control for other noted effects known to impact variability in operating cash fows, are the following: – (TRVi,t): trade receivables variability. Defned as: The percentage change in the ‘trade and other receivables’ balance for entity ‘i’ during year ‘t’ from year t–1. – (TRVi,t): trade payables variability. Defned as: The percentage change in the ‘trade and other payables’ balance for entity ‘i’ during year ‘t’ from year t–1.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

This page intentionally left blank

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Chapter 16

The Effect of Micro Entrepreneur Arrangements to City Culture in Terms of Urban Esthetics and Street Peace in the Urbanization Process Ahmet Fidan

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Abstract The urbanization process that develops in parallel with the increase in population, get volume in vertical level on the ground today just like the underground expansion of urban spaces in antique ages, in parallel with the intensifcation of spatial expansion, leading to new problems and research questions in urban spaces. Because the increase in the number of people per square meter as a result of vertical concentration on the ground makes the streets or the land we step on become a more rentable market. While this market has been flled with classical artisan businesses so far, street economy actors serve the population (consumer) where artisans are not suffcient for meeting the demand in highly populated streets. This situation confronted law enforcement and street sellers in cities for decades or may be centuries, and urban peace and harmony often deteriorated. In the integrated urban areas, in addition to a series of urban problems, the registration of the informal economy and the adaptation of the street economy actors to the urban identity and esthetics have become the problems that await priority solutions. Street economy is an aesthetic and ergonomic fact of living cities, in accordance with this microeconomic reality, sustainable legal regulations are essential. Such that, these legal regulations should be established on a solid basis not only in certain countries but also in all countries in the world. Keywords: Street vendors regulations; urban life; local inspection; local management; governance; micro entrepreneurship; urban aesthetics; street economy Uncertainty and Challenges in Contemporary Economic Behaviour, 259–271 Copyright © 2020 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-80043-095-220201017

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

260    Ahmet Fidan

1. Introduction In general; states, particularly central and local-scale city administrators, end up rejecting the street economy and its actors, as in the past century when capitalism objected against the unions but then internalized them as it failed to cope with, at frst and resisted them with prohibiting/banning practices and painful law enforcements. In urban life, however, this has neither been prevented nor has a sustainable climate of peace been achieved in the streets. For this reason, street economy bears an extremely important function in urban life in terms of both registering the economy and achieving the human dignity and contemporary civilization level as well as urban esthetics in order to maintain the economy and the democratic rights and undertakings of individuals in wellbeing and peace.

2. The Indispensability of a Street Economy in the Context of the Urbanization Process and a Chronic Problematic City In the context of the emergence of a climate of peace in the streets, anticipating the beneft of registering the unregistering economy and making it an indispensable part of urban life was just like the continuous movement and organization of the working class during and after the industrial period. Ensuring street peace will be discussed within the context of the assumptions we will make in a few sub-headings to ensure that urban life is sustainable in peace and tranquility. These are: ⦁⦁ The indispensability of street economy in the context of urbanization process

and the chronic problematic of the city.

⦁⦁ Street economy in sustaining urban life esthetics and ergonomics. ⦁⦁ Regulatory requirement regarding entrepreneurship activities of micro entre-

preneur actors in urban areas.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

⦁⦁ Generic research on micro entrepreneurs in metropolitans.

The urbanization process that develops in parallel with the increase in population gets volume in vertical level on the ground today just like the underground expansion of urban spaces in antique ages, in parallel with the intensifcation of spatial expansion, leading to new problems and research questions in urban spaces. Because the increase in the number of people per square meter as a result of vertical concentration on the ground makes the streets or the land we step on become a more rentable market. While this market has been flled with classical artisan businesses so far, street economy actors serve the population (consumer) where artisans are not suffcient for meeting the demand in highly populated streets. This situation confronted law enforcement and street sellers in cities for decades or may be centuries, and urban peace and harmony often deteriorated. In the integrated urban areas, in addition to a series of urban problems, the registration of the informal economy and the adaptation of the street economy actors to the urban identity and esthetics have become the problems that await priority solutions.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Effect of Micro Entrepreneur Arrangements to City Culture    261 The terms “micro entrepreneurship,” “street economy,” “micro entrepreneurship” newly entering the literature, which we use to refer the street economy and the actors engaging in this, scrap dealers, street sellers, street vendors, street traders, peddlers and hawkers, shoe shiners, etc., together, are not only the very serious gaps of informal economy in urban geographies but also cause visual disturbances within the contemporary modern buildings (today’s situation) of cities. These terms used interchangeably by most of the authors (Bernadus, Utami, & Liliana, 2018; Bromley, 2000; Suharno & Dini, 2018). Street economy will serve not only for registering the informal economy, but also registering and controlling it in order to achieve urban esthetics and sustainable urban life, street peace and human street ergonomics, which is the fact that we deal with. What is important here is that, this situation is becoming clear and defnite now. The state authorities have failed to cope with or prevent informal economic activities regardless of location through the use of law enforcement on central and local basis for centuries and even measures such as deterrent fnes did not (could not) succeed in any cases. Under these circumstances, the state, together with its central and local authorities and law enforcement, is obligated to make these actors recognized and registered primarily by imposing a low taxation regime, to have them registered and to offer facilities that are in compliance with the urban esthetics and street ergonomics. When we look at the situation in the context of cities all over the world, the economic activities in the cities where the population is concentrated correspond to 30% of the economic activities carried out in urban spaces. It is not possible to stay indifferent to the facts that such a big economic activity is carried out informally in an uncontrolled and cluttered manner and method, away from institutionalism, esthetics, symmetry and harmony, in terms of the urban life quality. In reverse logic, recognition of the street economy provides a two-unit added value rather than a one-unit value. On the one hand, the activities of street peace and street economy are controlled and recorded while on the other hand, the positive outcomes of the issue regarding the quality of urban life and urban esthetics are seen. This two-way added value will enable potential consumers living in urban areas to receive more effcient and quality services.

3. Street Economy in Sustaining Urban Life Esthetics and Ergonomics For the past 30–35 years, investigations of the informal sector have focused mostly on the economic role of unregulated and undocumented work in national and global economies. Street Economy is based entirely on the economic activities of street vendors. A street vendor is a person who offers goods or services for sale to the public without having a permanently built structure but with a temporary static structure or mobile stall (or head-load). (NASVI, 2018) The largest openaccess dictionary in the world, Wikipedia, has examined the street vendors under the title of informal economy (Wikipedia, 2019).

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

262    Ahmet Fidan C. Cross emphasized his book, Street Vendors and the State in Mexico City, brings to the forefront the political infuence of Mexico City’s street vendors in the implementation of state policies which affect them. Using ethnographic and comparative historical methods, Cross (1998) details four case studies of recent interactions between street vendor organizations and the state and compares three distinct periods of Mexico City’s government’s relationship to its street vendors. We, as the pioneers of the Street Economy trend, who have been trying to introduce the concept of Street Economy to the literature recently, need to emphasize that we are referring to the concepts of micro entrepreneurship, micro entrepreneurship and related legal, physical and sociological studies within the concept of street economy. Street economy is a multinational sector which has been excluded, ignored, rejected by well-heeled individuals and entrepreneurs, tried to be destroyed by central and local government institutions continuously but failed and needs to be incorporated into economic dynamism. Street economy is a sector that should be regulated and accredited not only in country a, country b, country c, but in all cities and countries around the world. As the saying goes, this situating can be described as “if you can’t beat them, join them.” Yes, this process has no unethical dimensions. As long as this sector is legally recognized and registered by the countries of the world and PEACE on the STREET is ensured in a sustainable way. The frst scientifc fndings on the street economy is the CROSS book. Secondly, Krishna Prasad Timalsina wrote the article titled “Urban Informal Economy: Livelihood Opportunity for Poverty or Urban Governance Problems, Investigation of Kathmandu’s Street Vending Activities” (Timalsina, 2011). Thirdly, The Street Vendors Act 2014 came into force in May 1, 2014 in India. Act of the Parliament of India enacted to regulate street vendors in public areas and protect their rights. On the other hand, Sirkeci’s work has been a continuation of activities on this subject. His activities in Turkey, started with the Editorial Letter sent to the Kent Akademisi Journal by Mr Osman Sirkeci (Sirkeci, 2018) and after that stage, it has started to be publicity in the community with the Kent Akademisi Journal, where I work as the chief editor and supported as a concept defned in the scientifc community Same way, Mr. Sirkeci, stressed the importance of the street economy in Turkey and wrote in every environment (Sirkeci, 2019). These works have yielded results, and Izmir Metropolitan Municipality Street Economy Directorate was established under the leadership of Sirkeci, again under the theoretician of Kamuran Elbeyoğlu. In the following process, it is necessary to achieve the academic broadness and depth of the street economy on one hand and to fll the gaps in the sub-felds of this sector on the other. In October 2018, Malta Conference was organized within the scope of the 3RD Democracy Symposium as the result of the cooperation between the pioneers of street economy and leading stakeholder organizations. This was the frst international meeting of the Street Economy Movement (Fidan, 2018). In the Malta Conference, one of the pioneers of the movement, Ahmet FİDAN, discussed the priorities of urban esthetics and ergonomics (Altinova, 2018).

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Effect of Micro Entrepreneur Arrangements to City Culture    263 The topic of introduction and mission setting of the street economy in the most appropriate manner depending on the urban identity was addressed and then, the gaps on the details of legal regulations of central and local authorities at the point of the street economy’s supporting role for street peace and global peace were identifed and it was ensured to fll those gaps in legal platform on regulatory/law basis.

4. Regulatory Requirement Regarding Entrepreneurship Activities of Micro Entrepreneur Actors in Urban Areas Contemporary cities and urban administrators, eventually end up rejecting the street economy and its actors, as in the past century when capitalism objected against the unions but then internalized them as it failed to cope with, at frst and resisted them with prohibiting/banning practices and painful law enforcements. In urban life, however, this has neither been prevented nor has a sustainable climate of peace been achieved in the streets. For this reason, street economy bears an extremely important function in urban life in terms of both registering the economy and to achieve the human dignity and contemporary civilization level as well as urban esthetics in order to maintain the economy and the democratic rights and undertakings of individuals in well-being and peace. Firstly, Street Vendors Act 2014 came into force in 2014 in India, Secondly, however, the street vending has become a major challenge for the local authority in Kathmandu. The increasing number of street vendors has deteriorated the environment of the city creating pollutions and congestions disturbing traffc (Apil & Sherstha, 2019). For these reasons, a legally rooted regulation is expected for street sellers in Nepal very soon. As a fourthly, Izmir Metropolitan Municipality Street Economy Directorate was established in Izmir in Turkey.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

4.1. Legal Regulations Regarding the Economic Activities of Micro Entrepreneurs Setting off from the fact that the economic activities of micro entrepreneurs or micro entrepreneurs cannot be destroyed for centuries and even for thousands of years and based on this reality, the necessity of registering these economic activities at this point is indispensable. In that case, we have to content ourselves with at least the requirements that should be made in legal terms. It is necessary to impose tax for the micro entrepreneurs but to design the procedure and conditions required for this registration in a bearable format free from formalities in order to ensure that the system is realistic in terms of the dynamism and sustainability of the economic life. In order to recognize and accept micro entrepreneurship or micro entrepreneurs by both central and local public administrations all around the work, job descriptions and scope should be prepared in general and in each sub-entrepreneurship branch. Although it may be possible to cover micro entrepreneurs in general as ­peddlers, we cannot except from these types of entrepreneurs to carry out these

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

264    Ahmet Fidan long-term activities, and therefore their legal relations with the system must have some facilities for leaving the system unavoidably. The entrepreneur should be able to terminate the initiative without additional costs or procedures when he/ she cannot or will not continue to operate. There should not be any obligations that will last for more than one week and cancelation or contract termination fees not exceeding the maximum value of the banknote in the currency of the country. The informal order of mafa on micro entrepreneurs’ entry into the sector should be prevented by the state’s law enforcement agencies. In this sense, market entrances and exits must be established by law in accordance with FULL COMPETITION CONDITIONS. The position of Micro Entrepreneurship in the hierarchy of norms in the legal order should be done at the level of the laws or rules of the countries. In other words, resolving this activity at regulation or directive level is not suffcient to ensure the sustainability of street peace and to protect the legality.

5. Infrastructure Regulations for the Economic Activities of Micro Entrepreneurs

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

It will be able to categorize these issues as follows: First of all, the economic activities of micro entrepreneurs should be recognized without hesitation by top and side actors in the sector. Initiatives and activities related to these activities of micro entrepreneurs should not be perceived by the concept of horizontal justice in the fair distribution of income, opportunity and rent, but should be systematized according to the principles of horizontal and vertical justice. Obstacles to the passage or belonging of each micro entrepreneur to the upper main sector groups in their feld of activity should be removed and the economic and legal system should encourage the fow of these entrepreneurs to the upper sector groups.

6. Solutions for the Rehabilitation of the Economic Activities of Micro Entrepreneurs Perhaps one of the most important issues for micro entrepreneurs is the necessity of ensuring that they are recognized by households in urban areas and are visible in the context of analysis and cultural diversity, away from the perception of alienation. The duty of the central administration institutions on this issue is to prepare audiovisual materials in the related media platforms and subliminal rehabilitation of the households in the form of a public spot. Micro entrepreneurs are also required to shape themselves in the format appropriate to their URBAN IDENTITY and URBAN VISION in parallel with their acceptance by the legal and social system. In this context, it should be considered normal that these actors should be dressed according to the sectors in which they operate or the anticipate the color and color of the clothing accordingly.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Effect of Micro Entrepreneur Arrangements to City Culture    265 Micro entrepreneurs should be seen as living representatives of the city and the trainings should be given according to the dynamism of the city at regular intervals and the language, subliminal message, jargon, gesture and mimic patterns they will use in customer relations should be shown or at least recommended. Consultancy offces for micro entrepreneurs must become a standard practice at metropoles (in megapoles and metropoles, central, local government and nongovernmental organization representatives, independent management in the form of presidency, provincial level local government units representative, in the form of district level cities (population less than 100,000).

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

7. Generic Research on Micro Entrepreneurs in Metropolitans Our studies were conducted at urban areas and metropoles in Turkey due to the limitation obligation in terms of the population and in this context, observations, interviews and surveys were conducted with approximately 300 entrepreneurs in total at Istanbul, Turkey, Ankara, Izmir, Adana, Antalya, Konya, Samsun, Ordu provinces. Observation and interview were conducted on micro entrepreneurs, while the survey was conducted on randomly selected population in the provinces where observation and interview were conducted. The observations were made completely in the cosmopolitan areas of the cities, without notifying the actors and their problems and problems were analyzed. The observation process of our study lasted approximately six months and the study was set based on these six-month impressions. In our interview-type research, at least 10 entrepreneurs from each province were interviewed based on the micro entrepreneurs listed below. In our surveytype research, 100 people were interviewed in Ankara and Istanbul, 50 people were in İzmir and 10 people were in Adana, Antalya and Konya, Samsun and Ordu with random sampling method cosmopolitan areas. In addition, similar crossover inquiries were carried out on the same quantity of peddlers on products or services. Within the framework of our research, micro entrepreneurs commonly seen at urban areas particularly in Turkey and around the world have been observed.   (1)   (2)  (3)   (4)   (5)   (6)   (7)   (8)  (9) (10) (11) (12) (13)

Stationary peddlers Traveler peddlers Recyclers Street musicians Roadside food and goods vendors Sexual service providers Roadside glass wipers Overlock service providers Peddlers Mobile Tinsmith Mobile shoe polishers Fortune tellers Painters and cartoonists

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

266    Ahmet Fidan In these contexts, we can list the problems experienced by these entrepreneurs as follows in the order of importance:

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

(1) Future guarantee. (2) Fear of imposition of fne. (3) Exposure to attacks, harassment, insults and so on. (4) Individual hygiene problem. (5) Product and/or service hygiene problem. In our interviews with the street vendors, the frst three problems of each of them are expressed in a snap. The answers given to the questions we asked to the service providers (street vendors) are as follows: The responses of street vendors to the problems in judicial context and the approaches of street vendors to the problems related to age and education levels can be revealed by field researches that will be conducted separately in the future. However, in general terms, the higher the level of education is, the scope and quality of the product sold or the product offered become better or higher. On the other hand, it was observed that younger street vendors were more mobile and older street vendors were less mobile. The future concerns of young street vendors and the future concerns of older street vendors make a difference of at least 10 units. A series of questions addressed to those procuring goods and services from street were about how they saw street vendors were seen in the urban texture and how they would like to see them. One of the dramatic fndings that draw our attention within the scope of our research is that although the street vendors have mentioned the hygiene problem for the product provider and the product/service, these problems are not parallel with the consumption tendencies. In other words, people do not hesitate to shop easily, even if they do not fnd street vendors and products hygienic or safe.

8. Conclusions Fig. 1’s conclusion: Fear of punishment is higher in male street vendors, but less in female street vendors, because local government law enforcement agencies are a bit more tolerant of female street vendors. In the other column, female street vendors expressed less of the hygiene problem in their products than men, as they were cleaner than men’s. Fig. 2’s conclusion: A large number of customers who buy products or services from street vendors see street vendors’ livelihoods and future concerns as equivalent. Equally important to those who beneft from goods and services are the concern of street vendors being exposed to police or law enforcement attacks.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Effect of Micro Entrepreneur Arrangements to City Culture    267

Fig. 1.  Responses of Micron Entrepreneurs to Their Problems per Gender.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Regarding hygiene, people who naturally buy the product from the hygiene and quality of the products or services are more concerned than those who sell the product. However, while consumers complain about hygiene and quality, they continue to consume goods and services from them.

Fig. 2.  Gender Distributed Answers Given by Those Procuring Goods/Services from Street Vendors on Street Vendors.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

268    Ahmet Fidan

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Fig. 3.  Perception and Level of Perception of Those Procuring Goods and Services from Street Vendors. Fig. 3’s conclusion: While an important part of the participants considered street vendors necessary in urban life against the questions posed both to the street vendors and those who beneft from the goods and services, they also envisaged them to be regulated legally and esthetically. On the other hand, the same group (seller and buyer) thinks that street sellers are an indispensable and very important part of the employment defcit in the society. In the last column, they asserted that 15% of the group should be tackled without compromising the weight of the street vendors, and that they spoil the peace and esthetics of the city. When we address open-ended oral questions to this group to support their views, they say that this is due to the fact that street vendors are not disciplined, disorganized, and do not guarantee product and service. However, if street vendors, as micron entrepreneurs, provide local government agencies with registration and authorization, it will be inevitable that they stand behind the product they sell and have to protect the quality and safety of their goods or services.

9. Suggestions and Evaluations Based on the observations and interviews we have obtained on the street economy, which makes around 30 trillion-dollar contribution to the world economy (Sirkeci, 2019). We can say the following based on the objective of making urban life, street peace and ergonomics sustainable: (1) In the context of the emergence of a climate of peace in the streets, the registration of informal economy, systemizations of the situations, which contradicted with urban esthetics and urban peace until today by the urban

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Effect of Micro Entrepreneur Arrangements to City Culture    269 administrations in a manner that brings a separate identity and diversity to the cities, become organic mobile electronic system integration and source of movement of the street, which will lead to the development of urban life quality while maintaining the dynamism of the streets. (2) The recognition, systematization, internalization of micro entrepreneurship by urban administrations and state will have a positive impact on the street peace, improvement of urban life quality, multiculturalism, urban esthetics and urban life ergonomics. (3) It is among our fndings that micro entrepreneurship arrangements, belonging to the urban life, embracement, citizenship awareness, preservation of urban furniture and urban textures, etc., have extremely advanced-dimensional added value. (4) The last and interesting one is that although the street vendors have mentioned the hygiene problem for the product provider and the product/service, these problems are not parallel with the consumption tendencies. (5) Another interesting result is that although the seems to have the perception that street vendors cause disturbances and ruin the street peace, people living in the cities want to see these street vendors as part of the multiculturalism of the city but anticipate introduction of legal regulations rather than an uncontrolled situation. Since both the redefnition of statute and regulation are seen as problems that back each other, our solution proposal is in parallel to this.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Every social problem, even if it does not attain legal regulation for the moment, is delayed sooner or later. Legislation and regulations for street vendors and street economies, the majority of Turkey and other countries will form the basis of time. On the one hand, we will create the literature with the academic studies we do in the scientifc community, and on the other hand, we will base the concept of Street Economy with cultural activities. Thus, Street Economy will take its place in daily economic life as an integral part of urban esthetics and urban culture. So much so that the color of urban identity will be the color of street vendors’ mobile vehicles, ethnographic products, and clothing will be the uniform of street vendors. (6) We hope that our study will contribute to the fair distribution of world cultural values and the level of contemporary civilization for academicians and those interested in practical life.

References Altinova, G. (2018). Malta’da Sokak Ekonomisi Kongresi; Kültürler Arası Bir Diyalog Köprüsü. Retrieved from https://www.altinovagazete.com/maltada-sokak-ekonomisi-kongresikulturler-arasi-bir-diyalog-koprusu-makale,613.html. Accessed on February 5, 2019. Apil, K. C., & Shrestha, E. (2019). Street vendors in Kathmandu. Final Seminar Presentation 2014, No. 2, Urban Informal Sector. Bernadus, D., Utami, C. W., Liliana, D. (2018). Factor analysis of ownership behavior at family business: The case of Indonesia. International Journal of Economics and Business Administration, 6(2), 27–38.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

270    Ahmet Fidan

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Bromley, R. (2000). Street vending and public policy: A global review. International Journal of Sociology and Social Policy, 20(1\2), 1–28. Cross, J. C. (1998). Informal politics: Street vendors and the state in Mexico city. Stanford, CA: Stanford University Press. Fidan, A. (2018). Malta Sokak Ekonomisi Kongresi. Retrieved from http://www.yazarportal.com/ maltada-sokak-ekonomisi-kongresi-kulturler-arasi-bir-diyalog-koprusu/146094/. Accessed on February 10, 2018. NASVI. (2018). Defning street vendors. National Association of Street Vendors of India – NASVI. Retrieved from http://nasvinet.org/newsite/defning-street-vendors/ Accessed on December 12, 2019. Sirkeci, O. (2018). The manifest of the global street economy. Kent Akademisi Dergisi, 11(1), 211–213. Retrieved from https://dergipark.org.tr/download/article-fle/ 457462. Accessed on February 1, 2019. Sirkeci, O. (2019). Global Sokakların Dünya Ekonomisine Katkısı 30 Trilyon Dolar. Türkiye İnteraktif Köşe Yazarı Gazetesi ve Yazar Portali. Retrieved from http:// www.yazarportal.com/global-sokaklarin-dunya-ekonomisine-katkisi-30-trilyondolar/148915/ Suharno, P., & Dini, I. (2018). The infuence of work stress, working cost, compensation and work discipline on employee’ productivity. International Journal of Economics and Business Administration, 6(4), 62–75. Timalsina, K. P. (2011). An urban informal economy: Livelihood opportunity to poor or challenges for urban governance, study of street vending activities of Kathmandu metropolitan city. International Journal of Politics and Good Governance, 2(2.2), 1–13. Quarter II, ISSN: 0976-1195. Wikipedia. (2019). Informal economy, informal sector. Retrieved from https://en.wikipedia.org/ wiki/Informal_economy Accessed on January 10, 2019.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Effect of Micro Entrepreneur Arrangements to City Culture    271

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Appendix

Fig. A1.  A Letter Sent by a Street Economy Activist (Female Entrepreneur or Market Woman) to Her Son Who Was Studying at University. Note. The woman had hidden 5,000 TL in it. “Dear son, I sacrifce this for you. I salute you very much, kiss you in your eyes, I am ready for the last (this) I have saved this for my death, but I am sending as you are in trouble. No need to get depressed.” She placed the money she earned as a seller in the market in this letter and covered it with a nylon bag and buried it in the butter she produced for the market.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

This page intentionally left blank

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Chapter 17

A Study of Kosovo’s Fiscal Policies and Tax System: Key Issues to Align with European Standards Gani Asllani, Simon Grima and Sharon Seychell Abstract Purpose: This research study addresses the key issues of the fscal policies and tax system in Kosovo to align with the contemporary tax principles and requirements of the European Standards.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Design/methodology/approach: In order to explain the fscal policies and tax system in Kosovo, the authors provide an overview of the fscal policies and tax system in the country. The authors focused on the building and development of the tax system, the role of the tax policy in economic development and the macro-fscal stability. The methodology approach is based on primary and other secondary sources, such as academic literature, the reports provided by the Ministry of Finance and international institutions, tax law and other available important statistical data. Findings: The chapter substantiates and confrms the need for tax reforms in order to have an adequate tax system oriented to indirect tax, changing the structure of tax collection for border tax into domestic tax, simplifcation of the legal procedures, improvement of management and audit systems, reducing the informal gray economy and to have a gradual growth-friendly fscal policy. Positive fscal performance continued, although there is still room to improve the execution of revenues. Practical Implications: This chapter highlights the fact that Kosovo needs to intensify their fscal reforms to build a sustainable tax system in order to meet the criteria of the EU tax system and policies.

Uncertainty and Challenges in Contemporary Economic Behaviour, 273–284 Copyright © 2020 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-80043-095-220201018

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

274    Gani Asllani et al. Originality/value: The authors defne the needs for strengthening the budget revenues, taking strong and coordinated steps to improve the management and collection of the revenues from indirect and direct taxes. Moreover, the authors show that the Kosovo tax system is broadly in line with the EU acquis and needs reform in order to have a sustainable tax regime. Keywords: Fiscal policy; taxation; economic development; fscal instruments; Kosovo; European Union JEL classifcation: D41; F12

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

1. Introduction After 1999, Kosovo started to build the tax policy oriented to the market economy, which includes the development and implementation of modern concepts of taxation. The Kosovo tax systems aim at being in compliance with EU acquis and advanced international standards. Initially, based on the international support, Kosovo had managed to successfully end the emergency phase (i.e., the stage of creating institutions) and is now in the process of sustainable economic development (Asllani, 2011). Time and knowledge were needed for the creation of new fully operational institutions and for them to comply with legislation. In particular, the complex issue was the establishment of the fscal policy which would have two key roles: the collection of adequate budget revenues in order to cover public expenditure and supporting economic development (Grima & Caruana, 2017; Mintz, 2004; Rupeika-Apoga, Zaidi, Thalassinos, & Thalassinos, 2018). By the end of the 2003, Kosovo, with its revenues, was capable of funding the essential needs of its public administration and social transfers, with the ongoing trend of increasing budget revenues (Statovci & Asllani, 2017). It has built and continues to build a simple tax system, and with the application of low tax rates and free mode of trade (barriers and customs restrictions), it is among the most liberal countries in the region. Such policy shall impose the need for rapid preparation of businesses to enter in a free competition with other competitive economies and to increase productivity (Mintz, 2005; Keen, 2004; Pociovalisteanu, Thalassinos, Tirca, & Filho, 2010; Thalassinos & Kiriazidis, 2003; Thalassinos & Politis, 2011). Kosovo’s fscal system is relatively new, a system that is based on direct and indirect taxes. Its development had started in 1999 with the regulation on customs, value added tax (VAT), then in other direct and indirect taxes. Tax legislation and customs legislation in Kosovo is almost harmonized with EU directives and pretends to meet contemporary requirements. Tax revenues currently provide a good source of funds for state functioning, as well as economic and social functions. Since January 2005, tax policies begun with the initial tax amendments and with the application of new taxes such as corporate income tax, as well as, personal income tax and this continued until the end of 2008 (Ministry of Finance of Kosovo, the annual reports 2002–2010). In January 2009, a new law came into effect, with a deduction in tax rates in Kosovo, and the tax rates on dividends, interest, rent, gambling, capital gains, sale of intangible property, etc., were reduced from 20% to 10%. However, the value-added tax (VAT) went through a slight increase from 15% to

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

A Study of Kosovo’s Fiscal Policies and Tax System    275 16%, which caused a small increase of prices in the country. Reforming the tax system is one of the most important segments for countries in a socio-economic transition, led by the aspiration to join the European Union. In recent years, there has been a continuous reform in the tax policy in Kosovo. Currently, the tax system in Kosovo is considered to be quite simple and harmonized. However, much remains to be done regarding the treatment of some issues of strategic importance, such as further development of an IT system, the harmonization of payment procedures for the revenues used by banks and tax authorities, the development and utilization of effcient taxpayer registration system, and improvement in the risk management of the system.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

2. Literature Review The tax system is an important element of the macroeconomic system which should provide a fundamental fscal basis which intends to collect income to cover public costs and other expenses provided by law. The state will not be able to exist and function without payment of tax obligations and economic development cannot be imagined without a fscal policy and the proper tax system. According to Dabrowski and Tomczynska (2001) the tax is part of the existence of the state and the payment of taxes is obligatory. The purpose of the tax is to collect revenue for the state that are necessary for the fnancing of their functions (Komoni, 2008). Gallagher and Babić (2005) state that the fscal policy is oriented at improving the economic environment and usually include steps that simplify the tax system, while at the same time, eliminate the numerous taxes which give low performance and eliminate unproductive taxes. The objective of the fscal policy is to provide needed revenues by the application of fscal instruments which cover public expenditures (Jurković, 1991). Through fscal policy, the state can directly affect the economy. When a state decides about a tax collection system, transfers, goods and services it buys, defcit fnancing, it is in fact engaging in fscal policy (Bronchi & Burns, 2000). Bird (1990) argues that the effciency of a tax system is not determined only by appropriate legal regulation, but also by the effciency and integrity of the tax administration. According to Kaldor (1980), a good tax system is an output of good tax policies, legislation, the effciency and integrity of tax administration. In tax reforms, there is a close correlation between successful tax policy and effcient tax administration. In other words, there is no good tax policy without effcient tax administration and ineffcient tax administration poses serious constraints to tax revenue mobilization (Tanzi & Pellechio, 1995). Without the permanent reorganization of the tax administration and almost daily improvements in methods of its management, it is impossible to expect that tax reforms could be realized successfully (Bejaković, 2001; Ministry of Finance, 2016).

3. Contemporary Requirements Set Out in the Field of Taxation and the EU In the last few decades, the theoretical knowledge, empirical research and practical experience in the feld of taxation on the one hand and the movements and integration

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

276    Gani Asllani et al. on the other hand (the EU) have had an important impact on changing the tax system and the measures in the economic policy, namely the requirements which set the modern conditions and economic policy. Some of the most important tax requirements are: Tax harmonization – The harmonization of the tax system and the measures of the economic policy are integral for a better and more rapid implementation of the goals formulated by the economic community, leading to full integration (Jelicić, 1997). Experience has shown that for the successful functioning of these communities is necessary to unify and approach the tax system and tax policies. Member States of the European Community aim not only to have a reconciled system, but also to have a unifed system of customs, and have taken great steps in harmonizing the rules regulating the matter of taxation of turnover and application of specifc tax (excise). However, although all EU Member States have built a sales tax in the form of VAT there are still differences in tax rates, tax exemptions and the share of VAT in total tax revenues within the countries. Following in the example of the EU, there are other states wishing to become members of the EU, which have trade relations developed with the EU countries and with the countries in transition try to reform their taxation system and tend to quickly and successfully adapt this consumption tax. Tax neutrality – The tax system should strive to be neutral, so that decisions are made on their economic merits and not for tax reasons. The role of taxation should be limited during the implementation of fscal goals, so that the state economic policy measures not to damage the taxpayers. Reducing the tax burden – In last decades, more outspoken theoretical demands are coming from politicians, businesses, various associations of interests and the public in general for the reduction of tax burdens, namely in the social production taxes. The request for the creation of “saver” state, requesting the restriction of any assignment of powers of the state, to reduce the loading of the economy, in order to promote productivity, increase employment and reduce infation is present in all modern states (Gallagher & Babić, 2005). The criteria that the EU members shall have to meet under the agreement of the Maastricht treaty to become a member of the European Union, particularly in reducing the defcit to 3% of revenues and of total debt to 60% of social production, in the long run, will affect the withholding tax load growth not only in the EU but also in other modern countries. Simplicity in tax – In an attempt to prove that the tax should be as simple as possible, the taxpayers should easily fnd and understand the tax regulation. Indeed, a tax system should remove the so-called “inferior” taxes in order to be more precisely regulated, with the view to exclude any interpretation by the tax offcers. Tax is a logical consequence of increasing democratization of the state. The taxable person requires having clear written obligations which are understandable. Simplicity in the tax burden by reducing tax constitutes the best tax reform.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

A Study of Kosovo’s Fiscal Policies and Tax System    277 Priorities of consumption tax in relation to income taxes – For half a century, experts have discussed if consumption is the best base for taxation or incomes. It is noted that consumption tax is fairer than income tax, the possibility of tax evasion is smaller in the case of consumption tax and it is logical to tax consumption rather than income. In favor of consumption tax, we should mention that people work and generate income, not for what they have, but for what you consume to meet their needs. Consumer orientation encourages taxation of savings, with this affecting their investments and this enables the creation of new jobs and overall economic development. There are also these of the opinion that taxation of consumption is more in line with the principle of payment of tax by economic powers (Bronchi & Burns, 2000). In the undoubted advantage associated with consumption taxation, one needs to also highlight some weaknesses of consumption taxation in relation to income. Economic effciency – The requirement that a tax system be effcient arises from the nature of a market economy. Although there are many examples to the contrary, economists generally believe that markets do a fairly good job in making economic decisions such as consumption, production, and fnancing. Thus, they feel that tax policy should generally refrain from interfering with the market’s allocation of economic resources. Taxation should entail a minimum interference from individual decisions. It should not discriminate in favor of, or against, particular consumption expenditures, particular means of production, particular forms of organization or particular industries. Fairness in taxation – The long-term task to be undertaken by the tax authorities is equitable distribution of the tax burden, the greater realization of horizontal and vertical equity and the greater equality in taxation. Equality in taxation is the highest principle on which tax must be supported in the future.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

4. The Tax System in the EU The fscal system of each member state of the European Union is a fundamental exclusivity of national sovereignty. In the EU, the harmonization of taxes refers to coordinating a tax system between Member States, as such a way to avoid national tax measures that may adversely affect the functioning of the internal market in the European Union, including free movement of goods, services and capital and the possibility of distortion of competition. Each member state has the right to have its tax system but also should apply changes (the tax rate, the tax base) in accordance and harmonization with European regulations. The Directive 6 of the European Union on the VAT represents the regulation uniformly to all Member States. It sets standard rates of VAT at 15%, while it allows to have two degrees at 5% and 0% rate provided by the Annex, although the maximum rate is not provided. The basic duty of EU tax policy in recent years is closely related to the development of the internal market, the strengthening of monetary union and economic integration (European Commission, 2019). Regulation of the internal market in the early 1990s defnes a legal framework for the indirect taxes (VAT and excise), while the area of direct taxes (income tax

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

278    Gani Asllani et al. and proft tax) is not defned in clearly on a legal base. To improve coordination and harmonization of tax policies among member countries, in 1997 the EU set out the main directions of tax policy (Commission of the European Communities, 1997), which encouraged the stabilization of tax revenues of Member States, the reduction of problems in the functioning of the internal market, employment, etc. The diversity of tax systems was a constant obstacle to the full realization of these objectives; the resolution of these differences should be consistent, in order to achieve continuous balancing of the tax systems of Member States. However, it has been concluded that, harmonization of these provisions is very diffcult to achieve in the area of taxation in general.

5. Changes in the Fiscal System in Kosovo

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

In 2015, the Government of Kosovo made several decisions for fscal adjustment (fscal reforms package two). Among others, these decisions are related to the approval of new drafts such as: (a) VAT; (b) corporate income tax and (c) personal income tax. The biggest changes regarding these draft laws are expected to occur in the draft for VAT. First of all, under Article 6 of the Draft Law on VAT, the threshold of VAT is reduced from € 50,000 in € 30,000. So, any business that perform turnovers over € 30,000 per year will be obliged to register for VAT and pay the portion that exceeds this amount. The budget would beneft from this reduction in the VAT threshold because a greater number of businesses will be obliged to pay VAT. On the other hand, this fscal policy will burden the new and small businesses, especially those services that have lesser levels of supplies. The legal base is Law no. 05/L-037 on VAT, entered into force on September 1, 2015, Administrative Instruction no. 03/2015 on the implementation of law no. 05/L-037 on VAT. The rate of reduction is from 16% to 8%. A person is obliged to register for VAT when a person independently runs an economic activity in Kosovo, in accordance with Article 4 of the Law. Turnover (including turnover of supplies exempt of VAT) means exceeds € 30,000 within the calendar year (Tables 1 and 2). Kosovo is making maximum efforts toward the harmonization of national legislation with the EU acquis. Based on this, the government of Kosovo, as carrier of the tax system, designs and implements the action plan with the aim of unifying the tax system with EU requirements. Customs and Taxes are the main areas of focus to be in line with the implementation of EU requirements and are based on action plans developed since 2009. With reference to Customs, since September 2012 the data processing system as well as integrated system of Kosovo tariffs – TARIC was implemented (Kosovo Customs, 2010). The taxes are harmonized with the Directive of the European Union, starting from VAT Law 03/L-146 and the completion of this law with the Law 04/L-108, Law 03/L-222 on Tax Administration and Procedures amended the Law 2004/48 and Law 03/L-162 on Corporate Income amending the Law 03/L-113 and 2004/51. Lately, from the beginning of the 2016 the process of the integration has started the process of unifcation of the Kosovo Tax Administration and Kosovo customs in one institution (Kosovo, 2015).

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

A Study of Kosovo’s Fiscal Policies and Tax System    279 Table 1.  Tax Structure and Tax Rate in Kosovo (in Euro). Type of Tax

Turnover Threshold

Value added tax

30,000.00

Tax Rate 8% and 18%

Corporate tax

Over 50,000.00

10%

Corporate tax: (manufacturing, trading, transport and service activities)

Below 50,000.00

3% and 9%

Personal income tax

Over 50,000.00

Annual income from 0 to 960

0%

Annual income from 960 to 3,000

4%

Annual income from 3,000 to 5,400

8%

Annual income from 5,400 to …

10%

Wage tax (monthly) Monthly income from 0 to 80

0%

Monthly income from 80 to 250

4%

Monthly income from 250 to 450

8%

Monthly income from 450 to …

10%

Personal income tax: (manufacturing, trading, transport and service activities)

Below 50,000.00

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Tax on rent, interest and property rights

3% and 9% 10%

Social contributions: (employer and employee)

5% and 5%

Property tax

1% to 0.15%

Customs tax

0% to 10%

Excise

Specifc and ad valorem

Source: Ministry of Finance, Kosovo.

6. EU Progress Report for Kosovo 2019 Kosovo is a potential candidate country for EU membership, since in 2016 it has signed “The Stabilization and Association Agreement” with the EU, therefore, should until the accession, process and adopt all acquis and make it an integral part of its domestic legislation. One the most important issue is taxation. The progress report highlights:

6.1. Customs The latest EU progress reports for Kosovo in the area of customs conclude that Kosovo Customs is at a moderate stage of preparation but needs improvement.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

– – 131.0 361

Border collection

Costumes duties

Excise tax

611.0

0.0

Domestic collection

Value added tax

Tax debts

403.3

130.0

514.6

179.2

693.8

0.0

1,227.4

2.2

2.0

Other direct tax 1,107

Property tax

Indirect tax

25.1

20.0

Tax on personal income

80.8 124.0

68.0 109

Tax on corporate income

0.0

Tax debts

0.0

232.1

198

Direct tax

1,421.1

1,269.0

1,596.5

1,472.0

2016

1.1 Tax revenue

2015

1. Total revenue

Description

432.3

126.0

557.6

198.3

756.1

0.0

1,315.2

2.3

22.4

136.9

75.3

0.0

238.0

1,495.7

1,681.6

2017

466.0

111.0

615.9

203.5

819.4

5.0

1,378.0

4.0

33.0

145.8

84.2

0.0

267.0

1,607.4

1,816.9

2018 Budget

Table 2.  Total Budget Revenues (Budget and Forecast in Euro) for Period 2015–2020.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

478.2

108.0

655.0

228.0

883.0

5.0

1,471.2

4.0

30.0

158.9

90.1

0.0

283.9

1,710.1

1,923.3

2019 Forecast

512.6

100.0

700.9

244.0

944.8

5.0

1,559.4

4.0

33.0

171.6

98.3

0.0

306.9

1,818.3

2,032.7

2020 Forecast

280    Gani Asllani et al.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

85.0 38.0 0.0 7.0 30.0 15.0 15.0 0.0

Fees, charges, and other from Central Level BO-s

Fees, charges, and other from local level BO-s

Of which: the revenues related to waste tax

Concessionary fees

Mineral realty fees

Revenue from mobile network license fess

Dividends

Income from interest

Source: Adapted from Medium Term Expenditure framework, 2019–2021. Ministry

189

1.2. Non-tax revenues

0.0

One-off revenue from POE debt −36.0

0.0

One-off revenues from the collection of tax debt

Tax refunds

3.0

Other indirect tax

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

0.0

0.0

0.0

30.6

7.7

0.0

47.7

92.4

175.3

−38.4

0.0

0.0

0.4

0.0

0.0

0.0

30.6

8.5

0.0

43.3

103.4

185.9

−57.6

0.0

0.0

0.8

4.5

0.0

0.0

33.0

10.0

0.0

55.0

100.0

202.5

−46.0

4.0

4.0

2.0

4.2

0.0

0.0

33.0

11.0

7.0

57.0

103.0

208.2

−53.0

3.0

5.0

2.0

4.2

0.0

0.0

33.0

12.0

7.0

57.0

103.2

209.4

−56.0

4.0

4.0

2.0

A Study of Kosovo’s Fiscal Policies and Tax System    281

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

282    Gani Asllani et al. There has been some progress in enforcing customs measures (European Commission, 2019). In order to fulfll the recommendation, Kosovo should work in two directions: frstly, further align customs legislation, including the customs and excise code, with the Union customs code and secondary, the excise acquis and step up coordination between Kosovo Customs, law enforcement agencies and other relevant institutions on fghting the informal economy and customs fraud. In the area of customs legislation, the customs and excise code is largely aligned with the Union customs code, but further alignment is still needed. Further efforts are needed to address corruption more effectively. The level of informality in the economy is still high and further coordination efforts are needed between relevant institutions and law enforcement bodies to fght the informal economy and customs fraud. Customs’ performance in the fght against illegal cross-border activities continues to show progress.

6.2. Taxation Referring to the progress report in of the taxation area, Kosovo has made some progress in collecting tax revenues, reducing the gray economy and the secondary tax legislation. Progress report emphasizes the need for improvement in many directions, such as legislation aspects, capacity building and law enforcement. On indirect taxation, the VAT system is broadly aligned with the acquis. Kosovo’s standard rate of 18% and the reduced rate of 8% are in line with the acquis. The tax administration should further strengthen its administrative and professional capacity. However, tax evasion and informality continue to hamper the economy and further efforts are needed, in particular on risk-based feld inspections and inter-institutional cooperation (European Commission, 2019).

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

7. IMF Report for Kosovo 2018 – Fiscal Policy The report prepared by the IMF staff for Kosovo concludes that the Kosovo fscal authorities are fully committed to safeguarding macro-fnancial stability while preserving the credibility of the fscal rule and keeping public debt low. At the same time, concerted efforts are being made to improve the composition of the budget by containing unproductive current spending (IMF, 2019). The implementation of a gradual and growth-friendly fscal consolidation strategy in the last three years has kept current spending in check, while creating room for high priority spending and growth-enhancing capital investments. Despite low tax rates and a low labor tax wedge, total revenues have increased steadily in recent years. Notwithstanding the progress, there is scope to further mobilize domestic revenues by broadening the tax base, including through addressing existing tax gaps for VAT and personal income tax, and continuing to fght informality. Specifc tax and custom administration reforms are being implemented in line with the IMF recommendation, including improvements in the areas of organization and performance management, compliance risk management, fling enforcement, debt collection and tax audit.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

A Study of Kosovo’s Fiscal Policies and Tax System    283

8. Conclusion The development of the tax system and fscal policy shall be adopted with the economic fow of the country. The reforms should be adopted based on tax practices of the countries in the region and in line with the new trends of the reforms developed in the European Union. The tax reforms include the establishment of the tax policies, modernization of tax administration with new taxes, in particular VAT reforms oriented toward the improvement of the economic environment which includes steps toward simplifying tax system and at the same time elimination of taxes with low effciency and elimination of unproductive complexity of different taxes. There is still need for improvement and advancement, both in terms of management and the simplifcation of legal procedures, in order to reach the optimal tax system.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

References Asllani, G. (2011). The impact of fscal policy to the Kosovo economic development. Acta Universitatis Danubius. Œconomica, 7, 123–137. Bejaković, P. (2001). Improving the tax administration in transition countries. Commission Staff Working Document, Institute for Public Finance, Zagreb. Bird, M. (1990). Expenditures, administration and tax reform in developing countries. Bulletin for International Bureau of Fiscal Documentation, 44, 263–267. Bronchi, C., & Burns, A. (2000). The tax system in the Czech Republic. Working Papers No. 245, OECD Economics Department, OECD Publishing. Commission of the European Communities. (1997). Towards tax co-ordination in the European Union. A package to tackle harmful tax competition. Brussels, 01.10.1997 COM(97) 495. https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX :51997DC0495&from=IT Dabrowski, M., & Tomczynska, M. (2001). Tax reforms in transition economies. Warsaw: CASE – Center for Social and Economic Research. European Commission. (2019). Commission staff working document, Kosovo*, 2019. Report (p. 61). Retrieved from https://ec.europa.eu/neighbourhood-enlargement/sites/near/ fles/20190529-kosovo-report.pdf Gallagher, M., & Babić, A. (2005). Tax simplifcation for jobs and growth. USAID ESP Project in Croatia. Retrieved from https://pdf.usaid.gov/pdf_docs/PNADG620. pdf. Grima, S., & Caruana, L. (2017). The effect of the fnancial crisis on emerging markets: A comparative analysis of the stock market situation before and after. European Research Studies Journal, 20(4B), 727–753. IMF. (2019). Country report No. 18/30 (p. 58). Retrieved from http://www.imf.org Institute for Development Research “Riinvest” Pristine. (2005). Jelicić, B. (1997). Public fnance (pp. 176–182). Zagreb: Informator. Jurković, P. (1991). Designing a tax system to promote structural change. The role of the tax reform in central and eastern European economies. Paris: OECD. Kaldor, N. (1980). The role of taxation in economic development. In N. Kaldor (Ed.), Essays on economic policy (Vol. 1, pp. 225–254). London: Duckworth. Keen, M., & Mintz, J. (2004). The optimal threshold for a value-added tax. Journal of Public Economics, 88(3–4), 559–576.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

284    Gani Asllani et al.

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Komoni, S. (2008). Public fnance (p. 45). Pristine, RKS Kosovo: University of Prishtina Press. Kosovo Customs. (2010). Kosovo Customs Tariffs. Retrieved from www.Dogana.rks-gov.net. Ministry of Finance, Kosovo. (n.d.). Finance, Transfers and Tariffs. Retrieved from https:// mf.rks-gov.net/. Mintz, J. M. (2004). Growth and taxes, some implications for developing economies. In J. Alm & J. Martinez-Vazquez (Eds.), Public fnance in developing and transitional countries (p. 207). Northampton, MA: Edward Elgar. Ministry of Finance (2016). Public Finance Management Reform Strategy (PFMRS) of Kosovo 2016–2020. Retrieved from http://www.kryeministri-ks.net/repository/docs/ Public_Finance_Management_Reform_Strategy2016-2020.pdf Pociovalisteanu, M. D., Thalassinos, I. E., Tirca, A., & Filho, L. W. (2010). Trends and challenges in the energy sector of Romania in the post-accession to the European Union. International Journal of Environmental Technology and Management, 12(1), 3–15. doi:10.1504/IJETM.2010.029957.18 Statovci, B., & Asllani, G. (2017). Fiscal policy of Kosovo, taxation and reforms. International Journal of Management Excellence, 9(3), 1127–1131. Tanzi, V., & Pellechio, A. (1995, February). The reform of tax administration. IMF WP/95/22. Washington, DC: IMF. Thalassinos, I. E., & Kiriazidis, T. (2003). Degrees of integration in international portfolio diversifcation: Effective systemic risk. European Research Studies Journal, 6(1–2), 119–130. Thalassinos, I. E., & Politis, D. E. (2011). International stock markets: A co-integration analysis. European Research Studies Journal, 14(4), 113–129.

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Index

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Note: Page numbers followed by “n” indicate foot notes. Ability-signaling hypothesis, 53–54 Accounting (see also Behavioral accounting), 78 system, 80 Accounting decisions (see also Nonaccounting decisions), 53–54 Agency theory, 83, 118–119 AHP model, 162 Akaike Information Criteria (AIC), 133–134 Anglo-Saxon Model, 120 Anonymity hypothesis, 56–57 ARAS method, 162 Asset quality, 164 Asymmetric causality test, 146–147, 149–150 cointegration test, 148–149 data, analysis, and fndings, 151–155 between developed and emerging markets, 146 literature, 147–148 methodology, 148 Audit frms, 54 Augmented Dickey Fuller test (ADF test), 66 fndings, 69 unit root test, 67, 133, 152–153 Banking system, 160 Banks, 160 Banks Listed in Borsa Istanbul (BIST), 161 banks, 160–161 data set, 164–165 fndings, 170–175 GRA, 165–168

limitations, 179 literature review, 161–164 results, 175–179 TOPSIS, 168–170 Behavioral accounting, 78 agency theory, 83 contingency approach, 82 development, 78–79 equality theory, 81–82 expectation theory, 81 factors infuencing behavioral dimension of accounting, 79–81 goal setting theory, 82 neuroaccounting, 84–86 private sector in Turkey, 77 research in World and Turkey, 83–84 theories in, 81 Behavioral fnance theory, 128 Big-Four audit frms, 54 “Bim” service, 92 Bonus contract hypothesis, 54–56 Borsa İstanbul manufacturing industry frms, 38–39 Brazil, Russia, India, China and South Africa (BRICS), 23–24 Business, 140 climate change disrupt business of bank’s customers, 140–141 Capital adequacy, 164 fow, 146 Capital Expenditure (CAPEX), 222 Capital expenditure to sales ratio (CAPEXS ratio), 223

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

286   Index Cash fow volatility (CFV) (see also Earnings volatility (EV)), 218, 246 modelled impact on Maltese operational CFV, 244–246 POMs for, 255 reduction in, 223–225 Causality analysis, 152 Central Bank of Malta (CBM), 197 Chi-square (x2) distribution, 68 test statistics, 25 Climate change, 139 confusing regulators, 143 damaging physical collateral, 140 disrupt business of bank’s customers, 140–141 disrupt operations of fnancial institutions, 140 effect on fnancial sector, 139–140 effect on fnancial system, 142 harsher working conditions, 141 higher disclosure requirement and costs, 142 increased insurance premiums and reduction in coverage, 141 leading to capital fight, 143 loss of offce branch networks, 142 severe climate change leading to increased insurance claims and liabilities, 141 uncertainty destabilizes fnancial markets, 143 Cluster analysis, 3, 7–8 sampling, 8 Collaborative intervention theory of fnancial inclusion, 101 Collapse avoidance hypothesis, 57–58 Collective Investment Schemes, 230 Commodity price risk (CPR), 234 Community Echelon theory of fnancial inclusion, 98–99 Confdence, 129–130 indices, 128–129, 132 Consumption price index (CPI), 25

Continental Europe Model, 120 Contingency approach, 82 Corporate governance, 119–120 models, 123–124 Corporate risk management, 218 Costs of fnancial distress (CFD), 220 Country-level investor protection, 120 Crash model, 25 Critical mass of operations, 226–227 Culture, 79 shock, 199 Cumulative logit link function, 224 Cumulative seasonal decomposition method, 151 Customs, 279–282 Data pre-processing, 166 set, 164–165 Decision(s), 80 decision-making process, 160 matrix, 166 Derivative hedging instruments (DHIs), 218 accounting treatment applied for DHIs and Hedging relationships, 242–243 comparison of results with previous studies and deliberation, 246–249 descriptive statistics pertaining to variables, 238 determinants of Maltese DHI usage, 236 impact on DHI usage and HA on local earnings and operational CFV captures, 243 Mann–Whitney-U test outcomes, 239 modelled impact on Maltese EV, 243–244 modelled impact on Maltese operational CFV, 244–246 number of DHI identifcations, 237

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Index    287

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

prevalence of Maltese DHI usage, 232–236 rationale for DHI usage, 219–220 signifcant determinants at individual proxy variable level, 240 signifcant determinants resultant from Logit model, 240–242 Dickey–Fuller-Generalized Least Squares unit root test (DF-GLS unit root test), 152–153 Direct fxed capital investment (GSSO), 66, 68 Dow Jones Industrial Average index (DJIA index), 148, 151, 153–155 Earnings volatility (EV) (see also Cash fow volatility (CFV)), 218, 224 POMs for, 255 reduction in, 223–225 Economic agents, 22 confdence index, 129 effciency, 277 indicators, 128 migration indicators, 3 sensitivity index, 130 Economic policy uncertainty index (EPU index), 22, 32 Effective hedgers (EH), 227n10 Equality in taxation, 277 theory, 81–82 Error correction models (ECM), 63, 134 Ethical understanding, 81 European Economic and Social Committee (EESC), 198 European Markets Infrastructure Regulation (EMIR), 230 European Parliament (EP), 196 European Securities and Markets Authority (ESMA), 230

European Union (EU), 277 immigration cluster analysis results, 9–12 progress report for Kosovo, 279–282 Facial Coding (FaCs), 85 Filipino live-in care workers in Malta, 198 choosing Malta, 201–204 contextual setting, 197–199 life beyond work, 210–211 methodology, 199–201 motivation and employer’s relations, 207–208 social integration, 208–210 working conditions, 204–207 Filipino migrant community, 197 Final prediction error (FPE), 133–134 Financial Crisis Inquiry Commission (FCIC), 186–187 Financial inclusion, 90 case studies, 104–105 empirical data, 104 foundations, 91–93 theories of fnancial inclusion benefciary, 93–98 theories of fnancial inclusion delivery, 98–102 theories of fnancial inclusion funding, 102–104 theory in, 90–91, 109–115 Financial/fnance climate change disrupt operations of fnancial institutions, 139–140 entity, 230n12 information, 160 integration, 146 literacy theory of fnancial inclusion, 101–102 markets, 146 policy, 62 potential costs of fnancial distress, 220–222 reporting, 49, 51 skills, 50

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

288   Index

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

stem, 142 system regulators, 143 Foreign currency risk (FCR), 234 Forensic accounting theory (see also Behavioral accounting), 49–50, 52–53 ability-signaling hypothesis, 53–54 accounting decisions, 53 literature review, 50–52 materiality hypothesis, 53 non-accounting decisions, 54–58 Fraud detection, 49–50, 52 schemes, 51 Functional magnetic resonance imaging (fMRI), 85 G2P social cash transfer, 96 Galvanic skin response (GSR), 85 Gearing ratio (GR), 222 Global fnancial crisis (2008), 22–23, 62 on cause of crisis, 186–187 liquidity, 191 long before fnancial crisis, 185–186 preparing for next, 192–193 questioning economics profession, 190–191 refections after fnancial crisis, 188–189 rethinking economics of fnancial regulation, 189–190 shortly before and during, 186 Sober refection, 193 Global risk perception, 23, 25, 29 Goal setting theory, 82 Granger casualty/causality test, 64–65, 135 Gray relational analysis models (GRA models), 161, 165–168, 173–174 Gray relational coeffcient matrix, 167 Gross domestic product (GDP), 6, 63 per capita, 7 Hannan Quinn criteria (HQ), 133–134 Hatemi-J asymmetric causality test, 154

Hedge accounting models (HA models) (see also Behavioral accounting), 218–219 impact on local earnings and operational CFV captures, 243 reporting volatility reduction and adoption of HA regime, 227–228 Hedge fund activism, 120–122 Hedging, 218 characteristics of DHI users, 219–220 data and sampling frame, 230 derivatives, 220 detailed variable descriptions, 256–257 determinants of Maltese DHI usage, 236–249 effective reduction of entity’s corporate tax rate, 220–222 explanatory note on use of POMs for EV and CFV, 255 frms, 227 HA models, 218–219 hypothesis formulation for factors affecting Maltese entities’ DHI usage, 220 mitigating ineffcient investment, 222 mitigation of agency problem and critical mass of operations, 226–227 model specifcations, 230–232 non-normality of data collected for variables under study, 254 parsimonious model selection, 253 potential costs of fnancial distress, 220–222 prevalence of Maltese DHI usage, 232–236 reduction in EV and CFV, 223–225 reporting volatility reduction and adoption of HA regime, 227–228

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Index    289

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

research fndings and deliberation, 232 research methodology outline, 228–230 Humanitarian migration (see also International migration), 14 indicators, 3 IMF Report for Kosovo 2018, 282 Immigration, 197 Impulse–response functions, 33 Imputed tax rate (ITR), 221 Income tax rate (ITR), 220 Income–expenditure structure, 164 Index returns of developed markets, 146 Indian stock market, 148 Industry/industrial production index (IP), 25, 66 Ineffective hedgers and speculators (IS), 227n10 “Inferior” taxes, 276 Infation, 29 Insolvency rules, 40 Interbank (I), 25 rate, 33 settlement infrastructure, 142 Interest channel of monetary transmission mechanism, 62 Interest coverage ratio (ICR), 39, 42–44 Interest rate risk (IRR), 218, 234 Interest rates, 66, 129–131 International Labour Organisation (ILO), 196 International migration, 2–4 data–selected indicators, 6–7 EU-10 immigration cluster analysis results, 9–12 indicators, 6 research methods, 7–9 SEM models and results, 12–19 theoretical approaches of, 4–6 International Organization for Migration (IOM), 4

Investors, 160 aim of study, 131 data methodology and fndings, 131–136 hypotheses, 131 literature, 129–131 sentiment, 128–129 Irrational behavior of investors, 128 Jarque–Bera coeffcient, 132, 151 Johansen cointegration, 149, 153 analysis, 134 test, 148, 154 Juselius cointegration method, 149, 153 “Just Vote No” campaigns, 123 Kosovo’s fscal policies and tax system changes in fscal system in, 278–279 contemporary requirements set out in feld of taxation and EU, 275–277 EU progress report for, 279–282 IMF Report for Kosovo 2018, 282 literature review, 275 tax system in EU, 277–278 VAT, 274–275 Kurtosis coeffcient, 152 Labour code, 205 Lee–Strazicich unit root test, 25–28 Legal Entity Identifer (LEI), 230 Likelihood Ratio (LR), 133–134 Liquidity, 164, 191 Live-in care workers, 196 Logit model, 229, 240–242 explaining Maltese DHI usage, 230–231 M-Pesa, 92 Macroeconomic policy uncertainty, 22 variables, 129 Malta Stock Exchange (MSE), 227 Mann–Whitney-U test, 229, 240

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

290   Index Market participants, 128 price of banks, 160 Materiality hypothesis, 53 MAUT method, 162 Maximum likelihood estimation method (MLE method), 9, 134 Micro entrepreneur actors entrepreneurship activities of, 263–264 infrastructure regulations for economic activities of, 264 solutions for rehabilitation of economic activities of, 264–265 Micro entrepreneurship, 260 Migration, 2, 4 integrative study, 3 Mobile technology, 93 “Modelo Peru”, 92 Monetary policy, 22, 62 Monetary transmission mechanism, 62–63 ADF unit root test, 67 data series, 66 fndings of ADF unit root test, 69 fndings of unit root test with Zivot–Andrews structural breakup, 69–70 literature review, 63–66 stability analysis, 66 Toda–Yamamoto causality test, 68, 70–72 in Turkey-based study, 66 Zivot and Andrews unit root test, 68 Morgan Stanley Capital International Emerging Market Index (MSCI Emerging Market Index), 151, 153 Multiple criteria decision making methods (MCDM methods), 161–162, 165

‘Nagelkerke Pseudo-R2’ model, 231 National Statistics Offce (NSO), 206 Natural hedging techniques, 218 Negative shocks, 149–150 Neuroaccounting, 84–86 Neuromarketing, 86 Non-accounting decisions, 54 anonymity hypothesis, 56–57 bonus contract hypothesis, 54–56 collapse avoidance hypothesis, 57–58 Non-contractual hedging techniques, 218 Non-normality of data collected for variables under study, 254 Non-parametric statistical hypothesis tests, 229n11 Non-zombie frms, 45 “One-share, one-vote” principle, 119 One-way Mann–Whitney-U hypothesis tests, 229 Ordered logit model, 230–231 Organizational culture, 80 Pakistan Stock Exchange (PSX-100), 148 Parsimonious logit model, 231 Phillips–Perron unit root test (PP unit root test), 133 PMJDY scheme, 92 Policy resolutions, 62 uncertainty, 22–23 Positron emission tomography (PET), 85 PPE to sales Ratio (PPES Ratio), 223 Proftability ratios, 163–164 Property plant and equipment (PPE), 222 Proportional odds model (POM) (see also Structural equation modelling (SEM)), 229–231 for EV and CFV, 255

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

Index    291 for Maltese EV, 232 for Maltese operational CFV, 232

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Re-fnance, 38 Real effective exchange rate index (REER), 25 Real gross domestic product (RGDP), 66 Reference serial matrix, 166 Registrar of Companies (RoC), 230 Risk, 218 management, 218 SAW method, 162 Say-on-Pay reforms, 123 Schwarz Bayesian data criteria (SBC), 70 Schwarz criteria (SC), 133–134 Seemingly Unrelated Regression method (SUR method), 71 Shareholder activism, 118–120, 123–124 corporate governance models and shareholder activism, 123–124 hedge fund activism, 120–122 “just vote no” campaigns, 123 say-on-pay, 123 shareholder proposals, 122 types, 120 Shareholder proposals, 122 Social integration, 208–210 Social isolation, 199 Social wellbeing, 197 Society, 92, 96–97 Steady state typography (SST), 85 Stock returns, 163 Street economy, 260–261 in sustaining urban life esthetics and ergonomics, 261–263 Structural equation modelling (SEM), 3, 7, 9 models and results, 12–15 results for EU-10 countries, 18–19 Structural VAR analysis (SVAR analysis), 64

Sustainable economic development, 274–275, 277 Tax, 276 administration, 275 changes in fscal system in Kosovo, 278–279 in EU, 277–278 harmonization, 276 neutrality, 276 reforms, 275 revenues, 274 system, 275 Taxation, 275–276, 282 Theory of ‘Psychological Contract’, 199 Threshold VAR method (TVAR method), 23–25 Toda–Yamamoto causality test, 68, 70–72 method, 64–65 Todim Fuzzy method, 162 TOPSIS method, 161, 162, 165, 168–171 Traditional accounting, 79 Traditional fnance theory, 128 Turkey behavioral accounting research in, 83–84 Turkish economy, 65, 129 Uncertainty, 22, 24, 218 destabilizing fnancial markets, 143 United Nations High Commissioner for Refugees (UNHCR), 7 Urban culture, 269 Urban life esthetics, street economy in sustaining, 261–263 US Economic Policy uncertainty impact on developing countries data and methodology, 24–25 empirical results, 25–33 literature, 23–24 policy uncertainty, 22–23 US stock markets, 148

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing

292   Index Vector autoregression model (VAR model), 63–64, 134, 148, 150, 153 Vector error correction model (VECM), 134–135 Volatility clustering, 223 Volunteering, 209

Zivot and Andrews unit root test, 68 with Zivot–Andrews structural breakup, 69–70 Zombie frms, 38–39 literature review for, 39–42 methodology, 42–44 results, 44–46

Copyright © 2020. Emerald Publishing Limited. All rights reserved.

Zero hypothesis, 134 Ziraat Bank, 161–162

Uncertainty and Challenges in Contemporary Economic Behaviour, edited by Ercan Özen, and Simon Grima, Emerald Publishing