Working Time and Workers' Preferences in Industrialized Countries: Finding the Balance

Working Time and Workers’ Preferences in Industrialized Countries Finding the balance

Globalization, demographic changes, increasing numbers of women in the labour market: all these changes have resulted in working time arrangements which often look vastly different than they did several decades ago. These changes have significant ramifications both economically and sociologically. This volume examines the changing nature of working time in industrialized countries. It analyses many aspects of this intriguing sphere including: • • •

recent trends in working time legislation; working time trends in specific enterprises; the effects of these changes on ‘work–life balance’.

With its insightful contributions into labour and the work-related arena, this volume will greatly interest those researchers and academics working on labour economics, employment issues and more specifically working time. The policy ramifications that this book uncovers mean that it will also be a useful reference for policy-makers. Editor Jon C. Messenger is a Senior Researcher with the Conditions of Work and Employment Programme at the International Labour Office, Geneva. Contributors Dominique Anxo is Director of the Centre for European Labour Market Studies at the University of Göteborg, Sweden. Colette Fagan is a Senior Lecturer at the University of Manchester. Sangheon Lee and Deirdre McCann are with the Conditions of Work and Employment Programme at the International Labour Office, Geneva.

Routledge studies in the modern world economy

1 Interest Rates and Budget Deficits A study of the advanced economies Kanhaya L. Gupta and Bakhtiar Moazzami 2 World Trade after the Uruguay Round Prospects and policy options for the twenty-first century Edited by Harald Sander and András Inotai 3 The Flow Analysis of Labour Markets Edited by Ronald Schettkat 4 Inflation and Unemployment Contributions to a new macroeconomic approach Edited by Alvaro Cencini and Mauro Baranzini 5 Macroeconomic Dimensions of Public Finance Essays in honour of Vito Tanzi Edited by Mario I. Blejer and Teresa M. Ter-Minassian 6 Fiscal Policy and Economic Reforms Essays in honour of Vito Tanzi Edited by Mario I. Blejer and Teresa M. Ter-Minassian 7 Competition Policy in the Global Economy Modalities for co-operation Edited by Leonard Waverman, William S. Comanor and Akira Goto 8 Working in the Macro Economy A study of the US labor market Martin F.J. Prachowny 9 How Does Privatization Work? Edited by Anthony Bennett 10 The Economics and Politics of International Trade Freedom and trade: volume II Edited by Gary Cook 11 The Legal and Moral Aspects of International Trade Freedom and trade: volume III Edited by Asif Qureshi, Hillel Steiner and Geraint Parry

12 Capital Markets and Corporate Governance in Japan, Germany and the United States Organizational response to market inefficiencies Helmut M. Dietl 13 Competition and Trade Policies Coherence or conflict Edited by Einar Hope 14 Rice The primary commodity A. J. H. Latham 15 Trade, Theory and Econometrics Essays in honour of John S. Chipman Edited by James C. Moore, Raymond Riezman, James R. Melvin 16 Who Benefits from Privatisation? Edited by Moazzem Hossain and Justin Malbon 17 Towards a Fair Global Labour Market Avoiding the new slave trade Ozay Mehmet, Errol Mendes and Robert Sinding 18 Models of Futures Markets Edited by Barry Goss 19 Venture Capital Investment An agency analysis of UK practice Gavin C. Reid 20 Macroeconomic Forecasting A sociological appraisal Robert Evans 21 Multimedia and Regional Economic Restructuring Edited by Hans-Joachim Braczyk, Gerhard Fuchs and Hans-Georg Wolf 22 The New Industrial Geography Regions, regulation and institutions Edited by Trevor J. Barnes and Meric S. Gertler 23 The Employment Impact of Innovation Evidence and policy Edited by Marco Vivarelli and Mario Pianta 24 International Health Care Reform A legal, economic and political analysis Colleen Flood 25 Competition Policy Analysis Edited by Einar Hope 26 Culture and Enterprise The development, representation and morality of business Don Lavoie and Emily Chamlee-Wright

27 Global Financial Crises and Reforms Cases and caveats B. N. Ghosh 28 Geography of Production and Economic Integration Miroslav N. Jovanovic´ 29 Technology, Trade and Growth in OECD Countries Does specialisation matter? Valentina Meliciani 30 Post-Industrial Labour Markets Profiles of North America and Scandinavia Edited by Thomas P. Boje and Bengt Furaker 31 Capital Flows without Crisis Reconciling capital mobility and economic stability Edited by Dipak Dasgupta, Marc Uzan and Dominic Wilson 32 International Trade and National Welfare Murray C. Kemp 33 Global Trading Systems at Crossroads A post-Seattle perspective Dilip K. Das 34 The Economics and Management of Technological Diversification Edited by John Cantwell, Alfonso Gambardella and Ove Granstrand 35 Before and Beyond EMU Historical lessons and future prospects Edited by Patrick Crowley 36 Fiscal Decentralization Ehtisham Ahmad and Vito Tanzi 37 Regionalisation of Globalised Innovation Locations for advanced industrial development and disparities in participation Edited by Ulrich Hilpert 38 Gold and the Modern World Economy Edited by MoonJoong Tcha 39 Global Economic Institutions Willem Molle 40 Global Governance and Financial Crises Edited by Meghnad Desai and Yahia Said 41 Linking Local and Global Economies The ties that bind Edited by Carlo Pietrobelli and Arni Sverrisson 42 Tax Systems and Tax Reforms in Europe Edited by Luigi Bernardi and Paola Profeta

43 Trade Liberalization and APEC Edited by Jiro Okamoto 44 Fiscal Deficits in the Pacific Region Edited by Akira Kohsaka 45 Financial Globalization and the Emerging Market Economies Dilip K. Das 46 International Labor Mobility Unemployment and increasing returns to scale Bharati Basu 47 Good Governance in the Era of Global Neoliberalism Conflict and depolitization in Latin America, Eastern Europe, Asia and Africa Edited by Jolle Demmers, Alex E. Fernández Jilberto and Barbara Hogenboom 48 The International Trade System Alice Landau 49 International Perspectives on Temporary Work and Workers Edited by John Burgess and Julia Connell 50 Working Time and Workers’ Preferences in Industrialized Countries Finding the balance Edited by Jon C. Messenger

Working Time and Workers’ Preferences in Industrialized Countries Finding the balance Edited by Jon C. Messenger

First published 2004 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Simultaneously published in the USA and Canada by Routledge 270 Madison Ave, New York, NY 10016 Routledge is an imprint of the Taylor & Francis Group This edition published in the Taylor & Francis e-Library, 2005. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” © 2004 International Labour Organization All rights reserved. No part of this book may be reprinted or reproduced or utilized in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the copyright holders. The designations in ILO publications, which are in conformity with United Nations practice, and the presentation of material therein do not imply the expression of any opinion whatsoever on the part of the International Labour Office concerning the legal status of any country, area or territory or of its authorities, or concerning the delimitation of its frontiers. The responsibility for opinions expressed in signed articles, studies and other contributions rests solely with their authors, and publication does not constitute an endorsement by the International Labour Office of the opinions expressed in them. Reference to names of firms and commercial products and processes does not imply their endorsement by the International Labour Office, and any failure to mention a particular firm, commercial product or process is not a sign of disapproval. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data A catalog record for this book has been requested ISBN 0-203-34247-X Master e-book ISBN

ISBN 0-415-70108-2 (Print Edition)

Contents

List of figures List of tables List of boxes List of contributors Preface Acknowledgements

Introduction: working time in industrialized countries

xi xiii xv xvi xviii xxi

1

DOMINIQUE ANXO, COLETTE FAGAN, DEIRDRE MCCANN, SANGHEON LEE AND JON C. MESSENGER

1

Regulating working time needs and preferences

10

DEIRDRE MCCANN

2

Working-hour gaps: trends and issues

29

SANGHEON LEE

3

Working time patterns among industrialized countries: a household perspective

60

DOMINIQUE ANXO

4

Gender and working time in industrialized countries

108

COLETTE FAGAN

5

Working time at the enterprise level: business objectives, firms’ practices and workers’ preferences JON C. MESSENGER

147

x 6

Contents Implications for working time policies

195

DOMINIQUE ANXO, COLETTE FAGAN, SANGHEON LEE, DEIRDRE MCCANN AND JON C. MESSENGER

Bibliography Index

212 228

Figures

2.1 2.2 2.3 2.4 2.5 2.6 2.7

2.8 3.1 3.2 3.3 3.4 3.5 3.6

3.7 3.8 3.9 3.10 3.11 3.12 3.13 3.14

Annual working hours in selected countries, 1999, total employment Effects of changes in legal standards Effects of negotiated changes Effects of changes in regulation techniques Trends in excessive hours Involuntary part-time work in selected countries, 2000 Relationship between part-time employment and involuntary part-time employment, 2000, selected OECD countries Trends in involuntary part-time work: UK vs the Netherlands Distribution of couples by working arrangements, all couples, 18–64 years old Gender disparities in employment rates, male employment rates and female rates Employment rates, married and cohabitant households, fathers and mothers Lone mothers’ employment rates and differential between cohabitant mothers and lone mothers Average weekly working time, couples, dependent employees Gender differential in weekly working time in single household, and between mothers and lone mothers, dependent employees Working time patterns, males, singles Working time patterns, females, singles Working time patterns, males, dual earners Working time patterns, females, dual earners Working time patterns, lone mothers Working time patterns among dual earners, Australia Working time patterns among dual earners, UK Working time patterns among dual earners, USA

30 33 36 38 42 52

55 55 65 71 73 74 77

78 79 80 81 82 83 84 85 86

xii 3.15 3.16 3.17 3.18 3.19 3.20 3.21

4.1 4.2

Figures Working time patterns among dual earners, France Working time patterns among dual earners, Italy Working time patterns among dual earners, Germany Working time patterns among dual earners, the Netherlands Working time patterns among dual earners, Sweden Impact of young pre-school children on labour participation and working time, singles Impact of young pre-school children on labour participation and working time, cohabiting and married females The incidence of long weekly working hours in selected industrialized countries Working time policy objectives and elements to promote gender equality in employment and work–life balance

87 88 89 90 91 94

95 115 137

Tables

2.1 2.2 2.3 2.4 3.1 3.2 3.A1 3.A2 3.A3

3.A4

3.A5 4.1 4.2 4.3 4.4 4.5 4.6 4.7

Gaps in working hours, 1998, EU countries 40 Marginal part-time work: incidence and changes 48 Composition of marginal part-time work 50 Involuntary part-time work and the quality of work 53 Employment rates by household types and gender, 18–64 years old 70 Weekly working hours among couples and single households, dependent employees, 18–64 years old 75 Mean values for LIS samples of singles, individuals between 18–64 99–100 Mean values for LIS samples of couples where the age of individuals is between 18–64 101–2 Children and disposable income: percentage disposable income differentials, single households, baseline; no children and age 18–30 103 Children and disposable income: percentage income differentials, married and cohabitant households, baseline; no children and age 18–30 103 Tobit with selection, dependent variable hours of work 104–5 The usual weekly hours of work of the employed by gender in selected industrialized countries 112–13 Mean weekly hours in selected industrialized countries 114 The usual weekly volume of hours of the employed, by occupational group and gender, EU-15 119 Weekly work schedules of the employed in the EU-15 by occupation and gender 120–1 Work schedules of full-time wage and salary workers in the USA, by occupation and gender, 2001 122 Fixed and variable start and finishing times, EU-15 127 Flexible schedules of full-time wage and salary workers in the USA, by occupation and gender, 2001 128

xiv Tables 4.8

4.9 4.10

Employed men’s and women’s preferences for working time adjustments by their current working hours, EU-15 plus Norway The volume of actual and preferred hours of the employed and ‘job seekers’, EU-15 plus Norway Employed men’s and women’s assessment of the compatibility of their working hours with family and other commitments, EU-15

131 132

134

Boxes

5.1 5.2 5.3 5.4 5.5 5.6 5.7 5.8

Long hours of work in the Australian mining industry Long hours of work for professionals and managers Very short hours of work: ‘zero hours contracts’ ‘Involuntary’ part-time work in the French retailing industry The use of annualized hours in seasonal industries ‘Net annual working time’: a new model for standard hours Maximum flexi-time: discretionary working hours systems in Japan Meeting workers’ preferences over the life cycle: flexible options for older workers in the Netherlands

159 162 166 168 175 176 182 184

Contributors

Dominique Anxo is Director of the Centre for European Labour Market Studies and Professor in Economics at the Department of Economics, University of Göteborg, Sweden. His research interests fall broadly into the areas of labour economics, industrial relations, gender and time-use studies and evaluation of employment policy. He has, in the last few years, been involved in labour market research at both the national and international level and has acted on several occasions as expert and consultant for the Swedish Government as well as for the European Commission and the ILO. Colette Fagan is a Senior Lecturer in Sociology and Co-director of the European Work and Employment Research Centre, at the University of Manchester, UK. Her research interests focus on gender relations and employment, with particular interests in working time and international comparative analyses. Her publications include Part-time Prospects: An International Comparison (edited with Jackie O’Reilly, 1998, Routledge), Gender, Employment and Working-time Preferences in Europe (2001, Office for Official Publications of the European Communities) and Gender, Jobs and Working Conditions in Europe (with Brendan Burchell, 2002, Office for Official Publications of the European Communities). Sangheon Lee is an economist and Research Officer with the International Labour Office, Conditions of Work and Employment Programme in Geneva. His research is focused on the issues of working time, work organization and, more broadly, personnel economics. His publications include ‘Seniority as an employment norm: the case of layoffs and promotion in the US employment relationship’ (Socio-Economic Review, forthcoming). He holds a PhD in economics from the University of Cambridge. Deirdre McCann is a labour lawyer and Research Officer with the International Labour Office, Conditions of Work and Employment Programme in Geneva. Her research focuses on the regulation of working

Contributors xvii time, non-standard working arrangements and workplace violence and harassment. She holds a DPhil in law from the University of Oxford. Jon C. Messenger is a Senior Research Officer with the International Labour Office, Conditions of Work and Employment Programme in Geneva, with the lead responsibility for its sub-programme on working time and work organization. He specializes in policy research on working time and work organization, with a particular interest in issues relating to working time flexibility and non-standard work arrangements. Prior to joining the ILO, he worked at the US Department of Labor (USDOL) in Washington, DC, for fifteen years, where he served as Team Leader for Research on employment and training.

Preface

From an American perspective, until very recently working time never seemed to be more than an afterthought in discussions of labour issues and labour market policies. Even now, with changes to US overtime regulations in the news, the focus is not on the number of hours that people work, but rather how much they will be paid for working those hours. Indeed, to the extent that we in the USA considered working time in policy development at all, it was as one potential impact of successful policies to increase employment and earnings. For example, if hours or weeks worked in paid employment increased for a particular group of workers, from an American perspective, that was nearly always regarded as an unalloyed good – a result to be applauded and even replicated if possible. The idea that more working time could be a problem was, well, something that a handful of academics (most notably, Juliet Schor, author of The Overworked American (1992) and other volumes) were concerned about, but it was certainly not considered to be an important matter for public policy. My immersion in the voluminous literature on working time in European Union (EU) member states has not only opened my eyes to ‘classic’ concerns about working time, such as excessive hours of work and its effects on health and safety, but also led me to question other aspects of working time as well. Why, for example, were so many individuals reporting that they felt a ‘time squeeze’ when data on hours of work seemed to indicate few, if any, changes in average weekly hours of work? While the full answer is a complex one and is treated in more detail elsewhere (see, for example, Jacobs and Gerson 2000), the basic answer is quite simple: women’s labour force participation is much higher than it was in most industrialized countries even 20 years ago, and that means that each family unit or ‘household’ is working more hours, even if the working hours of individuals had remained unchanged. In fact, however, the working hours of individuals weren’t just standing still, as data on average hours of work seemed to indicate. Some workers were working much longer hours, while others were working fewer hours as part-time work expanded – even more so in many EU member states,

Preface xix most notably the Netherlands. In other words, the assumption of a relatively uniform or ‘standard’ working week for most (if not all) nonmanagerial workers was no longer valid; an increasing diversity in the workforce was being complemented by an increasing diversity in the hours being worked. And this growing diversity in hours worked was not limited just to the number of hours worked. Other factors, including advances in information and communications technologies and new patterns in consumer demand for goods and services (often referred to by the shorthand, the ‘24-hour economy’), were also changing the timing of when workers needed to be available for work. If the workforce was becoming increasingly diverse, and the hours being worked were also diversifying, then what did these developments mean for the cherished ‘40-hour workweek’ in the USA or its equivalent in other industrialized countries? Did these sorts of common standards for working time make sense anymore when workers’ individual situations were becoming less and less common? We at the ILO thought that it made sense to consider these questions by looking at what workers say that they want in terms of their working time – their ‘preferred’ hours of work, so to speak – compared with their actual situations – that is, the days and times when they are currently working. This approach allows us to identify ‘gaps’ between workers’ ‘preferred’ hours of work and their current hours. For example, some workers may want to work only 40 hours a week, but have to work 50 hours given their workload; others may want to work only during weekdays, but they are required to work weekends, or vice versa. In taking this approach, we have followed in the footsteps of the pioneering work on workers’ needs and preferences regarding working time by the European Foundation for the Improvement of Living and Working Conditions, to whom we acknowledge a debt of gratitude. However, we have also extended this broad conceptual approach, not only to investigate additional countries within the industrialized world, but also by considering working time from a number of different viewpoints: the national level; the enterprise level; and also from two unique perspectives – those of households and gender. Most importantly, this report provides an international comparative analysis of working time in the truest sense of the term; the individual chapters do not present national cases; rather, each chapter adopts a comparative perspective covering most of the major industrialized countries. Though different researchers author five of the six main chapters, this book is the result of a team effort. Indeed, the Introduction to the report and also its concluding chapter, ‘Implications for working time policies’, were the fruit of a two-day intensive discussion among all of the contributors and are truly the product of the entire team. In conclusion, we are pleased to offer you this report on working time in industrialized countries, as seen through the unique lens of workers’ needs and preferences regarding their working time. I should emphasize,

xx

Preface

however, that this monograph, while important in its own right, is but one part of a broader ILO research effort on working time involving a series of studies, including studies in the developing world and in those countries still making the transition to market economies. Thus, this study can also be seen as an important first step towards an analysis of working time around the world. Jon C. Messenger Senior Research Officer International Labour Office Geneva

Acknowledgements

I would like to offer my thanks to each of the working time colleagues who contributed to this report, Dominique Anxo, Colette Fagan, and my working time colleagues at the ILO, Sangheon Lee and Deirdre McCann. Their efforts are greatly appreciated, not only for their excellent work on their own individual chapters of this volume, but also for their creative ideas and input into the Introduction of the report and the concluding chapter on ‘Implications for working time policies’. It was a truly collaborative team effort, and the final product is greatly enriched as a result. I also want to express my appreciation to a number of individuals in the International Labour Organization who also made important contributions to this report. First, I would like to thank François Eyraud, Director of the ILO Conditions of Work and Employment Programme, for his support and encouragement throughout the long process of bringing this publication to final fruition. Without his strong support, I am certain that this report would never have been published. I would also like to thank Jesica Seacor, Alison Irvine, Charlotte Beauchamp, and Rosemarie Beattie of the ILO Publications Office for their invaluable advice regarding publication matters. And I am particularly grateful to my colleagues Mariela Dyrberg and Pernilla Melin, who deserve special thanks for all of their assistance with editing, formatting, and proofreading the manuscript. Several members of the Routledge team also deserve thanks, including Terry Clague and Gail Welsh, all of whom have helped to move this manuscript through the various stages of the publication process. Finally, I would like to thank my wife, Laura, for her exceptional understanding and patience when I did not always ‘find the balance’ in my personal working time!

Introduction Working time in industrialized countries Dominique Anxo, Colette Fagan, Deirdre McCann, Sangheon Lee and Jon C. Messenger Working time changes and challenges Working time issues have been at the heart of political and social debates since the Industrial Revolution. The focus of the debate, however, has changed over time. After the First World War, a number of legislative measures were introduced in order to regulate working time. The main objective of these initial laws on working time, which introduced the goal of an 8-hour working day, was to combat the adverse effects of long working days on employees’ mental and physical health in order to diminish the high numbers of industrial accidents and to standardize employers’ practices as regards working time. During the period of economic prosperity following the Second World War, as working conditions began to improve and incomes started to grow, there was a change in the focus of the debate on working time. In particular, the concerns about working time and health widened to include more general welfare issues, i.e. the distribution and trade-off of productivity gains and economic growth between increased income and/or leisure. In the context of full employment and sustained growth, most industrialized countries experienced a substantial reduction of actual working time. The rising imbalances and slackening-off of growth brought about by the first oil crisis in the early 1970s led to a lively debate in a number of European countries about ways of reducing unemployment by a general reduction in working time. In some European countries, like France and Germany, trade union organizations and governments took a favourable attitude towards a reduction of working time, hoping for net job creation or at least to preserve existing jobs in order to curb rising unemployment (Anxo and O’Reilly 2002). In addition, while working time reduction was being discussed in relation to unemployment in some countries and sectors, in many cases the focus was on the use of working time flexibility rather than reductions as a means of stimulating economic growth and job creation. This marked the beginning of the big ‘flexibility debate’ that has dominated much of the activity around working time policies since the early 1980s.

2

D. Anxo et al.

During the past few decades in most industrial societies, the historical trends towards a progressive standardization of working time have given way to a diversification, decentralization and individualization of working hours. This is the result of pressures from a number of sources over this period. First, profound changes in household and demographic structures, and in particular, the increased feminization of the labour force and the related shift from the single male breadwinner household to dual earner households have created new needs and also new challenges for patterns of working time and household composition. The process of globalization and the resulting intensification of competition, the associated development in information and communications technologies, and new patterns of consumer demand for goods and services in the ‘24-hour economy’ have had a great impact on production methods and work organization. Modifications in consumer behaviour and product diversification have meant that an increasing number of enterprises have gradually abandoned traditional Taylorist methods of mass production. The introduction of new methods of flexible production (just-in-time, lean production, etc.) has been accompanied by a gradual abandonment of traditional ways of adjusting employment and by a much more flexible organization of work and working time, and also by an increased variation in the hours that people work. These changes often make the conventional weekly standard of working time obsolete, and thus are forcing enterprises to rethink the ways in which they structure working time. For example, in a growing number of enterprises, strict control of working hours is being replaced by performance monitoring. These modifications in working time appear to give employees much more freedom in choosing their working hours, in that more diversified forms of working time appear to be on offer. However, it should be borne in mind that the development of some forms of working time also seems to exacerbate the duality between the primary and secondary segments of the labour market, as well as increasing gender segregation. In particular, certain forms of flexible working time are often associated with lower pay levels and with less stable employment relationships. Thus, in parallel with the upturn in flexible forms of working time, there has also been an increase in fixed-term and other temporary forms of employment contracts. These trends reflect the transition from a relatively standardized structure of work organization and working time patterns to more complex and more diversified structures (Anxo and O’Reilly 2002). Second, these working time developments have also meant an increase in long and ‘unsocial’ working hours (i.e. work in the evenings, at nights and on weekends) for increasing numbers of workers, particularly in those countries that are in transition towards a ‘24-hour society’. While some workers may actively seek out long or ‘unsocial’ hours of work, many others may have little choice but to follow these working time patterns if they are the only ones which are on offer to them. Third, in a number of

Introduction 3 economic sectors (e.g. the retail trade), tensions have emerged over the extension of opening hours between the interests of individuals as workers and individuals in their role as consumers. Over the same period, on the supply side, individuals have increasingly sought working time schedules which allow them to combine employment with other activities, most often with domestic and care responsibilities. This development has emerged primarily from the ‘feminization’ of the labour force, as women’s increasing participation in paid work has ushered in a shift from male breadwinner to dual-earner households. Most visibly, part-time work has expanded across the industrialized world. In many countries, this expansion has accounted for a significant proportion of the employment of women, while in some countries there is also an expanding part-time workforce of students and older persons approaching retirement. The result of this increase in part-time work for some workers and the lengthening of the working week for others has been a general trend towards the polarization of working hours. Nonetheless, the primary objectives of traditional measures on working time – the concern for health and safety and the preservation of ‘leisure’ time – have remained prominent in regulatory measures which limit working hours. However, beginning in the 1980s when ‘flexibility’ began to dominate the discourse on working time policies, these types of measures were increasingly accompanied by those which facilitate the diversification of working time. Depending on the legal and regulatory framework of the particular country, these measures can be initiated through legislative changes, by way of collective bargaining, at the individual level, or as a combination of all of these techniques. Their common goal, however, is the removal or liberalization of restrictions on unsocial hours and on the variation of working hours.

Needs and preferences: issues and approaches These changes in enterprises’ working time arrangements, in conjunction with labour supply changes – particularly the integration of an increasing number of women into paid work – are contributing to a growing concern about conflicts between paid employment and individuals’ personal responsibilities (e.g. caring for family members), and point to the increasing importance of assisting workers to better balance work and personal life. As Supiot (2001: 84) notes, it needs to be acknowledged that ‘time must be envisaged not only as working time, as a measure of the exchange of work for pay, but also as a subjective experience, that is to say, as time in workers’ lives’. As a consequence, insofar as working time tends to be increasingly heterogeneous and individualized, it is important to try and establish the extent to which such developments reflect the needs and preferences of individual workers. This concern with heterogeneity in working time arrangements and the

4

D. Anxo et al.

working time needs and preferences of workers is in many ways a departure from the traditional approach to working time regulation, but this departure is an elaboration rather than a radical break. From the outset of the Industrial Revolution, working time regulation has always been rooted in concerns about health and safety and the preservation of ‘free time’ – time away from employment. Traditionally, the predominant concerns have been the negative impacts of long hours, night work and certain shift patterns on workers’ health and safety. Related public safety issues have also been drawn into the debate, for example, in relation to the working hours of transport workers. The need to preserve ‘free time’ has also been present since the earliest campaigns for working time reductions, initially expressed in terms of the need for time for physical recuperation and for ‘moral development’ through religious instruction,1 and subsequently elaborated to include time for leisure and family life, most recently expressed in terms of the ‘work–life balance’. Hence, the core focus remains the same – to promote health and safety and work–life balance – but the issues involved have expanded and thus require a shift in perspective. In particular, the regulatory measures that developed under the traditional approach to working time regulation emphasized homogeneity. These measures were largely designed in relation to one ‘ideal type’ of worker: men working full-time in primary or manufacturing activities, who were implicitly assumed to have fairly homogeneous needs and preferences. By comparison, women and children were precluded from the labour required by the ‘ideal type’ worker through protective legislation that often included limits on their working hours. It is the diversification in working time arrangements found in enterprises, in conjunction with more diversity in the workforce – or at least an awareness of this diversity – (by age, care responsibilities for children and elderly parents, etc.) that makes it imperative to incorporate individuals’ working time needs and preferences into debates about working time regulation. These new concerns and challenges were echoed in several ILO discussions as early as the 1970s (Evans 1975; Maric 1977). In particular, while recognizing the changes in the social and human aspects of working hours, it was suggested that: [the] problem of working hours goes beyond the setting of statutory limits and involves also the scheduling and distribution of hours in accordance with two principles, i.e. a relaxation of standard patterns and a degree of freedom of choice, accepted by society and regarded as basic to job satisfaction. (Maric 1977: 4) However, it is only fairly recently that workers’ needs and preferences began to be seriously considered by enterprises and in government policies. For example, some companies began to realize the value of introduc-

Introduction 5 ing quality part-time work so as to induce qualified women with family responsibilities to enter or continue in paid employment. As is discussed by Messenger in Chapter 5, a better consideration of workers’ needs and preferences can be part of a successful management strategy for enhancing firms’ competitiveness. Working time policies in many industrialized countries have also demonstrated creative and innovative ways of better accommodating workers’ needs in achieving different economic and social goals. In some countries, for example, workers have a statutory right to reduce their working hours to part-time and to resume full-time hours at a later period while remaining in their current job. For example, this option has been available in the Swedish parental leave system for many years, and recently a number of countries, including the Netherlands, have introduced a legal entitlement to request part-time hours in order to meet their personal needs. Another example is Germany, which has been a pioneer in increasing flexibility in organizing working time, such as flexi-time and time banking schemes – initiatives that have been developed largely through collective bargaining. The French laws establishing a 35-hour workweek have also encouraged workplace negotiations, so as to ensure that workers’ needs and preferences are appropriately considered in the process of implementing this reduction in working time. Alternatively, the UK provides an example in which companies’ voluntary initiatives are emphasized, for instance, through the ‘work–life balance’ campaign. A similar situation exists in the USA, where flexi-time has become increasingly widespread in individual enterprises. In Japan, the government has been developing guidelines to improve the quality of part-time work so as to create a better environment for voluntary take-up of part-time work. Finally, the developments in different EU countries regarding working time policies have recently been consolidated in the EU-level initiatives concerning indicators on the quality of work, with the suggestion that workers’ ability to combine working and non-working life is one of the key dimensions of quality of work (European Foundation 2002). However, any analysis of workers’ needs and preferences regarding working time is a difficult proposition and should be undertaken with great care. In fact, different views have been expressed about how workers’ preferences on working time should be analysed. In economic theory, workers’ preferences are important elements in determining their labour supply, and, more importantly, it is often assumed that workers’ preferred working time corresponds to their actual working time due to the effective role of the labour market in matching workers’ and employers’ preferences (Ghez and Becker 1975). This assumption has often led economists to believe that workers’ preferences can be induced from actual (often called ‘revealed’) preferences, and led them to question the usefulness of any information on workers’ ‘subjective’, self-reported preferences. However, as continuing concerns about ‘under-employment’ suggest, the

6

D. Anxo et al.

belief that workers’ preferred working time corresponds to their actual working time is not well supported by the reality of labour markets in which workers’ preferences regarding their working hours remain unfulfilled. The presence of a mismatch or ‘gap’ between actual and preferred working time – which is clearly demonstrated throughout this volume – points to the need for obtaining information on workers’ needs and preferences, in order to develop better working time policies. Recent EU data on gaps between actual and preferred working time suggests considerable room for policy initiatives regarding the labour market in general and working time in particular (Bielenski et al. 2002; Fagan et al. 2001a). In essence, this information provides an overall indication of policy developments that workers would like to see, and some statistical analyses indicate that information on workers’ current working time preferences can even be used to predict their future working hours (for example, Euwals 2001). Thus, workers’ preferences provide an indication of individuals’ future actions and suggest the types and directions of policy interventions that can help workers to undertake these preferred actions. At the same time, it is also clear that the working time preferences expressed by respondents should be interpreted very carefully, in part because they could differ significantly depending on the structure and wording of working time preference questions in surveys. For example, if a question refers explicitly to wage adjustments that would occur with changes in working hours, respondents are less likely to report preferences for reduced working hours. In addition, the reported preferences tend to be sensitive to the circumstances in which workers work and live (cf. Hakim 2000; Fagan 2001b, 2001a). As Bielenski et al. succinctly note: On the one hand, preferences express individual desires for change; on the other hand, however, these desires are influenced by objective factors within which individuals plan their lives. Thus, preferences are usually compromises between what is desirable and what is feasible. (2002: 16) One often-quoted example in this regard is the observation that mothers tend to prefer shorter hours when public child-care services are not readily available, implying that the extent of child-care services would be an important factor determining mothers’ preferences regarding working time. Other economic and social factors, including the wage structure and tax and benefit systems, are also known to affect working time preferences. At the same time, working time preferences are neither fixed nor static. They are changing across different stages of life, and the historical evidence also demonstrates continuing shifts in working time preferences, as seen in the historical changes in expectations regarding what is considered to be ‘healthy’ working time.

Introduction 7 The interest in workers’ needs and preferences in this report is not designed to resolve these methodological and conceptual issues, but rather to place them in context as a ‘social phenomenon’ and then see how policy measures would be helpful in addressing them. By ‘social’ it is meant to emphasize that the existence of gaps between actual and preferred working time is not a purely individual matter, and achieving success in realizing workers’ needs and preferences requires strong social support. More specifically, while recognizing the roles of various social factors in shaping workers’ preferences, the focus of the analysis in this book is placed on the social mechanisms creating working time gaps – the gaps between workers’ actual and preferred working time. Policy initiatives are therefore aimed not only at reducing the existing working time gaps, but also at broadening the range of working time options available to workers (or ‘what is feasible’), thereby making their choices less constrained and thus more meaningful. This report also takes a cautious approach in estimating the extent of working time gaps. For example, some supplementary objective information, such as the health effects of long working hours and the quality of part-time work, is used in estimating working hour gaps in Chapter 2. Various structural constraints on workers’ preferences are also explicitly considered throughout the book, particularly in Chapters 4 and 5.

Overview of the report The focus on ways of balancing work and life in this report is an important step towards the elaboration of the ILO’s goal of ‘decent work’ in the area of working time. As decent work should be an issue for all and ‘the goal of decent work is best expressed through the eyes of people’ (ILO 2001a: 7), the concept of ‘decent working time’, which is further developed in the conclusion of this report, requires going beyond the vague and abstract concept of the ‘average worker’. The focus should instead be on individual workers’ concerns in different and varying contexts at different stages in the life cycle, as revealed in their daily working lives. For this reason, the report investigates various initiatives taken by governments, employers and workers to ensure that workers’ needs and preferences are reflected in their working time arrangements. These measures represent attempts to more comprehensively integrate family-work and the time demands of care responsibilities, as well as gender equality, into working time policies. And as will be discussed in detail throughout this report, these developments have fostered debates across the industrialized world about the best ways to respond to the heterogeneous nature of working time needs and preferences; to better coordinate work, family and public life; and to balance the needs of consumers and workers. These initiatives have also raised questions about the relationship between collective and individual needs and how they can be balanced in regulatory measures.

8

D. Anxo et al.

Addressing workers’ needs and preferences inevitably involves the issue of how to coordinate different (and often conflicting) needs and preferences among different types of workers. Key factors which can induce significant differences in working time needs and preferences are unquestionably family structure and gender (discussed in Chapters 3 and 4), and coordination can be achieved either at the national level (discussed in Chapter 1) or at the enterprise level (discussed in Chapter 5), yet in many cases may require an effective combination of the two (which is discussed in the concluding chapter, Chapter 6).2 Following the Introduction, in Chapter 1, Deirdre McCann reviews recent trends in working time policies in industrialized countries, outlining different regulatory models and their implications. This chapter then discusses the range of policy goals that underlie recent initiatives, highlighting their inter-relationships and the ways in which they shape and are influenced by the regulatory models within which they are situated. In Chapter 2, Sangheon Lee explains the actual situation of working hours and the existence of ‘gaps’ between actual and preferred working hours. This chapter reviews available data on working hours for individual workers, and identifies two types of ‘gaps’ between actual and preferred working hours – working hour surpluses and working hour deficits – which indicate an increasingly important area for future working time polices. Working time from the perspective of families, using households as the unit of analysis, is investigated by Dominique Anxo in Chapter 3. This chapter investigates the incidence of different types of working time patterns among different types of households (singles, lone parents with children, couples with no children, and couples with children); assesses the main factors affecting the distribution of working time among the different types of households, in the context of different societal characteristics, including different regulatory frameworks; and considers the extent to which the prevailing pattern of working time is meeting household income and welfare needs and preferences. Colette Fagan focuses on the increasingly important gender dimension of working time in Chapter 4. This chapter compares the current pattern of working time arrangements in selected industrialized countries by gender and occupation along the following dimensions: volume of hours, schedule, and autonomy. The working time preferences of men and women are then considered, based on data regarding individuals’ perceptions of their current hours of work and working time arrangements. In Chapter 5, Jon Messenger investigates how the working time trends outlined in the previous chapters have manifested themselves at the enterprise level. This chapter uses enterprise case studies to illustrate the range of working time practices which have emerged in particular industries and firms, while also considering the extent to which such arrangements fit with workers’ needs and preferences regarding working time. In addition, the chapter examines some enterprise cases that illustrate innovative

Introduction 9 working time arrangements in which firms have explicitly attempted to balance business requirements with workers’ needs and preferences regarding working time. Based on the findings and policy suggestions from each of these chapters, the concluding chapter proposes and elaborates the concept of ‘decent working time’, consisting of five key dimensions: healthy working time, ‘family-friendly’ working time, gender equality through working time, productive working time, and choice and influence regarding working time.

Industrialized and developing countries This report covers only industrialized countries. Such limited coverage is primarily related to the existing disparity in hours of work between developing and industrialized countries, which makes it extremely difficult to undertake meaningful comparisons around the world. The gap between these two groups of countries is also reflected in the fact that the issue of workers’ needs and preferences has not taken the central stage of working time debates in many developing countries, while there has been a lively discussion of this issue in industrialized countries. Consequently, developing countries typically have not addressed issues concerning workers’ preferences regarding their working time. However, we expect that those issues discussed in this book will concern some developing countries and countries in transition to market economies in the future, as an increasing number of these countries are becoming interested in different working time patterns such as flexi-time to improve workers’ ability to balance work and family life. It is thus hoped that these countries will also find this book helpful in developing better working time policies. In addition, the ILO is currently investigating working time trends and issues that exist in the developing countries and countries in transition, and the results of this important research will be forthcoming in the near future.

Notes 1 The efforts of the religious organizations and other philanthropic organizations concerned with the moral welfare of the working class were very influential in the early campaigns for working time reductions. 2 We expect that the diversification of working time preferences might make it harder for trade unions to establish ‘collective’ preferences based on which collective negotiations are undertaken. While this issue is not fully discussed in this report, a useful discussion is provided in Ozaki (1999). Nonetheless, it is suggested in this report that, given the changing environments, collective bargaining should pay more attention to the possibility of establishing collective entitlements to some type of a portfolio of working time options, which can better accommodate the heterogeneity of individual working time preferences. See the Conclusion of this report (Chapter 6) for a further discussion of this issue.

1

Regulating working time needs and preferences Deirdre McCann

Introduction The gap between individuals’ working time preferences and their actual working hours highlighted by recent research has always been one of the themes underlying policy approaches towards the regulation of working time. In recent years, however, the need to more strongly reflect workers’ preferences in their work schedules has become more prominent as countries across the industrialized world have embarked on regulatory agendas which seek to realize this goal. This chapter reviews traditional methods of incorporating workers’ needs in working time regulation and highlights the tendency of regulatory trends to undermine this kind of conciliation. It then outlines the range of possible approaches towards responding to workers’ needs, focusing on those which promote individual influence over working hours, and indicating the kinds of questions they raise for the future of working time regulation.

Traditional working time regulation and the impact of flexibilization Late twentieth-century trends in working time regulation represented a break from a history of progressive reduction of weekly working hours, which began in the nineteenth century and was realized through the imposition of limits on normal hours and pay premia additions for work during ‘unsocial’ hours (evenings, nights and weekends) (Bosch et al. 1994; Fagan and Lallement 2000; Supiot 2001). The labour movement’s demand for the 8-hour working day, embodied in the ILO’s first Convention in 1919,1 was gradually translated into the emergence of ‘standard’ working hours which, although the precise norm varied, were common to all industrialized countries by the 1960s (Bosch 1999b; Fagan and Lallement 2000). Indeed, the goal of limiting working hours remained prominent in employee demands and in working time policies until the 1980s (Supiot 1996). These hours reduction policies stemmed partly from a concern for the needs of workers for both health and safety protection and for ade-

Regulating working time needs 11 quate time outside of paid work, usually conceptualized as ‘leisure’ time in the gendered assumption that it would not involve other forms of labour. Although the concern for ensuring adequate non-work time has been to some extent lost as work has seeped into unsocial hours, the long-standing concern for the preservation of workers’ health remains a dominant theme of policy discourses, even if often inadequately realized in substantive measures. Although working time reductions have partly resulted from the introduction of various forms of leave, the postponement of the transition from school to paid work and, in many countries, the shortening of working life through the reduction of the pension age, the need to ensure workers’ health is most strongly reflected in measures which continue in the tradition of imposing limits on weekly working hours. The Preamble to the European Union’s Working Time Directive,2 for example, enunciates the protection of workers’ health as one of its primary aims, a rationale subsequently reinforced by the European Court of Justice in response to the UK government’s contention that the Directive was an employment creation measure in disguise.3 Moreover, the Directive also encourages the social partners to take account of workers’ health in their negotiations on working time, although it does not require them to do so. In fact, the EU Working Time Directive’s 48-hour limit on working hours was already exceeded in most EU member states on its adoption; its provisions on rest breaks and night work have been more influential. Where hours limitations are strong, however, they can contribute to the preservation of workers’ health (Spurgeon 2003), and this traditional policy theme remains prominent in certain countries. In Japan, for example, there is an increasing concern for the overall quality of life in its long-hours culture. The Japanese government has attempted to convince firms to adhere to the Japanese Labour Standards Law’s4 40-hour limit and to discourage employees from working long overtime hours (Japan Institute of Labour 2001). It is also worth noting that, in addition to the concern for preserving workers’ health, a minor theme in traditional working time measures is the recognition of the need for family time. Most notably, prohibitions on night work for women reflected to some extent an, albeit discriminatory, drive to conserve family time, embodied in more modern laws in the shape of limitations on night work which cover all workers (Supiot 2001). In the last two decades, these traditional themes of working time regulatory policy became more muted as the necessity of ensuring labour market flexibility came to dominate policy discussions throughout the industrialized countries. The diversification of working hours championed by flexibility discourses demanded regulatory change of a form intimately related to the working time regime of the country in which they were to be introduced. In the market-driven regimes of the USA and the UK, working time changes have tended to be determined at the individual level, while in Australia the deregulation and decentralization of the

12

Deidre McCann

regulatory regime have overlapped with the promotion of flexibility, in response to employers’ calls for both labour market deregulation and changes in working time arrangements at the workplace level (Campbell 1997). In contrast, in France, the state’s traditional leading role in the regulation of working time has been reflected in the facilitation of flexibility through legislative change. And in Denmark, in which legislation has traditionally been a peripheral feature of the industrial relations system, flexibility has been left in the hands of the social partners and effected through collective bargaining. Despite these differences in regulatory regimes, similar techniques have been adopted to facilitate the diversification of working hours. Transposed into legal form, ‘flexibility’ has involved the relaxation of restrictions on varying and individualizing working time schedules and on work during unsocial hours introduced with the goal of increasing capital utilization and extending opening hours. These legal changes facilitated the kinds of working time arrangements – long hours, compressed work weeks, variable daily shift lengths, part-time work – promoted as essential to productivity and competitiveness. The traditional flexibility tool of extending the hours of the existing workforce, for example, has been easily achieved through the use of overtime hours, since they are usually subject to weak restrictions. But to avoid the costs of overtime premia, the introduction of reference periods over which working time limits can be averaged has been a popular technique in more highly regulated regimes, allowing statutory or collectively bargained maximum limits on working hours to remain applicable, but to be averaged over specified time periods. In the Netherlands, for example, the legislation’s limit of an average of 40 hours per week is calculated over a period of 13 weeks, while the 48-hour limit in the UK regulations is attached to a 17-week reference period. And a number of countries have introduced legislation which includes reference periods as long as a year, including Finland and France. The late twentieth-century quest for working time flexibility to which these kinds of legal measures were a response was substantially employerdriven, although also linked to consumer demands for product diversification and longer opening hours. The needs of workers, the possibility that they might need flexibility to be equated with the ability to discharge domestic obligations or pursue other interests, was not a dominant concern in most countries until very recently. Indeed, although the diversification and individualization of working time patterns have in some cases been compensated for by reductions in hours, and the individual consequences of working time flexibility have varied according to the legal regimes in which they are embedded, the forms of working time flexibility which have been adopted have often conflicted with the goal of ensuring that workers’ needs and preferences are reflected in their work schedules. Most obviously, long hours, especially those that are unpredictable or scheduled at short notice, tend to detract from the ability of workers to

Regulating working time needs 13 organize their domestic and caring responsibilities and to adequately schedule leisure pursuits. And when introduced through the use of reference periods, these long hours tend to be combined with a demand that they be worked during periods which the labour movement had struggled to have classified as ‘unsocial’. Moreover, there are signs that the transition towards working time flexibility may have created environments in which demands can be made for longer hours irrespective of legal limits, exemplified by reports from labour inspectors in France of a growing tendency for firms to demand that their employees be available beyond their scheduled working hours without receiving overtime pay (Supiot 2001). In addition, forms of work which entirely evade the conventional regulation of working time have become more widespread, most obviously those in which individuals are required to work as-and-when they are required, for whom periods during which they are ‘on-call’ are difficult to reconcile with the available definitions of ‘working hours’. Together with the threat of long hours, flexibility has too often reinforced existing labour market disadvantage, particularly when it has taken the shape of part-time working arrangements. These have become synonymous with disadvantageous terms and conditions across most industrialized countries, especially where very short hours are worked (see Lee, Chapter 2), due to part-time workers being disproportionately found in lower-status, low-paid and female-dominated sectors and jobs (Fagan and Lallement 2000; OECD 1994; O’Reilly and Fagan 1998). Not only does this gender-based labour market segregation have implications for benefits and wages, it also appears to influence part-timers’ career patterns, given the evidence that women who transfer to part-time work in order to accommodate child-raising often experience downward occupational mobility, particularly where the transfer required them to change jobs or where it was preceded by an exit from the labour market (Blossfeld and Hakim 1997; Fagan and Lallement 2000). At the same time, since many higher-status jobs are structured to require long or non-standard working patterns, and therefore preclude an extensive involvement in family and domestic work, many women are excluded from male-dominated sectors and occupations, while men’s more limited involvement in the domestic arena is reinforced and their ability to devote a significant amount of time to their families and non-work interests is curbed (see Fagan, Chapter 4). There is particular concern for part-time workers in countries in which flexibility has been viewed primarily as a method of reducing labour costs. In these jurisdictions, regulatory regimes establish hours-thresholds for social security contributions and do not mandate pay premia for, or impose limits on, the overtime hours worked by part-timers. Under these kinds of regimes – which have included those in Austria, Germany, Ireland and the UK (European Commission’s Group of Experts 1998) – the incentive structure underlying the system of social security contributions has encouraged employers to offer short-hours contracts. Moreover,

14

Deidre McCann

in many countries, the combined impact of social security, taxation and working time regimes has tended to favour an asymmetrical distribution of hours between married men and women, under which part-time work is performed by a woman married to a male full-timer, reinforcing traditional gender roles in the division of paid and household labour (European Commission’s Group of Experts 1998; Figart and Mutari 2000). This ‘one-and-a-half earner’ household model is partly attributable to both incentives for employers, stemming from the way in which labour costs are structured under social security and working time regulations, and incentives for workers derived from both taxation regimes and rules of entitlement to social security benefits. To illustrate, in Japan, part-time work has been encouraged by the provision of significant tax incentives to both businesses who hire part-time workers and to the workers themselves (Houseman and Osawa 2000; Kezuka 2000). Married part-timers who earn less than a specified limit are regarded as dependants of their spouses, and are not required to pay social security and health insurance premiums while remaining entitled to receive health insurance and a basic pension. These regulatory regimes, then, operate in tandem to create a financial incentive for married women to work on a short-hours basis, and a significant proportion of this group of workers have been found to adjust their hours to this end.5 Labour law regimes have contributed to the disadvantageous working experience of part-time work through their differential treatment of parttimers with respect to entitlement to employment rights, combined with their failure to prohibit the exclusion of part-time workers from, or their disadvantageous treatment under, employer-provided benefits. As a result, throughout the industrialized world, benefits such as pay increments, training, promotion and pension rights have been awarded primarily to those who work on a full-time basis. In Japan, for example, firms are not legally required to provide health insurance to employees who work for less than three-quarters of full-time hours (Houseman and Osawa 2000). And in the United States, the rules which govern both pension provision and employer-provided health insurance schemes both contain hours-thresholds for eligibility (Houseman and Osawa 2000; Lo 1996). Houseman and Osawa (2000) have found that, even controlling for demographic and job characteristics, workers in all kinds of ‘non-standard’ jobs in the USA, including part-time workers, are significantly less likely than permanent full-time workers to be eligible for membership in employer-sponsored health insurance and pension plans. Even in highly regulated labour markets, however, part-time workers have been subject to lower levels of legislative protection. In Denmark, for instance, until relatively recently, a minimum number of working hours was required for entitlement to certain legislated employment rights, including notice periods, sick pay and maternity pay, although collective agreements tended to ensure comparable protection for excluded workers.

Regulating working time needs 15

Recent trends: integrating workers’ needs and preferences In recent years, the primacy of employer-oriented flexibility has dwindled to some degree as conventional flexibility concerns have been accompanied by a growing recognition that workers’ needs and preferences are not being reflected in their working time arrangements. This transition towards a recognition of workers’ interests has been reflected in the emergence of regulatory measures geared towards ensuring that workers’ preferences can have at least some degree of influence on their working time. These instruments complement the traditional protective measures discussed above and tend to be aimed either at facilitating the combination of paid and unpaid labour or allowing individuals a degree of influence over the amount and distribution of their working hours. Among the most significant aspects of this trend towards integrating workers’ needs and preferences into their working arrangements is the increasing recognition in working time policy of the interrelationship between the workplace and family life. This perspective sharply contrasts with that which underlies the main thrust of traditional working time laws, which have tended to be grounded in ‘a conception of the workplace as a discrete and bounded sphere of social and economic activity in which its participants are fully and exclusively engaged’ (Conaghan 2002: 55–6). This conventional understanding of the role of paid labour in workers’ lives is exemplified by the legal norms which have propped up the ‘standard’ workweek and which have assumed a model of the worker as a fulltime male breadwinner. The policy shift towards work–family harmonization has accompanied the increased participation of women in the labour force and the resulting difficulty in maintaining the artificial separation between work and home which is based on the assumption underlying the male breadwinner model – that a worker’s family responsibilities are confined to the provision of financial support (Conaghan 2002). At the same time, the decline in birth rates has ushered in fears about the future of welfare states and the hope that work–family policies can contribute to ensuring population stability (Guerrina 2002). The outcome of these factors has been the growing prominence of work–family concerns in regulatory policies, including through the introduction and strengthening of laws on maternity protection and discrimination and an increased focus on child-care provision. In addition, however, work–family policies are being developed which incorporate working time initiatives. The EU’s employment strategy, for example, which has as one of its main goals the continued integration of women into European labour markets (see Rubery et al. 2001) incorporates the promotion of measures which entitle workers to periods of time out of the labour force to give birth, be involved in child-rearing or care for other family members, most notably maternity, parental and family leave. This chapter, however, is primarily concerned with measures of a more recent pedigree

16

Deidre McCann

which have the goal of allowing individuals to respond to family needs and other aspects of their lives on a more regular basis. These kinds of initiatives are particularly valuable in that they can both recognize that parents may want to directly participate to a substantial degree in caring for their children (Fredman 1998) and allow individual workers some influence over the scheduling of their paid labour. Part-time work: promoting quality and combating gender inequality Given the dominance of the male breadwinner model in the arrangement of working time, women in most industrialized countries have used parttime work to enable them to organize their working lives in order to combine paid labour with their family responsibilities, particularly where limited child-care has been available to them. Part-time work has been, then, both a response to and a facilitator of women’s inequitable responsibility for caring and domestic work. As discussed above, it helps to structure the ‘one-and-a-half earner’ households common to most industrialized countries (European Commission’s Group of Experts 1998; Ginn and Arber 1998). The establishment of part-time work as the default work–family reconciliation measure for working mothers, then, has obvious implications for gender equality in that it risks reinforcing the assumption that women are primarily responsible for child-care and domestic work. Moreover, the implications for gender equity of the preservation of the traditional division of labour is exacerbated by the poorer terms and conditions which tend to attach to part-time work and the resultant reinforcement of both occupational segregation and the gender-based wage gap. Despite these potential problems, however, in recent years, part-time work has increasingly been promoted as a potential solution to the dilemma of realizing the work–family balance. This promotion of parttime work as a response to workers’ needs raises obvious concerns given its association with poor working conditions and gender inequality. On the other hand, part-time work has facilitated the entry of women into the labour market, for example in Sweden, and can be conceived of and developed as a transitional measure for this purpose. Moreover, it is possible to envision a scenario in which part-time work is of a quality comparable to that of full-time work, and is also ‘de-gendered’, in that both men and women combine shorter hours with domestic responsibilities. Regulatory measures addressing discrimination against part-timers which have been enacted across Europe in recent years represent a response to the question of how to improve the quality of part-time work. Some have amended existing legislation which discriminated against parttime workers. In the UK, for example, the coverage of unfair dismissal and redundancy rights has been extended to cover part-time workers. More

Regulating working time needs 17 powerfully, the removal of such blatantly discriminatory provisions has been accompanied by the introduction of legislated rights for part-timers not to be discriminated against in their terms and conditions of work. This equality-focused approach to part-time work is embodied in the ILO’s 1994 Part-Time Work Convention.6 At the national level, the Netherlands adopted the equality approach some years ago as part of its efforts to integrate the protection of part-time workers into its gender equality policy (Pfau-Effinger 1998; Plantenga 1992, 2002). An early Dutch measure removed minimum hours thresholds for entitlement to the statutory minimum wage and to holiday pay, ensuring that both were available to all employees irrespective of their hours. More recently, in response to continuing concerns about the occupational segregation and poorer working conditions of part-timers, the Dutch government introduced legislation in 1996 specifically providing for a broad-ranging right to equality for part-time workers in areas negotiated by the social partners including basic wages, overtime payments, bonuses, holiday pay and training.7 The combined efforts of the government and the social partners have resulted in the comparatively high standard of working conditions for part-timers in the Netherlands (see Pfau-Effinger 1998) and the Dutch equality approach has been influential, having eventually been extended to other EU member states by the 1997 Part-Time Work Directive.8 The effectiveness of any right to equality for part-time workers will of course depend on the content and structure of the legislative scheme in which it is embedded. The most comprehensive coverage is offered by legislation which requires equality in all spheres, including basic wages, other forms of work-related income such as overtime and shift work premia, and access to training and promotion. For instance, given that part-timers are generally required to compare their treatment to those of full-time workers, the range of comparisons allowed by a particular equality law is significant. Where occupational segregation produces part-time workforces, part-timers can risk being unable to identify a full-timer with whom to compare their treatment, and thereby are prevented from bringing a claim for equality. Among the less restrictive approaches is that offered by the ILO’s Part-Time Work Convention which provides that, where there is no full-time comparator in the same establishment or enterprise as a parttimer bringing a claim for equality, they should be able to compare their terms and conditions with an individual employed in ‘the same branch of activity’.9 In addition, since part-time equality legislation usually affords employers the opportunity to justify the differential treatment of part-timers on objective grounds, the kinds of defences which will be permitted to equality claims are also significant. These defences can prove limiting unless carefully phrased and sensitively interpreted by adjudicatory bodies. With respect to pay differences, for example, even where a part-time worker performs the same or similar work as full-timers, there is a risk that their

18

Deidre McCann

lower pay will be justified on the grounds that it results from their lack of seniority or qualifications (Murray 1999). In these kinds of cases, an assessment can be made of whether there is any link between the length of service or qualification under consideration and the acquisition of a level of knowledge or experience which justifies the pay differential.10 In addition, equality legislation can also be strengthened by the inclusion of a burden of proof provision which requires that the employer, rather than the part-timer, identify the grounds for any less favourable treatment. Although the move towards legislated equal treatment of part-time workers is strong in Europe, there have been no comparable developments in other industrialized countries. However, legislation on part-time work of a different type has been introduced in Japan, where the 1993 Part-Time Workers Law11 has the stated aim of improving the welfare of part-time workers and more effectively utilizing their skills. The law empowered the Ministry of Labour to develop policy measures, require reports from employers and provide them with advice and guidance on the working conditions of part-time workers, and the Ministry issued guidelines in 199412 that encourage employers to ensure adequate working conditions for part-timers, including with respect to hours, wages, and education and training (see Kezuka 2000; Lo 1996). The Act also provides for subsidies to be awarded to firms which improve their systems for managing part-time workers, for example, by providing health assessments or training. Employers are required, however, only to endeavour to appoint an individual to manage the employment of part-time workers and to maintain a ‘balance’ with full-time workers by securing proper working conditions, and implementing education and training and improving their welfare. Neither the legislation nor the Ministry guidelines constitute binding legal duties. Even if powerful laws mandating the improved treatment of part-time workers are introduced, however, it is necessary to be sensitive to the implications of part-time work for gender equality. In this regard, Dutch work–family policies are of some interest in that they promote a combination of paid and unpaid work equally shared between men and women (Plantenga 2002). In this ‘combination model’, part-time work is envisioned as a method of advancing gender equality by promoting shorter hours for both men and women and the equal distribution of unpaid work (ibid.). The model remains an aspiration, as it is primarily women who combine part-time paid work with care in the Netherlands (Plantenga et al. 1999; Plantenga 2002). The combination approach is valuable, however, in that it envisions a ‘degendering’ of part-time work, offering the seeds of an approach towards working hours which is not grounded in the conventional division of labour. This approach partly underlines the trend towards measures that advance workers’ influence over working time, which are discussed in more detail below.

Regulating working time needs 19 Collective hours reduction as a preference measure Since the 1980s, the collective reduction of working hours has often been implemented as an employment creation measure, most notably in the introduction of the 35-hour week in the German metalworking and printing industries in the mid-1990s (see Seifert 1999) and the most prominent recent example of legislated reductions, the French 35-hour ‘Aubry’ Laws of 1998 and 2000. In addition to any employment creation effects, however, the reduction of weekly hours has the potential to facilitate the coordination of paid work with other responsibilities and interests. For, as discussed in more detail by Fagan in Chapter 4, collective reductions can be harnessed to the goal of recognizing and valuing family labour (see also Bruegel et al. 1998; Figart and Mutari 2000). This linkage has been recognized in the employment policies of a number of countries, including in the Netherlands ‘combination model’ (Bruyn-Hundt 1996; Plantenga 1997, 2002). Moreover, collective hours reductions have been argued to have the potential to contribute to advancing gender equality. Although these reductions can take different forms, for example, the increasing of holiday or leave entitlement, perhaps the primary focus for gender equality is the reduction of weekly working time, since this approach allows individuals to respond to domestic responsibilities on a daily basis. Workweeks of 40 hours and beyond are inevitably difficult to combine with significant domestic and caring labour. The hours demanded in much of men’s paid work in industrialized countries, then, prohibits their full participation in these other forms of work. If gender equity is conceived of as a model in which both men and women participate in paid employment and caregiving (Fraser 1997), it has been argued that this balance can be most effectively realized within a labour market in which both work in full-time jobs but for comparatively short hours in a ‘double three-quarter’ model of substantial part-time employment or short full-time working hours for both (Plantenga 2002). And it is apparent that shorter hours would at least create the conditions needed for a more equitable sharing. Although a resulting change in gender time-allocation in the household cannot be assumed, reducing working hours would at least provide the environment in which it is possible, and could be introduced in conjunction with other measures such as those aimed at reducing the gender-based wage-gap or strengthening parental leave schemes. This association between reduced weekly working hours and gender equality, however, is not always present. More specifically, the implementation of working time reductions in conjunction with non-standard work schedules has been seen to detract from their potential for work–family balance (see Fagan, Chapter 4). Unfortunately from this perspective, the goal of reducing working hours is often tied to the promotion of flexibility in current employment policy. Indeed, the trade-off of shorter hours for

20

Deidre McCann

increased flexibility has been a prominent strategy in collective bargaining across industrialized countries since the 1980s. Revealingly, research among French workers shows that although the vast majority feel the overall quality of their lives to have improved since the 35-hour week was introduced, they are less satisfied with their working conditions, many complaining not only of an intensification of workloads but also of irregular and unpredictable schedules (National Economic Planning Agency 2001). Care is needed, then, in the ways in which hours reductions are implemented if they are to be harnessed to the goals of responding to workers’ needs and preferences and advancing gender equality. Worker-choice rights: employee-oriented flexibility? The policy goal of assisting employees to realize an adequate balance between their paid work and other facets of their lives underlies a further trend, emerging in working time regulatory policies across the industrialized world, towards enabling employees to exercise some influence over their hours of work. For as the contested terrain of flexibility has become increasingly conceptualized as incorporating the interests of employees, the question of what precisely they want from their working time arrangements has become a major focus. The literature on preferences reviewed in the Introduction and by Lee (Chapter 2) and Fagan (Chapter 4) has revealed the complexity of workers’ preferences and their specific dissatisfactions with their current working schedules, as well as highlighting the kind of changes they would like. The pressing question is how these preferences can be most effectively realized. Of course, even in labour markets dominated by employer-oriented flexibility, some workers have been able to ‘exploit’ the individualization of working hours to meet their own needs (European Commission’s Group of Experts 1998; Fagan, Chapter 4). But the underlying current of these kinds of individual strategies has often been one of resignation, of making the most of the available forms of flexibility, which, as has been seen in the above discussion of part-time work, can often result in substantial limitations on the career progression of the workers concerned and in reductions in their income during their working lives and beyond. Moreover, this strategic exploitation of working time diversification has not been possible for all or even for significant numbers of workers, as is clear even from the limited research available on the gap between workers’ preferences and their actual working hours. Traditionally, any influence over working time arrangements has been exercised mainly through informal negotiations between individual workers and their employers. Recently, however, there has been a growing impetus in many industrialized countries towards developing measures which allow workers’ preferences to be mirrored in their work schedules. At the national level, in the past few years worker influence over working

Regulating working time needs 21 time has begun to make an appearance in legislative measures which are among the most interesting developments in the regulation of working time in recent decades, and which represent a shift within working time policy from a substantive towards a procedural approach to the realization of workers’ needs and preferences. First, legislation that allows collective agreements to implement or adapt working time standards can be used to enhance workers’ influence, at least to some extent. Increasingly, employers’ and workers’ representatives are being empowered by legislative measures to implement their provisions or even to set standards in the areas which they cover, a technique common in regulatory regimes which rely heavily on collective bargaining, but which appears to be spreading to other jurisdictions (Ozaki 1999; Sisson and Marginson 2001; Supiot 2001). In fact, these developments in the field of working time can be seen as forming part of a wider trend towards labour legislation which outlines principles and establishes procedures but contains few detailed substantive provisions and is mirrored in the relationship between ‘framework’ collective agreements negotiated at industry level and implemented through company-level bargaining (Sisson and Marginson 2001; Supiot 2001, 2002). This kind of interaction between legislation and collective bargaining is particularly common in the working time arena, however, in which legislation will often embody a rule or principle, and leave the social partners to determine the specific ways in which it will be applied. France, for example, has witnessed this kind of interplay between legislation and collective bargaining by virtue of its hoursreduction legislation, which allows detailed implementation of its general principles to take place through negotiation at the sectoral and firm levels. Of particular interest are legislative measures which allow for collective agreements to derogate from their provisions. Perhaps the best known among these is the EU Working Time Directive, which permits derogations from its standards on night work, daily rest, the weekly day of rest, and rest breaks. Indeed, the Directive is unusual in that it contemplates derogations at the level of the individual which would allow employees to ‘opt out’ of its 48-hour weekly limit. This provision was included in the Directive at the behest of the UK government, which has since been the only EU member state to make the individual opt-out universally available.13 The French Aubry Laws also permit a number of derogations from their standards. Collective agreements can, for example, raise the annual overtime limit or introduce weekend shifts which avoid the Sunday-rest rule. Similarly, in the realm of collective bargaining, there has been a trend identified in Germany towards company-level derogations from agreed limits on weekly working hours (Supiot 2001). The second and more direct way in which working time legislation can facilitate workers’ preferences is by allowing individual workers the opportunity to exercise some influence over the length and arrangement of their hours. A number of countries, for example, allow reductions in

22

Deidre McCann

working time specifically for carers, an approach first introduced in Sweden. Its legislation on parental leave entitles workers to convert their leave into part-time work, and they also have the right to reduce their working hours to three-quarters of full-time hours in order to look after children under 8 years of age. A number of countries have since followed Sweden’s example by allowing working time reductions for carers. Some, including Austria, Denmark and Finland, offer reduced working hours to parents. Others have enacted more expansive legislation. In Spain, for example, the working time reductions offered to parents are also available to workers who care for a relative. These kinds of provisions are in line with the ILO’s Part-Time Work Recommendation14 which suggests that, where national or firm-level conditions permit, workers should be able to transfer to part-time work in ‘justified cases’, such as during pregnancy and to care for a young child or disabled or sick member of their immediate family.15 Another way in which certain workers can exercise an influence over their working hours is through progressive retirement schemes (see EIROnline 2001c). Introduced by legislation in a number of European countries, including Austria, Finland and Spain, these measures allow workers to reduce their working hours as they approach retirement age, while receiving an income which generally consists of payment for the part-time hours they work, together with a pension or other form of income such as unemployment benefits. The legislation usually imposes certain eligibility requirements, for example, that the individual has worked for a specified number of years or made sufficient social security contributions, and the permissible hours reductions vary, although they generally range from 25–75 per cent of normal hours. In contrast to these measures, which are confined to specific groups of workers, a universal approach towards the individual right to influence working hours has been pioneered in the Netherlands. This approach builds on the striking degree of consensus between Dutch workers’ and employers’ organizations over the need for decentralization of decisionmaking on working time which has existed since the early 1990s (Visser 2003). This consensus was initially reflected in the increase in ‘à la carte’ provisions in collective agreements, under which employees are permitted, within certain limits, to ‘trade’ working time for wages or vice versa. Many agreements, for example, have included the right to work on a part-time basis, usually by providing that requests to work part-time should be granted unless there are significant reasons for refusing them (Fajertag 1999).16 And by 1999, the peak Dutch union and employer organizations had reached an agreement within the national bipartite Labour Foundation on the need to increase the choices for individual employees in collective agreements and had produced a policy document which promoted such ‘customized’ conditions of employment, although phrased as a recommendation to unions and employers rather than as a binding obligation.

Regulating working time needs 23 The Dutch legislation on the right to request a change in working hours, the Act on the Adaptation of Working Time,17 was introduced in 2000 with the stated aim of allowing employees the opportunity to adapt their working time in accordance with their needs and to contribute towards the realization of a more equitable balance between work and family responsibilities (Jacobs and Schmidt 2001). The Act provides employees with a right to request a reduction or an extension of their working hours or to have their existing hours redistributed, allowing them to work to a different schedule. This law, which is one of the measures developed to advance the Dutch ‘combination model’ of work–family balance discussed above, has since been adopted as a model for legislation in Germany and is also reflected to some degree in recent UK legislation. This recent trend in working time regulatory policy, then, involves a shift towards ensuring at least some degree of direct employee influence over working hours. Like part-time equality measures, however, much depends on the way in which the right to change working time is implemented. For example, the strongest form of worker choice legislation would, like the Dutch law, permit workers to request not only an increase or decrease in their hours but also changes in their scheduling. In addition, this kind of legislation usually provides a right to employees to request that their hours be changed, which should then be seriously considered by the employer. As a result, a debate has arisen over the range of justifications which should be accepted from employers for refusing employee requests. In the Netherlands, the legislation takes the approach of providing that, as long as no specific reason stands in the way, employers must grant employee requests, and listing a number of ‘serious interests of the enterprise or the service’18 which will be permitted to defeat such requests, while under the German legislation, ‘operational reasons’19 can justify a refusal to meet employee demands (see Jacobs and Schmidt 2001; Schmidt 2001). Whatever formulation is adopted, however, if this kind of legislation is to be effective in realizing individual preferences over working time, the adjudicatory bodies who determine in individual cases which grounds for refusals are permissible need to interpret the legislation with care. Where worker choice legislation is effectively crafted and implemented, however, it would appear to have the potential to operationalize individuals’ time preferences and thereby to help realize a workable work–family balance. In addition, although the outcomes of the enactment of this kind of legislation are not yet clear, it may also have the potential to advance gender equality by allowing for shorter hours to become more widely available (see Fagan, Chapter 4). For rather than compelling women who would prefer to work on a part-time basis to take whatever work is available, often at the cost of reduced wages and limited career progression, this type of legislation can enable them to work shorter hours in their current jobs, and thereby make part-time work available across a

24

Deidre McCann

wider range of occupations and sectors and focus it on the life-phases in which it is most needed. These laws also make available at least a formal right for men to choose to work on a part-time basis, challenging traditional assumptions about male and female working patterns.20

Worker choice: challenges for regulatory techniques The incorporation of the principle of worker choice into working time regulation raises compelling questions about the relationship between the twin regulatory techniques of legislation and collective bargaining and the future of working time regulation. And although these questions are not amenable to easy solutions, the recent introduction of individual preference measures lends them a degree of immediacy. In fact, working time regulatory regimes have always varied in the extent to which they rely on legislation or on collective bargaining, and have been classified as ‘statutory’, when primarily based on legislative measures (France, Italy, Spain), and ‘negotiated’ when operating mainly through collective bargaining (Denmark, Germany, the Netherlands, Sweden) (Anxo and O’Reilly 2000). This classification should not, however, mask the degree of linkage between regulatory techniques in the contemporary regulation of working time. There often exists a complex and shifting interaction between bargaining and legislation. For example, although collective bargaining has long been a source of regulation on topics which have not been addressed by legislation, since the early 1970s, the social partners in a number of industrialized countries have been increasingly involved in the formulation of laws. Moreover, collective negotiations can precede legislative measures, developing approaches which legislation can then adopt or build on, or may instead need the impetus of prior legislative innovations (Ozaki 1999; Supiot 2001). In recent decades, the shift towards favouring firm-level collective bargaining has formed part of a broader trend towards the decentralization of industrial relations regimes. Among the most extreme examples, Australia has witnessed a progressive decentralization of its industrial relations regime during the 1980s and 1990s from a system of industrial and occupational awards21 towards the primacy of enterprise-level bargaining (Campbell 1997; Campbell and Brosnan 1999; Wailes and Lansbury 1999). Europe, too, has undergone a similar transition as collective bargaining systems centred on national or industry-wide bargaining (continental Europe) and company-wide agreements (the UK) have, since the 1980s, shifted towards the firm level (Supiot 2001).22 In the working time arena, the French Aubry Laws, by obliging employers and unions to negotiate, have prompted a significant increase in both sectoral and, most significantly, firm-level bargaining, thereby shifting collective bargaining towards the firm level and reinforcing a trend towards the decentralization of the French regulatory regime (Jefferys 2000).

Regulating working time needs 25 These newly-forged relationships between legislation and bargaining, and the related decentralization of working time regimes towards the firm level, have influenced the distribution of actual working hours towards more diversity, as can be seen from the discussion by Lee in Chapter 2. In addition, however, these trends offer the potential for ensuring that working time regulation is responsive to worker needs and preferences, as discussed above, while raising questions about the legitimate roles of legislation and collective agreements in the regulation of working time. On the one hand, the devolution of deliberations over working time arrangements to the firm level can be viewed as creating a conducive environment for worker choice over hours scheduling to be effectively realized. It is widely accepted that collective bargaining is particularly appropriate for regulating working hours; that legislation cannot be sufficiently detailed or exhaustive and, most significantly, is inevitably ill-suited to ensuring the incorporation of the needs and preferences of individual workers (Ozaki 1999; Supiot 2001). And it is at the firm level where the principle that individual preferences should be reflected in working time arrangements can be translated into detailed work schedules, as discussed by Messenger in Chapter 5. Moreover, collective bargaining can validly be seen as a dynamic technique, better suited to adapting to changes in an ongoing process than its more static legislated counterpart (Supiot 2001). Finally, collective bargaining has also been hailed as having the potential to advance gender equality, to the extent that it can allow women the opportunity to identify their own needs and preferences and have them acted upon (Dickens 2000; Supiot 2001). On the other hand, the increase in regulation at the firm level raises a number of concerns for the direction of working time regulation, not least of which is the fear that it could result in legislation being drained of all strength. Derogation provisions are particularly worrying in this regard. The EU Working Time Directive, for example, is often argued to be of limited effect, given that much of its content can be disregarded in collective agreements or even by individuals (see, for example, Conaghan 2002). This concern with respect to the relationship between legislation and collective agreements can also be raised in relation to that between industry and company-level collective agreements, with the risk that the protections of the former will not be strongly adhered to in the latter (Supiot 2001). In ensuring individual influence over working hours, then, there is a need to carefully consider the relationship between the regulatory techniques of legislation and collective bargaining, to ensure that worker influence does not detract from strong legislated protection. It is necessary to ensure, for example, that the most fundamental legal rights can only be built upon in collective bargaining. Coupled with this fear of weak regulation is the concern that the regulation of working time, including the power to determine how best to realize workers’ preferences, will be devolved to institutions in which unions are

26

Deidre McCann

not recognized, or negotiated by non-union employee representatives or by individuals whose bargaining power is weak. The kind of individual entitlements offered by worker choice laws are likely to be best effected in circumstances of collective strength: workers with limited bargaining power are susceptible to being compelled to forgo their employment rights. That these workers, including many in non-union and low-pay establishments, would be unlikely to realize any control over their working schedules within a too-devolved regime offers an argument for the existence of strong legislated substantive rights, in combination with procedural entitlements to influence working time schedules. Relatedly, there may also be a need to re-think exclusions from the entitlement to protection under working time legislation. Entitlement to leave arrangements taken on a part-time basis, for example, is often conditional on a period of continuous, full-time work, criteria which many workers may be unable to meet. Moreover, it is clear that laws which provide support for trade unions, such as those on recognition and the right to strike, can also contribute towards strengthening worker influence in the area of working time. Given the growing trend towards worker influence over working time, traditional concerns about the representativeness of trade unions also acquire new resonance. Who is involved in negotiations on working time? Which voices are heard and which interests represented? Despite the potential advantages of firm-level bargaining for ensuring that gender equality is advanced through working time arrangements, the continuing under-representation of women and their interests by many trade unions remains a concern. Within bargaining units, the powerful voices may be those of men or workers without family responsibilities. There is therefore a risk that these workers will agree to working time arrangements which are detrimental to women or do not enable carers and those with significant domestic responsibilities to adequately merge their paid work with their domestic lives. To fully represent the interests of women and workers with care responsibilities, then, it is necessary to ensure that bodies which speak for them are truly representative. It could, for example, be specifically required that both of these groups be represented on negotiating bodies. Moreover, shifting perspective, it is clear that the arrangement of working time involves the interests not only of workers, but of society as a whole. How can this general interest be expressed in working time laws at the same time as they allow individuals to assert their individual preferences? The group of experts who reported to the European Commission in 1998 on the future of work and labour law in Europe were particularly vocal in their concern that the individualization of working time, while welcome, may, if not carefully designed, have the potential to undermine what their report terms ‘community time patterns’ – such as night-time and Sunday rest periods and midday breaks – and the social institutions which

Regulating working time needs 27 they support, including both families and trade unions (Supiot 2001). Their concerns highlight the need for a comprehensive approach to working time that allows individuals control over their work schedules, but also preserves time devoted to community life and does not undermine labour standards. In drafting legislation or conducting bargaining, for example, governments and the social partners could respond to the need to preserve time traditionally dedicated to family and community life by taking account of the opening hours of shops and services. These kind of concerns reinforce the need for a high degree of care to be devoted to achieving an appropriate balance between the provisions contained in legislative measures and those which can be the subject of collective negotiations.

Conclusion The reduction of weekly working hours, which began in the nineteenth century and was fuelled in part by a concern for preserving the health of workers and allowing them sufficient time outside of their paid labour, remains present in some regulatory measures, most notably in limits on weekly hours. In more recent decades, however, these goals have often been subordinated to that of facilitating a diversification of working hours in order to advance flexibility. This policy shift has entailed a number of regulatory changes, including the relaxation of restrictions on weekly hours and the averaging of hours over longer reference periods. Primarily employer-driven, these measures have often exacerbated the conflict between working time schedules and the needs and preferences of individual workers, while reinforcing gender inequality in the labour market. More recently, however, there has been a growing recognition of the value of reflecting workers’ interests in the arrangement of their working hours, including by enabling them to secure sufficient time to carry out their family obligations and other forms of unpaid labour, and to pursue other interests and commitments. Although these goals can be achieved to some degree through part-time work, it has tended to reinforce gendered assumptions about the appropriate division of work-based labour, and is often associated with poorer terms and conditions for women workers. However, recent attempts have been made to reconfigure part-time work as a family-friendly measure, in combination with initiatives aimed at securing equality for part-time workers. The reduction of weekly working hours offers another approach to the realization of workers’ needs and has the potential to contribute towards generating favourable gender outcomes. In conjunction with these two conventional pathways towards reflecting workers’ needs in working time arrangements, a growing number of regulatory measures are being designed which, to a greater or lesser degree, allow individual workers influence over the amount and scheduling of their working hours. These initiatives, while potentially beneficial, raise concerns for the future of

28

Deidre McCann

working time regulation that are centred around a consideration of the appropriate interaction and balance between regulatory techniques.

Acknowledgements My thanks to Dominique Anxo, François Eyraud, Colette Fagan, Sangheon Lee, Jon Messenger and Jillian Murray for their comments on an earlier draft of this chapter.

Notes 1 Hours of Work (Industry) Convention, 1919 (No. 1). 2 Council Directive 93/104/EC of 23 November 1993 concerning certain aspects of the organization of working time. 3 Case C-84/94 United Kingdom of Great Britain and Northern Ireland v Council of the European Union [1966] ECR I-5755. For an alternative view of the policy goals underlying the Directive see Murray (2001: 199–203). 4 Law No. 49 of 7 April 1947. 5 A 1999 Ministry of Labor Survey found that 42.3 per cent of part-time workers have adjusted their hours in this way, Kezuka (2000: 9). 6 Part-Time Work Convention, 1994 (No. 175). 7 Act of 3 July 1996. 8 Council Directive (EC) 97/81 concerning the Framework Agreement on parttime work concluded by UNICE, CEEP and the ETUC [1998] OJ L14/9. 9 Article 1(c). 10 See, for example, the decision of the European Court of Justice in C-1/95 Hellen Gerster v Freistaat Bayern [1997] IRLR 699. 11 The Law concerning Improvements in Employment Management of Part-Time Workers, Law No. 76 of 18 June 1993. See Lo (1996). 12 The Basic Policy Measures to Deal with Problems of Part-Time Workers. 13 Working Time Regulations 1998, Regulation 5. A number of other EU member states have recently taken steps to make use of the opt-out in their health sectors. See Commission of the European Communities (2003). 14 Part-Time Work Recommendation 1994, No. 182. 15 Ibid., Paragraph 20. 16 A survey from May 1995 found that more than half of the 400 collective agreements surveyed included these kind of provisions and that they applied to 80 per cent of the workers covered by the agreements. Fajertag (1999: 152). 17 Wet op de aanpassing van de arbeidsduur, Act of 19 February 2000. Translation by Jacobs and Schmidt (2001: 372). 18 ‘[Z]waarwegende bedrijfs- of dienstbelangen’. Ibid., Article 2(5). Translation by Jacobs and Schmidt (2001: 378). 19 ‘[B]etriebliche Gründe’, Act on Part-Time Work and Fixed-Term Contracts (Gesetz über Teilzeitarbeit und befristete Arbeitsverträge), Section 8(4). Translation by Jacobs and Schmidt (2001: 378). 20 On the influence of workplace cultures on working time decisions, see Fagan, Chapter 4. 21 This model has been depicted by Macintyre (1989) as a distinct ‘Australasian model’ of labour regulation. 22 The strength of the decentralization has varied between different countries. In the Netherlands, for example, there has been no significant decline in industrywide bargaining, see Supiot (2001).

2

Working-hour gaps Trends and issues Sangheon Lee

Introduction With changes in working time policies, notable changes in actual working hours have occurred, and the extent and scope of these changes are increasingly complex. While national average figures on the length of working hours suggest a rather stable or moderate declining trend in the 1990s in many industrialized countries, significant changes are found in the distribution of individual working hours. From a comparative perspective, there are considerable differences across countries in the way working hours are distributed among workers. In some countries, working hours are so varied among individual workers that virtually no ‘standard’ or ‘normal’ working hours can be identified. Certainly, this change was induced to varying degrees by workers’ diversifying needs and preferences. Yet it is in this process of diversification where the gaps between actual and preferred working hours have been most clearly revealed. Following up the discussion of changes in working time policies in Chapter 1, this chapter will examine the actual situation of working hours in the industrialized countries. Given the diversifying trends of individual working hours, it will investigate the distribution of individual working hours and link its national variations to different regulatory and institutional frameworks, and thereby to different methods of regulating working time. Then, this chapter will move to discuss individual working hours in detail by investigating how workers’ actual working hours differ from their preferred hours. In doing so, two kinds of working-hour ‘gaps’ – namely working-hour surpluses and deficits – will be highlighted. Trends for each of these types of working hour gaps will be examined based on excessively long working hours and marginal/involuntary part-time work, respectively. This chapter will conclude by drawing policy implications regarding working-hour gaps in the industrialized countries. Policy implications resulting from the analysis in this chapter will be more fully developed in Chapter 6, taking into account other aspects of recent working time developments.

30

Sangheon Lee

Trends in working hours Before examining different dynamics of changes in working hours, it is useful to review the overall level of working hours in the industrialized countries. This comparison can be made using one increasingly popular indicator for working hours, annual working hours, which reflects changes and variations in paid holidays and contractual status (ILO 2002a).1 As Figure 2.1 suggests, there are considerable variations in annual working hours among OECD countries, ranging from 2,500 hours in the Republic of Korea to 1,400 hours in Norway in 1999. Overall, relatively high working hours are found in the USA, New Zealand, Australia and Japan, along with middle- and low-income OECD countries (e.g. the Czech Republic, Mexico and Iceland). By contrast, average working hours are significantly lower in Norway, Germany, France and Switzerland. In 1999, the average worker in Norway was working around 500 hours less (or 13 weeks or so less on a 40-hour workweek basis) than their counterparts in the USA, Australia and New Zealand. It is often reported that the speed of reductions in working hours in the industrialized world has slowed down since the mid-1970s (Evans et al. 2001; Lehndorff 2000). This trend appears to persist in many countries. In fact, average annual working hours did not change much in most countries during the 1990s, with variances normally within 2 per cent of the average annual hours at the beginning of the decade. Nevertheless, due to changes in labour laws or collective agreements on working time, some significant reductions in working hours were achieved in the 1990s in some countries, such as Japan (9.3 per cent), West Germany (3.9 per cent) and France (3.2 per cent). By contrast, an increasing trend in hours of work is found in Sweden and the USA (the bar columns are shaded for these countries in Figure 2.1). However, any international comparison of these average figures needs to be done carefully, as these figures may not reflect the actual working 2,600

2,497

2,400 2,088

2,000 1,800

1,596 1,597 1,634 1,635

1,600

Korea

Czech Republic

Mexico

Iceland

United States

Australia

New Zealand

Japan

Spain

Finland

Canada

United Kingdom

Sweden

Italy

Switzerland

France

Germany

1,200

1,921

1,483 1,503

Norway

1,400

1,840 1,842 1,864 1,871 1,873 1,785 1,815 1,719 1,765

1,395

West Germany

Hours

2,200

Figure 2.1 Annual working hours in selected countries, 1999, total employment (source: OECD Labour Market Statistics).

Working-hour gaps

31

hours of individual workers. A meaningful international comparison based on average annual hours requires that the distribution of individual working hours must be similar across countries, otherwise, such comparisons would be misleading. In addition, cross-country variations and trends over time are often associated with the level of and changes in part-time work in many countries (OECD 1998). Thus, recent changes in working time policies are not properly captured by these average figures. To better understand the changes in working hours, therefore, data on the distribution of working hours by specified time bands is essential. In this regard, the distribution of annual working hours would be desirable, but such data has yet to be developed for time-series analysis and crosscountry comparisons. Thus, weekly working hours data will be used here. Importantly, as will be shown, these data are helpful in capturing general patterns and changes in working time in the industrialized countries. Furthermore, weekly data are important in examining the effects of working hours on the work–life balance (see Fagan, Chapter 4). The selection of those countries reviewed in this chapter is based on the fact that the distribution of working hours varies dramatically, to a large extent depending on institutional and regulatory frameworks regarding working time (Anxo and O’Reilly 2000). The articulation of different levels of regulation differs from one country to another, and these differences can be characterized according to three general types: (1) stateinitiated regimes, in which statutory regulation and state intervention are crucial (e.g. France); (2) negotiated regimes, in which collective agreements at industry and plant level are essential (e.g. Germany); and (3) market-based regimes, in which working time regulation is weak and agreements on working time are normally reached through enterpriselevel or individual negotiations (e.g. the USA) (see McCann, Chapter 1 and Anxo, Chapter 3). From this perspective of institutional and regulatory frameworks regarding working time, three major changes need to be examined: (1) changes in legal standard hours of work; (2) negotiated changes (based on collective agreements); (3) and changes in the regulatory framework.2 Changes in legal standards The long-term historical trend of shorter working hours, as illustrated by Maddison (1995), is to a large extent attributable to changes in legal normal hours as well as in paid annual leave. Since the Factory Act of 1833 in England, almost all countries have placed restrictions on the length of working time. The rationale for this state intervention stems mainly from the following factors: (1) the negative effects of long working hours have social, health and safety dimensions that are sometimes ignored by individual workers and employers (hence, there is a potential for negative externalities); (2) workers with weaker bargaining positions are forced to

32

Sangheon Lee

work longer hours; and (3) the reduction of working hours needs to be undertaken simultaneously across firms, since initiatives by individual employers to reduce working time can simply mean a reduction in their competitiveness, particularly against those companies with long working hours. The legal imposition of a standard number of normal hours is thus seen as a realistic solution in this context. This may mean that working time is a ‘currency’ which should not be wasted, but long hours should not be used as the major source of a firm’s competitiveness. While the need for such legal intervention is widely recognized, its effects on actual working hours have been rather controversial. Empirical studies indicate that these effects are often positive, but the extent of the effects varies based on different legal approaches, divergent economic and social circumstances (e.g. the labour market situation), and reactions from different labour market actors (Anxo 1999; Costa 2000; Hart 1987; White 1987). More importantly, the real impact of changing legal standards on actual working hours tends to depend on the nature of the prevailing institutional and regulatory framework in the country. While legislative changes are often merely an ex post recognition of well-established practices in countries in which collective bargaining plays a central role (e.g. Denmark and Sweden), they tend to be the major method of workinghour changes in some other countries (e.g. France). The direct and obvious effects of changes in statutory normal hours are well illustrated by the cases of Portugal, Japan and, most recently, France – countries in which working hours for the majority of workers have been adjusted to new legal standards. In Portugal, statutory normal weekly hours were reduced to 40 hours in 1996, with an emphasis on the potential effects on job creation. The effects of the legal change in Portugal are clearly revealed in Figure 2.2a, which shows that the working-hour band with the highest incidence shifted from 45–49 hours to 39 hours. This change was accompanied by an increase in part-time workers, most of whom normally work 30–34 hours. Thus, the overall pattern in this country is towards shortening and standardization of working hours for both full-time and part-time workers, although the incidence of part-time remains relatively low compared to other European countries (Eurostat 2001). A similar change is found in Japan where legal normal hours were gradually reduced to 40 hours per week in the 1990s, in order primarily to improve the quality of working life and discourage the culture of long hours which has long existed in Japan. Figure 2.2b demonstrates that this government initiative has been effective in curbing long hours: the effect is particularly noticeable in the time band of 43 to 48 hours, which appears to be one of the target groups for the legal change. Compared with Portugal, two facts are noteworthy. First, whereas part-time has also increased in Japan, the increase tends to be concentrated more on short-hours parttime work (or ‘marginal’ part-time work, see below). In addition, although

70 60 50 40 30 20 10 0 1–9 10–14 15–19 20–24 25–29 30–34

35

36

37

38

employees (%) 1987

39

40

41

42

43

44

45–49 50⫹

employees (%) 2000

(a) Portugal: male and female

35

CHANGES

30

1993

25

2000 20

%

15 10 5 0 1 to 14

15 to 29

30 to 34

35 to 42

43 to 48

49 to 59

60 and over

⫺5 ⫺10 hours

(b) Japan: male and female 70 60 50 40 30 20 10 0 1–9

10–14

15–19

20–24

25–29

30-34

35

employees (%) 1987

36

37

38

39

40

41

42

43

44

45-49

50+

employees (%) 2000

(c) France: male and female Figure 2.2 Effects of changes in legal standards (sources: (a) and (c) from Eurostat, author’s own calculations; (b) Japan Statistics Bureau).

34

Sangheon Lee

the size of the reduction in legal normal hours is four hours in both Japan and Portugal, the effect on actual working hours is relatively small in Japan, compared with Portugal. These differences appear to be associated with the presence of ‘service’ (unpaid) overtime, which suggests that the corporate culture of rewarding total commitment to the company remains firm in Japan (Hippo 1993; Rengo 2001a).3 Most recently, France introduced a new legal standard of 35 hours with the specific aim of boosting employment. Figure 2.2c shows that the effects of this legal change on actual working hours are identifiable. The general pattern is that an increasing proportion of workers are beginning to work 35 hours rather than the old standard of 39 hours. Thus, the prediction would be that the peak of the distribution would be shifted to 35 hours, as the new principle has just begun (in 2002) to be applied to small companies as well. Compared with Portugal and Japan, it is interesting to note that the change in working hours has taken place only in the time band of 38–49 hours, without affecting other time bands; for example, there was no significant change in part-time work. Although this legislative change has affected actual working hours in France considerably, it is not clear whether this change in actual working hours has induced the creation of new jobs.4 Overall, the effects of these legal changes on working hours are notable but not evenly distributed among workers in the countries considered. In fact, these countries have in common the result that changes in legal standards do not reduce the proportion of ‘over-worked’ employees. If the focus is placed on ‘excessive hours’, defined as 50 hours or more per week (see next section), their incidence has actually increased in Portugal (by 1.0 percentage point) and in Japan (by 1.4 percentage points). This result may be seen as disappointing, particularly in Japan where these workers – who are much more vulnerable to karoshi (death from over-work) – were the primary target group for the government initiatives regarding reduction in working time. Other countries have changed their legal normal hours of work. However, the effects of these changes are not large, particularly for those countries that encourage collective bargaining as the main vehicle for the reduction of working hours. For example, legal normal hours changed in the late 1990s in Belgium (from 40 to 39 hours in 1999) and Italy (from 48 to 40 hours in 1997), but there is no clear indication of the effects of these changes on actual working hours. In these countries, legal changes do not precede collective agreements on working time, but rather tend to provide ex post legitimacy to the normal hours that have already been established through collective agreements. Negotiated changes Recent experience in the industrialized countries implies that collective agreements could be a more effective method for reducing actual working

Working-hour gaps

35

hours in some countries than changes in statutory norms. In fact, collective agreements, if they are well coordinated, could be an alternative way of handling the externalities related to working hours. It is also known that collective agreements can better address the potential problems (e.g. those of financing and implementation) that would arise from shorter working hours (White 1987). This explains in part why many industrialized countries have placed an increasing emphasis on collective bargaining in determining working hours. However, the effectiveness of this approach can be limited when union density and coverage are not sufficiently high, and it could result in an increasing disparity in working hours across different groups of workers. Negotiated changes in working hours are clearly illustrated by the cases of Denmark, Germany and the Netherlands. They share a common feature in that safeguarding jobs or employment creation is the primary focus of changes in working hours. However, it should be noted that differences in bargaining structures between these three countries are reflected in different distributional patterns of working hours. Denmark is well known for its centralized and coordinated bargaining structure and the high coverage of collective agreements under the leading roles of the LO (Danish Confederation of Trade Unions) and the DA (Danish Employers’ Confederation) (ILO 1995). In recent years, the union coverage has further increased to 83 per cent in 2000.5 As Figure 2.3a suggests, these characteristics have created a working-hour distribution similar to the ones in those countries in which statutory intervention is strong and the majority of workers work the statutory normal hours. In addition, the effect of negotiated changes in working hours is very similar to that of state-initiated changes. In Denmark, normal working hours were reduced to 37 hours in 1991 through centralized collective agreements. This change is clearly reflected in the distribution of actual working hours, in which the peak has shifted to 37 hours (Figure 2.3a). One accompanying change in this context was an increasing proportion of workers working 30–34 hours, which is now higher than the proportion working 20–24 hours. A less centralized structure of collective bargaining on working time is found in Germany, where collective agreements are normally concluded at the industry level, although some sectors, such as the metal industry, play a leading role in setting normal hours. Hence a wider spread in working hours might be expected, particularly owing to disparities in the collectively agreed norms between the various bargaining areas (Anxo and O’Reilly 2000). In 2000, collective agreements on working time ranged from 35 to 40 hours, which corresponds to the two-peaked distribution of working hours in Figure 2.3b. Since the introduction of a 35-hour workweek in the metal industry to safeguard jobs in the mid-1990s, some noticeable increase has taken place in the incidence of workers working 35 hours per week, while the importance of 40 hours declined. Thus, the

70 60 50 40 30 20 10 0 1–9

10–14 15–19 20–24 25–29 30–34

35

36

37

38

employees (%) 1987

39

40

41

42

43

44

45–49

50⫹

employees (%) 2000

(a) Denmark: male and female 70 60 50 40 30 20 10 0 1–9 10–14 15–19 20–24 25–29 30–34

35

36

37

38

employees (%) 1987

39

40

41

42

43

44

45–49 50⫹

employees (%) 2000

(b) Germany: male and female

70 60 50 40 30 20 10 0 1–9

10–14 15–19 20–24 25–29 30–34

35

employees (%) 1987

36

37

38

39

40

41

42

43

44

45–49 50⫹

employees (%) 2000

(c) Netherlands: male and female Figure 2.3 Effects of negotiated changes (source: Eurostat, author’s own calculations).

Working-hour gaps

37

general trend in Germany would be a shift in normal hours from 40 hours to 38 hours per week, and then possibly to 35 hours. Another change is the increase in the share of workers with short hours (particularly very short hours), who are covered by collective agreements to a much lesser extent than other workers in Germany. In the Netherlands, the principle of 36 hours was introduced by collective agreements in the mid-1990s. The resulting changes in actual working hours are reflected in Figure 2.3c, which suggests a significant increase in the incidence of those working 36 hours per week. Working hours in the Netherlands are more complicated, however. First of all, some agreements allow for the possibility of working longer than the established average of 36 to 38 hours per week (typically 40 hours); the incidence of those workers working 40 hours per week remains the highest at around 30 per cent. Second, due to the country’s consistent policy of encouraging female participation in the labour market through opportunities for part-time jobs, the overall incidence of part-time work has increased even further. Obviously, there is also a significant gender gap in the distribution of working hours. Although details of this phenomenon are discussed by Fagan in Chapter 4, it should be noted that: (1) the increasing popularity of part-time work, particularly among female workers, has created a situation where a norm in working hours does not exist for women workers; and (2) in the case of male workers, working hours appear to be bifurcated into norms of either 36 or 40 hours per week, thanks to the increasing incidence of workers with a 40-hour workweek. Changes in regulation techniques There are also some countries that have experienced changes in the techniques used for regulating working time. Good examples of these types of changes are found in Australia and the UK, which are seemingly moving in different directions. In Australia, trade unions and employers’ associations have been fragmented, but, due to the critical role of so-called ‘industry awards’, hours of work were effectively determined ‘quasilegislatively’ (Dawkin and Baker 1993). However, a recent deregulation drive in Australia has partially dismantled this centralized mechanism of working-hour determination and placed more focus on enterprise and individual agreements. It is sometimes argued that the collapse of the award system ‘has already produced a major shift, which has moved Australia from a relatively regulated labour market system . . . towards a deregulated system much closer to the US model’ (Campbell and Brosnan 1999: 354). Figure 2.4a shows the significant effects that these institutional changes have generated on actual working hours. In the late 1980s, Australia had a working hours distribution pattern similar to those in countries with strong legislative roles (i.e. a sharp peak at 40 hours per week). However, by

38

Sangheon Lee

2000, the distribution had flattened, becoming similar to that of the UK (Figure 2.4b). This result is mainly due to the fact that the incidence of both long hours and short hours has increased significantly. The close association between this dramatic change and the institutional changes is illustrated by the finding that the increasing number of enterprise and individual agreements under the new legislation, the Workplace Relations Act of 1996, tended to provide for longer working hours (Peetz 2001). 40.00

35.00

30.00

%

25.00

20.00 YEAR OF 2000 15.00

10.00 YEAR OF 1987 5.00

0.00 1 to 15

16–29

30–34

35–39

40

41–44

45–48

49 and over

hours

(a) Australia: male and female

20

18

16

14

12

10

8

6

4

2

0 1–9

10–14

15–19

20–24

25–29

30–34

35

employees (%) 1987

36

37

38

39

40

41

42

43

44

45–49

50⫹

employees (%) 2000

(b) UK: male and female Figure 2.4 Effects of changes in regulation techniques (sources: (a) from Australian Bureau of Labour Statistics, Labour Force Survey (6203.0); (b) from Eurostat, author’s own calculations).

Working-hour gaps

39

The UK provides another interesting example of changes in regulation techniques. The UK had been known for its absence of legislative regulations on working time. As noted in Chapter 1, however, the UK was required to move away from this traditional approach to working time and to introduce the Working Time Regulations in 1998 in order to implement the EU Working Time Directive, which stipulates a maximum workweek of 48 hours including overtime. Although the full effects of this new regulation on actual working hours are yet to be realized, Figure 2.4b indicates that this law so far has not changed the general pattern of working hours in the UK. In particular, the law has not been successful in curbing long hours. Such disappointing results arise at least in part from the UK law itself. Unlike other EU countries, the UK tried to minimize the impact of the Working Time Regulations by allowing individual opt-outs from the 48hour limit. As a result, the establishment of individual agreements is the preferred way of responding to this new law (Neathey and Arrowsmith 2001). In fact, it is reported that 81 per cent of UK companies adopted individual agreements that allow for long working hours (i.e. over 48 hours per week) since the inception of the new working time law (Goss and Adam-Smith 2001). This new institutional experiment in the UK suggests that established practice is still powerful enough to minimize the impacts of legal intervention in the area of working time.

Working-hour gaps Standard economic theories often assume that workers’ current levels of working hours correspond to their needs and preferences, but this is not the case for many workers. In fact, their ability to exercise choice over their working hours is to large extent constrained by many factors that are beyond the control of individual workers, including the structure of the labour market itself (Euwals 2001; Kahn and Lang 1992; Schor 1992; Stewart and Swaffield 1997). As Schor (1992: 129) put it, it is often the case that ‘competition in the labor market is typically skewed in favor of employers: it is a buyer’s market. And in a buyer’s market, it is the sellers who compromise’. This explains at least in part why active working time policies have been adopted in many industrialized countries to reduce the gap between current and preferred working hours (see McCann, Chapter 1). How large, then, are these working-hour gaps in the industrialized countries? A useful insight to this question is provided in the European Survey on Employment Options for the Future (1998). When asked the question, ‘Provided that you could make a free choice so far as working hours are concerned and taking into account the need to earn your living, how many hours per week would you prefer to work?’, workers revealed different preferences depending on their current working hours. As Table 2.1

40

Sangheon Lee

Table 2.1 Gaps in working hours, 1998, EU countries Current weekly working hours Under 20

20–34

35–9

40–9

Preferred changes in working hours Reduction (15 hours or more) 1 (5–14 hours) 5 (less than 5 hours) 1 Keep same hours 40

2 10 4 54

9 22 15 46

10 43 9 34

56 25 0 18

Increase (less than 5 hours) (5–14 hours) (15 hours or more)

7 20 26

3 17 10

4 4 0

1 3 0

0 1 0

100

100

100

100

100

6 46

12 27

31 4

53 3

81 1

Total Working-hours gaps Surplusa Deficitb

50 plus

Source: The 1998 Employment Options Survey; adapted from Table 31 in Fagan et al. (2001) Notes Percentage of workers under each band of working hours. a Proportion of workers who prefer to work shorter than current hours by no less than 5 hours. b Proportion of workers who prefer to work longer than current hours by no less than 5 hours.

shows, there is a notable polarization in workers’ preferences: a significant proportion of part-time workers want to increase their working hours while a higher proportion of full-time workers prefer to reduce their working hours. If only those workers who want to adjust their working hours by no less than 5 hours per week are considered, the most prominent groups of workers with working-hour gaps are part-time workers with very short working hours and ‘over-worked’ full-time workers. Table 2.1 indicates that, in 1998, 46 per cent of short part-time workers (those working less than 20 hours per week) were faced with working-hour deficits; by contrast, a working-hour surplus was reported by 81 per cent of full-time workers with longer than 50 hours per week. This pattern also appears to be the case in other industrialized countries, such as Canada (Drolet and Morissette 1997), Japan (Japan Statistics Bureau 2002) and the USA (Jacobs and Gerson 2000).6 This situation implies that the traditional concerns about long working hours still remain, and, with the increasing incidence of part-time work, new concerns about the working hours for part-time workers have also emerged in the industrialized countries (see also Fagan, Chapter 4, concerning differences by gender).

Working-hour gaps

41

Working-hour surpluses: excessive working hours If a worker desires to work less than his/her current working hours, he or she can be said to be working ‘excessive’ hours. This is a subjective method of identifying working-hour surpluses that may require policy interventions.7 Obviously, this method has the advantage of reflecting individual workers’ preferences, but it also tends to exaggerate the size of workinghour surpluses, considering the fact that market work is normally considered to be a disutility for workers. While some new approaches have been taken in ways that can better reflect the trade-off between more leisure and lower earnings, a more rigorous technique is yet to be developed to control for the vulnerability of individual preferences.8 Another method of measuring the extent of excessive hours relies on an ‘objective assessment’ regarding the negative effects of long hours on workers’ health and safety (and other aspects of workers’ well-being, such as family life). This approach underlies the imposition of legal standard hours in many labour laws. Recently, 48 hours was implicitly suggested as the threshold for excessive hours in the EU Working Time Directive (1993). Similarly, Spurgeon (2003) also concludes (from her extensive review of the literature regarding the health effects of long hours) that working hours longer than 48–50 hours per week could expose workers to potential health risks. As Spurgeon’s review indicates, the effects of such long hours can vary depending upon how the hours are organized (e.g. in relation to shift/night work and work intensity) and the nature of work, as well as the characteristics of individual workers (among other factors). While each method has its own merits and weaknesses, it is interesting to note that the threshold in the second method largely corresponds to that in the first method based on workers’ preferences. In other words, working hours greater than 50 hours per week are neither preferred by most workers, nor desirable from the perspective of health and safety. Thus, whenever data is available, a threshold of 50 hours will be used here to examine the extent of excessive working hours in the industrialized countries.9

Trends and variations Trends in excessive hours in the industrialized countries are reviewed in Figure 2.5. Three points deserve special attention. First, there are significant cross-country variations in the incidence of excessive hours, ranging from 1.4 per cent in the Netherlands to 28.1 per cent in Japan. From the perspective of institutional frameworks, the overall pattern underlying these variations would be that countries with relatively weak institutional and regulatory frameworks regarding working time, such as the USA, the UK and Australia, tend to have a much higher incidence of excessive hours than other countries, with the notable exception of Japan. By

42

Sangheon Lee 1.4

Netherlands

2.5

1.9 1.6

Sweden

2000

2.7 2.9

Austria Belgium Italy

2.4

Finland

2.3

Denmark

1987

3.8

2

4.2 4.5 5.1

3.7

5.3 5

Germany Portugal

4.3

5.3

France

5.7 5.9

Spain

5.8 5.4

Greece

4.6

6.2 6.2

Ireland

7.9 15.5

UK

13.2

Australia

20

15.3

20

USA

15.4

New Zealand

21.3

18

28.1

Japan

26.8 0

5

10

15

20

25

30

% of employees

Figure 2.5 Trends in excessive hours (sources: Developed by Dominique Anxo and Sangheon Lee, from EUROSTAT (special tabulations); Australian Bureau of Labour Statistics, Labour Force Survey (6203.0); New Zealand Statistical Office, New Zealand Official Yearbook; Japan Statistics Bureau, Annual Report on the Labour Force Survey; US Department of Labor (1999), Report on the American Workforce, Table 3.1). Notes 1 Department employees in non-agricultural sectors. 2 Excessive hours defined as 50 hours and more per week, except the USA and Japan (49 hours and more). 3 For Australia, Finland and Sweden, 1995 and 2000 data are used. 4 For the USA, 1979 and 1998 data are used. US data are not strictly comparable due to changes in survey methods. 5 For Australia, 1979 and 2000 data are used. 6 For Japan, 1993 and 2000 data are used. 7 For New Zealand, 1990 and 2000 are used.

contrast, countries with a strong tradition of collective bargaining on working time tend to have a very low incidence of excessive hours. The countries in this latter category have kept the incidence of excessive hours below 6 per cent. For example, it can be said that excessive hours hardly exist in the Netherlands, Sweden and Austria. The extent of excessive

Working-hour gaps

43

hours in countries in which statutory regulation dominates falls within the middle range, i.e. it is typically from 5 to 6 per cent of all workers. Second, there appears to have been an increasing trend towards excessive hours in many countries in the 1990s, irrespective of their institutional and regulatory frameworks regarding working time. In most countries (e.g. Spain), the increase is negligible, but some countries, particularly the USA and Australia, have witnessed significant increases in the incidence of excessive hours (Heiler 1998; Hetrick 2000). This trend is somewhat surprising, given the prevalent perception that major progress has been made in reducing excessive hours of work across the industrialized world. Finally, although detailed figures are not presented in Figure 2.5 (for additional information, see Fagan, Chapter 4), it should be emphasized that excessive working hours are primarily a male phenomenon, in contrast with part-time hours, which are primarily a female phenomenon. For example, among countries with a high incidence of excessive hours, male workers are about three times more likely than their female counterparts to be involved in excessive working hours (5.9 per cent for women and 24.2 per cent for men in the UK; 11.7 per cent and 27.0 per cent in the USA; 12.6 per cent and 38.3 per cent in Japan; 10.4 per cent and 30.1 per cent in New Zealand; 9.4 per cent and 28.4 per cent in Australia). This overall pattern is also the case in other countries with a lower incidence of excessive working hours. Causes of working-hour surpluses These variations are in part related to differences in the prevailing level of normal hours (beyond which an overtime premium is typically paid), determined by laws or collective agreements. Normal hours of work range typically from 35 to 40 hours across the industrialized countries. However, considering that working 50 hours or more typically involves some form of overtime in most countries, part of the cross-country differences in total hours of work may be ascribed to the dramatic differences across countries in the incidence of overtime. The countries with a higher incidence of excessive hours normally show a heavy reliance on overtime, especially to secure flexibility in working time in relation to changes in product market demands. In the UK, for example, overtime is somewhat ‘institutionalized’ and overtime payments constitute a considerable part of total wage income (Cranfield School of Management 1997). At the same time, overtime is not always paid, notably in the case of white-collar workers. According to Harkness (1999), unpaid overtime in the UK increased from 2.2 hours to 4 hours between 1988 and 1998. A trade union estimation reveals a much higher figure of 7.1 hours per week in 2000, which amounts to £23 billion per year (Trade Union Congress 2000). The USA has also witnessed an increase in overtime. Weekly overtime

44

Sangheon Lee

in the manufacturing sector increased from 3.3 hours in 1991 to 4.9 hours in 1998, and this increasing trend is obvious even considering the changes in overtime due to the business cycle (Hetrick 2000). This implies that employers have been increasingly relying on overtime rather than new recruitment in responding to increases in product demand. One common feature of these two countries, as noted earlier, is that there are no statutory limits on the length of overtime: while in the USA overtime can be undertaken as needed insofar as an overtime premium of 150 per cent is paid, the new working time regulation of the UK, although designed to limit excessive hours of work, allows for opt-outs from the 48hour maximum based on individual agreements. An increasing trend of overtime is also the case in Australia, particularly due to an increase in unpaid overtime, although the average amount of paid overtime hours has decreased somewhat (Campbell 1999). It is argued that this trend is to a large extent associated with the changes in the regulatory framework on working time in Australia (ACIRRT 1999). High levels of overtime are also found in Japan. Overtime fluctuated between two and four hours per worker per week in the 1990s, depending on the overall economic situation (OECD 1998, Table 5.5), which explains the prevalence of excessive actual hours despite the introduction of shorter normal hours in that country. By contrast, average hours of overtime stood at a low level in most European countries in 1995, typically less than 1 hour per worker per week (European Commission 2000b: 73–4). In some cases, overtime is voluntarily undertaken by workers. According to standard economic theory, workers will choose their length of working time, constrained by the trade-off between leisure and wages. In this understanding, long hours can be attributed to the result of mutually beneficial market exchanges regarding working time. However, it is noteworthy that the relative bargaining disadvantages of workers, especially during recession periods, may make it difficult for them to resist employers’ demands for longer hours. Restrictions on overtime by laws or collective agreements can be effective in this regard, which is an important source of cross-country variation in the incidence of excessive hours. As illustrated by Australia and the UK, individual employment contracts in the absence of regulation are more vulnerable to excessive hours than laws and other collective measures. In addition, evidence indicates that the presence of trade unions or collective bargaining machineries could reduce the incidence or length of overtime (Trejo 1993; Scheuer 1999; cf. Bell and Hart 1998). Not surprisingly, countries with higher union density are likely to have a lower incidence of excessively long working hours, which suggests the essential role of trade unions in discouraging long working hours. Relying on OECD data on trade unions (OECD 1997), our analysis indicates that the incidence of excessively long working hours is negatively correlated with union density. This correlation is strong and statistically significant (the Spearman correlation coefficient is 0.734).10

Working-hour gaps

45

Structural factors related to human resource management are often responsible for excessively long working hours, especially when they are seen as an indicator of workers’ competency, productivity and commitment to the firm (Eastman 1998; Landers et al. 1996). This could generate ‘competition’ for longer hours (which involves extensive unpaid or ‘service’ overtime) among colleagues, resulting in an ‘inefficient equilibrium’ with a high incidence of long hours. This situation may develop into a cultural dimension in which excessive hours become the norm, sometimes resulting in devastating effects on workers, such as karoshi in Japan (Hippo 1993; Spurgeon 2003). The prevalence of excessive hours may also discourage even qualified workers (particularly women workers with children) from continuing to work (Eastman 1998; Landers et al. 1996). The risk related to competition among workers in terms of working hours is known as being particularly high for professional workers, since the length of their working hours is often explicitly considered as part of their performance appraisal. As the professionalization of jobs proceeds (e.g. in the IT sector), there are reasons to believe that the incidence of excessive hours will increase if an appropriate regulating mechanism is not provided (ILO 2001b). Furthermore, when full-time regular employment requires additional labour costs, such as contributions to pension, health insurance and other non-wage benefits, firms may have an incentive to increase working hours rather than hiring new workers, thereby reducing their unit labour costs (Anxo 1999; Schor 1992; cf. Jacobs and Gerson 2000). This incentive is clearly revealed in employers’ reliance on overtime in order to meet market demands. At the same time, it is noteworthy that this employer strategy is often welcomed by workers – especially blue-collar full-time workers, who are more likely than other workers to be paid for their overtime work – because it increases their wage incomes at a higher rate due to the presence of an overtime premium (ILO 1995). Therefore, wages are closely associated with the length of working time. The income-induced incentive for extended hours of work is particularly strong for workers with low incomes (Low Pay Commission 2000, Appendix 4). Relatedly, some US evidence suggests that the presence/expectation of employment instability could lead to long hours when workers are employed, in order to ‘cushion the blow of depressed income when joblessness strikes’ (Bluestone and Rose 2000: 31). All of these findings imply that a reasonable level of income security is an important precondition for implementing a working time policy that successfully discourages excessive hours, which marketbased working time regimes normally lack. This may imply that the deterioration of the social welfare system or social ‘safety net’ could induce workers to resort to longer working hours as a buffer against future layoffs and thereby partially mitigate the increased uncertainty of future incomes (see Chapter 6).

46

Sangheon Lee

The distribution of incomes is also important, as is suggested by the presence of promotion-induced incentives for longer hours (with higher compensation), particularly among professional workers. Bell and Freeman (2001), for example, show from a comparison of the USA and Germany that the more unequal distribution of wages in the USA is associated with its long working hours (see also Bosch and Lehndorff 2001). Thus, if a country is serious about reducing working hours, accompanying changes in other aspects of labour market and welfare policies are of pivotal importance. Working-hour deficits: involuntary and marginal part-time workers Working-hour deficits constitute another dimension of working-hour gaps in many industrialized countries. As mentioned earlier, the issue of working-hour deficits has gained importance as part-time work has increasingly been introduced to save on labour costs and/or provide better opportunities for those workers who would otherwise be unemployed or remain out of the labour market (particularly women with family responsibilities). The importance of such working-hour deficits is associated with the fact that they are often linked with the low quality of parttime work, resulting in what might be called ‘segregation in the part-time employment market’ (Tilly 1996). Despite their increasing importance in many industrialized countries, it is difficult to measure the degree of working-hour deficits because of both conceptual difficulties and data problems. One well-known concept for measuring working-hour deficits is ‘time-related under-employment’, which is defined to exist when ‘the hours of work of an employed person are insufficient in relation to an alternative employment situation in which the person is willing and available to engage’.11 While this concept captures the main aspect of working-hour deficits, the way of measuring it often differs from one country to another. In some cases, as Table 2.1 implies, time-related under-employment includes workers who would like to increase their working hours, but only slightly. It might be argued that these workers do not present a priority for working time policies. Considering the available data, two proxy indicators of working-hour deficits will be reviewed separately in this section: marginal and involuntary part-time employment. Marginal part-time work refers to very short working hours, which are often known to be accompanied by lower wages and poorer working conditions, even compared with the overall conditions of part-time employment as a whole. By contrast, involuntary part-time workers are those who are currently working part-time, but would prefer full-time jobs. Obviously, involuntary part-time represents one approach to measuring time-related under-employment, and, at the same time, it may be this category of workers that suffers most significantly from

Working-hour gaps

47

working-hour deficits. Interestingly, a recent report has indicated that these two types of part-time workers tend to have lower job quality compared with other part-time workers (European Commission 2002b: Chapter 3). Marginal part-time employment It is known that very short working hours are often related to poor working conditions. In fact, there is some evidence that these jobs are ‘designed to meet employers’ requirements rather than labour supply preferences’ (Fagan 2001b: 253). Those workers who are working very short hours are also reported to be disadvantaged in terms of promotional opportunities and non-wage benefits (Walsh 1999). In terms of earnings, evidence indicates that hourly earnings for part-time workers with less than 20 hours per week appear to be lower than for other part-time workers (OECD 1999). Interestingly, a recent EU Working Conditions Survey reveals that the increase in part-time in Europe between 1995 and 2000 was largely ‘bottom-up’ (Bodin 2001): the increase resulted mainly from an increase in the lowest working time band (i.e. less than 10 hours), in which employment and pay conditions are generally unfavourable. Furthermore, these workers (particularly women workers) are exposed to ‘a significantly higher risk of unemployment’ (European Commission 2002b: 87). As pointed out earlier, it is also these groups of workers that want to increase their working hours most significantly (see Table 2.1). At the same time, the information on very short working hours can be used in understanding the actual extent of unemployment, given that standard statistical practices count workers who are working more than as little as one hour per week as being ‘employed’ (ILO 2002a). These concerns about very short working hours imply that part-time workers cannot simply be analysed as a homogeneous group. According to Hakim (1997), this type of part-time can be called ‘marginal’ part-time work.12 Thus, in some countries ‘marginal’ part-time workers can be considered to be ‘outsiders’ within a group of ‘outsiders’ (part-time workers), as compared with the ‘insiders’ (full-time workers). Although the threshold working hours for marginal part-time is not clear and normally varies across studies, this threshold typically ranges from 10 to 20 hours per week. This definition partly reflects national variations regarding the threshold for entitlements to social protection (see OECD 1999: Table 5.12). In this section, the threshold of 20 hours is adopted simply for the convenience of international comparison (ILO 2002a; Tijdens 2002).13 It should be emphasized that these marginal part-time workers are still a heterogeneous group. Some marginal part-time workers may have a strong preference for very short hours due to their personal situation. In other words, marginal part-time work is sometimes ‘chosen’ rather than ‘tolerated’. This is particularly the case for students with marginal

48

Sangheon Lee

part-time jobs who tend to be concentrated in the age group of 16–24 years (see the discussion below). As is the case with excessive hours, there are considerable variations across countries in the incidence of marginal part-time workers as a proportion of total employees. As Table 2.2 suggests, marginal part-time has not gained statistical importance in many industrialized countries, particularly Greece, Portugal, Spain, and Austria. One common feature of these countries is that part-time work is relatively uncommon: 6.3 per cent for Greece, 5.3 per cent for Portugal, 8.1 per cent for Spain, and 12.4 per cent for Austria (when part-time work is defined as working hours of less than 30 hours per week).14 Both the low level of social protection for part-time workers and the cost disadvantages of part-time against other types of ‘atypical’ employment appear to have contributed to the unpopularity of part-time work in these countries (Ruivo et al. 1998; cf. OECD 1999). It is also notable that these countries are known for their relatively strong state role in regulating working time (i.e. they are based on a state-initiated working time regime). High incidences of marginal part-time are found in the UK, New Zealand, the Netherlands and Australia.15 This situation is to be expected, as part-time work in general is widely used in these countries, normally accounting for over 20 per cent of total employment. Table 2.2 Marginal part-time work: incidence and changes (% of employees) Countries

Proportions (2000) (A) ⬍ 20 hours

Greece Portugal Spain Austria Finland Italy Sweden France Belgium Ireland Denmark Germany The UK New Zealand The Netherlands Japan The USA Australia

(B) 20 to 29 hours

1.7 4.6 1.8 3.5 3.2 4.9 3.5 8.9 4.3 4.7 5.2 8.5 5.3 9.4 5.4 9.4 7.7 12.5 7.8 11.5 9 6.4 9.2 8.6 12.5 9.9 15.6 9.6 17.7 13 ⬍ 15 hours 3.6 10.5 5 12.3 12.9 13.2

Sources and notes: See Figure 2.5.

A/(A ⫹ B) (%)

■

Changes (% points)

(A)

27 ⫺0.6 34 ⫺0.7 39.5 ⫺1.3 28.2 ⫺0.2 47.8 ⫺1 38 ⫺1.6 36.1 ⫺0.7 36.5 ⫺0.7 38.1 ⫺3.3 40.4 ⫺4.3 58.4 ⫺1 51.7 ⫺6.2 55.8 ⫺0.1 61.9 ⫺2.6 57.7 ⫺1.6 15 to 29 hours 25.5 ⫺0.9 28.9 ⫺0.1 49.4 ⫺4.4

(B)

Periods

⫺1.3 ⫺0.5 ⫺2.6 ⫺1.5 ⫺0.3 ⫺3.6 ⫺2.2 ⫺1.6 ⫺2.7 ⫺5.7 ⫺4.2 ⫺1 ⫺1.2 ⫺1.1 ⫺3.7

1987–2000 1987–2000 1987–2000 1995–2000 1995–2000 1987–2000 1995–2000 1987–2000 1987–2000 1987–2000 1987–2000 1987–2000 1987–2000 1990–99 1987–2000

⫺0.8 ⫺0.4 ⫺4.5

1993–2000 1979–98 1979–2000

Working-hour gaps

49

To show the extent to which marginal part-time contributes to the overall level of part-time employment, the ratio of marginal part-time as a proportion of total part-time (defined as less than 30 hours per week) is calculated for each country in Table 2.2. Obviously, the relatively high incidence of part-time work in the UK, New Zealand, the Netherlands and Australia is associated with a greater reliance on marginal part-time. In these countries, 50 per cent or more of total part-time work falls within the category of marginal part-time work. Interestingly, however, some of those countries in which collective negotiation plays an essential role (e.g. the Nordic countries and Germany) also show a heavy reliance on marginal part-time work.16 One common feature of marginal part-time work in the industrialized countries is the predominance of female workers in this type of work, as is the case for part-time work as a whole. While there are some crosscountry variations (see Table 2.3), in the EU around 80 per cent of marginal part-time work was taken by women in 1999. It can thus be said that the gender bias underlying part-time work overall is also echoed in marginal part-time employment (see Fagan, Chapter 4, for a further discussion of this issue). Another important feature regarding marginal part-time work is the over-representation of young workers among marginal part-timers. As noted earlier, in many countries with a high incidence of marginal parttime, people aged between 15 and 24 years are highly integrated into the labour market while they are continuing their studies (see Table 2.3). Certainly, working hours of less than 20 a week are relatively well received by such workers. In Denmark, for example, this young age group accounted for 64 per cent of all marginal part-time work in that country in 2000, which is associated with high employment rates for this cohort. The high incidence of young workers among marginal part-timers also applies to other countries in which marginal part-time work is frequent, such as the UK and the Netherlands.17 One notable exception is Germany where marginal part-time work is rather proportionately distributed among different age groups. In Germany, female second-earners appear to be the functional equivalent of student workforces in Denmark, the Netherlands and the UK (Lehndorff 2001). Involuntary part-time work Involuntary part-time work constitutes another dimension of working-hour deficits. The involuntariness of part-time work is normally determined depending on whether part-time workers have taken their current jobs because they could not find a full-time job. In this sense, it is this type of part-time workers who may have the largest gap between their preferred and actual working hours. Furthermore, in some cases, involuntary part-time employment is a working-hour-related transition strategy for the

64.0 47.9 46.0 40.2 38.0 29.7 23.2 13.2 12.0 11.9 11.2 10.5 9.0 6.9

14.2 19.9 18.5 13.7 15.9 17.3 26.3 37.3 20.9 22.3 23.2 31.4 23.3 22.8

25–34 6.4 12.3 11.2 19.8 22.7 22.9 24.2 24.8 21.2 25.4 30.2 29.4 32.0 32.1

35–44 7.3 8.1 10.2 17.7 16.9 17.1 18.2 18.0 21.6 26.6 20.3 19.8 27.3 28.8

45–54 8.1 11.8 14.1 8.6 6.4 13.0 8.1 6.8 24.4 13.9 15.2 8.8 8.4 9.4

55–64 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0

Total

Sources: EUROSTAT raw data; OECD Labour Market Statistics; EU Labour Force Survey 2000.

Denmark Sweden Finland Ireland The Netherlands The UK Spain Greece Portugal France Germany Austria Belgium Italy

16–24

Composition of MPT by age (% distribution)

Table 2.3 Composition of marginal part-time work (EU, 2000)

62.3 69.4 71.8 77.3 76.0 81.7 86.4 74.0 85.6 82.3 84.8 88.8 83.3 63.9

Female share of total MPT (%)

67.1 36.9 45.4 47.8 68.4 55.9 31.8 26.9 41.9 28.3 46.1 52.5 30.3 26.1

Employment rates for the 15–24 age group

Working-hour gaps

51

unemployed – especially when they are desperate to become employed to meet their financial needs (Caputo and Cianni 2001; Farber 1999).18 In addition, there is some evidence that involuntary part-time work is associated with a lower quality of part-time employment, particularly lower wages. For example, Barrett and Doiron (2001) found that in Canada, average wages for involuntary part-time workers are around 18 per cent lower than those for their counterparts voluntarily working part-time. Involuntary part-time employment is also known to be more common among the socially disadvantaged, such as younger, less educated, women and black workers (Stratton 1996). Other evidence also indicates that involuntary part-time work may negatively affect mental health (Dooley et al. 2000). Basic data on involuntary part-time work is available for many industrialized countries, reflecting its increasing importance in these countries. Yet, the definition of involuntary part-time varies across countries, which makes a strict cross-country comparison very difficult. Figure 2.6 shows the share of involuntary part-time employment among total part-time employees in selected countries where involuntary part-time work is defined as above.19 As in the discussion of marginal part-time work, the threshold of a 30-hour week is used to determine part-time work. As Figure 2.6 demonstrates, the overall incidence of involuntary part-time work is considerable in many OECD countries, such as Finland, France, Greece, Portugal, Spain, and Sweden (over 20 per cent in each country), while the ratio is much lower in the Netherlands, the USA, the UK and Norway (below 10 per cent). While it might be assumed that male part-time workers are less likely than their female part-time counterparts to accept their current jobs (hence, they would exhibit a higher incidence of involuntary part-time work), Figure 2.6 suggests that such a reverse gender gap is the case only in some countries, such as the UK, Germany, Ireland, and France. The opposite situation is found in Denmark, Finland and Greece. It is noteworthy that, as involuntary part-time workers have a preference for full-time employment, their proportions tend to be positively related to unemployment rates. In 2000, for example, countries with higher unemployment rates, such as Spain (14.1 per cent), Finland (9.7 per cent), and France (9.5 per cent) show a higher incidence of involuntary part-time employment, while the figures are lower in countries with better labour market performance in terms of unemployment, such as the Netherlands (2.3 per cent), Norway (3.5 per cent), the US (4.0 per cent), and the UK (5.6 per cent).20 It is also known that this correlation also applies to the historical trends of involuntary part-time employment in many industrialized countries (Stratton 1996; Tilly 1996). From a policy perspective, the key questions that arise from this comparison are as follows: (1) how can involuntary part-time employment be reduced? and (2) does a lower incidence of involuntary part-time employment imply a reduced need for policy initiatives? To examine these questions, a closer look at part-time workers’ preferences is essential.

52

Sangheon Lee

38.98

Portugal

41.61 31.31 33.5

Greece 26.92 26.65 26.15 28.77

Spain

Finland

25.15 26.69

21.88

24.24

France

23

29.33

23.38 23.48 22.96

Sweden 19.63 19.67 19.52

Italy

18.3 18.32 18.21

Belgium Ireland

12.71

9.36

24.42 11.21 12.96

Denmark 7.05

10.71 9.66

Germany

17.29

Total

9.51 9.52 9.35

Norway

Women

8.17 6.43

UK

Men 16.37

6.1 5.56 7.23

USA

3.65 3.27 4.87

Netherlands 0

5

10

15

20

25

30

35

40

45

Figure 2.6 Involuntary part-time work in selected countries, 2000 (source: OECD Labour Market Statistics 2002).

The preferences for full-time work among involuntary part-time workers are likely to be shaped mainly by two factors. One factor is the presence of economic disincentives/discrimination against part-time work, and the other is the sociological/psychological factors underlying workers’ preferences. First, it is conceivable that the relatively low quality of parttime work as compared with full-time work discourages workers from favouring part-time work, while providing incentives for employers to offer part-time work. Thus, it might be said that a higher quality of parttime work should be associated with a lower incidence of involuntary parttime work. However, this conjecture is not well supported by Table 2.4, which summarizes earnings for part-time workers and the overall perceptions of social protection for part-time workers compared with full-time

3.3 (1) 5.6 (2) 6.4 (3) 9.7 (5) 13 (6) 9.4 (4) 23.5 (8) 23 (7) 26.7 (9)

4.9 (1) 7.2 (3) 16.4 (4) 17.3 (5) 7.1 (2) 24.4 (9) 23 (8) 21.9 (6) 21.9 (6)

Male 3.7 (1) 6.1 (2) 8.2 (3) 10.7 (4) 11.2 (5) 12.7 (6) 23.4 (7) 24.2 (8) 25.2 (9)

Total 93.1 (1) 62.5 (8) 69.6 (7) 87.5 (4) 76.4 (6) NA 92.3 (2) 81.7 (5) 90.2 (3)

Female

■ 69.8 (6) 44 (8) 54.2 (7) 78.4 (2) 74.2 (4) NA 88.7 (1) 73.2 (5) 76.5 (3)

Male 73.2 (5) 54.3 (8) 58 (7) 82.5 (3) 74.2 (4) NA 87.2 (1) 73 (6) 82.6 (2)

Total

Hourly earnings, 1995a (Median hourly earnings for full-time workers ⫽ 100)

■ 66.7 (1) NA 22.9 (8) 52.2 (3) 45.6 (5) 27.1 (7) 47.8 (4) 61.4 (2) 44.4 (6)

Perceptions of part-time protection (%, 1998)b

Sources: a OECD Employment Outlook 1999, Table 1.5. b Gasparini (2001), Table 1. The figures indicate the proportion of respondents who said ‘No’ to the question ‘Do you think that in general part-timers are worse off than full-time workers so far as protection by employment law and social security is concerned?’. The figures are re-estimated excluding the respondents who said ‘Don’t know.’

The Netherlands The USA The UK Germany Denmark Ireland Sweden France Finland

Female

Involuntary part-time employment (% of total part-time employment, 2000)

Table 2.4 Involuntary part-time work and the quality of work (country ranks in brackets)

54

Sangheon Lee

workers, along with the incidence of involuntary workers. Overall, the Netherlands and the Nordic countries demonstrate the possibility of a positive link between a high quality of part-time work and low involuntary part-time employment; however, a completely different picture is found in the UK and the USA. In these latter countries, part-time workers were earning around half of what their full-time counterparts earned per hour, and most people perceived that part-time work had disadvantages in terms of social protection, yet the rates of involuntary part-time in these countries are relatively low. Partly due to these two contrasting cases, the correlation coefficients between the three columns in Table 2.4 were not significant. Thus, from a comparative perspective, the relative quality of part-time work in a certain country may not fully explain the difference in the extent of involuntary part-time work in the country concerned. The lack of a consistent relationship between involuntariness and quality of part-time work is to a large extent related to the fact that parttime workers’ responses to the question of ‘voluntariness’ depend on their perception of the acceptability of part-time work, which tends to be ‘adjusted’ to their sociological, psychological, cultural and institutional environments. As Fagan (2001b: 243) succinctly notes, ‘the degree of dis/satisfaction with current practice, and the preferences for some alternative, will partly depend on people’s perception about what the feasible alternatives are’.21 More specifically, it is entirely plausible that, apart from the conditions of full-time work (e.g. whether or not long hours are involved) and the type of child-care services available, workers’ views on part-time work are more or less shaped by their observations of what their ‘reference groups’ (i.e. people like them) are doing. In other words, workers’ perceptions about ‘alternatives’ can be influenced by what their friends or neighbours are normally doing in choosing employment options. For example, if a worker finds that many of their neighbours or friends are working on a part-time basis, then they are more likely to regard part-time work as acceptable than those workers surrounded by friends who are in full-time employment. This explains in part why countries with a higher incidence of part-time employment tend to have a lower incidence of involuntary part-time employment, as is demonstrated in Figure 2.7 (the correlation coefficient is ⫺0.812, significant at the 1 per cent level).22 This relationship demonstrates the complexity of policy interventions aimed at reducing involuntary part-time work, especially when a considerable number of ‘disguised’ involuntary part-time workers exist. It should be understood that the discussions above do not deny the importance of policy interventions regarding involuntary part-time work. That good policies can matter is well illustrated by the Dutch experience. Although the overall proportion of involuntary part-time employment in the Netherlands is now the lowest in the industrialized world, this ratio had been relatively high until the 1990s – much higher than the ratio in the UK (see Figure 2.8). It should be noted that, in the late 1980s and early

Figure 2.7 Relationship between part-time employment and involuntary part-time employment, 2000, selected OECD countries (source: OECD Labour Market Statistics).

Figure 2.8 Trends in involuntary part-time work: the UK vs the Netherlands (source: OECD Labour Market Statistics).

56

Sangheon Lee

1990s, considerable policy changes were undertaken to improve incentives for part-time work, particularly by removing discrimination in social security rights between full-time and part-time workers. These changes include the 1990 Pensions and Savings Act, which outlawed the use of an hour-limit for entry into occupational pension funds (Visser 2002: 33), and the extension of minimum wage protection to short-hour part-time workers during the 1990s. The incidence of involuntary part-time workers in that country has decreased dramatically since these active policies were implemented, as is shown in Figure 2.8. This experience in the Netherlands is in sharp contrast with that of the UK where, until recently, few policy interventions concerning the quality of part-time employment were made and involuntary part-time employment has been slightly increasing.

Conclusion Changes in working time in the industrialized countries are multidimensional and increasingly complex. Underlying the decreasing trends in national average hours of work in many countries is the diversification of working hours among individual workers, the degree of which obviously differs from one country to another depending largely on their regulatory and institutional frameworks. Along with these changes, concerns are increasingly expressed regarding the gap between workers’ actual and preferred working hours. Although there are still methodological problems with estimating these gaps, some key aspects of them are discussed in this chapter: working-hour surpluses (excessively long working hours) and working-hour deficits (marginal and involuntary part-time work). The analysis of excessively long working hours has indicated that in some countries the proportion of the workforce that is affected is very low, but in other countries, particularly those that have rather weak regulatory frameworks, the ratio of affected workers is much higher and even increasing. When it is intended to discourage long working hours, it is important to note that long hours are often linked to specific incentive structures (e.g. wage gains, promotions, job security, etc.), which in turn are shaped by the overall labour market structure. If this kind of an incentive structure is in place, it will not be easy for individual workers to address their own working-hour surpluses. In this case, the difference between extended hours based on voluntarism (i.e. workers’ own ‘rational choice’ to work longer hours for additional income) and by force is only a subtle and thin one. Thus, collective and coordinated approaches, along with individual initiatives, are expected to play an important role in reducing workinghour surpluses. In some cases, part-time work is one way of addressing working-hour surpluses, a theme which has been widely echoed in various debates on ‘family-friendly’ working time. Yet, part-time work itself is often the cause

Working-hour gaps

57

of a different type of gap between preferred and actual working hours: working-hour deficits. It has been shown that part-time work is only ‘tolerated’ (rather than welcomed) by many part-time workers. This observation sometimes causes a concern that working-hour deficits might add a new dimension of labour market segregation to the already existing segregation among full-time workers (Tilly 1996). However, the policy implications of this development in working-hour deficits are complex. Certainly, measures towards improving the quality of part-time work, such as those stipulated in the ILO Part-Time Work Convention, 1994 (No. 175), would be helpful in addressing working-hour deficits. They include a pro-rata approach with respect to basic wages for part-time workers and the requirement that they receive conditions equivalent to comparable fulltime workers in the areas of paid annual leave and public holidays, sick leave and maternity protection. While the analysis in this chapter shows that social, cultural and psychological factors could be important in determining the effectiveness of these policy efforts, these active policy measures may well contribute to making the overall environment more favourable to creating voluntary, high-quality part-time work. Finally, one might argue that an over-working society is less likely to suffer from working-hour deficits, but we are currently witnessing the coexistence of working-hour surpluses and deficits in some countries. This situation would imply the possibility of working-hour redistribution – a socalled ‘time transfer’ from workers with excessive working hours to those whose working time is less than desired. This approach is worthy of consideration in some countries (see Drago 2000), but it requires effective and intensive coordination among different labour market and economic policies (e.g. training/education, income, social security policies). Thus, a far broader set of policies would be required to implement such an approach. Another key consideration in this type of approach should be the gender bias underlying these working-hour gaps: working-hour surpluses mainly concern male workers, while working-hour deficits predominantly affect female workers. This implies that a redistribution of paid working hours should also involve a ‘time transfer’ from male to female workers (the gender dimension of working time is addressed by Fagan in Chapter 4).

Acknowledgements The author would like to thank Dominique Anxo, Peter Auer, François Eyraud, Colette Fagan, Adriana Mata-Greenwood, Jon Messenger, Deirdre McCann and Catherine Saget for their helpful comments.

58

Sangheon Lee

Notes 1 There have been lively debates on what should be measured for actual working hours. For example, questions have been raised about the relevance of weekly working hours, as various reference periods for the calculation of average working hours have been introduced (Mata-Greenwood 2001). These debates have often led to a more fundamental question, i.e. What is the correct concept of working time in today’s society? (Supiot 2001: Chapter 3). 2 Other institutional factors, such as social insurance systems, also affect the distribution of working hours. See Anxo in Chapter 3. 3 According to the 1999 survey by the Japanese Trade Union Confederation (Rengo), 30.5 per cent of overtime was unpaid. The ratio was highest at 66.6 per cent in the financial and insurance business (Rengo 2001a). 4 An optimistic prediction by the National Economic Planning Agency is that this legal change will create around 500,000 jobs in France, once it is fully implemented (EIRO 2001: 4). 5 While the proportion of workers who are not covered by collective agreements is relatively low in Denmark, those workers are more likely to work longer and receive lower overtime compensation (Scheuer 1999). 6 However, the data from non-European countries are not directly comparable, because different wordings are used. The difference mainly comes from (1) if survey questions refer to changes in incomes according to changes in working hours (if this is the case, the incidence of workers’ preferences for reduced working hours tends to decrease); (2) if so, how these income changes are incorporated in the survey question. 7 This method is based on the concept of ‘inadequate employment related to excessive hours’, referred to as the ‘situation where persons in employment wanted or sought to work less than they did during the reference period, either in the same job or in another job, with a corresponding reduction of income’ (ILO, Resolution Concerning the Measurement of Underemployment and Inadequate Employment Situations, October 1998, Paragraph 17). 8 See note 6. 9 Undoubtedly, working less than 50 hours per week could still be seen as ‘excessive’ in many cases. For example, working couples with children may find that 45 hours or more per person does not provide sufficient time for child-care (see Anxo in Chapter 3). Similarly, even a 40-hour workweek might be physically infeasible for older workers. It is also important to note that more than half of workers working between 40 and 49 hours per week have reported workinghours surpluses (see Table 2.1). 10 This result is unchanged even when the threshold of long working hours is reduced to 45 hours. 11 ILO Resolution concerning the measurement of underemployment and inadequate employment situations (October 1998, Paragraph 7). 12 This group of part-time workers is often also called ‘secondary part-time’ (Tilly 1991). 13 However, our analysis indicates that the use of other thresholds such as 15 hours per week does not change the main points in this section. 14 The threshold or ‘cut-off point’ of 30 hours is adopted because the contractual working hours for full-time workers are close to 35 hours per week in many of these countries (OECD 1997). It should also be noted that these figures on part-time are different from those based on ‘spontaneous response by the respondents’, such as, for example, EU statistics on the incidence of part-time work (see Eurostat 2001: Table 12.) 15 Note that the hours threshold of marginal part-time for Australia is 15 hours.

Working-hour gaps

16

17

18

19 20 21 22

59

According to ILO (2002a: Table 6a), around 18 per cent of workers worked less than 20 hours per week in 2000. It should be remembered that the figures for marginal part-time work may be different from those calculated from national statistics, because the cut-off of 30 hours per week is used for the definition of part-time work in Table 2.2. Countries where a higher cut-off is normally used (such as Sweden) may have lower incidences of marginal part-time work than that reported in Table 2.2. When students account for a large proportion of marginal part-time workers, it may be difficult to draw any policy implications. Some argue that the inclusion of the student workforce in the standard employment statistics will bring about a statistical distortion (Hakim 1998: 174–5). At the same time, a Dutch study indicates that student part-time workers tend to replace low-paid unskilled workers (Wielers and van de Meer 2001). There are few studies on how successful these involuntary workers are in making the transition to full-time work after a period of part-time work, but a significant proportion of these workers seem to drop out of the labour market (Anxo and O’Reilly 2000; O’Reilly and Bothfeld 2002). In Canada, about 20 per cent of involuntary part-time workers are found to be unemployed or out of the labour market after 18 months in their jobs (Noreau 2000). ‘Time-related under-employment’ is often used to refer to involuntary parttime employment in some countries such as New Zealand and Australia. The Spearman’s coefficient for 14 selected OECD countries is 0.622, significant at the 5 per cent level. It is also possible that workers, after they unwillingly accept part-time work, will ‘convince themselves that they actually prefer part-time hours’ (Tilly 1996: 4). The exclusion of the USA increases the correlation coefficient to ⫺0.900.

3

Working time patterns among industrialized countries A household perspective Dominique Anxo

Introduction Households and families are seldom considered to be the relevant unit of analysis for the study of working time. Yet, despite the growing interest in finding new forms of employment and working time patterns favouring a better reconciliation of paid work and family commitments, there has been remarkably little comparative research into the determinants of participation and working time arrangements at the household level. Most studies on working time rely on individual data, disregarding the distribution of working time within and across households with different demographic, socio-economic or life-cycle characteristics. This relative dearth of research applies both within industrialized countries, where the impact of changing male and female participation on life-cycle patterns of welfare has not been widely studied, and also within the comparative research literature where there should be rich scope for analysing the impact of the different welfare state systems, employment and working time regimes on household participation and working time patterns over the life cycle. One of the most salient features and persistent trends in advanced economies is the increased feminization of the labour force, the related shift from the single male breadwinner household towards dual-earner households, and the increased diversification of household structures. Despite these common trends implying a significant reduction of the gender employment gap, there are still large differences in the patterns of female labour market integration and the extent of male breadwinner or dual-earner couples among advanced countries. Furthermore, among dual earners, large cross-country disparities still persist regarding working time patterns between spouses. In this context, one important issue is to contrast the incidence of various working time patterns among different types of households and in different societal regimes, and to assess the main factors affecting the distribution of working time among household types (singles, lone parent families, couples and couples with children) and across countries. The link between the patterns of household labour market integration,

Working time: a household perspective 61 both in terms of participation and working time arrangements, and the generation of individual income and welfare (in terms of social inclusion or exclusion) is another motivation for researching the patterns of participation at the household level. Independent of the country context, individuals will have different needs and preferences depending on different life phases and household structures, and thus will face different constraints, given the institutional framework and the prevailing working time patterns in a particular country. Determining the extent to which the current pattern of participation is meeting individual and household lifecycle needs and preferences, and the extent to which these needs are met through a balanced gender distribution of paid work, a dominant male breadwinner regime or by the state through tax and benefit systems, appears to be crucial to assess the necessary welfare, working time and labour market policy reforms. The main objective of this chapter is therefore to identify and compare household labour market integration and working time patterns across industrialized countries and analyse the potential link between the extent of labour market participation and the prevailing welfare and working time regimes. We intend, in particular, to examine how family formation and household composition affect both male and female labour supply and working time patterns. Special focus will be placed on the prevailing legal possibilities or institutional barriers to adapt working hours over different life phases. Time is a scarce resource and modifications in household structure over time may, depending on the national context, have different effects on a household’s time allocation between paid work and other time-consuming activities. Our hypothesis is that a part of the crosscountry disparities in the patterns of household labour market integration may be ascribed to institutional factors such as the design of family policy, the tax and benefit systems, the availability and cost of child-care facilities, gender wage differentials and firms’ recruitment strategies (selectivity). In other words, there are grounds for thinking that the specificity of a country’s institutional framework may affect and shape the choice that a household’s members can make regarding the extent of their labour market participation and the overall gender division of labour. The prevailing working time regimes, in particular the range of available working time options over the life course, may also favour or hinder a better balance between work and other social activities. The lack of opportunities to temporarily reduce or increase working time over the life course is also one factor explaining the current gap between actual and preferred working time (see Fagan, Chapter 4). Therefore, institutional barriers preventing men and women from adapting their working time in order to cope with various events/risks over the life course need to be identified and removed in order to foster not only gender equal opportunity but also a more efficient and optimal resource allocation. This chapter is structured as follows. After an analysis of the overall

62

Dominique Anxo

societal characteristics and the ‘societal logic’ of the selected countries, we investigate the cross-country patterns of household labour force participation and working time patterns, focusing on household structure and composition. Then, using a standard econometric model (Tobit), we analyse the impact of socio-economic variables on household labour market participation and working time. Finally, the major policy implications and conclusions of our study are provided.

The societal characteristics of the countries selected One reason for the relative paucity of comparative studies on working time arrangements at the household level is the lack of suitable data across countries. Some attempt has been made to overcome this problem by the creation of a data bank of harmonized data sets known as the Luxembourg Income Study.1 The main advantage of this dataset, which is the one used in this chapter, is that it provides information on most of the key variables that are of relevance for this study, such as duration of working time, working time patterns and income distribution, enabling a systematic comparison of working time arrangements among households with different demographic, socio-economic and life-cycle characteristics. The selection of industrialized countries and time period (mid-1990s) in this chapter has been driven principally by data availability. Nevertheless, the selected sample of 8 industrialized countries, 6 countries belonging to the European Community (France, Germany, Italy, the Netherlands, Sweden and the United Kingdom) and two non-European countries that are part of the OECD area (Australia and the USA), covers a wide spectrum of the potential societal systems, with variations in labour market characteristics, welfare state regimes, gender relations, levels of labour market participation and working time patterns. In this section we present an overall summary of the characteristics of these societal systems. Four broad interrelated dimensions have been employed to characterize these societal regimes: (1) the nature of the welfare state; (2) the current gender contract; (3) the employment regime; and (4) working time regime. Along the lines developed by Esping-Andersen (1990, 1999) the various welfare state regimes have been classified according to the degree of decommodification, i.e. the extent to which individuals have to rely on the market, and the nature and extent of family policy. Following EspingAndersen, we distinguish four broad welfare state regimes: (1) AngloSaxon liberal (Australia, the UK and the USA); (2) continental conservative (France, Germany and the Netherlands); (3) Mediterranean (Italy); and (4) Nordic social democrat (Sweden). The various regimes are expected to produce different patterns of labour market participation, whether in terms of employment rates or in terms of working time patterns. The employment and institutional regime of a country as a whole, and

Working time: a household perspective 63 its specific gender relations, shape opportunities and constraints, and influence the choice households can make regarding the extent of their participation in the labour market. The gender allocation of time between paid work and other activities may be affected by a set of institutional and economic factors. To illustrate: the design of the tax and benefit system, for example, in the case of a joint taxation system with high marginal rates, may raise barriers and disincentives, affecting the gender division of labour and constraining the labour supply of second earners within households. The extent to which households may shift or outsource domestic tasks (such as the availability and cost of child-care, elder care, etc.), are also other examples of factors which may affect households’ allocation of time between paid work and other activities. The regulatory framework regarding working time may also have an impact on labour supply and affect both the duration and the distribution of working time (see Anxo and O’Reilly 2000; Lee, Chapter 2). For instance, the lack of opportunities to temporarily work reduced working time (such as part-time work) or the absence of flexible working time arrangements at the firm level may constrain women’s participation in the labour market. The institutional rules used to regulate working time reflect different industrial relations’ traditions and the differences in bargaining systems, that is, differences in their degree of centralization and coordination (see Anxo and O’Reilly 2002; McCann, Chapter 1). These differences can have a significant impact on the length and distribution of working time. Essentially, regulation can take place at four levels: (1) at the national level through the universal application of statutory legislation; (2) at the branch or industry level through collective bargaining applied to a range of firms or sectors; (3) at the plant or company level through localized collective agreements; and (4) at the purely individual level through the employment contract concluded between an employer and employee.2 The relevance, the co-existence, and the strength of each of these levels of regulation vary considerably between the countries examined in this study. Some countries rely essentially on collective agreements at industry and plant level (Germany, the Netherlands and Sweden), while in other countries the role of statutory regulation and state intervention appears to be crucial in accounting for the parameters of differences in the length and distribution of working time (France and Italy). Finally, in countries with relatively weak labour market regulation (Australia, the UK and the USA), working time is essentially determined through enterprise-based collective agreements, employment contracts and market forces. Hence, the profile of the working time distribution in a given country is partly a reflection of the relative strength and degree of articulation of the various levels of regulation, which in turn are intimately related to the type of industrial relations regime, the regulatory role of the state in the labour market, and the degree of centralization and coordination of the bargaining process in the country.

64

Dominique Anxo

We may expect countries characterized by a strong statutory system of regulation (France and Italy), or a highly centralized and coordinated bargaining system with high coverage rates (Sweden), to have a significant concentration of employees around the statutory or collectively agreed working time norm. These countries would, therefore, have a narrower distribution of working time patterns. This implies that highly centralized and coordinated bargaining systems may produce working time patterns similar to those countries relying on statutory norms (that is, a high concentration around a standard norm). In contrast, we might expect a relatively wider variation in the distribution of working time in countries where the statutory norm is weaker and where negotiation at industry level is prevalent (Germany and the Netherlands). This is because negotiated working time standards may vary significantly between bargaining areas. However, in countries with a strong tradition of collective bargaining, the impact of the regulation system on the distribution of working time depends to a large extent on whether working time is de facto an issue in the bargaining process. The distribution of working time may vary among countries with similar regulation and bargaining regimes because of differences in the general orientation of working time policy, which might reflect distinct priorities and objectives of the two sides of industry. Finally, countries with no or weak statutory working time regulation, an extremely decentralized bargaining process and/or an individualization of working time through the employment contract and market forces may be expected to exhibit the largest variation in working time across industries and individual workers (Australia, the UK and the USA). The following developments examine in more detail the ‘societal logic’ underlying the patterns of household labour market participation and working time arrangements, using the above-mentioned classification of welfare state regimes. As mentioned previously, the institutional regime of a country as a whole, and its specific structure of gender relations, shape the choices that household members can make regarding the extent of their participation in the labour market. It is therefore important to identify to which extent the institutional framework can explain crosscountry disparities in household labour market attachment and working time patterns. The Scandinavian countries, here represented by Sweden, are a good illustration of the so-called Nordic social democratic welfare states, emphasizing the principles of egalitarianism, de-commodification and individualization (Esping Andersen 1999). In the whole spectrum of social policies, individualization has been a key part of the Swedish universal welfare state. The basic principle of the institutional model is entitlement based on citizenship. The individual, and not the family, has for many years been the unit not only of taxation but also of social benefits as social rights. Sweden, like the other Scandinavian countries, stands out as providing one type of societal system based on high employment rates with a

Working time: a household perspective 65 small gender gap, a high incidence of dual-earner households (see Figure 3.1), extensive and generous family policies, strong welfare support systems both for child-care and parental leave, and egalitarian wage structures, including low gender wage inequality. Individualized taxation systems in a context of high average and marginal tax rates reinforce the dual breadwinner model. As far as working time is concerned, the high union density and the relatively high degree of centralization and coordination of collective bargaining produce a rather low dispersion of working time and a high concentration of dependent employees around the standard full-time norm (40 hours). Some gender differences persist with a relatively high share of women working part-time, but in contrast to other countries with high part-time

Figure 3.1 Distribution of couples by working arrangements, all couples, 18–64 years old (source: LIS and author’s own calculations).

66

Dominique Anxo

rates like the UK or the Netherlands, many women in Sweden work long part-time hours (more than 24 hours a week) and receive income compensation for working reduced hours; the parental leave system3 allows for partial paid leave to be taken, thereby reinforcing women’s bargaining power and status as a significant breadwinner even when they are temporarily not participating on a full-time basis in the labour market. The overall political context characterized by gender mainstreaming, and high female involvement in the political process and institutions (government bodies, parliament and labour market organizations), creates a favourable institutional set up conducive to a more balanced gender division of labour and responsibilities. Australia, the United Kingdom and the United States belong to the liberal Anglo-Saxon welfare state regimes with a clearly lower degree of de-commodification and high dependency on the market. Furthermore, those countries exhibit neither a strong family system nor a strong welfare state system; consequently, all individuals are constrained to rely to a greater extent on the market for support even during periods of high responsibility for dependants. But some differences among these countries have to be stressed. After Sweden, the USA is the country with the next highest share of dual-earner households (see Figure 3.1). The USA stands out by having primarily dual full-time earners, and a medium dispersion of working time with a relatively high incidence of long hours of work for both men and women. The logic of the dual-earner household in the USA is, however, quite different from that in the Scandinavian countries. In the USA the gender pay gap is high but so is the overall level of wage dispersion. High participation rates among single and married women appear to be related both to the weak family and welfare support system, and to high levels of educational attainment and high returns to education. In contrast to other industrialized countries, the low incidence of part-timers is related to the low access of part-timers to health insurance benefits (see McCann, Chapter 1), discouraging US women from taking up part-time jobs. In contrast to the United States, there is in Australia, and to a lesser extent in the UK, more evidence of the resilience of the male breadwinner model (see Figure 3.1), with a large proportion of men working long hours and for relatively higher pay, while women are supplementing male earnings through working as part-timers (with a large incidence of marginal part-time, see next section) both during and after the main period of responsibility for children (see Rubery et al. 1998a, 1999). One of the striking differences is the larger dispersion and the higher gender polarization in the distribution of working time in Australia4 and the UK, as compared with the USA. The male distribution of working time exhibits the highest cross-country variation, and is heavily skewed with a relatively high incidence of long working time. Even though the dispersion in female working time is also high, the female distribution of working time, on the other hand, is significantly flatter with a relatively high share of marginal part-

Working time: a household perspective 67 timers (see Anxo et al. 1999, and next section). In the last decade some policy measures in the UK have been undertaken to improve child-care facilities and to provide more parental leave, although still on an unpaid basis (see Rubery et al. 1998a, 1999). The three countries are similar in that the public provision of child-care facilities is low and that family policies of employers to facilitate parental leave and/or flexible hours are more important than those of the state. Similarly, the individualized tax system can be considered relatively neutral for labour market participation because of low overall and marginal rates favouring the use of long working hours. But the exemptions from social security contributions for low earners may be one factor explaining the high level of marginal part-time work in the UK (see also McCann, Chapter 1 and Fagan, Chapter 4). Italy belongs to the so-called Mediterranean conservative welfare state regime, far less egalitarian and de-commodified than the Nordic social democratic welfare states. Italy displays a low level of public provision of child-care for young pre-school children (younger than 3 years old) and limited welfare support (social assistance and family benefits).5 One of the central features of the Italian welfare state remains the central role of family as a provider of care and as the relevant locus of solidarity (Bettio and Villa 1998). Among those countries reviewed in this chapter, Italy is the country with the lowest overall and married female employment rates6 and the highest occurrence of traditional male breadwinner households (40 per cent of couples, see Figure 3.1). This situation implies a high economic dependency of wives on their husbands. Furthermore, Italy has one of the lowest fertility rates among industrialized countries and an increased marital instability, reflecting the increasing tensions between the changing employment patterns of young well-educated generations of Italian women and an institutional framework that encourages women to devote themselves to taking care of dependants. Overall, Italy displays a relatively low dispersion of working time and a high concentration of dependent employees around the standard norm, reflecting, as in France, the predominant role of the state in the regulation of working time. The limited availability of part-time jobs or flexible working time arrangements, coupled with the shortage of services to substitute for the informal care work of women, reinforces the male breadwinner system. France, Germany and the Netherlands belong more to the so-called continental welfare state model than the Nordic or Anglo-Saxon models already described. These models were based primarily on the notion of a male breadwinner system with the state providing support through tax and benefits for a non-working spouse. Compared with the Anglo-Saxon liberal welfare states, the labour market of the continental welfare state is, however, more regulated. Within these three countries, however, there are considerable differences in the form and shape of the labour market, working time regime and welfare system.

68

Dominique Anxo

Germany is taken as the prime example of the traditional strong male breadwinner welfare state (35 per cent of couples, see Figure 3.1). The high levels of tax subsidies to the family through the tax splitting system, the high wages for men employed in manufacturing, and the low availability of child-care for children under 3 years of age, all support the notion of the housewife system of child-care and the male breadwinner system of family provisioning (Anxo et al. 1999; Rubery et al. 1999). There are many elements of this model still evident in Germany but some changes are occurring. First of all an increasing number of women are entering the labour market, despite the strong male breadwinner welfare model remaining largely intact. Like Italy, however, Germany has marked regional variations in women’s labour market integration; these regional variations are rooted in the pre-unification process. While the male breadwinner arrangement still dominates West Germany, the dual full-time earner model was dominant in East Germany. The weakening of the dualearner model in East Germany is principally related to the rapid fall in women’s employment and the large increase of female unemployment in East Germany after unification. As in Italy, many women are also choosing not to have children, as Germany displays a low fertility rate in contrast, for example, to the Nordic countries or, indeed, the UK. Overall, the distribution of working time in Germany displays a medium dispersion, partly reflecting variations in negotiated working time standards between bargaining areas (see Anxo et al. 1999). However, a large share of German men still work long hours (see the next section) while German mothers tend to follow British mothers with a high incidence of part-time work. Women’s employment rates and the incidence of dual-earner households is higher in France, but in contrast to Germany and the Netherlands (see below), the incidence of female part-timers in France is notably lower and there is more evidence of mothers working full-time when they have children than is the case in at least Germany and the Netherlands. From the age of 3, more than 80 per cent of children in France are covered by public child-care arrangements. In France maternity leave amounts to 16 weeks and the parental leave system, which was introduced in the early 1980s, does not generally provide compensation for lost earnings, unless it is coupled with the parental child-rearing allowance, which is a relatively low, flat-rate sum. The relatively high coverage rates of public child-care and the relatively short maternity leave create strong incentives for women to quickly return to the labour market after maternity leave, but in contrast to other countries with high provision of public child-care like Sweden, this re-entry occurs for a large proportion of women on a fulltime basis (see Anxo et al. 2002). Globally, the French working time distribution displays a higher concentration of employees around the statutory norm, reflecting the predominant role of the state concerning working time regulation, due in particular to relatively low levels of unionization, and consequently, a weak tradition of collective bargaining. In contrast to

Working time: a household perspective 69 Germany, however, the non-fully individualized taxation system in France probably has only a relatively weak impact on female labour supply, as there are only limited opportunities for men to transfer income to their non-working spouses. The Netherlands may be said to stand between the continental and the Nordic models because access to transfers and benefits is based more on citizenship, similar to the Nordic model, than on the employment record, as in France and Germany. However, in other respects, the Dutch system is quite different. Even though the incidence of the traditional male breadwinner does not significantly depart from that in France, the Netherlands displays, however, the lowest share of dual-earner households in which both spouses work standard hours and the highest share of dual earners in which the male works full-time standard hours and the spouse marginal part-time hours (see below). This situation is partly related to the very limited provision of child-care facilities and to the fact that women in the Netherlands are expected to a large extent to provide the care. The distribution of female working time is negatively skewed7 and very flat, with a high dispersion of working time. The male distribution of working time is strongly peaked around the standard working time, with an extremely low incidence of long hours. In contrast to the UK, however, part-timers tend to be relatively well paid, and do not face a major additional downgrading of their hourly pay levels (Fagan et al. 1995). Also in dual-earner households, the short hours for women are combined with only medium length, not extra long, average working hours for men, as for instance in Australia and the UK (see next section).

Household employment and working time patterns This section examines cross-country disparities in employment rates and working time patterns by household type. One objective here is to analyse how marital status (single and cohabiting/married couples) and household composition (in particular, the presence of children) affect household labour market integration in the different societal regimes described in the previous section. Gender employment gap by household type As shown by Table 3.1, the patterns of labour market integration by household types (singles/couples, lone parents/couples with children) vary widely across countries. In all countries, however, average employment rates are higher among couples than among single households. Obviously, these disparities in employment rates across household types reflect structural effects, in particular, differences in the demographic structure between couples and single households. To illustrate: the large employment gap between Swedish married/cohabiting males (females) and single

69 81 56 73 59 87 54

Couples Couples (all) Males (all) Females (all) Males no child Females no child Fathers Mothers

69 80 58 64 53 89 62

66 69 64 68

France 1994

Source: LIS and author’s own calculations.

60 71 52 40

Singles Singles (all) Males Females Lone Mothers

Australia 1994

66 78 54 69 58 87 50

64 71 60 51

Germany 1994

53 70 35 51 27 87 42

49 61 43 49

Italy 1995

65 75 55 66 57 87 53

63 77 55 29

The Netherlands 1994

Table 3.1 Employment rates by household types and gender, 18–64 years old

84 87 83 82 80 91 85

54 53 55 65

Sweden 1995

66 72 60 66 64 77 56

51 55 49 35

United Kingdom 1995

83 91 75 86 76 94 73

82 86 79 77

USA 1997

Working time: a household perspective 71 males (females) should not be interpreted as specific difficulties in the labour market integration of single males (females), but merely reflects the relatively larger share of young individuals among single households (see Tables 3.A1 and 3.A2 in the Appendix). In other words, a large part of the discrepancy in employment rates between household types may be ascribed to individuals’ different phases over the life course (i.e. these individuals are in a transitional phase from education to the labour market or family formation). Overall, the gender employment gap is less pronounced between single men and women than between cohabiting men and women. As shown by Figure 3.2, independent of household type the most egalitarian gender distribution of labour market participation is found in Sweden. Swedish single women exhibit even higher employment rates than their male counterparts. In the other countries, single males there have on average significantly higher employment rates than their female counterparts. Two of the Anglo-Saxon liberal countries, the UK and the USA, and two of the conservative continental regimes, France and Germany, display a relatively small gender employment gap among single households (between

Figure 3.2 Gender disparities in employment rates, male employment rates and female rates (source: LIS and author’s own calculations).

72

Dominique Anxo

5 and 7 per cent). The highest gender disparity among single households is found in the Netherlands (22 per cent), Australia (19 per cent) and Italy (18 per cent). The highest employment rate among married/cohabiting women is found in Sweden (83 per cent). American married/cohabiting women also display a relatively high degree of labour market integration (75 per cent), while Italy exhibits the lowest employment rates among married cohabiting women (35 per cent). In the remaining countries married women’s employment rates range from 54 per cent up to 60 per cent. As expected, children have a gender-differentiated impact on employment rates. Not surprisingly fathers display higher employment rates than mothers (see Figure 3.2), but the cross-country differences are significant. The lowest employment gap between fathers and mothers is found again in Sweden (6 per cent), followed, but at a much higher level, by two of the Anglo-Saxon liberal countries, the USA and the UK (21 per cent). The highest gender employment gap between parents is found in Italy (45 per cent), followed by the three continental conservative regimes, Germany (37 per cent), the Netherlands (34 per cent) and France (27 per cent). Also worth noticing is that, independently of country and household types, fathers display on average higher employment rates than cohabiting males without children (see Figure 3.3). There are grounds for thinking that the positive correlation observed between the male labour supply and children reflect the fact that obtaining paid employment is a precondition for deciding to have children for married/cohabiting men (see Anxo et al. 2002). Conversely, married/cohabiting mothers tend to have lower employment rates than similar women without children. Among countries with overall high female employment rates, such as Sweden and the USA, the gap between mothers and childless cohabiting women is, however, very low. Sweden, as mentioned previously, has since the early 1970s implemented a set of specific policy measures that support and strengthen mothers’ labour market commitments (a generous and flexible parental leave system with employment guarantees and a highly subsidized childcare system). As far as the USA is concerned, the lack of institutional support for combining work and family and the lower degree of decommodification and social transfers, combined with a relatively higher level of family instability (higher divorce rates), create strong incentives for mothers to continue to work. With regard to lone mothers’ labour market integration, the variation in employment rates is wider and ranges from 29 per cent in the Netherlands to 77 per cent in the USA (see Figure 3.4). France and Sweden have relatively high employment rates among lone mothers, above 65 per cent. Also worth noticing is that lone mothers in France and the USA exhibit higher employment rates than cohabitating mothers. However, the conditions supporting this may be quite different given France’s higher degree

Working time: a household perspective 73

Figure 3.3 Employment rates, married and cohabitant households, fathers and mothers (source: LIS and author’s own calculations).

of de-commodification and widespread child-care facilities, whereas American lone mothers may to a greater extent be forced to work, given the lower level of social welfare benefits available to them. Average working hours Table 3.2 shows average weekly working time for dependent employees (18–64 years old) by gender and household type. Comparing average weekly working time among household types within and across countries presents the same methodological difficulties as employment rates, since average figures may conceal large differences in the age distribution and also other structural effects. The cross-country differences in average weekly working time between couples and single households may therefore partly reflect disparities in the age structure of the sub-population (single/couples). This is particularly true for the Netherlands and Sweden

74

Dominique Anxo

90

77

80 68

70

65

60 51

49

50 40 40

30

35 29 24 21

20

20 14 10 4 0 ⫺1 ⫺6

⫺10

⫺4

⫺20 NL

UK

Australia

Lone mothers’ employment rates

Italy

Germany

Sweden

France

USA

Mothers/lone mothers differential

Figure 3.4 Lone mothers’ employment rates and differential between cohabitant mothers and lone mothers (source: LIS and author’s own calculations).

with an over-representation of young singles (see Table 3.A1 and 3.A2 in the Appendix). Since in these two countries single households are on average younger than individuals living in couples, we should expect lower average weekly working time, reflecting the different phases over the life course already mentioned above. This is true for Swedish single males and females but only for Dutch single males. Actually, Dutch single women work on average longer than their married counterparts. Therefore, family formation in the Netherlands seems to have a clear negative impact on Dutch female labour supply that goes beyond differences in the demographic structure among household types, reflecting a more traditional gender contract and gender division of paid work. As mentioned also in the previous section, the transition from being single to being married or cohabitant seems not to affect Dutch women’s labour force participation rate, but notably reduces their average working time. Since in the remaining countries, the age distribution among household types does not diverge notably, the comparison by household types is more reliable. Independently of household type, males in the three countries belonging to the Anglo-Saxon liberal regime and also in Germany exhibit on average longer weekly working time than in the other countries. Among female employees, American, and to a lesser extent French and Italian, women also display longer average weekly working time. This is

27.8 35.7 36.5 37.8 41.1 42.2 42.5 43.0

37.5 40.2 40.5 41.7 42.4 43.9 44.0 43.4

Married males

Source: LIS and author’s own calculations.

SW IT NL FR AUS UK USA GE

Single males

9.7 4.5 4.0 3.9 1.3 1.7 1.5 0.4

Married males–singles males 27.3 34.2 30.7 33.4 34.6 32.3 38.7 36.7

Singles females

31.7 34.0 25.2 34.5 30.6 23.9 36.7 31.8

Married females

Singles men–women

0.5 1.5 5.8 4.4 6.5 9.9 3.8 6.3

Married females– singles females ⫺4.4 ⫺0.2 ⫺5.5 ⫺1.1 ⫺4.0 ⫺8.4 ⫺2.0 ⫺4.9

Table 3.2 Weekly working hours among couples and single households, dependent employees, 18–64 years old

5.8 6.2 15.3 7.2 11.8 20.0 7.3 11.6

Couples men–women

76

Dominique Anxo

partly linked to the impact of the regulatory system on the working time distribution, with a high concentration of employees around the standard full-time norm and a comparatively low incidence of female part-time work. Owing to the large incidence of marginal part-time in the Netherlands and the UK, the shortest weekly working time is found among British and Dutch married women (see below). Except for Swedish single households, in which there are no significant gender differences in weekly working time, single male employees in the remaining countries work on average longer hours than their female counterparts. Among dependent employees, married females tend to work shorter hours than employed single females. The working time differential between married and single females is particularly important in the UK, where married women work on average eight hours less per week than their single counterparts. Among continental conservative regimes, German married/cohabiting women also work significantly shorter hours than their single counterparts. As in the Netherlands, family formation in all these countries has negative effects on female labour supply; however, this is not in terms of labour force participation, since in almost all the countries cohabiting married women have higher employment rates than single women, but merely in the form of a reduction in average weekly working time. Two exceptions are worth noticing. The first exception is Italy, where family formation still takes essentially the form of a withdrawal of women from the labour market and no changes in weekly working time for those women still in employment. The second exception is Sweden, where family formation is positively related to female labour force participation and also has a positive effect on working time. As far as the impact of children is concerned, fathers not only display higher average employment rates, but also work slightly longer weekly hours than married/cohabiting males without children (see Figure 3.5). The weak but positive impact of children on males’ working time might be partly ascribed to the associated reduction in household income: in order to compensate for reduced income, partly related to the reduction in the female labour supply, cohabiting/married men tend to work longer hours. Not surprisingly fathers also work longer average hours than mothers in all countries. Among employed parents, the largest gender disparity in working hours is found in the Netherlands (18 hours). Dutch employed mothers have very short average hours (19 hours) due to the large incidence of short part-time work in this country. In two of the Anglo-Saxon liberal countries, namely the UK and Australia, the gap in working hours between employed parents is also important (15 and 18 hours respectively). Among lone mothers, average weekly working hours for dependent employees vary from 26 hours (UK) to 39 hours (Italy). Employed lone mothers work, on average, longer hours than married/cohabiting employed mothers in all countries except the UK where the difference is

Working time: a household perspective 77 50

43.9

43.7

41.7 41.8 40.5

40

38

31.7

32.9

34 31.8

23.9

25.2 19.1

HOURS

25

44.2

26.7

27.4

31

34.5 30.6

30

34.4

37.5 35

44

35.4

40.2 40.7

44.7

41.4 36.7

43.4

42.4 42.8

31.2

45

20

15

10

5

0 Australia

France

Germany

Italy

NL

Sweden

UK

USA

COUNTRY Males

Females

Fathers

Mothers

Figure 3.5 Average weekly working time, couples, dependent employees (source: LIS and author’s own calculations).

negligible. Australia, Sweden and France present almost no differences in working hours between lone mothers and employed cohabiting mothers. It is also worth noticing that lone mothers’ working hours are relatively longer in the Netherlands and Italy. In these countries, lone mothers work between seven and eight hours more than their female counterparts living in couples (see Figure 3.6). The Dutch and Italian lone mothers’ situation is also interesting since they on average exhibit relatively low employment rates but when they work, they work long average working hours. In other words, while the degree of labour market integration of lone mothers is low in the Netherlands (29 per cent) and Italy (49 per cent), those lone mothers in employment work relatively long hours compared to cohabiting mothers and also compared with lone mothers in the other industrialized countries. For Italy, this is probably related to the specificity of the working time regime with, as mentioned above, a relatively low incidence of part-time work and lower working time flexibility. The low spread of working time in Italy, therefore, has a tendency to produce lower female employment rates and also relatively long average weekly hours among lone mothers and parents.

78

Dominique Anxo

Figure 3.6 Gender differential in weekly working time in single households, and between mothers and lone mothers; dependent employees (source: LIS and author’s own calculations).

Working time patterns by household type Working time patterns among single households Average weekly working hours may conceal large differences in the distribution of working time. We have classified working time arrangements into four broad working time categories: (1) ‘marginal’ part-time (less than 16 hours per week); (2) standard part-time (16 to 34 hours per week); (3) standard hours ranging from 35 to 42 hours to take into account the crosscountry heterogeneity in the regulatory framework regarding standard working time; and (4) long weekly hours as 43 hours and above. As shown by Figures 3.7 and 3.8 (single households) and also by Figures

Working time: a household perspective 79 100%

4 13

17

29 80%

37

33 38

38

70

60% 61 73

40%

64

58 53

56

58

20% 9

0%

13

6

7

1

0

3 1

0

Australia

France

Germany

Italy

19

13 8 NL

Sweden

7

8

2

1

UK

USA

COUNTRY

Marginal part-time 1-15 hours

Part-time 16-34 hours

Standard 35-42 hours

Long hours 43+

Figure 3.7 Working time patterns, males, singles (source: LIS and author’s own calculations).

3.9 and 3.10 (couples), the gender distribution of working time varies notably across countries. Overall, men are concentrated in the upper tail of the working time distribution (standard and long working hours), while females’ working time distribution displays a much higher dispersion. Some cross-country differences have to be stressed. Among employed single males, marginal and standard part-time is relatively frequent in the Netherlands and Sweden, reflecting the above-mentioned larger proportion of young single males combining part-time with other activities (mainly education). The incidence of marginal part-time is also significant among British and Dutch single women. At the other end of the working time distribution, the incidence of long hours among employed single males is particularly important in the three Anglo-Saxon liberal countries (from 33 per cent up to 38 per cent) and also in Germany (38 per cent). To a lesser extent it is also the case in the two other continental conservative countries, France (29 per cent) and the Netherlands (17 per cent). If, overall, the incidence of long hours is more prevalent among single males than employed single females, the incidence of long hours is also

80

Dominique Anxo 0

100% 5 13 19

17

14

18

14

19

38 80%

51 62

49

60% 50

63

59

58

63

40%

28

20%

29

21 22

22

25

17

14

3

0% Australia

14

16

10

France

9

7

4

2 Germany

Italy

NL

Sweden

UK

USA

COUNTRY

Marginal part-time 1–15 hours

Part-time 16–34 hours

Standard 35–42 hours

Long hours 43+

Figure 3.8 Working time patterns, females, singles (source: LIS and author’s own calculations).

important among single females in the three Anglo-Saxon liberal countries. Except in Sweden, where the incidence of long hours is negligible, the proportion of single females working long hours is also important in two of the continental conservative countries, France and Germany, and in Italy as well. Working time patterns among dual-earner households (couples) In dual-earner households, the incidence of male part-timers is low overall, the highest incidence being found in Sweden, Italy and the Netherlands. In those three countries, standard working time is clearly the dominant working time arrangement among married/cohabitant males, with a relatively low incidence of long working hours. The remaining countries display a very low incidence of male part-time work but a much larger incidence of long hours. As for single households, the incidence of long hours for males in dual-earner households is particularly high among the liberal Anglo-Saxon countries (between 37 up to 46 per cent) and also in Germany (42 per cent) (see Figures 3.9 and 3.10).

Working time: a household perspective 81 100% 8 21

22

29 37

80%

42

46

43

60%

82 70 40%

70

66 59 55

50

54

20%

0%

8

4 1

9

7

3 1

2 1

0

2

2

2 1

3 1

Australia

France

Germany

Italy

NL

Sweden

UK

USA

COUNTRY

Marginal part-time 1–15 hours

Part-time 16–34 hours

Standard 35–42 hours

Long hours 43+

Figure 3.9 Working time patterns, males, dual earners (source: LIS and author’s own calculations).

In the liberal Anglo-Saxon regimes, the large incidence of long working hours among males living in dual-earner households is probably related to the relatively large spread in the earnings distribution and the high returns to education, linked with low average and marginal tax rates. In those countries with relatively weak regulatory systems, the relatively large incidence of low paid jobs, together with the high returns to education, foster the emergence of a culture of long weekly working time. As far as Germany is concerned, the high incidence of long hours might be ascribed to regional differences with a higher incidence of long hours in East Germany,8 partly reflecting the large differences in the earnings distribution between West and East Germany. Not surprisingly, females in dual-earner households exhibit a larger spread in working time patterns than their male counterparts. But once again large cross-country disparities exist. The incidence of long hours among married/cohabitant women is significant in two of the Anglo-Saxon countries, the USA (15 per cent) and Australia (13 per cent), and highest among German women (17 per cent). The UK presents a somewhat lower incidence of long hours among married women (9 per cent) a level similar to France’s. Again, Swedish and Dutch married women display, like their

82

Dominique Anxo

100% 13

3

6

9

2 9 15

17

90%

24 80% 51 70%

42

40 58

58 44

60% 60 47

50%

40%

30%

34

34 41

29 20%

33

28

20

26 10% 14

15

11 5

0% Australia

France

Germany

6

6

2 Italy

NL

Sweden

UK

USA

COUNTRY

Marginal part-time 1–15 hours

Part-time 16–34 hours

Standard 35–42 hours

Long hours 43+

Figure 3.10 Working time patterns, females, dual earners (source: LIS and author’s own calculations).

male counterparts, the lowest incidence of long hours. At the other end of the working time distribution, marginal part-time is highest among Dutch cohabiting women (26 per cent of females living in dual-earner households), followed by Australian and British married women (14 and 15 per cent, respectively). Standard part-time is also clearly predominant among Dutch and Swedish women (47 and 41 per cent), and ranges between 28 and 34 per cent in France, Germany, Australia, and the UK. In the USA and Italy, on the other hand, standard weekly working time clearly dominates (60 per cent of females living in dual-earner households). As shown by Figure 3.11, the incidence of long hours among lone mothers in the Anglo-Saxon liberal countries and Germany is particularly important. In all of these countries, except Sweden, the proportion of lone mothers working long hours is even larger than the proportion for married women. Here two extremes are worth noticing. On one hand, in the USA, with a low degree of de-commodification and relatively weak family and public child-care systems, the situation of lone mothers is characterized by both a high level of labour force participation and also on average long working hours, implying that a majority of them must rely to a greater extent on the market for income generation and the provision of child-

Working time: a household perspective 83 100%

1.6 6.4 16.5

12.5

11.8 15.7

16.5

18.9

90%

80% 42.1 56.7

70%

35.9

39.4 47.1

60%

59.9 57.5

65.7

50%

40%

30%

28.6 39.6

26.2

35.3 30.4 20% 26.3 10%

20.1

23.8

17.9

15

11.9 1.3

0% Australia

France

6.8

6.4

3.5

Germany

Italy

NL

Sweden

2.8 UK

USA

COUNTRY

Marginal part-time 1–15 hours

Part-time 16–34 hours

Standard 35–42 hours

Long hours 43+

Figure 3.11 Working time patterns, lone mothers (source: LIS and author’s own calculations).

care services, and that opportunities to combine work and family commitments are limited. On the other hand, we find the Swedish social democratic system, with strong welfare support systems both for child-care and parental leave allowing high labour market integration of lone mothers, but with greater possibilities to adapt and reduce working time with limited loss of income,9 implying that the Swedish system favours a better reconciliation between work and family responsibilities. Working time arrangements between spouses, married/cohabiting couples As mentioned above, large disparities exist among industrialized countries regarding the extent of female labour market integration. Except for Sweden and the USA, where dual-earner households constitute a large majority (more than 70 per cent of couples), the male breadwinner model is still prevalent in some countries like the Italian Mediterranean regime (40 per cent) and also to a lesser extent in the conservative continental welfare states (see Figure 3.1). Working time distribution between spouses provides a good illustration

84

Dominique Anxo

of the prevailing gender contract and complements the analysis of labour market integration performed previously. As shown by Figures 3.12 and 3.13, the various welfare state regimes produce a rather contrasting gender division of paid work. The three Anglo-Saxon liberal market-oriented countries display a large incidence of couples where both spouses work long hours (between 5 and 8 per cent of dual-earner households). In the USA, the most common working time pattern remains the combination in which both spouses work standard full-time, and also couples in which the husband works long hours and the wife standard hours (58 per cent of all dual-earner couples). The corresponding figures for Australia and the UK are 39 per cent and 41 per cent, respectively. The main difference between the USA and the two other Anglo-Saxon liberal market-oriented countries is the relatively higher incidence of American couples in which both spouses work fulltime. The spread in working time patterns and the gender polarization of working time also appear to be more pronounced in the UK and in Australia. Among the continental conservative countries, working time patterns between spouses also present large disparities. In France, the dominant working time patterns remain the ones in which both spouses work stan-

25

20

15

10

5

Male long hours Male standard working hours

0 Female marginal Female part-time part-time Female standard working hours

Male part-time Male marginal part-time Female long hours

Figure 3.12 Working time patterns among dual earners, Australia (source: LIS and author’s own calculations).

Working time: a household perspective 85

25

20

15

10

5

Male long hours Male standard working hours

0 Female marginal Female part-time part-time Female standard working hours

Male part-time Male marginal part-time Female long hours

Figure 3.13 Working time patterns among dual earners, UK (source: LIS and author’s own calculations).

dard normal working time (40 per cent of dual earners) and in which the husband works standard full-time and the wife part-time (18 per cent of dual earners). Compared to the Anglo-Saxon liberal countries, the working time distribution among French couples exhibits less variation, partly reflecting the features of the prevailing working time regulatory regime (State-dominated regulation, see above). Worth also noticing is that Italian couples exhibit the same patterns of working time arrangement as in France. Hence, one of the major differences between the two Latin countries is the higher incidence in Italy of the male breadwinner model and the related lower female employment rates. The two remaining continental conservative countries, Germany and the Netherlands, also constitute two polar cases regarding the division of paid work between spouses (see Figures 3.17 and 3.18). German working time arrangements are rather similar to those found in the Anglo-Saxon liberal countries, with a relatively high gender polarization of working time and a very high share of couples where both spouses work long hours (11 per cent). The high incidence of German couples where both spouses work long hours is certainly related to the previously noted regional earnings

86

Dominique Anxo

40

35 30 25 20 15 10 Male long hours 5

Male standard working hours

0 Female marginal Female part-time part-time Female standard working hours

Male part-time Male marginal part-time Female long hours

Figure 3.14 Working time patterns among dual earners, USA (source: LIS and author’s own calculations).

disparity between Eastern and Western Germany. Dutch dual-earner households present a rather different gender division of paid work, with a working time distribution clearly skewed to the left, and the lowest proportion of couples in which both spouses work full-time (18 per cent), as well as a relatively high incidence of couples in which the husband works full-time and the wife works either standard or marginal part-time (50 per cent). There are reasons to believe that the prevailing Dutch situation reflects some form of catching up, and thus represents a transitional phase between the previously dominant male breadwinner model and a progressive integration of women into the labour market through the development of part-time work and working time flexibility. Finally, Sweden appears to be the country with the most evenly distributed intra-household gender division of paid work, be it in terms of labour market participation or in terms of working time. The bulk of Swedish couples are found in two main working time arrangements: both spouses working full-time (44 per cent) and men working full-time and women working standard part-time (34 per cent). To sum up, our descriptive analysis clearly shows that the patterns of household labour market participation and working time arrangements

Working time: a household perspective 87

45 40 35 30 25 20 15 10 Male long hours 5

Male standard working hours

0 Female marginal Female part-time part-time Female standard working hours

Male part-time Male marginal part-time Female long hours

Figure 3.15 Working time patterns among dual earners, France (source: LIS and author’s own calculations).

vary considerably among the selected industrialized countries. These differences are partly related to the design of the prevailing welfare state and employment regimes. The Nordic social democratic regime, here represented by Sweden, is characterized by overall high employment rates, the highest incidence of dual-earner households, and extremely low gender disparities in labour market integration. In contrast to other industrialized countries, family formation has a clear positive impact on female labour market participation. Sweden’s flexible and generous parental leave system, coupled with a highly subsidized and extensive child-care system, makes it possible for parents to better reconcile paid work with family commitments. Compared to the other industrialized countries, Sweden also exhibits a relatively low gender polarization of working time, with an extremely low incidence of excessive working hours and marginal parttime work. Low income inequality, including low gender wage differentials, in a context of high average and marginal tax rates, discourages the use of long working hours, while qualification rules for some benefits (e.g. working time thresholds), such as unemployment insurance, discourage the use of marginal part-time work.

88

Dominique Anxo

45 40 35 30 25 20 15 10 Male long hours 5

Male standard Working hours

0 Female marginal Female part-time part-time Female standard working hours

Male part-time Male marginal part-time Female long hours

Figure 3.16 Working time patterns among dual earners, Italy (source: LIS and author’s own calculations).

In the liberal Anglo-Saxon welfare state regimes, in which workers are more dependent on the market and alternatives to labour market income are more limited, we also found a high degree of overall labour market integration, but compared with Sweden, there is a somewhat lower level of female labour market integration. Compared with the other industrialized countries, the incidence of long working hours is very high. Relatively large wage inequality, coupled with high returns to education in a context of low taxation, fosters a tradition of long hours. In the three countries the incidence of long working hours appears to be high in the two tails of the wage distribution. On one hand, due to the low degree of de-commodification and relatively high incidence of low-paid jobs, low-skilled/low-paid workers have a higher propensity to work long hours to secure a decent income. On the other hand, highly educated workers also have clear incentives to work long hours due to relatively high returns to education, low marginal and average tax rates, and also the many opportunities to outsource part of their domestic tasks to the market.10 However, notable disparities exist between the USA and Australia and the UK. The latter two countries present more evidence of a resilience of the male breadwinner model, lower female employment rates and also a

Working time: a household perspective 89

30

25

20

15

10

5

Male long hours Male standard working hours

0 Female marginal Female part-time part-time Female standard working hours

Male part-time Male marginal part-time Female long hours

Figure 3.17 Working time patterns among dual earners, Germany (source: LIS and author’s own calculations).

much higher gender polarization of working times. In the three countries, family formation has a positive impact on men’s labour supply and also increases the incidence of standard and long working hours. On the other hand, family formation has a weak impact on female labour supply in the USA and Australia, while in the UK female labour market participation increases significantly. In the USA the female distribution of working time is hardly affected by marital status, but the incidence of long hours among cohabitant/married women becomes lower than among single females. In Australia and the UK, the incidence of long hours is also reduced when women get married, but the share of female employees in standard and marginal part-time work increases significantly. The conservative continental welfare state regimes, France, Germany and the Netherlands, stand out as providing a societal system with relatively lower employment rates and also larger gender disparities. But, as with the Anglo-Saxon regimes, some differences must be stressed. The spread in working time distribution and the gender polarization of working time are clearly higher in Germany and in the Netherlands compared with France. A part of these differences may be ascribed to disparities in

90

Dominique Anxo

35

30

25

20

15

10 Male long hours

5

Male standard working hours 0 Female marginal Female part-time part-time Female standard working hours

Male part-time Male marginal part time Female long hours

Figure 3.18 Working time patterns among dual earners, the Netherlands (source: LIS and author’s own calculations).

working time regimes and also to differences in the public provision of care. In France, the coverage rates of public child-care are clearly higher, and the incidence of part-time work is much lower. As in the other industrialized countries, family formation has a clear positive impact on male labour force supply and a negative impact on female labour supply, both in terms of labour participation and working time. Compared to single women, the share of married/cohabitant women working standard hours and long hours becomes significantly lower. Worth also noticing is the large incidence of long hours in Germany, which is probably related to large regional differences in wage distribution and skill levels between Eastern and Western Germany, and also reflects different traditions rooted in the pre-unification situation. Finally, the Mediterranean welfare state regimes, represented here by Italy, display the lowest female employment rates and the highest incidence of the traditional male breadwinner model. Family formation and the presence of children have a clearly negative impact on female labour market integration, but essentially in terms of lower labour force participation. The relatively low spread in the Italian working time distribution and the low incidence of part-time work have a tendency to reduce the

Working time: a household perspective 91

45 40 35 30 25 20 15 10 Male long hours 5

Male standard working hours

0 Female marginal Female part-time part-time Female standard working hours

Male part-time Male marginal part-time Female long hours

Figure 3.19 Working time patterns among dual earners, Sweden (source: LIS and author’s own calculations).

gender gap in working hours for dependent employees. The relatively low public provision of child-care facilities, coupled with the prevailing rigidities in the working time regime, still constitutes a barrier to women’s labour market integration and supports a traditional gender division of labour.

The socio-economic determinants of labour force participation and working hours within households: an econometric approach While the descriptive analysis performed in the previous sections of this chapter reveals some interesting features, the observed discrepancies in the patterns of labour market integration and in the gender allocation of paid work among the selected industrialized countries may conceal important structural differences in household composition and characteristics. The main objective of this section is to control for these factors, and thus shed some light on the determinants of the gender allocation of paid work within households by using a standard econometric approach (specifically, a Tobit model with selection). This section is concerned with

92

Dominique Anxo

questions such as how marital status, age, educational attainment, and the number and age of children can affect the gender division of paid work among single households and between spouses in those couples in married/cohabiting households in the eight countries. Obviously, disparities in welfare state traditions and in regulation and other institutional factors may hinder or favour a more equal gender division of labour. Hence, besides the above-mentioned individual and household characteristics, special attention is devoted to country differences in employment regulation (the extent of working time flexibility), the nature of family policy (parental leave, etc.) and the public and private provision of household and care services (child-care facilities, etc.). A detailed presentation of the estimation results is reported in the statistical appendix to this chapter (Table 3.A5). Interesting facts that emerge from the estimations are described below. Independent of marital status and countries, the impact of age on labour supply, be it in terms of participation in the labour force or working hours, points in the same direction: labour market participation increases with age but at a decreasing rate. Generally, the impact of age on labour market participation does not seem to be dependent on marital status. An interesting exception is the Netherlands, where labour market participation for married and cohabitant females starts declining very early compared with that of single women, still implying some form of early withdrawal of Dutch married females from the labour market. Overall, the average age in which labour market participation starts decreasing ranges from 35–36 years old for males (the UK and the USA) up to 42 years old (Sweden). Gender significantly affects labour market participation among single households. Overall, to be a single woman reduces, ceteris paribus, the extent of paid work, be it in terms of participation or working time. The most notable exception is Sweden where no gender differences are identified, implying a rather high degree of gender equal opportunity in this country among singles. The largest negative gender impact is found in Italy, Germany and the USA (see Table 3.A5 for details). The decomposition of the total marginal effect also reveals interesting differences. To be a single woman in Italy mainly affects the likelihood of working, while the situation is reversed for American and German single women for whom the major effect is a reduction of working time for those who are working. Among couples, time spent on paid work by each spouse may be affected by the labour market participation of the spouse. As far as the interaction effects are concerned, our results indicate that there is a marked gender complementarity of market work between spouses in Australia, Sweden and the UK. In the first two countries, most of this effect is due to an increase in the likelihood of working, while in the UK a larger share of the total effect is due to an increase in working hours. In other words, in the UK the longer one of the spouses works, the longer their

Working time: a household perspective 93 partner is likely to work. The impact of the spouse’s working hours is much lower in the remaining countries. Also worth noticing is that in Italy the husband’s working hours do not have any significant effect on his wife’s labour market participation, while in the Netherlands a husband’s working hours reduce his wife’s likelihood of working. Independent of marital status, gender and country, individuals’ own educational attainment11 increases their likelihood to work. Not surprisingly, the extent of labour market participation is correspondingly lower or higher among people with lower or higher levels of educational attainment. The quantitative impact of educational attainment on participation nevertheless varies among countries. Among single households, Italy, the UK and the USA show the highest labour market participation differentials between people with low and high educational levels. To illustrate, in Italy, highly educated singles work on average 16 hours more per week than those with low levels of education; the corresponding figures for the UK and the USA being 14 and 13 hours, respectively. The decomposition of the total marginal effect reveals that educational attainment influences mainly labour force participation rates in Italy, while in the UK and the USA it affects both participation and average working hours. More precisely, compared to singles with low educational levels and given that they work, highly educated American and British single men work on average three hours more a week than single men with a low level of education. Among couples, educational attainment may affect either partner’s allocation of paid work, and this occurs mainly through one channel: education affects earnings opportunities and thereby each individual’s comparative advantage or bargaining position. In other words, we may expect that the higher the differential in educational attainment between husband and wife, the higher the degree of specialization and the disparity in the gender division of work. The results of the estimation show clearly that the extent of labour market participation of married and cohabiting women is also strongly affected by their educational attainment. As for single households, most of the effect may be ascribed to labour force participation rates and not to working hours (given that they work), confirming that educational attainment remains a determining factor for female labour market integration and gender equal opportunity. As far as the interaction effect is concerned, educational attainment of the wife weakly affects her husband’s labour supply. Assuming a probable high level of assortative mating, highly educated couples in Australia and the USA work on average more than couples with low or medium levels of education. In other countries, on the other hand, the relationship is reversed: in couples in which the wife is highly educated, the husband reduces his labour supply (essentially working hours), but the quantitative impact remains low. Except in the Netherlands and the USA the education level of the husband does not affect the wife’s labour market participation. In the Netherlands, women married to highly educated men exhibit a higher

94

Dominique Anxo

level of labour market participation, while in the USA we have the reverse situation. Not surprisingly, the age and number of children affect individual labour supply, particularly for females. Overall, the impact of children is a decreasing function of the children’s age. Our estimation also confirms that the presence of young pre-school children is a strong source of gender differentiation, but large variations exist across countries (see Table 3.A5). Globally, young pre-school children do not affect (Australia, Germany, Italy and Sweden) or reduce weakly (France, the Netherlands and the UK) married or cohabiting males’ labour supply, be it in terms of labour force participation rates or working hours. In the USA, young children marginally increase married men’s labour market participation (approximately one hour on average), the impact being evenly distributed between increased labour force participation rates and increased working time. Figures 3.20 (lone mothers) and 3.21 (married and cohabitant females) decompose the total effect of young pre-school children (less than 3 years old) on labour force participation and working time. On the other hand, the presence of young pre-school children has a dramatic effect on females’ labour supply, both for married/cohabitant females and for lone mothers. As shown by Figures 3.20 and 3.21, the impact of young children on the 100% 2 90%

6

1 1

6 2 9

80%

70%

60%

50% 18 40%

23

8 4

20 5 17

30%

20%

10%

0% Germany

UK

Australia

Netherlands

France

Italy

USA

Country

Effect on labour force participation

Effect on working time

Figure 3.20 Impact of young pre-school children on labour participation and working time, singles (source: LIS and author’s own calculations).

Working time: a household perspective 95 100% 1

90%

1

7 80%

6

4 7 8 3

70%

%

60%

50% 5

40%

4

20 30%

8

6 8 7

20%

2

10%

0% Germany

Australia

Netherlands

UK

USA

France

Italy

Sweden

Country

Effect on participation

Effect on working time

Figure 3.21 Impact of young pre-school children on labour force participation and working time, cohabiting and married females (source: LIS and author’s own calculations).

married/cohabitant female labour supply, be it in terms of participation or working hours, is on average lower compared to lone mothers. An interesting exception is Germany, where the impact of children is of the same order of magnitude for married/cohabitant females as for lone mothers, implying that the strong negative impact of young pre-school children is independent in this country of marital status. Among other countries showing a large impact of children on the married female labour supply, we found the three Anglo-Saxon liberal countries and also the Netherlands. All these countries are characterized by a relatively low provision of public child-care facilities, confirming the central role of public child-care on gender equal opportunity, in general, and female labour supply, in particular. Other interesting features are that the impact of young children on married/cohabiting women’s working hours is high in the UK (55 per cent of the total effect), Australia, (45 per cent) and the Netherlands (around 40 per cent). This situation can be explained by the relatively substantial opportunities to work part-time in those countries, and it reflects the fact that ‘critical events’ such as the birth of a child, are met to a greater extent than in Germany by working time flexibility and to a lesser extent by a complete withdrawal of women from the labour force. Among these countries with a relatively low impact of young pre-school children

96

Dominique Anxo

on female labour supply, namely France, Italy and Sweden, the large impact of children on labour market participation in Italy and France is worth noticing since more than 80 per cent of the effect may be ascribed to a reduction of the likelihood of work, as in Germany. Conversely, in Sweden, the majority of the effect may be ascribed to a reduction of working time for those in employment, with a very weak impact on labour force participation rates. This contrasting impact of young pre-school children on the female labour supply may largely be ascribed to the specific features of the generous and flexible Swedish parental leave system, which favours a strong labour force attachment for women and a better reconciliation between work and family commitments.

Conclusion As discussed earlier, the extent of labour market integration and working time patterns varies considerably across industrialized countries, household types and gender. Family formation and young children have a strong gender-differentiated impact on labour market participation and working time patterns. Overall, to live as a couple and have children has a small positive impact on the male labour supply, be it in terms of participation or average weekly hours. On the other hand, the impact of family formation and young children on female labour market integration differs considerably among the various welfare states and employment regimes. In the Nordic social democratic regimes, family formation has a clear positive impact on female labour supply and the impact of young children on female labour market integration remains limited. Conversely, in continental and Mediterranean conservative regimes such as Italy, Germany and the Netherlands, family formation and motherhood are still associated with a sharp decline in women’s labour market participation. The relatively low public provision of child-care facilities for young pre-school children in those countries supports the notion of the ‘housewife’ system of child-care and underpins a male breadwinner system of family provisioning. Furthermore, in some countries such as Italy, working time rigidities, in particular, the low availability of part-time jobs, restrict female labour force participation and reinforce the male breadwinner model. Between these two extremes our comparative analysis shows that, in the AngloSaxon welfare state regimes, the negative impact of family formation on women’s labour force participation remains limited. Except in the USA, where marriage and children have a weak effect on female working time distribution, the impact of family formation and children in Australia and the UK takes the form of a large increase of part-time work, in particular marginal part-time work. The incidence of long hours appears also to be more prominent among the Anglo-Saxon liberal welfare state regimes. Weak labour market regulation regarding working time and relatively large wage inequality,

Working time: a household perspective 97 coupled with high returns to education in a context of low taxation, foster a tradition of long hours in these countries. With the notable exception of the USA, where part-time jobs remain limited, the Anglo-Saxon liberal regimes also display a high gender polarization of working time. The incidence of couples where the male works long hours and the female works marginal part-time is also significantly higher in the Anglo-Saxon liberal regimes than in other industrialized countries. The incidence of long hours is significantly lower in the Mediterranean and continental conservative regimes (with the notable exception of Germany) and lowest in the Nordic countries, which display the lowest wage and income inequality. Several policy implications might be drawn from our comparative analysis. In the Anglo-Saxon regimes and also in Germany, the large incidence of long working hours ought to be reduced. As well as the negative impact of excessive hours on workers’ health, a culture of long hours may have also a detrimental impact on individuals’ family and personal lives, making it difficult to balance paid work with other activities. Given the prevailing unbalanced gender division of domestic and caring activities, employment regimes favouring the use of excessive working hours may also reinforce gender inequality by restricting women’s career opportunities. If long hours lead to improved job prospects and greater job security, it may be to the detriment of those who are not able to work long hours – in particular women. Several complementary measures could be envisaged. In addition to regulatory measures to limit the use of excessive working hours, the reduction of wage inequalities appears to be one of the options to curb excessive hours. Encouragement to introduce more familyfriendly working time patterns at the firm level may also be a good instrument to better adapt working time to households’ varying needs and to reduce the incidence of long hours among parents. More globally, the wide range of households’ needs as regards working time patterns, related in particular to household structure and composition, points in the direction of a more flexible adaptation of working time over the life cycle. While statutory limitations on long working hours or a general reduction of standard working time may, for countries with a high incidence of long hours, be a means of curbing excessive hours, the need for working time flexibility over the life cycle, reflecting various household situations and working conditions, cannot be satisfied only by standardized or statutory regulations, giving little room for individual differentiation. Hence, the heterogeneity of household needs and preferences requires more options to adapt working time over the life cycle, that is, a differentiation and variability in working time patterns which favour a better reconciliation between paid work and other social activities. Policy measures have to be undertaken, therefore, to extend the range of working time options in order to better adapt working time to various events/risks over the life cycle. Some recent national initiatives, such as in the Netherlands, to secure, through statutory provision, the individual right to

98

Dominique Anxo

modulate working time over the life cycle appear promising. In our view, increased individual freedom in time allocation over the life cycle not only has to be guaranteed through the application of a universal citizen right, as in the Netherlands, but also has to be complemented by an integrated system of income redistribution and income transfers. The many opportunities to adjust working time over the life cycle, through various forms of income-compensated legal absenteeism and temporary reductions in working time, as in Sweden, allow a more flexible management of work and family constraints. On the whole, as stressed above, family commitments do not have a major impact on Swedish men’s and women’s participation in the labour market, and thus Sweden’s flexible and generous parental leave system with complete employment guarantees appears to be an efficient tool to secure women’s labour market integration and employment continuity, as well as to improve gender equal opportunities. Finally, as stressed above, the detrimental impact of children on female labour market integration in many industrialized countries is largely related to the lack of child-care facilities for young pre-school children. Policies aimed at increasing and developing subsidized child-care facilities should therefore be encouraged. Removing rigidities in the prevailing working time regimes, in particular by increasing the range of legal options to adapt working time over the life course without large income losses, appears, in our view, to be crucial not only in terms of promoting gender equal opportunities, but also in order to secure a better and more efficient allocation of time and resources over the life course.

17.8 34.6 0.48 0.05 0.11 0.26 0.1 42.5 0.18 0.24 0.25 0.33 0.71 0.13 0.17 0.57 0.33

12,219 17,223 14,683 11,512 11,774

Total unit earnings Total unit gross income Disposable income after taxes Adjusted disposable income Sample size

10,273 16,124 15,375 12,658 1,400

23.2 33.4 0.36 0.02 0.14 0.4 0.08 41.7 0.25 0.23 0.23 0.3 0.58 0.13 0.29 0.38 0.25

Mean

Mean

Female variables: Hours worked per week Hours worked per week employed Hours ⫽ 0 Hours 1–15 marginal part-time Hours 16–34 part-time Hours 35–42 standard Long hours 43– Age Age 18–30 Age 31–40 Age 41–50 Age 51–64 Education low Education medium Education high Number of children < 18 Lone mothers

France 1994

Australia 1994

13,488 19,124 13,886 11,932 11,682

22.0 36.7 0.40 0.04 0.1 0.35 0.11 42.6 0.27 0.21 0.18 0.34 0.08 0.74 0.18 0.38 0.27

Mean

Germany 1994

9,664 16,644 16,644 11,778 11,651

15.5 34.2 0.57 0.01 0.11 0.26 0.06 49.1 0.07 0.16 0.28 0.49 0.73 0.21 0.06 0.29 0.21

Mean

Italy 1995

Table 3.A1 Mean values for LIS samples of singles, individuals between 18–64

Appendix

10,741 18,265 12,763 11,256 11,650

17.4 30.7 0.43 0.09 0.16 0.29 0.03 39.4 0.33 0.23 0.19 0.25 0.17 0.63 0.19 0.30 0.19

The Netherlands 1994 Mean

8,608 15,487 11,808 10,339 2,024

17.7 27.3 0.45 0.05 0.16 0.34 0.00 36.6 0.44 0.17 0.17 0.22 0.29 0.48 0.22 0.35 0.22

Mean

Sweden 1995

10,793 18,836 15,426 11,658 1,021

15.7 32.3 0.51 0.07 0.11 0.24 0.07 40.7 0.27 0.26 0.21 0.26 0.69 0.21 0.10 0.82 0.46

Mean

UK 1995

24,522 30,580 25,049 18,625 8,828 continued

30.7 38.7 0.21 0.03 0.11 0.5 0.15 41.1 0.23 0.26 0.26 0.25 0.16 0.53 0.31 0.78 0.41

Mean

USA 1997

13,420 16,129 15,067 11,862

18,731 14,633 13,994 11,444

30.4 43.0 0.29 0.01 0.02 0.41 0.27 38.4 0.34 0.29 0.17 0.20 0.08 0.73 0.18 0.03 0.03

Mean

Germany 1994

Notes All income measures in USD/year adjusted by PPP. Self-employed are excluded. a The adjusted disposable income is equal to disposable income/(family size)0.5.

Source: LIS and author’s own calculations.

18,224 16,058 14,913 11,530

Total unit earnings Disposable income after taxes Adjusted disposable incomea Sample size

28.4 37.8 0.31 0.00 0.05 0.44 0.20 39.2 0.31 0.26 0.20 0.22 0.6 0.1 0.3 0.05 0.04

Mean

Mean

29.4 41.1 0.29 0.01 0.04 0.40 0.26 40.1 0.24 0.28 0.22 0.26 0.52 0.31 0.17 0.09 0.06

France 1994

Australia 1994

Male variables: Hours worked per week Hours worked per week employed Hours ⫽ 0 Hours 1–15 marginal part-time Hours 16–34 part-time Hours 35–42 standard Long hours 43Age Age 18–30 Age 31–40 Age 41–50 Age 51–64 Education low Education medium Education high Number of children ⬍ 18 Lone fathers

Table 3.A1 continued

12,695 19,548 14,940 11,331

23.2 35.7 0.39 0.00 0.08 0.44 0.08 44.4 0.18 0.23 0.21 0.38 0.61 0.27 0.12 0.05 0.03

Mean

Italy 1995

15,737 14,800 14,145 11,389

23.5 36.5 0.36 0.08 0.06 0.39 0.11 38.4 0.33 0.28 0.19 0.20 0.15 0.59 0.23 0.05 0.03

The Netherlands 1994 Mean

9,609 10,492 10,333 2,287

18.0 27.8 0.47 0.04 0.10 0.37 0.02 33.5 0.54 0.17 0.13 0.16 0.28 0.53 0.17 0.03 0.02

Mean

Sweden 1995

16,257 16,356 14,389 11,572

23.4 42.2 0.45 0.01 0.04 0.29 0.21 39.4 0.31 0.25 0.20 0.24 0.64 0.20 0.16 0.12 0.07

Mean

UK 1995

34,388 30,335 25,119 5,776

36.6 42.5 0.14 0.01 0.07 0.50 0.28 38.8 0.29 0.28 0.25 0.18 0.14 0.52 0.35 0.20 0.13

Mean

USA 1997

17.0 30.6 0.45 0.08 0.18 0.22 0.07 40.4 0.17 0.32 0.29 0.21 0.60 0.15 0.16

34.2 42.4 0.19 0.01 0.03 0.41 0.36 43.0

Male variables: Hours worked per week Hours worked per week employed Hours ⫽ 0 Hours 1–15 marginal part-time Hours 16–34 part-time Hours 35–42 standard Long hours 43– Age 33.2 41.7 0.20 0.00 0.04 0.52 0.24 42.4

20.2 34.5 0.42 0.03 0.16 0.34 0.05 40.1 0.23 0.30 0.27 0.18 0.67 0.13 0.20

Mean

Mean

Female variables: Hours worked per week Hours worked per week employed Hours ⫽ 0 Hours 1–15 marginal part-time Hours 16–34 part-time Hours 35–42 standard Long hours 43– Age Age 18–30 Age 31–40 Age 41–50 Age 51–64 Education low Education medium Education high

France 1994

Australia 1994

34.5 43.4 0.21 0.01 0.02 0.42 0.33 43.8

17.2 31.8 0.46 0.07 0.18 0.21 0.08 41.4 0.21 0.31 0.22 0.25 0.11 0.76 0.13

Mean

Germany 1994

29.2 40.9 0.29 0.00 0.04 0.49 0.17 46.4

12.2 34.1 0.64 0.01 0.11 0.22 0.02 43.0 0.13 0.30 0.29 0.27 0.65 0.28 0.07

Mean

Italy 1995

31.3 40.5 0.25 0.01 0.05 0.52 0.16 42.0

13.9 25.2 0.45 0.14 0.25 0.15 0.02 39.7 0.16 0.33 0.29 0.22 0.19 0.64 0.17

The Netherlands 1994 Mean

Table 3.A2 Mean values for LIS samples of couples where the age of individuals is between 18–64

32.3 37.5 0.14 0.03 0.08 0.70 0.06 43.0

26.0 31.7 0.18 0.05 0.35 0.40 0.01 40.8 0.23 0.28 0.27 0.22 0.21 0.51 0.28

Mean

Sweden 1995

31.5 43.9 0.28 0.01 0.02 0.37 0.31 43.0

17.9 23.9 0.40 0.09 0.20 0.25 0.05 40.8 0.23 0.28 0.25 0.23 0.65 0.25 0.10

Mean

UK 1995

40.0 44.0 0.09 0.01 0.03 0.53 0.34 42.3 continued

27.5 36.7 0.25 0.04 0.14 0.45 0.11 40.3 0.21 0.31 0.28 0.19 0.12 0.54 0.34

Mean

USA 1997

4,953

24,435 29,721 16,736

3.4 1.7 1.1

3,216

36,261 28,077 16,381

3.1 1.7 0.9

0.13 0.31 0.22 0.34 0.12 0.69 0.18

Mean

Germany 1994

Notes: All income measures in USD/year adjusted by PPP. Self-employed are excluded. a The adjusted disposable income is equal to disposable income/(family size).0.5

Source: LIS and author’s own calculations.

2,736

35,663 31,022 17,388

Total unit earnings Disposable income after taxes Adjusted disposable incomea

Sample size

3.4 1.7 1.1

Household variables: Number of persons in unit Number of earners in unit Number of children ⬍ 18

0.16 0.30 0.29 0.25 0.69 0.12 0.19

Mean

Mean

0.12 0.29 0.31 0.29 0.48 0.35 0.18

France 1994

Australia 1994

Age 18–30 Age 31–40 Age 41–50 Age 51–64 Education low Education medium Education high

Table 3.A2 continued

3,443

20,267 25,270 13,661

3.6 1.4 0.9

0.06 0.26 0.30 0.38 0.61 0.31 0.08

Mean

Italy 1995

2,812

31,815 25,996 15,346

3.1 1.5 1.0

0.19 0.29 0.27 0.25 0.18 0.60 0.22

The Netherlands 1994 Mean

5,199

32,188 28,323 17,004

3.0 1.9 1.0

0.17 0.26 0.28 0.29 0.25 0.46 0.28

Mean

Sweden 1995

2,754

33,159 29,002 16,957

3.2 1.5 0.9

0.17 0.28 0.27 0.29 0.68 0.19 0.13

Mean

UK 1995

19,343

58,553 50,309 28,586

3.4 1.9 1.1

0.17 0.30 0.29 0.25 0.14 0.50 0.36

Mean

USA 1997

Working time: a household perspective 103 Table 3.A3 Children and disposable income: percentage disposable income differentials, single households, baseline; no children and age 18–30 Age of head

Number of children

Number of children

Number of children

0

2

0

2

0

France 94 – 6 51 61 73 85

26 91 119

Germany 94 – 8 36 47 49 62

14 55 70

Sweden 95 – 43 70 143 84 163

63 178 201

1

18–30 31–40 41–50

Australia 94 – ⫺2 20 18 30 27

⫺4 15 24

18–30 31–40 41–50

Italy 95 – –5 3 –3 25 18

⫺41 ⫺39 ⫺26

18–30 31–40 41–50

UK 95 – ⫺13 8 ⫺6 10 ⫺4

⫺2 ⫺5 ⫺7

1

The Netherlands 94 – ⫺6 ⫺12 79 68 ⫺57 91 80 ⫺68 USA 97 – ⫺4 24 19 34 29

1

2

⫺6 16 26

Source: LIS and author’s own calculations. Note The estimated model is: Log (Earnings) ⫽ ␤0 ⫹ ␤1(child1) ⫹ ␤2(child2) ⫹ ␤3(child3⫹) ⫹ ␤4(age31–40) ⫹ ␤5(age41–50) Log (Disposable Income)⫽␤0 ⫹␤1(child1)⫹␤2(child2)⫹␤3(child3⫹)⫹␤4(age31–40)⫹␤5(age41–50).

Table 3.A4 Children and disposable income: percentage income differentials, married and cohabitant households, baseline; no children and age 18–30 Age of head

Number of children

Number of children

Number of children

0

0

3⬎

0

10 43 63

Germany 94 – 0 ⫺5 ⫺4 26 25 20 21 42 41 35 36

1

2

3⬎

1

2

18–30 31–40 41–50

Australia 94 – ⫺5 ⫺9 ⫺12 16 10 5 2 28 21 16 13

France 94 – 5 8 30 38 41 48 56 60

18–30 31–40 41–50

Italy 95 – 1 ⫺12 ⫺25 21 22 6 ⫺9 44 45 26 8

The Netherlands 94 – ⫺3 0 9 18 14 15 28 28 23 25 39

18–30 31–40 41–50

UK 95 – ⫺6 18 11 33 26

USA 97 – ⫺2 0 37 34 39 58 55 60

⫺9 ⫺7 7 9 22 24

1

2

Sweden 95 – 9 16 19 30 38 30 42 51

3⬎

20 42 56

⫺8 26 45

Source: LIS and own calculations. Note The estimated model is: Log (Earnings) ⫽ ␤0 ⫹ ␤1(child1) ⫹ ␤2(child2) ⫹ ␤3(child3⫹) ⫹ ␤4(age31–40) ⫹ ␤5(age41–50) Log (Disposable Income)⫽␤0 ⫹␤1(child1)⫹␤2(child2)⫹␤3(child3⫹)⫹␤4(age31–40)⫹␤5(age41–50).

Age Age square Hours spouse Children 0–3 (1 ⫽ yes, 0 ⫽ no) Children 4–6 (1 ⫽ yes, 0 ⫽ no) Children 7–12 (1 ⫽ yes, 0 ⫽ no) Children 13–17 (1 ⫽ yes, 0 ⫽ no) Education low head Education high head

France 3.59a ⫺0.05a 0.08a ⫺1.50a ⫺0.44 ⫺0.90 1.87a ⫺2.96a 1.90b

Australia

1.68a ⫺0.02a 0.20a 0.14 ⫺1.25 0.60 0.64 ⫺3.35a 3.59a

Married or cohabitant Males

Variables

3.32a ⫺0.04a 0.04a ⫺1.06 ⫺2.39b ⫺0.58 ⫺0.05 ⫺4.39a 4.57a

Germany

3.05a ⫺0.04a ⫺4.24a ⫺28.85a ⫺13.50a ⫺9.77a ⫺1.37 ⫺4.34a 3.53b n.a ⫺2.61a –

4.71a ⫺0.06a ⫺3.17a ⫺9.17a ⫺9.13a ⫺4.88a ⫺1.75 ⫺4.61a ⫺0.99 2.19a – –

1.65a ⫺0.02a ⫺2.37a ⫺19.51a ⫺12.17a ⫺9.46a ⫺0.19 ⫺2.35a 4.14b 0.34 – –

Age Age square Sex (1 ⫽ female) Children 0–3 (1 ⫽ yes, 0 ⫽ no) Children 4–6 (1 ⫽ yes, 0 ⫽ no) Children 7–12 (1 ⫽ yes, 0 ⫽ no) Children 13–17 (1 ⫽ yes, 0 ⫽ no) Education low Education high Metropolitan area (yes ⫽ 1, 0 ⫽ no) West Germany (yes ⫽ 1, no ⫽ 0) North Italy (yes ⫽ 1)

Germany

France

Singles

Australia

Variables

Tables 3.A5 Tobit with selection, dependent variable hours of work

3.24a ⫺0.05a 0.05a 0.97 ⫺0.65 0.75 2.40a ⫺6.24a 2.82a

Italy

5.13a ⫺0.07a ⫺5.22a ⫺7.42c 4.80 ⫺0.53 ⫺10.91a ⫺8.22a 8.26a 0.31 – 0.56c

Italy

4.76a ⫺0.06a ⫺0.05a ⫺2.15b ⫺2.08c ⫺1.30 0.67 ⫺16.74a 3.37a

The Netherlands

4.28a ⫺0.06a ⫺2.44a ⫺25.75a ⫺23.10a ⫺21.64a ⫺9.85a ⫺5.50a 6.67a n.a – –

The Netherlands

1.95a ⫺0.02a 0.18a ⫺0.29 ⫺0.69 ⫺1.82a 1.26b ⫺1.30a 0.59c

Sweden

2.64a ⫺0.03a ⫺0.52 ⫺0.17 1.14 ⫺0.72 1.63 ⫺3.58a 0.25 0.04 – –

Sweden

2.27a ⫺0.03a 0.30a ⫺2.09c 0.74 0.95 2.77a ⫺1.83b 4.99a

UK

2.91a ⫺0.04a ⫺1.11 ⫺26.15a ⫺22.53a ⫺12.18a ⫺6.37a ⫺8.31a 5.33a 2.91b – –

UK

0.98a ⫺0.01a 0.03a 0.59b 1.14b 0.04 0.59b ⫺2.72a 2.66a

USA

1.30a ⫺0.02a ⫺4.41a ⫺4.62a ⫺3.10a ⫺2.38a ⫺0.79c ⫺7.24a 5.80a 0.41a – –

USA

⫺0.78a ⫺1.23a 0.34 0.99a – – 0.15 ⫺0.01 0.48b n.a. ⫺2.28a –

2.80a ⫺0.04a 0.10a ⫺5.73a ⫺3.23a ⫺2.50a ⫺0.07 ⫺3.83a 2.90a ⫺0.35 ⫺0.48 0.16 1.69 – –

1.34a ⫺0.02a 0.25a ⫺14.44a ⫺11.71a ⫺4.39a ⫺0.62 ⫺2.41a 8.21a 0.44 0.11 0.48 0.39 – –

Age Age square Hours head Children 0–3 (1 ⫽ yes, 0 ⫽ no) Children 4–6 (1 ⫽ yes, 0 ⫽ no) Children 7–12 (1 ⫽ yes, 0 ⫽ no) Children 13–17 (1 ⫽ yes, 0 ⫽ no) Education low spouse Education high spouse Education low head Education high head House owner Metropolitan area (yes ⫽ 1, 0 ⫽ no) West Germany (yes ⫽ 1, no ⫽ 0) North Italy (yes ⫽ 1)

Notes Marginal effects are evaluated at sample means. a Indicates significance at 1% level. b Indicates significance at 5 % level. c Indicates significance at 10 % level.

France

Australia 2.35a ⫺0.04a 0.05a ⫺26.67a ⫺16.07a ⫺10.27a ⫺4.35a ⫺3.66a 3.98a 0.92b 0.03 ⫺1.36a n.a ⫺4.86a –

Germany

Married or cohabitant Females

⫺0.30 0.29 1.06a 0.23 – –

Variables

Education low spouse Education high spouse House owner Metropolitan area (yes ⫽ 1, 0 ⫽ no) West Germany (yes ⫽ 1, no ⫽ 0) North Italy (yes ⫽ 1)

2.63a ⫺0.04a 0.02 ⫺5.15a ⫺4.10a ⫺2.60a ⫺1.72b ⫺10.34a 6.30a 0.13 ⫺0.11 0.05 0.72a – 0.18

Italy

0.27 ⫺0.73c ⫺0.74a 0.77a – 0.41b

1.15a ⫺0.02a ⫺0.04a ⫺14.56a ⫺12.96a ⫺9.28a ⫺4.19a ⫺6.41a 7.51a ⫺1.13b 1.29a 0.05 n.a – –

The Netherlands

⫺0.59c ⫺0.04 0.66b n.a. – –

2.14a ⫺0.03a 0.22a ⫺5.23a ⫺6.60a ⫺4.99a ⫺3.06a ⫺3.99a 2.86a ⫺0.44c 0.04 0.07 1.34a – –

Sweden

⫺0.13 ⫺0.69a 1.13a 0.17 – –

2.21a ⫺0.03a 0.17a ⫺15.09a ⫺12.99a ⫺9.05a ⫺4.10a ⫺4.16a 1.71c ⫺0.31 0.62 ⫺0.22 0.01 – –

UK

0.25 ⫺1.13b ⫺1.53a ⫺0.67b – –

1.57a ⫺0.02a 0.02a ⫺9.99a ⫺8.41a ⫺4.41a ⫺1.52a ⫺8.39a 4.41a 0.79a ⫺1.30a ⫺0.21 0.25b – –

USA

⫺0.82a 0.57a 1.13a ⫺0.27b – –

106

Dominique Anxo

Acknowledgements I would like to thank Ellen Perez-Ducy de Cuello for assistance with a previous version of this chapter, and Colette Fagan, Sangheon Lee, Deirdre McCann and Jon Messenger for their perceptive comments. All remaining errors remain mine.

Notes 1 Launched in 1983, the Luxembourg Income Study (LIS) aims to promote comparative research on the economic and social status of the population in different industrialized countries. By using a common conceptual framework and by improving data comparability, LIS facilitates cross-country comparisons and ensures a relatively high degree of comparability (demographic and socioeconomic data have been harmonized by LIS). The LIS database contains over 70 data sets from 26 countries. 2 A fifth level at which regulation can take place is the supranational level through, for example, European Union (EU) Directives or International Labour Organization standards. However, the minimum criteria in these standards usually already exist in the national regulations of the countries examined here. A notable exception is the UK, where the recent introduction of the EU Directive on Working Time provides a minimum regulation that might influence both the length and the distribution of working time (see Lee, Chapter 2) 3 In Sweden, the prevailing parental leave system, with complete employment security, lasts for 16 months, and for the first 12 months the income replacement rate amounts to 75 per cent of previous earnings. In order to foster a more even gender distribution in the use of leave, two non-transferable months for fathers were introduced in 2002. 4 As stressed by Lee (Chapter 2), in the late 1980s, Australia had a working hours distribution characterized by a high concentration of employees around the statutory norms (40 hours per week). During the 1990s, the distribution of working time was similar to that of the UK due mainly to a significant increase in the incidence of both long hours and short hours. As noted by Lee, this tendency coincides with the overall deregulation of the Australian labour market and a clear tendency towards a decentralization of the bargaining process at the firm level. 5 Even though the extent of public child-care facilities is higher than in the Anglo-Saxon liberal regimes, this provision is best explained as a ‘pro-family’ Conservative provision rather than a ‘gender equality’ Social Democratic measure. 6 It should, however, be noted that Italy presents large regional variations in female labour market integration, with significantly higher female employment rates in the North. 7 That is, there is a high incidence of marginal part-time workers. 8 This is confirmed by the results of our estimations, see section called ‘The socio-economic determinants of labour participation and working hours within households: an economic approach’ and also Table 3.A5. Living in Western Germany decreases working time by more than 2 hours for married/cohabiting males and by 4 hours for married/cohabiting females. 9 It appears also that the negative impact of children on disposable income is higher in the USA than in Sweden, see Tables 3.A3 and 3.A4. 10 In the Anglo-Saxon liberal regimes, in particular the USA, the prevailing wage

Working time: a household perspective 107 distribution has encouraged the development of a generally low paid and labour-intensive private service sector for household- and care-related activities. The option for high-income households to outsource a part of these domestic tasks to the market makes it possible for both highly educated/paid males and females to work long hours. In the Nordic social democratic regimes, on the other hand, which are characterized by a much lower level of income inequality, an extensive provision of public services (child-care, elderly care) and high average and marginal tax rates, the incidence of long hours remains very low among both males and females. 11 The educational variables retained in the estimation are split into three different levels (compulsory, intermediate and higher-level education). The lowest educational level consists of compulsory elementary school (or less) and brief vocational training. The difference between low and intermediate levels is the completion of either higher vocational training or upper secondary school. The ‘high attainment level’ category includes individuals with college or university degrees. In the estimation the omitted category is the intermediate level.

4

Gender and working time in industrialized countries Colette Fagan

Introduction Working time arrangements in industrialized countries are changing along a number of dimensions. Some of these changes have been emerging for a number of decades, such as the expansion of part-time work, while others are more recent developments, such as the lengthening of the working week in some parts of the economy (Bosch 1999b; Evans et al. 2001; Lee, Chapter 2). The main forces driving these changes are enterprises’ requirements for more flexible modes of work organization to enhance their efficiency in the context of economic restructuring and globalization. However, there is another, subsidiary pressure in play from the labour supply. Some workers are seeking part-time or even ‘unsocial’ schedules such as night or weekend work in order to combine employment with other timeconsuming activities – such as women raising children, or young people participating in education. In contrast, some other groups of workers take jobs where very long hours or ‘over-working’ are the norm, perhaps in exchange for higher earnings. The gender implications associated with these working time developments are twofold. First, the restructuring of working time practices has occurred in parallel with a growing involvement of women in employment when they are raising children. Among couples the traditional ‘male breadwinner’ arrangement that was common in many industrialized countries is being eroded by the rise of ‘dual-earner’ households, and, in addition, a growing proportion of women are raising children in lone-parent households. The established gender division of labour in society is that women do most of the domestic work and put more time into care work than men, and this gender arrangement tends to channel women into working time arrangements that fit with their domestic commitments. The most notable example is part-time work, but in some countries it is also common for certain groups of mothers with young children to opt for evening and weekend schedules. In contrast, few men reduce their working hours to take on care responsibilities (Fagan 2001a).

Gender and working time 109 The second, related issue is that gender segregation is a persistent feature of the labour market, despite the growing presence of women in the workforce (Anker 1998; Fagan and Burchell 2002; Rubery et al. 1999). This segregation is associated with different working time schedules, so that women undertake most of the part-time work, while ‘over-working’ is mostly found in the male-dominated job areas. Most of the part-time jobs are concentrated in the lower-paid and lower status service jobs, and the opportunities to work part-time in many managerial and professional areas are limited (O’Reilly and Fagan 1998). Conversely, many of these higher status white-collar jobs are currently organized in ways that are generally incompatible with the time demands of family life due to the long hours or type of flexible hours required. Thus, the working time schedules associated with different jobs can operate to reinforce gender segregation. Very long hours or ‘over-working’ is one of the barriers to women’s entry to some male-dominated job areas and reinforces men’s lack of time involvement in their parental roles. In other parts of the economy the working time policies adopted by enterprises may mean that the only jobs open to women are part-time, particularly those with few labour market alternatives. For example, in the UK, part-time jobs with very short ‘marginal’ hours predominate in many female-dominated manual service jobs, such as cleaning or retail. So in this country the opportunities for women to obtain full-time work may be limited unless they have sufficient qualifications to enter higher occupational levels or are willing and able to enter male-dominated manual jobs. Thus large proportions of women may be under-employed in part-time jobs because of a lack of full-time alternatives. The diversification of working time practices – and perhaps workers’ working time preferences – in conjunction with the general trends of collective bargaining, decentralization and declining coverage – are creating new challenges for how working time standards are defined and established. Historically, state intervention and collective bargaining in relation to working time have centred upon full-time, daytime and weekday work schedules as the reference point of ‘standard hours’ (Bosch et al. 1994; Bosch 1999b). Shorter full-time hours and compensation for ‘unsocial’ hours (evenings, nights, weekends and rotating shifts) have been the focus for regulations, followed in more recent decades by measures to guarantee equal treatment for part-timers. Similarly, this ‘full-time standard’ has also been the reference point for the development of national social protection systems, and it is only recently that the importance of protecting periods of part-time work have begun to be accommodated in welfare state policies. The aim in this chapter is to analyse the gender differentials in working time arrangements in selected industrialized countries, and to assess the policy implications, with a particular focus on the issue of ‘work–family’ reconciliation or coordination. In the next section of the chapter the current pattern of working time arrangements are compared by gender

110

Colette Fagan

and occupational position along the dimensions of the volume, schedule and autonomy. Some national comparisons are also drawn. Working time preferences and the ‘work–family’ compatibility of work schedules are considered in the following section. The last two sections discuss the implications for working time policies, and summarize the conclusions of the chapter, respectively.

Gender and the component elements of working time The established gender division of labour in society, including gender segregation within employment, produces gender-differentiated working time arrangements. To evaluate the extent and form of gender differences in working time, it is helpful to distinguish between the different dimensions of working time, particularly the volume, schedule and type of variability (Fagan 2001c). The notion of ‘working time flexibility’ must also be clarified. Employers’ interests in working time flexibility for operational requirements are usually quite distinct from employees’ interests in obtaining more flexibility in the organization of their working time so as to make it more compatible with other parts of their lives. Employees may want working time arrangements that accommodate a variety of activities – as well as rest and leisure time – including care responsibilities (children, elders, the domestic work of running a home), education or other timeconsuming undertakings (political and civic commitments, small-scale self-employment, sporting or creative pursuits). Given this diversity in circumstances it is plausible that one working time arrangement may suit some people but not others. For example, some mothers of young children in the UK opt for evening or weekend work to fit in with their domestic arrangements if their partner or other family members are available to look after children during these periods, while others may find it difficult or impossible to organize child-care in order to go to work during these periods. Purcell et al. (1999) make a pertinent distinction between ‘unstructured’ and ‘structured’ flexibility, to which we can add the concept of ‘autonomous flexibility’. ‘Unstructured flexibility’ occurs when employees have little control over the schedule and often the volume of hours that they work. This may result from explicit contractual arrangements with their employer, in annualized or zero-hour contracts, for example, or in more informal workplace expectations such as overtime as and when necessary, or the ‘long hours cultures’ of many professional and managerial occupations. ‘Structured flexibility’ is found in working time arrangements that are predictable but ‘non-standard’, such as regular part-time schedules or rotating shifts. Such arrangements offer employees more control than ‘unstructured flexibility’ in the sense of predictable, regular and delineated patterns of work. Structured ‘non-standard’ schedules also offer labour market alternatives for people who are unable or unwilling to

Gender and working time 111 work ‘standard’ hours. The concept of ‘autonomous flexibility’ identifies those forms of working time flexibility that are geared to employees’ needs rather than organizational requirements, in the sense that these arrangements give employees some ability to vary or alter their working time in order to accommodate other activities. Examples include formal flexi-time or ‘time banking’ systems; contractual rights to switch between full-time and part-time hours; and extended leave entitlements such as parental leave.1 The volume of weekly working hours The gender gap in employment rates has fallen in recent years; however, men still commit more time to paid work than do women, whether measured over the day, week, year or lifetime. Conversely, gender inequality in the division of unpaid labour persists, for women commit more time to child-care and unpaid domestic work. However, there is a slow process of ‘lagged adaptation’ underway, whereby men’s relative contribution to child-care and domestic work has increased across recent generations in response to women’s rising time commitment to employment (Gershuny et al. 1994; Gershuny 2000). To date, this adaptation of the gender division of labour is still very limited. For example, in couple households with a child aged less than five years old and in which both adults are employed fulltime, on average, mothers spend just over twice as much time on childcare and other domestic work as fathers (OECD 2001).2 However, multivariate analysis shows that men’s relative contribution increases according to the number of hours that women devote to employment in some European countries, which suggests that a reduction in the gender working hours gap would facilitate a more gender-equal domestic division of labour (Anxo et al. 2002). Tables 4.1 and 4.2, and Figure 4.1 illustrate this gender difference in employment with a comparison of the volume of usual weekly hours worked by employed men and women in selected industrialized countries. Usual weekly working hours are important to consider given the regular time-demands of domestic life and hence ‘work–family’ reconciliation debates (for an international comparison of annual working hours, see Evans et al. 2001; Lee, Chapter 2).3 The spread of working hours is also relevant, for a polarization of working hours through an expansion of very short and very long work weeks seems to be emerging in some countries, such as Australia, the USA and Canada (Australian Bureau of Statistics 1999; Golden and Figart 2000; Wong and Picot 2001). In every country the profile for men is dominated by full-time hours. In contrast, employed women are more likely to work part-time or shorter full-time hours, and tend to have a more dispersed range of working hours than do men. The analysis also demonstrates national differences in the volume of weekly working hours, most notably in the rates of part-time

112

Colette Fagan

Table 4.1 The usual weekly hours of work of the employed by gender in selected industrialized countries Row % distribution of weekly working hoursa

EU countries Austria Belgium Denmark Finland France Germany Greece Ireland Italy Luxembourg The Netherlands Portugal Spain Sweden The UK

Men Women M&W Men Women M&W Men Women M&W Men Women M&W Men Women M&W Men Women M&W Men Women M&W Men Women M&W Men Women M&W Men Women M&W Men Women M&W Men Women M&W Men Women M&W Men Women M&W Men Women M&W

1 to 19

20 to 34

35 to 39

40⫹

1 7 4 2 15 8 7 11 9 4 7 5 2 9 5 3 18 9 1 3 2 3 13 7 3 8 5 *b 8 4 10 34 20 2 6 4 1 7 3 4 7 5 5 23 13

3 25 13 6 27 15 5 26 15 7 15 11 6 25 15 3 23 12 7 16 10 6 26 15 4 21 11 1 27 13 10 37 22 6 14 9 3 16 8 7 29 18 6 25 15

31 21 26 49 38 44 50 49 50 28 51 39 63 50 57 38 30 35 8 10 9 37 40 38 19 24 21 3 6 4 32 16 25 12 20 15 9 16 12 12 17 15 20 24 22

66 47 57 43 20 33 38 14 27 61 27 45 29 16 23 56 30 44 85 71 80 54 22 40 74 47 64 95 59 80 48 13 33 81 61 72 87 62 77 77 47 62 69 28 51

Gender and working time 113 Table 4.1 Continued Row % distribution of weekly working hoursa

Selected non-European Australia Men Women M&W Canada Men Women M&W New Zealand Men Women M&W The USA Men Women M&W Japan

Men Women M&W

1 to 19

20 to 34

35 to 39

40⫹

11 29 19 6 13 9 8 22 14 3 8 5

12 23 17 9 24 16 7 22 14 7 17 12 1–34 hours 14 41 25

15 16 16 16 28 21 4 9 6 4 9 6 35 or more 86 59 75

62 31 49 70 36 54 82 47 66 86 67 77

Source: OECD Labour Markets Indicators database (www.oecd.org). Notes a Data relate to the main job. The data are from the OECD for all persons in employment in 2001 except for Germany (2000) and the UK (2000). b ‘*’ indicates less than 0.5%.

work for women and, particularly among men, the proportion with very long full-time working hours. A substantial number of studies have shown that international differences in working hours are also found where more precise comparisons are made according to type of sector or occupation (e.g. Anxo and O’Reilly 2000; Bosch et al. 1994, 1997; Bosch and Lehndorff forthcoming; Boulin and Hoffman 1999; Golden and Figart 2000; Rubery et al. 1998b, 1999). This body of research has demonstrated that these national differences result from different institutional arrangements that impact on how working time is organized (see also Lee, Chapter 2). A key factor is the different statutory regulations and collective agreements on working time. The structure of employers’ non-wage labour costs also has a bearing. For example, hours or earnings thresholds in the structure of employers’ social security contributions can encourage employers to create ‘short hour’ part-time jobs to reduce these costs, while per capita rather than hourly-related costs create fixed costs that can deter the use of part-time contracts. Other institutional differences also play a role through their influence on labour supply, including wage levels and differentials, the structure of personal taxation, incentives or barriers to part-time work in the unemployment support system, and the availability and opening

114

Colette Fagan

Table 4.2 Mean weekly hours in selected industiralized countries Mean weekly hours Men

Women

All

Austria Belgium Denmark Finland France Germany Greece Ireland Italy The Netherlands Portugal Spain Sweden United Kingdom

41 40 39 41 40 41 45 42 41 37 42 42 39 43

35 33 33 36 34 32 40 33 35 25 38 37 34 31

38 37 36 39 37 37 43 38 39 32 40 40 37 38

EU–15

41

33

38

Source: The European Labour Force Survey – 2001 (special analysis), the Australian Bureau of Statistics, the US Bureau of Labor Statistics and the Statistics Bureau Japan, and Lee, this volume, as detailed in the notes below. Notes The data for the EU countries are from the European Labour Force Survey 2001. The US data are for 49⫹ hours and are from Current Population Statistics (CPS) from the US Bureau of Labor Statistics (www.bls.gov/cps). The Japanese data are for 49⫹ hours and are from the Labour Force Survey, table 5, from the Statistics Bureau Japan (www.stat.go.jp/english/data). The data for Australia are from the 1999 Social Trends Report, from the Australian Bureau of Statistics (www.abs.gov.au/austats). The data for New Zealand are for employees only, working 50 or more hours, from Figure 2.5, Lee this volume.

hours of child-care, the latter of which mainly impacts on maternal labour supply. These institutional frameworks influence the decisions and negotiations of employers and the workforce at the organizational level, where additional more specific sectoral and regional conditions affect the particular working time policies and innovations that develop. Long working hours remain prevalent, particularly for men, in many countries. This is despite the substantial reductions that were achieved over the last century, largely driven by collective negotiations and legislation. In some countries this downward trend continues, for example, in France where the 35-hour week has been introduced by legislation. Substantial reductions have also been achieved in some other countries, including Denmark, Germany, Portugal, the Netherlands and Japan since the 1980s (see Lee, Chapter 2). However, the incidence of long hours of work among full-timers has risen in some countries, most notably in the UK, the USA and Australia (Bosch 1999b; Evans et al. 2001; Lee, Chapter 2). A 48-hour upper-limit to weekly hours has been established by the

Gender and working time 115

Japan United States New Zealand Australia EU-15

United Kingdom Sweden Spain Portugal Netherlands Luxembourg Italy Ireland Greece Germany France Finland Denmark Belgium Austria 0

5

10

15

20

25

30

35

40

45

% of employed who usually work 48+ hours

Men

Women

All

Figure 4.1 The incidence of long weekly working hours in selected industrialized countries (Source: The European Labour Force Survey – 2001 (special analysis), the Australian Bureau of Statistics, the US Bureau of Labor Statistics and the Statistics Bureau Japan, and Lee, this volume, as detailed in the notes below. Notes The data for the EU countries are from the European Labour Force Survey 2001. The US data are for 49⫹ hours and are from Current Population Statistics (CPS) from the US Bureau of Labor Statistics (www.bls.gov/cps). The Japanese data are for 49⫹ hours and are from the Labour Force Survey, table 5, from the Statistics Bureau Japan (www.stat.go.jp/english/data). The data for Australia are from the 1999 Social Trends Report, from the Australian Bureau of Statistics (www.abs.gov.au/austats). The data for New Zealand are for employees only, working 50 or more hours, from Figure 2.5, Lee this volume.

European Union’s Working Time Directive, and regulations in most member states establish a more stringent limit on weekly hours (see McCann, Chapter 1). If we use this 48-hour threshold as one possible indicator of ‘over-working’,4 then Figure 4.1 shows that the incidence of very long weekly hours tends to be higher in the non-EU countries shown (Japan, the USA, Australia and New Zealand) compared to the overall figure for the EU member states. However, within the EU there are still around one in five employed men working hours at or in excess of this 48hour limit, as are 9 per cent of employed women, with the incidence of long working hours particularly prevalent for men in Ireland, Greece and the UK and for women in Greece, Spain and Portugal (see Lee, Chapter 2

116

Colette Fagan

for additional discussion of excessive hours of work). Men’s working hours are shortest in France, Sweden, Denmark and the Netherlands (the averages are between 37–40 per week). More than 60 per cent of employed men have usual weekly hours in the 35–39 hour range in France, as do around half of the employed men in Denmark and Belgium.5 In France, there has been a shift towards the lower end of the 35–39 hour category in line with the statutory reduction of the working week from 39 to 35 hours. Long working hours are less prevalent for women, but again there are national differences in average working hours, associated with the different rates of part-time work and ‘over-working’. Part-time employment has increased in recent decades in most industrialized countries, stabilizing or declining only in some countries where comparatively high rates have been reached (O’Reilly and Fagan 1998). Part-time hours can be defined using various thresholds,6 although it should be noted that in a few countries, such as Japan, the concept of parttime work is not tied to the volume of work undertaken but to the lower status of the worker within the firm (Houseman and Osawa 1998). In relation to working hours it is useful to distinguish between those jobs with ‘marginal’ or ‘short’ part-time hours (defined here as less than 20 hours per week) and those with longer or ‘substantial’ part-time hours (defined here as 20–34 hours per week), with some of the latter category working only a few hours less than some full-timers. This distinction is important for measuring the volume of hours worked; it is also important in some labour law or social protection systems where a specific hour threshold is used to define ‘marginal’ part-timers with fewer entitlements, for example, in Germany (European Commission 1999). As Lee argues in Chapter 2, there is also some evidence that in some countries short part-time hours are also marginal on the dimension of a poorer job quality relative to that of substantial part-time arrangements. Most part-timers are women. The rate of part-time work for women varies markedly between countries; for example, a full-time working week of 35 hours or more is much more common for employed women in the USA, Finland, Portugal and Spain than in the other countries shown. Taking ‘short’ and ‘substantial’ part-time hours together, the Netherlands has the highest rate of female part-time work by a long distance, followed by the UK, Australia, Sweden, and Denmark.7 Both short part-time (less than 20) and more substantial part-time hours are common in the Netherlands, the UK and Australia. The incidence of short part-time hours is rising rapidly in some other countries, notably Germany, Japan, Australia and Ireland (Lee, Table 2.2). Elsewhere women’s part-time work tends to be organized around longer weekly working hours (20–34). Some women with weekly working hours in the 20–34 hour range have full-time rather than part-time contracts; for example, many of the women working these hours in Greece and Italy are full-time public sector workers working less than 35 hours a week. Rates of part-time work have risen for men in

Gender and working time 117 recent years, but it is still rare in most countries and it is usually concentrated on students and those approaching retirement (Delsen 1998). Again, the Netherlands leads the way, where one-fifth of employed men now work less than 35 hours a week, and where part-time work for men has spread into the core working years. Hence, a large part of the explanation for the gender differential in the volume of weekly hours is the greater propensity of women to work parttime. Many select part-time work because they have care responsibilities. For example, around two-thirds of women and one-third of men working part-time in the European Union report that they do not want full-time jobs (European Commission 2000a: Table 38). Most of the women in the EU who have selected part-time work have done so because they have children or other domestic commitments, as have one-third of the minority of men who have selected part-time work (Fagan 2001a: Table 39). The rate of part-time work in the USA is lower than that found in many European countries, but again women are more likely to work part-time than men, and here around one in five part-timers have chosen this working arrangement due to child-care problems or family obligations.8 The decision to take part-time work for family reasons is influenced by the range of alternative work–family reconciliation measures available. For some it is the preferred option selected from a range of alternative forms of child-care or working arrangements; for others it is a tightly constrained decision or ‘notional choice’, particularly if they live in places where childcare services are limited, expensive or socially unacceptable. Other reasons for working part-time should not be overlooked. In particular, ‘involuntary’ part-time work – where full-time hours are preferred but unavailable – varies markedly for both sexes among countries, ranging from less than 10 per cent of part-timers in the USA, the UK, the Netherlands and Norway to at least one in five in Sweden, France, Spain, Greece and Portugal (see Lee, Figure 2.6). The majority of involuntary part-time workers are women, although in some countries the rate of involuntary part-time work is higher among the small proportion of men employed part-time. The proportion of part-timers who are students also varies markedly among countries, with particularly large proportions of the part-time workforce being students in Denmark (35 per cent), Finland (28 per cent), the USA and Ireland (19 per cent in each). In each country, men working part-time are much more likely to be students than are women in part-time jobs. Another reason for the gender differential in the volume of working hours is gender segregation, whereby the working time requirements often vary between ‘male-dominated’ and ‘female-dominated’ areas of employment. These working time differences between jobs are due to operational requirements plus labour supply considerations, and they play a role in perpetuating gender segregation in employment. For example, part-time work has often been created explicitly to recruit or retain women

118

Colette Fagan

(Beechey and Perkins 1987; Horrell and Rubery 1991), while the requirement to work long hours in particular jobs such as management help to preserve this area of employment as a largely male enclave (Wajcman 1998). More men than women are also self-employed – although the female share is increasing in a number of countries, particularly in the USA – and self-employed men typically work longer hours than employees, particularly the self-employed with their own employees (Fagan and Burchell 2002). Job differences in usual weekly hours of work are illustrated in Table 4.3, using occupational data from a recent survey of the EU member states.9 Men in management and in skilled agricultural and fishery jobs work the longest hours, while the shortest average hours for men are in clerical work and elementary occupations. The incidence of very long hours is also high for professionals and service and shop workers, of whom about one quarter of employed men usually work 48 or more hours per week, but these occupations also include a relatively high proportion of men working part-time. Overall, men are most likely to work less than 35 hours a week if they are in professional, service and shop work, elementary, or skilled agricultural work. Compared to men, women work shorter hours within each occupational group, but there are marked differences between women according to their occupational position. Women work particularly long hours if they are in the male-dominated areas of managerial work or agricultural work. Marginal part-time hours are most prevalent among women employed in elementary, service and shop work. The female-dominated occupation of clerical work offers the least exposure to either short part-time or long hours of work for both sexes. As we shall see later in this chapter, working time preference data shows that both ‘over-working’ and ‘short’ part-time work are unpopular among the workforce. The schedule of hours Working time schedules have become more diverse in industrialized countries in recent decades, associated with changing work practices including the spread of the ‘24/7 economy’. The proportion of the workforce with schedules that include evening, night or weekend work, or variable hours has increased in most industrialized countries, albeit slowly in Australia and some European countries over the 1990s (Australian Bureau of Statistics 1999; Evans et al. 2001). Similarly, in the USA there has been little change in the incidence of many types of schedules over the last decade (US Bureau of Labor Statistics 2002), but in the past 30 years pronounced changes have occurred, particularly with a decline in the 5-day week and an expansion in 7-day working (Golden and Figart 2000). The type of schedules worked is illustrated with data for the EU and USA in Tables 4.4 and 4.5. Around one-third of the European workforce

16 23 35 46 25 17 31 31 29 29

35–9 31 34 35 36 34 24 48 45 41 39

40–7 45 25 19 9 26 44 16 17 11 21

48⫹

2 14 11 9 20 5 10 3 28 14

Under 20 19 25 33 37 20 11 30 39 20 27

35–9 28 20 24 25 21 30 33 35 19 24

40–7

35 9 7 3 8 31 9 5 5 9

48⫹

48.7 40.5 40.6 38.7 41.8 47.4 41.4 41.3 36.9 41.6

Mean

44.7 32.3 33.7 33.0 31.1 42.9 35.5 36.1 27.6 33.2

Mean

15.6 12.5 10.6 7.8 14.3 17.1 8.7 10.3 11.8 11.9

Std Deviation

15.8 11.5 10.8 9.7 13.2 18.6 10.5 8.7 12.9 12.6

Std Deviation

Note The occupational classification used is ISCO Major occupational groups (1 digit data). The armed forces are not shown. The data relates to main job only.

16 32 25 26 31 23 18 18 28 26

20–34

% Distribution of usual weekly working hours (row)

8 18 11 9 15 15 5 7 19 11

Source: The European Working Conditions Survey, 2000.

(b) Employed women Managers and senior officials Professionals Associate professionals Clerks Service and shop workers Skilled agriculture and fishery Craft and related Operators and assemblers Elementary occupations All

(a) Employed men Managers and senior officials Professionals Associate professionals Clerks Service and shop workers Skilled agriculture and fishery Craft and related Operators and assemblers Elementary occupations All

Under 35

% Distribution of usual weekly working hours (row)

Table 4.3 The usual weekly volume of hours of the employed, by occupational group and gender, EU-15

M W All

M W All

M W All

M W All

M W All

M W All

Managers and senior officials

Professionals

Associate professional and technicians

Clerical

Service and sales

Skilled agricultural and fishery

19 11 17

12 19 17

45 60 55

30 33 32

27 39 33

16 18 17

Standard weekdays

4 2 3

3 2 3

10 6 7

15 8 11

16 10 13

14 10 12

Standard weekdays ⫹ long days

8 10 9

10 13 12

9 8 9

5 12 9

5 7 6

3 6 4

No long days

6 5 5

5 2 3

5 3 3

10 5 8

17 5 11

14 6 11

⫹ long days

Some weekend work

14 19 16

16 24 21

9 9 9

7 6 6

7 10 9

11 22 15

32 37 33

17 7 10

7 2 4

10 4 7

14 6 10

26 26 26

10 10 10

10 9 9

4 3 3

5 6 5

6 5 5

8 4 7

Regular weekend work Evenings/ nights No long ⫹ long days days

Table 4.4 Weekly work schedules of the employed in the EU-15, by occupation and gender (%)

7 7 7

27 24 25

11 9 10

18 26 22

8 18 13

8 8 8

Rotating shifts/night shift

M W All

M W All

Operators and assembly

Other (elementary) labour

28 37 34 43 32

34 47 40

19 35 22

34 38 35

9 5 8 3 8

3 2 3

7 5 7

9 4 8

7 10 8 12 8

11 10 11

7 11 8

7 10 7

10 3 5 1 7

4 3 3

10 1 8

10 6 10

9 14 13 14 11

11 14 12

4 7 4

7 18 9

12 7 9 3 10

6 7 5

7 * 6

10 3 9

6 6 4 8 6

6 6 6

6 5 6

4 5 4

19 18 19 16 18

25 18 20

40 36 39

19 16 18

Notes 1 The occupational classification used is ISCO Major occupational groups (1 digit data). 2 M ⫽ Men, W ⫽ Women, FT ⫽ Full-time (35 ⫹ hours per week), PT ⫽ Part-time (under 35 hours per week). 3 A ‘long day’ is defined as working 10 hours or more. 4 ‘Some’ weekend work includes 1–4 days per month, ‘regular’ weekend work is more than 4 days per month. 5 ‘Some’ evenings/nights includes people who report that that their work schedules do not always fall into daytime hours, but are not part of shift rotas. Evenings are defined as 6–10 p.m. and nights as 10 p.m.–5 a.m. in the survey. 6 Rotating shifts includes all those with an alternating shift pattern during daytime hours or day and night shifts or permanent nights. It excludes those with permanent morning or afternoon shifts.

Source: The European Working Conditions Survey, 2000.

Men Women Women – Full-time Women – Part-time All

M W All

Craft and related

83 87 87 85

Men Women Women with a youngest child under 6 years All

5 4 6 5

2 2 4 4 4 1 12 19 11 15 7 5 1 7 12 5 7 2

Evening

4 3 3 3

1 2 4 1 2 1 10 4 9 7 4 4 1 6 10 5 9 1

Night

3 2 2 3

1 1 3 4 1 1 13 7 6 2 5 2 1 5 5 5 3 1

Rotating or split shift

5 4 2 4

2 3 5 7 2 14 15 12 5 3 8 2 2 3 2 10 4 2

Irregular or ‘other’

Notes 1 Data relates to those aged 16 years and over who usually work full-time (35 hours or more per week) on their main job. 2 The schedules were defined in the questionnaire as: daytime (anytime between 6 a.m.–6 p.m.), evening (anytime between 2 p.m. and midnight), night (anytime between 9 p.m. and 8 a.m.), rotating (one that changes periodically from days to evenings or nights), split (one consisting of two distinct periods each day), employer-arranged irregular or ‘other’ schedule.

Source: Current Population Survey (CPS), derived from Tables 4 and 5 of the US Bureau of Labor Statistics, May 2001 supplement, USDL 02–225 ‘Workers on flexible and shift schedules in 2001’ http://www.bls.gov.cps.

94 92 84 84 91 83 50 58 69 73 76 87 95 79 71 75 77 94

Executive, administrative and managerial Professional Technicians and related support Sales Clerical and administrative support Services – private household Services – protective Services – food Services – health Services – cleaning and caretaking Services – personal Mechanics and repairers Construction trades Other precision production, craft and repair Machine operators and assemblers Transport and material moving Other (unskilled) labour Farming, forestry and fishing

Regular daytime

Table 4.5 Work schedules of full-time wage and salary workers in the USA, by occupation and gender, 2001 (%)

Gender and working time 123 have what can be defined as the traditional ‘standard’ schedule of daytime, weekday work without long daily hours (defined as people who said they never worked 10 hours or more per day). Another 8 per cent work daytime, weekdays but with some or regular long days (including 2 per cent who worked at least 6 long days a month). Weekend work, long days and shifts are prevalent. Over one-third have daytime schedules that include some or regular weekend work: 17 per cent with long days and 19 per cent without long days. Eighteen per cent work either rotating shift schedules or permanent nights. A small minority of 6 per cent work some ‘twilight’ evenings and nights without working either ‘long days’ or rotating shift schedules. There are some gender differences in work schedules in the EU. A larger proportion of women’s employment is organized into daytime work – including weekends – with no long days. Schedules with long days or some weekend work are more common for men. However, there is no gender difference in the incidence of regular weekend work (21 per cent of each sex), rotating shifts (18 per cent), or scheduled evening/night hours (6 per cent). The higher rate of part-time work among women accounts for most of the gender difference in work schedules. Part-timers rarely work long days of 10 hours or more, but they are roughly twice as likely as fulltimers to work ‘twilight’ evening and night slots, and only slightly less likely to work weekends or rotating shifts. Part-timers are disproportionately involved in these non-standard schedules relative to full-timers in some European countries, indicating the influence of national working time regimes on how part-timers are used to cover different working hours (Rubery et al. 1998b). There are also clear job-related differences in work schedules. The ‘standard’ working day (daytime, weekday, no long days) is most common for clerical workers, covering 45 per cent of men and 60 per cent of women employed in this occupation. Weekend work without long days features most for men employed in service and sales work (26 per cent), agriculture (22 per cent) and elementary occupations (22 per cent); for women it is particularly prevalent in service and sales work (37 per cent), and the incidence for women exceeds that of men for each occupational group except clerical work where there is no gender difference. Daytime work that includes weekends and long days is particularly found among managers and skilled agricultural workers of both sexes and also among male professionals. The incidence of rotating shifts or permanent nights for men is highest for operatives (40 per cent of men and 36 per cent of women), service and sales work (27 per cent of men and 24 per cent of women), men in elementary occupations (25 per cent), and women who are associate professionals (26 per cent), many of whom are in nursing and other health professions. The data for work schedules in the USA (Table 4.5) are not directly comparable with that for the EU because they only relate to full-time

124

Colette Fagan

workers (35 hours or more per week) and different definitions of schedules were used in the survey; in particular no distinction is made between daytime work on weekdays versus weekends (see Table 4.5). But there are parallels with the picture for the EU. First, there is little gender difference between full-timers in the incidence of schedules that fall outside of ‘regular daytime’ hours (defined as any time between 6 a.m. and 6 p.m. any days during the week). It is also of note that mothers of young children employed full-time in the USA undertake a similar range of work schedules to those of men and other women employed full-time. Second, the job-related differences are pronounced, with the lowest incidence of ‘regular daytime’ schedules found among full-time workers in service, operating and transport jobs. Other data from the same survey enhance the picture of daily schedules worked by the US workforce. While 85 per cent state they have a ‘regular daytime’ schedule, in practice, work schedules encroach into the early morning or the end of the day for many. In total 30 per cent of full-timers are at work by 7.30 a.m., 19 per cent do not usually finish work until after 6 p.m., and 15 per cent cannot specify a ‘usual’ finishing time because their work hours are so variable (US Bureau of Labor Statistics 2001, derived from Table 7). A similar pattern of encroaching work hours is found in Australia, where one-third of the workforce have work patterns that fall outside daytime hours (6 a.m.–6 p.m.) (Australian Bureau of Statistics 1999). The precise shape of these gender- and job-related differences in work schedules varies between countries, for example, in the extent and type of part-time work schedules, long hours working or shift patterns. The key points here are, first, that in every country, while men undertake most of the long working hours and women work shorter hours, both sexes are involved in a diverse range of work schedules encompassing weekend, evening, night and shift work. Second, that these work schedules are more prevalent in some occupational groups than in others. Where there is some diversity in the range of work schedules available within similar types of employment – for example, where nurses can choose between different shift patterns, or where secretaries can find either part-time or full-time jobs in the labour market – then there are more options open to the workforce. This can help to increase the labour market integration of groups with other demands on their time, such as mothers with young children. However, sectoral and company studies show that a diverse range of work schedules provides only a limited form of integration unless accompanied by other measures. For example, Perrons’ (1999) comparative study of retail in Europe showed that parttimers had fewer promotion prospects than those working standard fulltime schedules and that, where schedules were subject to unpredictable changes instigated by the employer, this undermined any ‘compatibility’ the schedules had with domestic responsibilities.

Gender and working time 125 Working time autonomy and variability Public debate is increasing about the need for more extensive work–family reconciliation policies to alleviate the difficulties that many employees face when managing the demands of their jobs with the time demands of family responsibilities. This reconciliation is likely to be facilitated for those workers who have some influence or control over when they work, or, as we discussed above, what can be termed ‘autonomous flexibility’. A number of aspects of working time contribute to enhancing workers’ autonomous flexibility, including the degree of influence they have over when they take their annual leave, entitlements to other forms of leave (maternity, parental, etc.) or working time adjustments, such as the right to work part-time for a specified period. Here the focus is upon the particular issue of whether workers are able to influence their starting and finishing times. Formal flexi-time systems are an important mechanism under which workers are able to vary their starting and finishing times within agreed time zones around a ‘core working hours’ period in which they must be present. A target number of working hours must be achieved within a predetermined period, but workers are allowed to carry ‘deficits’ or bank ‘surpluses’ within certain ranges. Surplus hours can be used to take full or half days off as ‘time off in lieu’, although this is usually conditional on the agreement of supervisors.10 These systems were first introduced in the 1960s and 1970s and are most widespread among white-collar workers, particularly in clerical and administrative occupations (Bosch 1997). ‘Time banking’ schemes extend the flexi-time principle to permit hours to vary to a greater degree over a longer settlement period, and while many are geared largely to employers’ needs as part of annualized hours contracts, some time banking schemes increase autonomous flexibility for employees. In recent years, a number of such schemes have been negotiated in Europe, particularly in Germany, Italy and Sweden, in a variety of sectors including metalwork, chemicals and banking (Arrowsmith and Sisson 2001). Where flexi-time is absent, informal flexible hours arrangements may operate between supervisors and some workers, for example, to retain key workers by accommodating specific family-related demands. Flexible hours are also a feature of some managerial, professional and craft occupations, in which the incumbents largely self-determine their own starting and finishing hours in order to complete their workloads. Both flexi-time and informal flexible arrangements are easiest to implement in jobs in which many of the tasks can be completed independently of the working patterns of other colleagues, and in which only a small part of the working time involves being available for direct contact with customers or other external service users. Finally, the other side of the coin is whether workers have variable starting and finishing times that are largely controlled by their employers’ requirements for them to work overtime or variable schedules.

126

Colette Fagan

Statistical indicators of flexible working schedules for both the EU countries and the USA indicate that the differentiation is more pronounced by occupational status, rather than gender per se (Tables 4.6 and 4.7). In the EU, 23 per cent of the workforce has some type of formal or informal arrangement for flexible working hours in which they are able to vary their starting and finishing times (Table 4.6). Another one in five (21 per cent) state that they work fixed hours and have some control over this arrangement. The remaining two-thirds have their hours set by their employers, including 11 per cent who work variable hours arrangements. Men are slightly more likely to vary their own starting and finishing hours, while women are slightly more likely to work fixed hours set by their employer. The data for the USA are collected using a different question format and thus are not directly comparable, but here 30 per cent of men and 27 per cent of women employed full-time have some form of flexible schedule, which includes 11 per cent of the full-time workforce with a formal flexi-time programme (Table 4.7). This follows a marked expansion in the incidence of flexible schedules in the USA in the 1990s (Golden 2000). Perhaps surprisingly, in the EU there are only negligible differences between women full-timers and part-timers in patterns of flexible schedules at this aggregate level of analysis. However, for the USA, Golden (2000) has shown that those who usually work very long full-time hours (50 or more) or part-time hours have more opportunity to vary or change their starting and finishing times than workers with standard full-time hours. He concludes that US workers can only obtain flexible hours if they opt out of standard full-time hours. Turning to occupational differences, in the EU, managerial and skilled agricultural workers are the most likely to work variable hours that they set themselves (both of these occupations encompass the majority of the self-employed, but this pattern also holds for employees). Conversely, those employed in craft, operative or elementary jobs are the most likely to be working hours that are set by their employers, whether on a fixed or variable basis. The data for the USA are not directly comparable, but again occupational differences are pronounced, with flexible schedules being most prevalent among full-timers in managerial and professional grades, sales and personal services. Both the European and the US data show that once occupational position is taken into account, the main gender difference is that among professionals and technical grades – as well as in sales in the United States – women have lower rates of flexible schedules. This gender difference among professionals is probably associated with segregated employment positions, whereby more women professionals are in the lower grades and in health, education and other ‘personal contact’ work, in which hours are more tightly scheduled to meet service requirements, than other professional areas such as IT or engineering. A recent review by the OECD (2001) concludes that there is less

Gender and working time 127 Table 4.6 Fixed and variable start and finishing times, EU-15 (%)

Managers and senior officials Professionals Associate professional and technicians Clerical Service and sales Skilled agricultural and fishery Craft and related Operators and assembly Other (elementary) labour Men Women Women – Full-time Women – Part-time All

M W All M W All M W All M W All M W All M W All M W All M W All M W All

Employer sets fixed hours

Fixed hours – personal influence

Flexible Employer schedules sets variable – personal hours influence

13 21 16 28 47 38 33 46 39 50 50 50 41 48 46 20 18 20 54 51 53 55 72 58 58 58 58

30 39 33 18 20 19 24 21 23 25 28 27 19 21 20 17 17 17 19 15 19 13 11 13 15 20 18

50 36 45 44 20 32 32 21 27 17 14 15 22 16 18 50 54 51 17 24 18 13 9 13 13 12 12

7 4 6 10 13 11 11 12 11 8 8 8 18 15 16 13 11 12 10 10 10 19 8 17 14 10 12

42 48 50 46 45

20 23 21 24 21

26 19 18 19 23

12 11 11 11 11

Source: The European Working Conditions Survey, 2000. Notes 1 The occupational classification used is ISCO Major occupational groups (1 digit data). 2 M ⫽ Men, W ⫽ Women, FT ⫽ Full-time (35 ⫹ hours per week), PT ⫽ Part-time (under 35 hours per week). 3 The item was derived from two questions: q26iv ‘Can you influence your working hours’ (Yes/No/Don’t know) and q18aiii ‘Do you work fixed start and finishing times’ (Yes/No/Don’t know). People who replied ‘no’ or ‘don’t know’ to q26iv are considered to have their hours set by their employer, and these may be fixed or variable depending on the response to q18aiii. People who replied ‘yes’ to both question items are defined as working fixed hours under their personal influence (for example, they may have options to vary their hours but choose not to). Those who replied ‘yes’ to q26iv and ‘no’ to q18aiii are defined as working flexible schedules under their personal influence.

128

Colette Fagan

Table 4.7 Flexible schedules of full-time wage and salary workers in the USA, by occupation and gender, 2001 % with flexible schedules

Executive, administrative and managerial Professional Technicians and related support Sales Clerical and administrative support Services – private household Services – protective Services – food Services – health Services – cleaning and caretaking Services – personal Mechanics and repairers Construction trades Other precision production, craft and repair Machine operators and assemblers Transport and material moving Other (unskilled) labour Farming, forestry and fishing All

Men

Women

All

% of all with a formal flexitime programme

49

42

46

17

46 37 44 24 – 16 24 21 15 31 20 17 18

26 28 36 25 35 14 22 16 15 38 18 – 16

36 31 41 25 35 16 23 17 15 36 20 17 18

14 12 12 12 10 8 7 8 5 15 7 6 7

11 20 12 21 30

7 22 10 18 27

9 20 12 21 29

5 7 5 6 11

Source: Current Population Survey (CPS), derived from Tables 2 and 3 of the US Bureau of Labor Statistics, May 2001 supplement, USDL 02–225 ‘Workers on flexible and shift schedules in 2001’, http://www.bls.gov.cps. Notes 1 Data relates to those aged 16 years and over who usually work full-time (35 hours or more per week) on their main job. 2 The question asked about flexible schedules was ‘Do you have flexible work hours that allow you to vary or make changes in the time you begin and end work?’ (Yes/No). If the respondent answered ‘yes’ a follow-up question about formal flexi-time programs was asked: ‘Is your flexible work schedule part of a flexi-time program or other program offered by your employer?’ (Yes/No).

national variation in the incidence of flexi-time than in a number of other ‘family-friendly’ working time arrangements in firms. In general, the highest levels of flexi-time are found in Australia11 and the USA, exceeding those found in most EU countries, while, in comparison, Japan ranks among the lowest. Furthermore, Bosch (1997) argues that the flexi-time arrangements that exist in Japan are scarcely used because they run counter to established organizational work cultures in which absence and lateness are severely penalized (see Takagi 1993); in contrast, the flexitime systems that exist in Europe, North America and Australia are generally widely used and popular with the workforce. For example, the USA

Gender and working time 129 may have one of the highest coverage rates for flexi-time systems, but Golden (2000) argues that surveys regularly show that many more of the workforce would like this arrangement.

Working time preferences and work-family compatibility In this section we consider men and women’s evaluations of their current working time arrangements using data on their working time preferences and their assessments of the work–family compatibility of their work schedules. This type of information provides an indication of how people feel about their current working time arrangements, and provides some insight into the kind of policy developments that they would like. Of course, such evaluations do not provide accurate predictions of how people would behave in some future situation, for competing priorities, constraints and other considerations will also influence labour supply decisions. In particular, when real wages are falling, or employment is becoming more insecure, people may be less willing to reduce their working hours than in more positive economic conditions. A recent survey of the EU-15 and Norway revealed that nearly two in three employed persons would prefer to work a different amount of hours than their present arrangement.12 Half (51 per cent) would prefer to reduce their hours, whether traded for lower current earnings or against future pay rises.13 Another 12 per cent would like to work longer hours. The available time series data suggest that the proportion of the European workforce that would like working time reductions has increased since the mid-1980s, even in those countries where such reductions have been achieved (Lehndorff 2000). Estimates made by Jacobs and Gerson (2000) suggest a broadly similar picture for the USA, although a slightly higher proportion of the US workforce wants to increase their hours. They calculate that nearly half of Americans would prefer shorter working hours, 90 per cent of whom want to work at least 5 fewer hours a week, while 17 per cent would prefer to work longer hours. Bell and Freeman (1995) also found that more Americans than Europeans wanted to work longer hours. They argue that the very dispersed income distribution in the USA provides the financial pressure that drives Americans to work long hours. This is echoed by Bluestone and Rose’s (2000) conclusion that falling real wages in the USA mean that many families are ‘running harder’ to maintain their standard of living. In contrast, the available data suggest that a smaller proportion of the Australian and Japanese workforces would prefer to adjust their hours, although such comparisons must be treated cautiously since measures of preferences are sensitive to differences in questionnaire design. In Australia, a government survey in 2000 indicated that 7 per cent of all employees would prefer to work fewer hours and 21 per cent would prefer more hours (Australian Bureau of Statistics 2002). However, according to

130

Colette Fagan

a national survey by the Australian Council of Trade Unions (ACTU), 79 per cent of employees want a limit set on weekly working hours, in the context that many of them are experiencing increasing workloads and a negative impact of working hours on their family life (ACTU 2003). In Japan 25 per cent of employed men and 20 per cent of employed women would like to reduce their hours of work (including stopping work), while only 7 per cent of employed men and 8 per cent of employed women would like to work longer hours (including taking on another job).14 In Japan, as in the USA, financial pressures to meet housing costs and maintain purchasing power dampen preferences for working time reductions, yet at the same time a high proportion of the Japanese workforce is dissatisfied with the balance of their time between employment and leisure (Seifert 1994). The European data show that employed men are even more likely to want to reduce their hours than are employed women; conversely, employed women are more likely to be under-employed and to want to increase their hours (Table 4.8). Overall, employed men are slightly more likely to have a preference to adjust their hours than employed women. This is partly because men are more likely to be working long hours, as we discuss next. It is also because having young children increases women’s propensity to exit employment, and one reason for this is that they have been unable to obtain working time arrangements that are compatible with their care responsibilities (Rubery et al. 1999). Preferences for adjustments to hours are clearly related to current hours of work. Very few full-timers want to work longer hours, and the proportion that wants to reduce their hours rises with the number of hours worked. Part-timers, particularly those in short hour jobs (20 hours a week or less) are the most likely to want to work more hours. The under-employment of these part-timers indicates that many of these jobs are designed primarily to meet employers’ requirements rather than labour supply preferences. The minority of men who work less than 35 hours a week are even more likely to want to increase their hours than are women with these working hours, while among those working longer hours, women are even more likely to want to reduce their hours than are men. The amount of adjustment that most people want to make is substantial, particularly when considered as a proportion of their current volume of work. Overall, there is a general tendency to prefer to exit the extremes of very short or very long hours of work and move into the middle ground of substantial part-time/short full-time hours in the 20–39 hour range. Some employed men and women who currently occupy this middle ground do want to move, but mainly switching between short full-time and parttime hours. Similarly, in the USA and Australia, the preference is to exit the extremes of very short and very long working weeks (Jacobs and Gerson 2000; Australian Bureau of Statistics 2002).

Gender and working time 131 Table 4.8 Employed men’s and women’s preferences for working time adjustments by their current working hours, EU-15 plus Norway (a) % who want to adjust their hours . . .

Current average weekly working hours are . . .

All

Under 20

20–34

35–39

40–49

50⫹

Men Reduce hours Keep same hours Increase hours Total %

5 34 61 100

15 48 37 100

41 49 10 100

61 36 3 100

80 18 2 100

57 34 9 100

Women Reduce hours Keep same hours Increase hours Total %

8 42 50 100

17 56 27 100

53 42 5 100

69 30 1 100

83 17 * 100

44 40 16 100

All Reduce hours Keep same hours Increase hours Total %

8 40 52 100

16 54 30 100

46 45 9 100

62 34 4 100

81 18 1 100

51 37 12 100

(b) Mean average adjustment in hours desired

Under 20 hours

20–34 hours

35–49 hours

Mean average addition desired Men (standard deviation) Women (standard deviation)

19 (10.7) 15 (9.4)

12 (5.9) 11 (6)

8 (6.2) 5 (3.4)

Mean average reduction desired Men (standard deviation) Women (standard deviation)

– –

7 (5.3) 7 (4.2)

8 (5.2) 10 (5.9)

50⫹ hours – – 20 (10) 20 (9.3)

Source: The Employment Options Survey, 1998. Note * indicates less than 5%. – not shown because base number too small for calculation to be reliable.

The net picture in the EU is that on average employed men would prefer a 37-hour week and employed women a 30-hour week; an average reduction of 6 hours and 3.5 hours respectively (Table 4.9). The standard deviation indicates that the spread of preferences around this average is wide for both sexes;15 however, it is clear that the general picture is that more of the employed would like to work short full-time or substantial part-time hours than currently do so. On average, ‘job seekers’16 have similar working time preferences to the employed. If working time preferences were realized in the labour market, then the average gender, occupational and national differences in the volume of hours worked would be reduced (Fagan 2001a).

100

43 (11.7)

Total %

Average (standard deviation)

36.7 (9.9)

100

8 18 32 33 8

Preferred

■

35.3 (9.9)

100

5 21 26 43 4 33.5 (12.2)

100

14 25 26 28 7

Actual

30.2 (10.0)

100

14 42 25 16 2

Preferred

Job-seeking men prefer . . . Employed women

Note ‘Job seekers’ are all those who want employment now or within the next 5 years.

Source: The Employment Options Survey, 1998.

3 6 22 44 25

Actual

Employed men

% distribution of actual and preferred weekly hours of work across hours categories

Under 20 hours 20–34 35–39 40–49 50 plus

Weekly hours

Table 4.9 The volume of actual and preferred hours of the employed and ‘job seekers’, EU-15 plus Norway

30.4 (9.6)

100

9 44 19 27 1

Job-seeking women prefer . . .

Gender and working time 133 Multivariate analysis showed that the current volume of hours worked was the most significant factor influencing working time preferences in Europe (ibid.). Once hours were controlled for, then among full-timers men were more likely to want to reduce their hours if they were a manager, highly educated (well-paid) or approaching retirement. Fatherhood had no significant influence on working time preferences. For women full-timers, the significant circumstances other than the volume of hours were having a young child or being a manager. Once hours were controlled for, then women working part-time were more likely to want to increase their hours if they had older or no children to care for, or were in low-income households. There is also substantial latent demand among full-timers for working time reductions through a move into part-time work in the EU-15 and Norway: 23 per cent of employed women and 19 per cent of employed men currently work full-time and would prefer to work part-time, as would 38 per cent of all job seekers. This latent demand for part-time work varied nationally, but is influenced by more factors than just the current availability of part-time work. For example, the demand for opportunities to work part-time was relatively low both in some countries where this form of employment is relatively under-developed (e.g. the Southern European countries) and in some countries where part-time work is already well established (e.g. Sweden). Full-timers wanted to work part-time in order to have more time for children and other domestic activities, for personal activities, and to reduce the strains of working; in other words, reasons similar to those given by those who do work parttime (see the discussion above regarding the volume of weekly working hours). Women were more likely to mention child-care, but so did a large minority of the men. All full-timers, regardless of their preferences for part-time work, perceived similar barriers to part-time work. The most commonly mentioned barriers were lack of opportunities for part-time work with their employer and career penalties, followed by inferior social protection entitlements and financial considerations (ibid.). Clearly one of the reasons why people may prefer to adjust their working time is to improve the fit between the time demands of their jobs and those associated with family and other commitments outside of employment. Table 4.10 explores this question using people’s answers in a European survey17 that asked, ‘In general, do your working hours fit in with your family or social commitments outside work very well, fairly well or not at all well?’ It should be noted that when ‘satisfaction’ questions such as these are asked in surveys, they typically produce positive responses. These positive assessments should not be read simply at face value. More probing explorations about satisfaction in qualitative studies enable people to make more considered and nuanced assessments that reveal higher levels of dissatisfaction (e.g. Burchell et al. 2002). Furthermore, respondents may consider that their work schedule is compatible

18 30 39 38 37 35 15 21 38 14 17 27 44 22

Daytime, regular weekend work, no long days Daytime, some weekend, including long days Daytime, weekday, regular long days Daytime, regular weekend, including long days

Some evenings/nights, but not rotating shiftsd Rotating shifts or permanent nightse

Low level of work intensity Some work intensity Higher level of work intensity

Fixes own start and finish times Start and finish times fixed by employer Varies own start and finish times Start and finish times varied by employer

All

20

12 20 22 36

15 20 32

31 32

24 34 25 34

9 17 12

Women, 35⫹ hours

9

6 9 10 13

5 11 14

13 18

8 8 – 35

4 5 6

Women, ⬍ 35 hours

19

12 16 22 36

13 19 32

28 31

18 28 33 37

6 10 11

Alla

Notes a ‘35⫹ hours’ refers to people working at least 35 hours per week and ‘⬍35 hours’ to those working less than 35 hours per week. Men working less than 35 hours per week are not shown separately, but are included in the overall ‘all’ column. b A ‘long day’ is defined as working 10 hours or more. ‘Some long days’ are 1–5 per month, ‘regular long days’ equal at least 6 per month. c ‘Some’ weekend work includes 1–4 days per month, ‘regular’ weekend work is more than 4 days per month. d ‘Some’ evenings/nights includes people who report that their work schedules do not always fall into daytime hours, but are not part of shift rotas. Evenings are defined as 6–10 p.m. and nights as 10 p.m.–5 a.m. in the survey. e Rotating shifts includes all those with an alternating shift pattern during daytime hours or day and night shifts or permanent nights. It excludes those with permanent morning or afternoon shifts. – indicates data not shown because of sample size limitations.

Source: The European Working Conditions Survey, 2000.

6 8 11

Men, 35⫹ hours

% who report that their working hours fit ‘poorly or not at all’ with family and social commitments

Daytime, weekday, no long daysb Daytime, some weekend work,c no long days Daytime, weekday, some long days

Schedule type

Table 4.10 Employed men’s and women’s assessment of the compatibility of their working hours with family and other commitments, EU-15

Gender and working time 135 with family life because it has been selected strategically in the context of having to arrange particular forms of child-care, or in relation to other considerations such as the work schedules of their partners, or because they have curtailed activities to accommodate their work schedules. If there was a change in their situation or their expectations about what ‘compatible’ standards are, for example, due to changes in child-care services, then it might be expected that the types of schedules assessed as ‘compatible’ would also change. Detailed and qualitative interviews would be necessary to obtain more considered opinions and to explore what people mean by ‘compatibility’ and what it is about their hours and other commitments that do or do not produce this sense of ‘compatibility’. Nonetheless, survey data provide some useful indications that have the advantage of permitting comparisons of the responses associated with various working conditions and domestic situations. The more time that is allocated to employment, the less there is available for other activities, hence full-timers are more likely to judge that their working arrangements are a poor fit with other commitments than are part-timers (Table 4.10). A similar pattern is found when people are asked about their satisfaction with the time they have for family and leisure pursuits rather than about ‘compatibility’ (Fagan 1996). This sense of incompatibility and dissatisfaction becomes more pronounced the longer the hours worked for both sexes, although what men and women mean when they say their hours are incompatible or unsatisfactory is likely to be different and coloured by the gendered division of care responsibilities in the home. The schedule as well as the volume affects the compatibility of working hours. On one hand, work schedules that spill into the evening, night and weekends can be considered disruptive to family life in that they present coordination difficulties with the daily schedules of raising children or creating shared ‘family time’. On the other hand, such schedules may offer alternative opportunities for the coordination of employment with family life. Men and women who work at least 35 hours a week are least likely to judge their work schedules to be compatible if they work long days or weekends on a regular basis, or some weekends in conjunction with long days, or evenings, nights or rotating shifts (Table 4.10). Conversely, daytime, weekday work without long days is the most compatible; although the inclusion of some weekend work or some long days only raises the incompatibility score by a few percentage points. Higher levels of work intensity also reduce the sense of ‘compatibility’. Women who work less than 35 hours a week are the least likely to report a ‘poor fit’ for each schedule, demonstrating that the compatibility of working arrangements depends upon the volume as well as the schedule of hours. Variable start and finish times are less compatible than fixed ones, particularly when the variation is set by the employer, but also when workers vary their own starting and finishing times. This seems paradoxi-

136

Colette Fagan

cal, for when workers have some autonomy to organize their working hours this might be expected to make their jobs more compatible with other activities. The explanation is likely to be found in the nature of the job, for this autonomy may be associated with a managerial or professional job that requires a commitment of long hours and thus the ability to influence starting and finishing times may provide insignificant relief from the other working time demands of the job. This issue is returned to below. Some workers do find ‘non-standard’ schedules to be compatible with their other commitments, for example, one quarter of men and women working rotating shifts report that this fits in very well with their other commitments. However, this rate of compatibility is much lower than that reported by non-shiftworkers. So it seems that the working time elements that contribute to a greater sense of work–family compatibility are regular, daytime schedules without long days, in other words, the ‘standard working week’ that has been the benchmark of industrial relations since the earliest negotiations about regulations. Yet this is in tension with many of the schedules that are being introduced to provide companies with more flexibility to cover variable or extended operating requirements. Multivariate analysis shows that long working hours, ‘unsocial’ schedules (long days or working during the evening/night) and high work intensity each have an independent, negative effect on men’s and women’s assessment of the degree of compatibility of their working time arrangements with their family and social life (Fagan and Burchell 2002). Of these three dimensions of working time, working ‘unsocial’ schedules (that involve evenings, nights or long days) had the most negative impact. Working time autonomy offers some respite, but this is relatively weak compared with the negative effect of working unsocial or long hours. Gender and occupational status are insignificant factors once the actual details of working hours and schedules are taken into account (Fagan and Burchell 2002).18

Working time policies to promote gender equality and ‘work–life balance’ The preceding analysis has demonstrated the gender differences in working time patterns in industrialized countries, particularly in the volume of hours worked, but also in the type of schedule and the degree of autonomy in relation to these schedules. Yet there are also gender similarities, for example, a shared preference to avoid or exit the extremes of very long or very short hours of work. Job-related differences are often more pronounced than gender differences, for example, the work schedules of women working in the male-dominated echelons of management are quite different to those of women in manual jobs. This suggests that particular working time arrangements, such as long hours or unpredictable schedules, may contribute to the process of gender segregation in employ-

Gender and working time 137 ment by creating barriers to occupational entry or progression for those with care responsibilities. On the basis of this analysis, working time policy must address five related objectives if it is to contribute to the promotion of gender equality in employment and work–life balance (see Figure 4.2). The first objective is to reduce working time barriers to labour market participation and thus contribute to raising women’s employment rate. The second objective is to address any particular working time obstacles to women’s entry to management and other male-dominated activities, thus contributing to the

Policy objectives 1 To reduce working time barriers to labour market participation and so contribute to raising women’s employment rate. 2 To address particular working time obstacles to women’s entry to management and other male-dominated activities, and so contribute to the reduction of segregation, particularly vertical segregation. 3 To develop working time arrangements to improve the quality of the reconciliation of employment and family responsibilities and ‘work–life balance’ more broadly. 4 To develop equal treatment between full-time and part-time workers including opportunities to make transitions between full-time and part-time hours at different life stages. 5 To adapt men’s working time patterns and increase their time involvement in parenting and other care activities and hence contribute to reforming the gender division of labour. Policy elements a) Collective reductions in full-time hours to tackle ‘over-working’ and ‘long hours cultures’ b) Increase the opportunities for good quality, part-time work • Encourage the creation of ‘substantial’ rather than ‘marginal’ part-time hours of work • Equal treatment in employment and social protection systems • Increase opportunities to work part-time in a wider range of jobs, including promotion opportunities • Increase opportunities for mobility between part-time and full-time work c) Improve the reconciliation of employment and family responsibilities and ‘work–life balance’ • Leave entitlements (Maternity, paternity, parental and other care-related leave) • Child-care services, including school opening hours, and other support systems (e.g. transport policy) • Rights to adjust between full-time/part-time hours • Autonomous flexibility arrangements (such as flexi-time and time banking, working from home) • Reduce other working time elements that have a negative impact on work-life balance and/or health (intense workloads, night-work and rotating shifts, unpredictable variations in schedules) d) Include specific incentives targeted at men to improve their use of ‘family reconciliation’ measures, such as non-transferable parental leave entitlements

Figure 4.2 Working time policy objectives and elements to promote gender equality in employment and work–life balance.

138

Colette Fagan

reduction of gender segregation, and particularly vertical segregation. Both of these objectives are about increasing women’s labour market integration, and the third has a more ambitious vision: to develop working time arrangements that improve the quality of the reconciliation of employment and family responsibilities and work–life balance more broadly. The fourth objective underpins the preceding three, and is concerned with developing equal treatment between full-time and part-time workers, including opportunities to make transitions between full-time and part-time hours at different life stages. The final objective focuses upon adapting men’s working time patterns and increasing their time involvement in parenting and other care activities, hence contributing to reforming the gender division of labour in households. The rest of this section discusses the policy elements that have the potential to meet these objectives, which are also summarized in Figure 4.2. Collective regulation to curtail long full-time hours or ‘overworking’ Long full-time hours are widespread in many countries, yet they are unpopular with most of the workforce and make it difficult to devote time to raising children or other time-consuming commitments. Regulations to curb long full-time hours in favour of more moderate arrangements have a key role to play in promoting work–life balance, alongside that of health and safety (discussed further in Chapter 6). However, it is also the case that while a general reduction in hours sets an important upper limit on the workplace demands that can be made on people’s time, it is too blunt a tool to resolve the complexity of work–life balance on its own. Rather, it is a foundation to support the efficacy of other measures tailored to accommodate the changing care responsibilities of workers at different stages in their lives (which are discussed later). A particular issue is how to regulate the long hours culture that is often associated with the ‘self-determined’ working patterns of a growing number of managers and new professions who largely fall beyond the scope of traditional working time regulations. The long hours in these occupations make it difficult for women with care responsibilities to advance, and hence reinforces gender segregation. Furthermore, this form of ‘unspecified’ working time may be spreading to other groups of workers as employers seek more flexibility from their workforce to meet evertighter deadlines, for example, in the ‘new economy’ firms. Some of these workers might feasibly be regulated through extending the coverage of existing collective agreements and legislation, but, for others, new ‘accounting currencies’ (Bosch 1999b) and new workplace mechanisms of monitoring and regulating workloads will be needed. This might include using time-budgeting methods to measure the required input for particular tasks in conjunction with incentives and targets for line managers to re-

Gender and working time 139 organize work methods and to replace the ‘long hours culture’ with ‘smart working’. For example, external consultants have been used in US companies such as the Marriott hotels chain and the high-technology Xerox Corporation. They worked with managers to identify and remove timewasting procedures, to create more productive structures in the working day, and to shift the work culture to focus on tasks accomplished rather than hours worked (Fletcher and Rapoport 1996; Munck 2001). For some groups it might also be more feasible to regulate the number of days worked rather than the volume of hours. This is the route taken in French working time legislation, although the French managerial union (CFE-CGC) argues that maximum weekly working hours should still be set for managers (IDS 2002). Where collective working time reductions are accompanied by the introduction of more non-standard or variable work schedules, for example, in some annualized hours arrangements, then this may introduce negative impacts on work–family compatibility that partly reduce the gains obtained due to working time reductions. This is because weekly hours may be a more important reference period for workers with the routines and timetables of domestic life with young children or elder relatives, yet such considerations may be less pressing for workers without such responsibilities. Workplace consultation can help to mitigate this, but only if the interests of those with care responsibilities are represented. In some areas of employment the ‘insiders’ may be predominately male, or young and childless, and hence may agree to working time arrangements that contribute to the exclusion of those with care responsibilities from the workplace. Increase the opportunities for good quality, part-time work Part-time work is one means of reconciling employment with other activities. In particular, ‘substantial’ rather than ‘marginal’ part-time hours are popular among the workforce. The key issue is that the quality of parttime work must be enhanced if it is to be used to promote gender equality (OECD 1994, 2001; O’Reilly and Fagan 1998). Equal treatment regulations in employment, non-wage benefits and social protection systems help to improve the conditions of part-time work. This regulation has recently been strengthened in law across the EU through the adoption of the 1997 Part-Time Work Directive.19 However, the coverage below certain hour thresholds is still poor in some EU countries, while in some non-EU countries, such as Japan and the USA, unequal treatment is widespread and part-timers receive comparatively few benefits. In Australia, part-timers have fared well historically but their situation is set to deteriorate with the decentralization of collective bargaining (Baxter 1998; Houseman and Osawa 1998; OECD 2001). It is also necessary to desegregate part-time work by increasing

140

Colette Fagan

opportunities to work ‘substantial’ part-time hours in a wider range of jobs. Under these conditions under-employment plus the loss of earnings and career advancement associated with part-time work is reduced, and part-time workers become integrated alongside full-timers in the workplace and the wider employment hierarchy (O’Reilly and Fagan 1998). Fiscal reform may be needed to remove cost incentives that favour the creation of marginal over substantial part-time working hours, perhaps in conjunction with regulations to establish minimum hour thresholds. One way of promoting opportunities for part-time work in a wider range of jobs is through disseminating ‘good practice’ examples in order to reduce information and design costs. Another mechanism is to introduce employee entitlements to request reduced hours in their current jobs as part of work–family reconciliation measures, with the right to resume fulltime hours at a later period. To date, mobility from part-time into full-time work is rare in most countries, with the exception of Sweden (Evans et al. 2001; O’Reilly et al. 2000) where this entitlement is built into the parental leave legislation (discussed below). Improve the reconciliation of employment and family responsibilities and ‘work–life balance’ In nearly every OECD country there has been an extension of statutory entitlements to maternity, paternity and parental leave in recent years. Child-care services have also expanded, and some countries have introduced new family leave arrangements to accommodate the care of sick children or other relatives (Moss and Deven 1999; OECD 2001). However, marked national differences still exist, and in general the most comprehensive leave packages are provided in the Nordic countries, while the lowest level of statutory provision is generally found in neo-liberal countries such as the UK and the USA, and also in Japan. Recent comparative studies of industrialized countries show that statutory entitlements are important, for in their absence voluntary provision by companies is limited and uneven (Arrowsmith and Sisson 2001; den Dulk 2001; OECD 2001). Individual entitlements to work reduced full-time or part-time hours, and to increase hours once more at a later stage, enhance the autonomous flexibility of employees by enabling them to adjust their hours without moving jobs. Also, this type of policy has the potential to contribute to a more general diversification of the range of jobs available on a part-time basis (see above) by introducing a ‘learning’ catalyst for companies. This option has existed for many years in the Swedish parental leave system. New legislation in the Netherlands has established the right for employees to reduce their hours to part-time, unless the employer can demonstrate that this is not viable for business reasons. The right to request part-time hours has also been introduced into law in some other European coun-

Gender and working time 141 tries, such as the UK, but in a diluted form that allows employers more grounds for refusing such requests. Other ‘autonomous flexibility’ arrangements also contribute to the reconciliation of employment and family responsibilities, including flexi-time and time-banking schemes, and opportunities to work from home. However, care-related entitlements to extended leave or other working time adjustments often have a low take-up rate, particularly when the policy is not underpinned by a statutory entitlement. This occurs when such policies are ‘symbolic statements’ that are undermined by conflicting workplace norms and practices (Lewis 1997). Such obstacles include a workplace culture that treats long working hours as a measure of productivity and organizational commitment, and hence necessary for job security or promotion, and where the use of alternative working time arrangements are opposed by line managers or are poorly implemented and incur resentment from colleagues because of the negative impact on their workloads. So to ensure that reconciliation measures are legitimated and used in organizations. a number of supporting measures are necessary. Such measures are also pertinent to the particular problem that take-up rates are systematically much lower for men than women, and so are discussed in the next section. Another issue is that being able to fulfil care responsibilities requires energy as well as time. Yet work intensity appears to be increasing, due to a combination of new information technologies and production methods and organizational reliance on ‘leaner’ workforces (Dhondt 1998; Gallie et al. 1998; Green and McIntosh 2001). The proportion of the workforce working at high speed increased in Europe over the 1990s, and the gender difference eroded as women ‘caught up’ with men (Fagan and Burchell 2002); work intensity has also risen in the USA and Australia (Bond et al. 1998, reported in Evans 2001; ACTU 2003). As discussed earlier, intense workloads, non-standard schedules and unpredictable variations in working hours each reduce the compatibility of employment with family life, as well as increasing work-related illnesses (see the discussion above, ‘Working time preferences and work-family compatibility’). Hence, these developments are a concern for policy for reasons of both health and safety and work–family reconciliation. Include specific incentives targeted at men to improve their use of ‘family reconciliation’ measures Very few men switch to part-time work or take parental leave when they are parents. It is important that reconciliation measures incorporate elements that address the organizational and attitudinal obstacles that deter male take-up. If more men are encouraged to make these changes, it will contribute to changes in the gender division of labour within the household, and in workplaces it will help to legitimate and ‘normalize’ the use of

142

Colette Fagan

these measures, instead of marginalizing those mothers who make working time adjustments (see above). Statutory entitlements improve take-up rates for both sexes, as does a reasonably high earnings-replacement rate for extended maternity or parental leave (Moss and Deven 1999; OECD 2001). Additional features which encourage men to use such measures are a non-transferable ‘daddy quota’ of leave and flexible options as to when and how the leave is taken, plus promotional campaigns and target setting. Here, the Swedish parental leave system provides a good example. It is a flexible system which allows parents to take leave full-time or to continue working part-time, supported by a generous income replacement rate, and since early 2002 a non-transferable ‘daddy quota’ of two months’ leave has been introduced. This quota has improved male take-up.20 Additional initiatives are required in workplaces to legitimate and promote the use of reconciliation measures by men as well as women, particularly in male-dominated workplaces which have had less exposure to such measures compared to many female-dominated workplaces. These initiatives include promotional campaigns and consultation with the workforce, in conjunction with training and the dissemination of good practice examples for those line managers and team leaders charged with implementing reconciliation measures. Targets and incentives for appropriate line managers to improve take-up rates could be developed to monitor and improve the situation, extending the application of established tools for measuring other aspects of organizational performance such as sales or health and safety records. Some companies with progressive personnel policies already use targets to measure and improve their performance in relation to the recruitment and promotion of women and other underrepresented groups, for example, the voluntary business initiative ‘Opportunity Now’ in the UK.21 A similar procedure could be developed where organizations audit the care responsibilities of their male and female workforce, revise their work–family policies in light of this audit and promote it in the organization, and set targets for take-up of these policies for each sex designated to the appropriate level of management. In other words, a new organizational norm has to be developed, whereby the use of family reconciliation measures is positively valued and encouraged for both men and women. A number of social actors have a role to play in making this normative shift, including the state, trade unions and employers’ organizations. Finally, reducing gender inequality in employment is also needed as a basis for increasing men’s time involvement in care responsibilities. The gender division of labour is more equal in couples where the woman has high earnings and qualifications that place her on a more equal economic footing with her partner, although normative attitudes towards the gender division of labour are also influential (Gershuny et al. 1994; Moss and Deven 1999).

Gender and working time 143

Conclusion The ‘male breadwinner’ division of employment and domestic responsibility structures women’s employment, particularly for mothers with young children. In most industrialized countries, women have a lower employment rate than men, and, when employed, generally work shorter hours in the context of doing most of the unpaid care and domestic work in society. In this chapter we have also seen that average gender differences are often less pronounced than job-related differences. For example, both men and women are involved in a diverse range of working time schedules in the labour market, working time autonomy is limited for many workers of both sexes, and rates of work intensity are similar. Both sexes also share a preference to exit or avoid very long or short hours of work, and there are similarities in the type of schedules that they identify as being most compatible with family life and other commitments. In recent years women’s employment rates have risen, but this has been concentrated on the better-educated, producing polarization among women. From a labour market perspective, improving work–family reconciliation measures will raise women’s employment rates and make better use of their skills in employment, will promote further progress towards gender equity, will raise the financial resources of households, and will help address low fertility rates which are exacerbating the projected shortfalls in labour supply as the working-age population shrinks (OECD 2001). From a broader perspective on social integration and citizenship, work–family reconciliation measures are needed so that people have the time available to develop and maintain social ties, networks and communities. This chapter has discussed the type of working time conditions that can promote gender equality and provide the basis for improving work–life balance. Currently, part-time work and many work–family reconciliation policies are gendered, for they rest upon women’s subordinate economic position and are disproportionately used by women. Such developments have facilitated women’s entry into employment, but to date appear largely to reinforce the gender division of labour, albeit in a new form. An optimistic scenario is that over time – perhaps across generations – such policies may provide the platform for a more radical transformation of the gender arrangement. For example, it appears that in Sweden and Denmark part-time work served as a transitional mechanism to integrate women into employment and that, for subsequent generations, part-time work has declined in favour of full-time employment for women. However, other policy interventions are required to speed up and ensure this process of change towards gender equality in economic resources and care work. In particular, action is needed to establish new workplace practices, whereby it is acceptable and positively valued for employees of either sex

144

Colette Fagan

to adjust their working time to accommodate family responsibilities, and for working arrangements in general to be made more ‘family compatible’. This will make it easier for women to enter and progress in maledominated areas of employment, and it will also provide a basis on which to get men more involved in child-care and other household tasks. Moss (1996: 23) argues that work–family reconciliation must be seen as a dynamic process in which ‘Equilibrium that equally meets the needs and interest of all parties is unobtainable yet constantly sought through a process of debate, review, negotiation and conflict.’ Intervention is thus required through a range of mechanisms and a variety of social actors. Statutory regulatory frameworks on working time through labour law, as well as in social protection and fiscal policy are crucially important. Additional supporting measures are required in workplaces for work–family reconciliation measures to be legitimated, developed and used. This includes information and promotional campaigns covering individual rights, the development and dissemination of ‘best practice’ across workplaces, and the negotiation of extra-statutory measures. Finally, state support to extend child and elder care services is a key piece of the work–family reconciliation jigsaw puzzle. In sum, working time reform, including work–family reconciliation, has to be integrated into debates about the modernization of work organization. The development of ‘gender mainstreaming’ may help to achieve this by stimulating the integration of gender equity issues into the design of all employment standards, including those regarding working time (Rubery 2002).

Acknowledgements Many of the survey data for the European Union presented in this chapter are from the ‘European Working Conditions’ and ‘Employment Options’ surveys of the European Foundation for the Improvement of Living and Working Conditions. I am grateful to the Foundation for permitting me to use these data for my analysis. The calculations and interpretation based on these surveys remain my responsibility. I would also like to thank my colleagues Hugo Figueiredo and Mark Smith, of the European Work and Employment Research Centre, UMIST, for assisting me with additional data collection.

Notes 1 Some arrangements for employees to do some or all of their work from home may also be included under ‘autonomous flexibility’ in the sense that working from home can offer temporal as well as spatial flexibility. 2 The inequality in the gender division of domestic labour is more pronounced when viewed from a societal rather than a two-adult household perspective because a growing proportion of child raising is located in female-headed lone

Gender and working time 145

3 4

5 6

7 8 9

10

11 12

13

14 15

parent households, and men’s contribution to the unpaid work in these households is minimal. A gender breakdown of annual working hours is available from Table 6b of the ILO’s volume (2002) Key Indicators of the Labour Market 2001–2. This 48-hour threshold has been adopted from the Working Time Directive for illustrative purposes only, and it might be appropriate to define ‘over-working’ using a lower threshold. Certainly this threshold was a matter of debate among governments, trade unions and employers’ associations in the European negotiations on the Working Time Directive. A similar pattern is found in Norway (see Fagan 2001a). The hour threshold used in official statistics, labour law and collective agreements varies between countries and sometimes within countries. A 30-hour threshold encompasses most forms of part-time work in many countries, while a 35-hour threshold prevails in some others, such as Sweden and the USA. Part-time work is also common for women in a number of other industrialized countries not shown here, including Norway and New Zealand. Derived from Current Population Statistics for 2001 (Household data, Table 20), US Bureau of Labor Statistics (http://www.bls.gov/cps) (accessed 15 January 2003). The European Foundation for Living and Working Conditions ‘The European Working Conditions Survey – 2000’ is a representative survey of the 15 EU member states. The sample size was approximately 1500 in each country; the data reported here for the EU-15 have been adjusted for country size. See Merllié and Paoli (2001) for further information including a discussion of the sample design and weighting. I have compared the volume of working hours recorded for each country in the European Working Conditions Survey with the results obtained in the larger sample of the European Labour Force Survey (European Commission 2000a), and the results for both surveys are very similar. For example, the core period might be 10–1600 hours with a 2-hour zone either side in which workers are free to decide when to start and finish. The permitted surplus or deficit might be 20 hours over a 4-week period, and surplus hours can be ‘cashed in’ for additional half or full days off with the agreement of the supervisor. In 2000, 38 per cent of all employees reported that they were able to work extra hours to take time off, an increase from 34 per cent in 1993 (Australian Bureau of Statistics 2002). The 1998 ‘Employment Options Survey’ was commissioned by the European Foundation for Living and Working Conditions and the Norwegian Royal Ministry of Labour. It is a representative survey of 30,000 people aged 16–64 years covering the 15 EU member states and Norway. For further details see the technical report of the fieldwork coordinators (Intratest Burke Sozialforschung 1998), or Fagan (2001a, appendix A.1). The analysis was derived from two open-ended questions: ‘In total, how many hours per week do you work at present – on average?’ (Question 55) and ‘Provided that you (and your partner) could make a free choice so far as working hours are concerned and taking into account the need to earn a living how many hours per week would you prefer to work at present?’ (Question 56). Derived from the Employment Status Survey 2001, Table 5, from the Statistics Bureau and Statistics Center, Japan (http://www.stat.go.jp/english/data/ shugyou/1.htm) (accessed 15 January 2003) The standard deviation indicates that approximately two-thirds of both populations fall within the range of plus or minus 10 hours, that is, 27–47 hours for employed men and 20–40 hours for employed women.

146

Colette Fagan

16 These were defined in the survey as all those who were not employed but would like a job now or within five years. Taken together, the employed plus ‘job seekers’ encompassed 90 per cent of all working-age men and 80 per cent of all working-age women in the sample. 17 The European Foundation for Living and Working Conditions ‘The European Working Conditions Survey – 2000’, see note 9. 18 As well as having negative effects on the issue of work–family compatibility, working arrangements that involve long hours, unsocial schedules or intense workloads also increase the incidence of work-related illness, independent from the effects of other working conditions (Fagan and Burchell 2002). Australian studies have also shown that having to work more hours than preferred, and a lack of influence over start and finish times were both independently associated with higher rates of job-related stress (reported in Evans et al. 2001). 19 The Part-Time Work Directive (97/81/EC, 15 December 1997) introduces equal treatment for part-timers on a pro rata basis with comparable full-time workers doing ‘similar work’ into Community Law. It also recommends that action should be taken by employers to facilitate part-time work at all levels in the establishment, including giving due consideration to requests by employees to switch from full-time to part-time work, or vice versa (it does not establish a right to request such working-time adjustments). It also recommends that member states take action to adapt social security systems to accommodate part-time work. The equal treatment principle is also established in the ILO’s Part-Time Work Convention 1994 (no. 175), where Article 7 states that ‘Measures shall be taken to ensure that part-time workers receive conditions equivalent to those of comparable full-time workers’. 20 In addition, some Swedish companies (e.g. Ericsson, Statoil, ABB, Telia) have introduced supplementary financial assistance to reduce the negative impact of the social insurance income ceiling on the parental leave income replacement rate for those wage earners who are above the threshold (often the men) (personal communication from D. Anxo). 21 See http://www.opportunitynow.org.uk for further details.

5

Working time at the enterprise level Business objectives, firms’ practices and workers’ preferences Jon C. Messenger

Introduction In this chapter, we turn our attention to the realities of working time at the enterprise level. By looking at the circumstances which exist in particular firms, it is possible to better understand the dynamics that are causing the working-hour ‘surpluses’ and ‘deficits’ described earlier in this volume – what the key factors are and what can be done about them. Moreover, it is not simply the number of hours which is a potential concern; it is also how those hours are arranged. For example, two workers with 40-hour workweeks may face very different situations if one works the same hours every week, while the other works a schedule which varies greatly from one week to another. Of course, enterprises do not implement working time practices in a vacuum. As we will see, the hours of work and working time arrangements which emerge are shaped by their competitive environment; this environment includes a variety of extra-organizational factors, such as the particular product and labour market(s) within which they operate and the industrial relations framework of the country, as well as intra-organizational factors, including organizational culture and policies (Oeij and Wiezer 2002). An important aspect of how enterprises respond to their environment is how they organize their productive resources, including their human resources, for the production of goods and services – that is their work organization. Even firms operating in the same market may adopt very different forms of work organization, with profoundly different results for their employees’ terms and conditions of employment, including their hours of work. This chapter will investigate how actual working time patterns have been implemented at the firm level in relation to changes in the organization of work and also considering the system of labour market regulation which exists in various countries. Following some background information on trends in work organization and the forces that appear to be driving these trends, this chapter will focus primarily on illustrating the range of actual working time practices that have emerged in specific industries and firms. Enterprise cases which illustrate long hours of work,

148

Jon C. Messenger

short hours or part-time work, and a variety of working time arrangements which promote ‘flexibility’ in how hours of work are organized (e.g. flexitime, annualized hours schemes, etc.) will each be reviewed in turn. For each set of cases, the chapter will analyse the working time practices which have emerged; the business objectives and strategies which underlie these practices; and the extent to which these working time practices fit, or do not fit, with workers’ needs and preferences regarding their working time. In addition, some innovative working time arrangements that explicitly attempt to balance enterprises’ business objectives with the needs and preferences of different types of workers will be examined. Finally, the last section of this chapter will summarize the key findings from the enterprise case studies reviewed and then draw out the most significant implications for creating enterprise policies that offer the possibility of simultaneously meeting both employers’ business objectives and workers’ needs and preferences.

Trends in work organization in the service and information economy Changes in the economy shaping changes in work organization In the last two decades, across the industrialized world there has been a marked trend away from the traditional systems of work organization based on mass production that emerged in the industrial era, and towards a variety of different approaches which have been broadly characterized as ‘flexible’ forms of work organization. While it is difficult to separate out the various causes of these changes in work organization, it appears that several factors have played a particularly important role. First, there has clearly been an intensification in competitive pressures in many economic sectors as a result of the increasingly global nature of competition. This intensification of competition has pushed firms to put greater emphasis on both innovation to bring new and improved products and services to market as quickly as possible, as well as a renewed drive to reduce unit costs of production. It also appears that this intensification of competition is continuing to increase, as illustrated by the expansion of the Internet, which has greatly increased the transparency of market transactions, allowing buyers to more easily seek out the ‘best deal’ and switch suppliers with the click of a mouse (Reich 2000). Under these circumstances, there is continuous pressure on firms to cut costs and increase quality at the same time, leading to efforts to maximize the utilization of fixed capital by, for example, increasing hours of operation and to minimize the number of employees required to produce any given amount of output. As a result, firms have sought various forms of flexibility to attempt to more exactly match both the numbers of workers and their functional skills with production requirements at any given time.

Working time in the enterprise 149 A second important factor driving changes in work organization is changes in market demand. More specifically, there has been a diversification in consumer tastes brought about, in part, by saturation in the many markets for mass-produced products in the industrialized countries and also by an increase in per-capita income in these countries (Piore and Sabel 1984; Phillimore 1989). This diversification in consumer tastes has led firms to focus on developing customized products and services designed to capture specialized ‘niches’ in particular markets. The need to produce a variety of products or services, each in relatively small volumes targeted to different customer groups, puts a premium on being able to rapidly switch products and even product lines in response to changes in customer buying habits. This environment puts a premium on flexible work organization as well – particularly those forms that provide the flexibility necessary to quickly redeploy workers from one product/service line to another as customer requirements change. A third factor is advances in microcomputer and communications technologies, which have enabled companies to re-engineer their production and service delivery processes in ways that successfully exploit the opportunities presented by more globalized markets and the diversification in consumer tastes and preferences. For example, the customer service functions of an enterprise, such as a bank or credit card company, no longer need to be handled in the same establishment, or even in the same country; these functions can be performed by call centres located nearly anywhere in the world. Rather than being a cause of the changes in work organization which have occurred, however, technology appears to be more of an ‘enabling’ factor – that is, the availability of these new technologies has made it possible for firms to more readily adapt their work organization to changes in their competitive environment (Milkman 1998). Yet another factor which has impacted the organization of work is women’s increasing rates of participation in the labour force. On the supply side, as many women shoulder a disproportionate share of responsibilities for household work, including care-related tasks (e.g. child-care and elder care), the traditional organization of work into units called ‘full-time jobs’ has not been very well suited to them. On the demand side, many firms found that there were advantages to restructuring work tasks in ways that would allow them to attract more women into their workforces – particularly women with family responsibilities (and especially those with young children) who might not otherwise have entered paid employment. The dramatic increase in part-time work in many of the industrialized countries, particularly in the European Union and Japan (see Lee, Chapter 2, for more information on this trend), appears to be directly linked to women’s entry into the labour market: 71.5 per cent of all part-time workers in industrialized countries are women (OECD 2001). At the same time, women’s increasing labour force participation also affects the timing and pattern of consumption; for example, the

150

Jon C. Messenger

fact of women working during daytime hours has generated an increased demand for various types of services (such as ‘take out’ food) and extended shop opening hours in the evening and on weekends (Presser 1999). Finally, there is substantial evidence that individual enterprises themselves have been asserting more influence in shaping employment practices, including changes in work organization. That is, even after taking into account the other factors mentioned above, individual firms appear to have a substantial and increasing amount of room for manoeuvre in determining what sorts of industrial relations and human resource management policies they will adopt. Given the full or partial deregulation of many industries in a number of industrialized countries during the 1980s and 1990s (e.g. telecommunications, banking), this situation is perhaps not terribly surprising, but it is important nonetheless. Obviously, the legal and regulatory framework regarding working time (along with other economic and social regulation) that exists in a particular country has an important influence on the degree to which enterprises are free to structure their organizational policies as they see fit. However, the general trend across the industrialized world is clearly in the direction of decentralization of economic decision-making to the level of individual firms (including collective bargaining), providing firms with more room to manoeuvre within any given regulatory framework. This factor would account for the enormous degree of variation in the industrial relations and human resource management policies across firms – even among firms in the same country and in the same industry. For example, one international study of changes in employment relations in 11 industrialized countries concluded that: In all the countries covered in this study, the individual enterprise has emerged as an increasingly important locus for strategy and decisionmaking on human resources and industrial relations. This implies that managers (sometimes in collaboration with local unions or works councils) have been the driving force for introducing change in employment practices in recent years. (Locke et al. 1995: 144, italics added) It must be noted that these are but a few of the most significant forces which appear to be driving the changes in work organization that have been observed in the past two decades. Certainly, there are other factors which have contributed to these changes as well, for example, public policies in many EU member states that are designed to promote the use of part-time work (see McCann, Chapter 1, for a further discussion of this point).

Working time in the enterprise 151 Overall trends in work organization The forces outlined above have had an enormous influence on developments in work organization which have occurred in the past three decades. Prior to the 1970s, the traditional approach to work organization – known as ‘Taylorism’ or ‘Fordism’ – was the dominant form virtually everywhere in the industrialized world. At its core, Taylorism/Fordism was about finding the ‘one best way’ to perform any particular task, finding the ‘best worker’ to perform that particular task, and then having that worker perform that particular task again and again. Each task was specialized and simplified as much as possible, resulting in a highly specialized division of labour with very narrow job classifications. In short, this ‘mass production approach’ was about developing standardized methods for the production of particular goods (or services) and then creating the bureaucratic infrastructure necessary to implement these methods in a manner designed to maximize the output of those goods or services. However, beginning in the 1970s, enterprises in a few countries began to experiment with alternative methods of work organization designed to promote more ‘flexible’ methods of production – a trend which accelerated and expanded into additional countries during the 1980s and 1990s. Perhaps the best-known articulation of this alternative paradigm of work organization was the concept of ‘flexible specialization’ presented in the book, The Second Industrial Divide (Piore and Sabel 1984). The basic idea of flexible specialization is, through techniques such as the use of computer-assisted design and manufacturing, to produce smaller batches of customized products designed to meet the needs of specialized market segments. Unlike traditional systems of mass production, the product lines in flexible specialization can be changed rapidly in response to changing product demand. Building on this concept, Phillimore (1989: 81) provides a comparison of flexible specialization and Taylorism/Fordism. Key differences identified include not only differences in the production concepts and the products produced, but also differences in work processes and workers’ skill requirements (e.g. ‘open-ended tasks’ vs. ‘fragmented and standardized tasks’); differences in relations with suppliers, (e.g. ‘stocks arrive “just-in-time” ’ vs. ‘stocks held “just in case” ’); and differences in organization and management (e.g. ‘managerial hierarchies’ in a ‘multi-divisional corporation’ vs. ‘flatter hierarchies’ with ‘decentralized production’). In describing the characteristics of what constitutes a production system based on flexible specialization, Atkinson and his colleagues (Atkinson 1984; Atkinson and Meager 1986) developed their now well-known model of the ‘flexible firm’. The ‘flexible firm’ model presupposes that firms seek three different types of flexibility: functional flexibility, numerical flexibility and financial flexibility. According to Atkinson (1984), ‘functional flexibility is sought so that employees can be redeployed quickly and easily

152

Jon C. Messenger

between tasks’. This implies that workers need to be multi-skilled, so that they can switch tasks – and even jobs – as business needs dictate. By contrast, ‘numerical flexibility is sought so that headcount can be quickly and easily increased or decreased in line with even short term changes in the demand for labour’ (ibid.). Finally, financial flexibility can promote both functional and numerical flexibility by linking employees’ pay more closely with prevailing market pay rates and especially by linking pay to performance. The result is that the ‘flexible firm’s’ workforce becomes divided into essentially two groups: ‘core’ and ‘peripheral’ workers. As Atkinson describes: The new structure involves the break-up of the labour force into increasingly peripheral, and therefore numerically flexible groups of workers, clustered about a numerically stable core group which will conduct the organization’s firm-specific activities. At the core, the emphasis is on functional flexibility; shifting to the periphery, numerical flexibility becomes more important. (ibid.: 29)1 A number of authors have built on the concept of the ‘flexible firm’. Particularly noteworthy is the work of the European Foundation for Living and Working Conditions, which explicitly added a fourth type of flexibility to the three types identified by Atkinson – ‘temporal flexibility’, that is, flexibility in working time (see, for example, Goudswaard and de Nanteuil 2000). Like numerical flexibility, temporal or working time flexibility is also designed to adjust the available labour force to fluctuations in market demand. However, with temporal flexibility these types of adjustments are made by varying the number of hours worked by a (generally) fixed number of workers over time, rather than by increasing or decreasing the number of workers employed by the firm. It is not clear to what extent enterprises have actually adopted the ‘flexible firm’ model. A number of researchers (Boreham 1992; Milkman 1998; Phillimore 1989) have concluded that, while aspects of flexible specialization have been adopted by some enterprises, the vast majority of enterprises continue to use some variant of the Fordist production model (often termed ‘Neo-Fordism’), modified to allow for the production of more customized products or the delivery of more customized services. In addition, while there has clearly been an overall trend towards fewer layers of management, many organizations nevertheless continue on with traditional command-and-control management practices, often aided by computer software that allows for more comprehensive tracking of processes and outputs. Thus, while important changes in work organization have certainly occurred in many enterprises, it is important not to get carried away by the often-sweeping rhetoric about those changes:

Working time in the enterprise 153 [T]his is not the ‘paradigm shift’ beloved of academics looking for a conceptual peg on which to launch their latest publication. Continuity is as pervasive as change . . . There is, however, no single or simple future workplace. Variation by firm, sector, or country co-exists with powerful structural tendencies to standardize through businessdefined ‘best practice’. (Warhurst and Thompson 1998: 19) Despite the ongoing debates on the nature and scope of changes in work organization, it is clear that, under the competitive pressures in the global economy, firms are increasingly seeking methods of organizing work that enable them to respond to their customers’ demands more rapidly and with higher quality goods and services. That search is also leading them to seek ways of organizing working time more ‘flexibly’. In general, this focus on working time ‘flexibility’ means that hours of work are becoming less fixed and more responsive to fluctuations in market requirements. Enterprises are seeking to fine-tune working time so that workers are available only as and when they are needed, with increasing use of working time arrangements that allow hours of work and the scheduling of those hours to vary substantially on a weekly and even a daily basis. The result is a diverse array of working time patterns that vary not only across countries and across industrial sectors, but also within companies in the same industry and even across workers in the same company depending upon the nature of the work performed. It is to these changes in working time in individual enterprises that we now turn.

Actual working time patterns: an analysis of selected enterprise cases In considering changes in working time patterns at the enterprise level, it is useful to begin by briefly recalling the institutional context in which these changes occur. First, there appears to be a link between the different types of institutional structures for the regulation of working time – that is, the working time regime – in a country and the actual working time patterns which emerge in that country. Specifically, there appear to be three basic types of working time regimes that exist in the industrialized countries: (1) state-regulated working time regimes, which are primarily based on national legislation; (2) negotiated working time regimes, which are based on negotiation and collective agreements between the social partners (employers and workers) at the national, sectoral, and/or enterprise levels; and (3) market-based working time regimes, which, while they may involve laws and collective agreements, are primarily the result of contracts (either implicit or explicit) between specific employers and individual workers (see Anxo and O’Reilly 2002; McCann, Chapter 1). Although there are considerable variations at the national level, in

154

Jon C. Messenger

general, state-regulated working time regimes (e.g. France) tend to produce a concentration in hours of work around the ‘normal’ hours of work established in statute and relatively small proportions of workers who are working either very long hours (more than 48 hours per week) or very short hours (less than 20 hours per week). By contrast, negotiated working time regimes tend to show more variation in the average hours of work, with a concentration in hours around ‘normal’ hours of work where collective agreements covering working time are negotiated primarily at the national level (e.g. Denmark, Sweden), and multiple clusters of working hours where collective agreements are typically negotiated at the sectoral and/or enterprise levels (e.g. Germany, the Netherlands). Finally, market-based working time regimes (e.g. Australia, the USA, the UK, and to some degree, Japan) tend to produce a wide dispersion in hours worked across individuals, with relatively high proportions of workers both with very long hours of work and also relatively high proportions of workers with very short hours (also called ‘marginal’ part-time work). These working time regimes, then, appear to serve a mediating role: that is, they interact with the driving forces which are shaping changes in work organization globally and ‘filter’ these forces to produce changes in the length of working hours and also changes in the arrangement of those hours. However, it must be emphasized that interactions between the driving forces shaping work organization internationally and the working time regimes of each country tend to be idiosyncratic, based on the competitive realities in different economic sectors and the particular mix of industrial relations and human resource management strategies that each enterprise adopts in response to those realities. For example, in capital-intensive industries such as automobile manufacturing, labour costs are typically much less important than the costs of plant and equipment; thus, the focus of firms in this industry is on increasing capital utilization, often by extending operating hours (see, for example, Anxo et al. 1995). However, in staff-intensive service sector industries, such as retail trade, labour costs are the single largest category of operating costs, and as a result, ‘the sector is dominated by strategies for reducing staffing costs’ (OECD 1995: 34). Obviously, the different realities in these different sectors may lead to very different types of working time arrangements. Moreover, the firms’ business objectives, and the business strategies used to pursue those objectives, may lead even firms in the same country and the same sector to utilize very different working time arrangements. For example, firms in the retail trade industry who choose to compete primarily on cost will have very different industrial relations and human resource manangement strategies – including different working time patterns – than firms in the same country and the same sector whose business objectives are focused on providing high levels of customer service. The evolution of these sectoral and firm-specific differences in actual working time patterns is likely to be even more pronounced in countries with market-based

Working time in the enterprise 155 working time regimes, because of the lack of institutional constraints on working time due to both the limited nature of relevant statutes and the absence of collective bargaining in many sectors and in many individual enterprises. Thus, it is imperative that we look not only at what is happening to working time at the national level in different countries, but also at the realities of working time ‘on the ground’ in individual enterprises. While it would be impossible to fully capture the diversity in working time patterns which exists in each country, let alone in each industrial sector of each country, it is nonetheless instructive to consider selected enterprise cases as a means of illustrating those actual working time patterns which are emerging in these different environments. This section builds on the discussion of working time trends in industrialized countries at the national level presented in Chapter 2 to illustrate how those macro-level trends are playing out at the micro-level: that is, in particular industries and especially in individual enterprises. At the same time, this section will also consider how well the weekly hours of work and the way in which these hours are arranged or scheduled – that is, the working time arrangements – fit with workers’ expressed needs and preferences. The case studies will draw examples from a variety of enterprises in different industries – particularly firms in the expanding service sector – but also cases involving manufacturing firms and the health care industry. This case study analysis will focus on three broad categories of working time which are linked to enterprises’ drive for increased flexibility, and have important implications for restructuring working time to account for workers’ needs and preferences. These three categories are as follows: •

•

Long hours of work, which will be defined as working an average of more than 48 hours per week.2 Long hours of work, typically involving overtime in excess of the standard workweek in the country, have long been an important method for employers to secure flexibility in working time, but also have obvious implications for workers, including health, safety and the ability to balance their work and non-work commitments. Short hours of work, commonly referred to as part-time work, which will be defined as working less than either 30 or 35 hours per week depending on the country.3 Part-time work is a particularly important area to address in the case studies because it provides both an important flexibility tool for employers and is also being used by many workers, primarily women, as a method of attempting to attain a balance between their work and their family responsibilities. Within the category of part-time work, the case studies will explicitly consider ‘standard’ part-time work in which the normal hours of work are ‘substantial’ (20 hours per week or more) compared with ‘marginal’ forms of part-time work (definitions of this concept may vary, but ‘marginal’

156

•

Jon C. Messenger part-time hours are usually defined as less than 20 per week), as well as the extent to which part-time work is voluntarily chosen or instead effectively imposed by a lack of full-time alternatives. The arrangement of working time, that is, alternative methods for scheduling a given amount of working hours. Working time arrangements can include a nearly infinite variety of shift patterns – from traditional, ‘standard’ daytime, weekday shifts or night shifts to rotating shifts (day, evening, night) to weekend working. However, this section will focus on certain ‘flexible’ types of arrangements for scheduling work that allow working time to vary over the day, week, month, or even an entire year (e.g. flexi-time programmes, annualized hours schemes) that have emerged in response to the needs of enterprises, but which may also offer opportunities for meeting workers’ preferences regarding the timing of their hours of work.

Long hours of work Long hours of work refer to hours that are typically longer than the established ‘normal’ hours of work in a particular country. As was discussed by Lee in Chapter 2, a number of industrialized countries have long working hours, or more precisely, these countries have major segments of their workforces that are working long hours. In general, those industrialized countries with highly decentralized, ‘market-based’ working time regimes, such as Australia, the UK, the USA, and (to some degree) Japan, are the ones which tend to have substantial portions of their workforces that are working very long hours. However, even in those countries in which working time is more tightly regulated – whether by statute or via collective agreements or both – long hours can still exist among some groups of workers. Long hours of work is one of the areas of working time in which workers have expressed strong preferences. Although these preferences do vary across countries, in general, the longer a worker’s hours of work, the more likely he or she will express a preference for reduced working time (see Lee, Chapter 2 and Fagan, Chapter 4, for an in-depth discussion of this issue). Thus, when considering the specific enterprise cases outlined below, it is important to keep in mind that, generally speaking, workers who are working 50 hours per week or more are likely to prefer a reduction in their working time. Moreover, the longer the hours that workers are currently working, the greater a reduction in working time they would typically prefer. The case studies reviewed in this section will illustrate how employers – even with flexible forms of arranging working time becoming more widespread – are continuing to rely on overtime as an important mechanism for extending operating hours in response to fluctuations in market demand. For example, the case of one company in the UK selling prepared and

Working time in the enterprise 157 frozen foods illustrates how unpredictable changes in market demand can lead to extensive use of overtime – and hence long hours of work – in a country with a market-based working time regime (Purcell et al. 1999). Frufood (this is a fictitious name) uses a fairly traditional fixed-shift system with three shifts of mostly full-time workers, which operates from 6:00–14:00, 14:00–22:00 and 22:00–6:00. The company’s primary customers are supermarkets, who often change their orders dramatically (on average 20–30 per cent) on very short notice – often only 24 hours in advance. Because many of the foods they handle are perishable, it is not possible to build up large inventories to respond to these volatile swings in demand. As a result, employees are routinely required to work 10 hours of overtime per week on top of their standard 40-hour workweek. In addition, employees typically receive little or no advance notice when they have to work overtime. The firm could demand large amounts of overtime on short notice because in the UK, with a market-based working time regime, there are no legal requirements (either in national statutes or in collective agreements) regarding the amount of advance notice required for overtime or indeed the amount of overtime that could be worked.4 Further, the relatively constant, yet unpredictable nature of the demand in this industry (as opposed to industries with more predictable peaks and troughs in demand) puts a premium on being able to respond immediately to changes in customer orders, and hence on the use of unplanned overtime. Thus, this case illustrates how unpredictable changes in customer demand can result in extensive recourse to overtime working – resulting in what is effectively a 50-hour workweek for most of the firm’s employees. While this was not a very positive result for workers, it also had negative effects for the enterprise as well, as both staff turnover and absenteeism were high. Interestingly, however, there is one group of workers at Frufood who seem to be largely satisfied with their working time: a small group of parttime workers, mostly mothers, who are working on the so-called ‘Mums’ shift’ from 10:00 to 15:00. According to this study (ibid.), this particular shift pattern was created specifically to attract mothers with children, with fixed hours during the school day and no required overtime. Despite the situation for the company’s workforce as a whole, turnover and absenteeism among this group was virtually nil. This result shows that creating working time patterns that meet the needs of particular workers can help to produce successful outcomes for both enterprises and workers. This is an important point to which we will return later in this chapter. Australia, along with the USA and Japan, has some of the longest hours of work in the industrialized world, with 20 per cent of workers working an average of 50 hours per week or more (see Lee, Chapter 2). As discussed earlier in this volume, an important part of the explanation for this trend is the strong move towards the deregulation of working time in Australia, with a shift away from the previous system – a regulatory structure with

158

Jon C. Messenger

industry-based ‘awards’ specifying normal hours of work, limitations on total hours, premium rates of pay for overtime, etc. – and towards a decentralized, market-based regime based on enterprise-level bargaining and particularly the extensive use of individual contracts. And the mining industry has the longest average hours of work of any industry in Australia – see the discussion in Box 5.1. The Australian mining industry also recorded the largest increase in hours of work of any industry in the country between 1989 and 1997 (Heiler and Pickersgill 2001; Heiler et al. 2000). Long hours of work, however, are not limited only to those industrialized countries with market-based working time regimes. While, as we saw in Chapter 2, average actual weekly hours of work in state-regulated and negotiated working time regimes are typically much closer to ‘normal’ hours, there is nonetheless a prevalence of long hours across the industrialized countries in certain industries, for example, information technology, and in certain occupational groups, such as managerial and professional staff (EIROnline 1997; EIROnline 2001d; TUC 2002; Jacobs and Gerson 2000; USDOL 2000; Fagan, Chapter 4). The problem of long hours appears to be particularly acute in certain sectors of the economy, and the information technology (IT) industry provides a particularly clear example of this reality. There are a number of characteristics of IT work that encourage long hours. One recent German study identified four ‘driving forces’ which result in long hours of work for IT workers (Voss-Dahm 2001). These driving forces include tight time and budget constraints resulting from numerical performance targets for IT projects; the large degree of uncertainty regarding the type and volume of work that will need to be done on any given project; the necessity for a high level of collaboration among IT team members, which requires a substantial amount of time for coordination; and the high level of motivation of the generally highly educated IT workforce, which leads them to disregard regulations on working time. A study on working time in the IT industry in the Netherlands – a country well known for its relatively short hours of work – reached a similar conclusion: The reliance on intrinsic motivation, in most cases supplemented with a performance-related wage system, together with de-standardization of working time and working hours, might translate into a rather ‘time-greedy’ employment relation in which it is very difficult to guard the lines between working time and personal time. (Plantenga and Remery 2001: 48) An interesting case of long hours of work in the IT sector comes from Denmark, a country with a negotiated working time regime and a collectively agreed 37-hour working week. A recent study revealed that in Denmark, engineers and information technology experts routinely work

Working time in the enterprise 159 Box 5.1 Long hours of work in the Australian mining industry Hours of work in the Australian mining industry are the longest of any industry in the country: an average of 51.7 hours per week, compared to an ‘all-industry’ average of 44.3 hours (Heiler and Pickersgill 2001: 23). Why are the hours of work in this industry so high? First, the move towards a more decentralized or ‘market-based’ working time regime has been an important factor. Second, there has been a marked increase in cost pressures on the mining industry due to lower commodity market prices for minerals, driving firms to seek ways of lowering their production costs. A third, particularly important, factor in explaining the long hours of work in the Australian mining industry is the move towards 12hour shifts and also multiple consecutive shifts in this industry. This change was actively sought by management as a means of extending operating hours to allow mines to engage in continuous production, thereby enabling them to reduce their unit costs of production. In theory, such longer daily shifts should not necessarily lead to long hours of work, but should be compensated by ‘compressing’ the same number of working hours into fewer shifts. However, the mining sites that have implemented 12-hour shifts are more likely to have long average hours of work; for example, one report (ILO 2002b) found that 49 per cent of the workers in the 12-hour shift sites worked over 49 hours per week, compared with just 31 per cent for all mining sites. Although it is difficult to determine exactly why the move to 12hour shifts produced this result, it appears that one key factor is that the number of shifts worked have not decreased sufficiently to compensate for the longer shifts. For example, one typical shift pattern is 4 ⫻ 12-hour shifts, followed by 4 days off (48 hours in 8 days), and in the non-coal mining segment of the industry, 60 per cent of mines use rosters that require more than 7 shifts to be worked in a row (ILO 2002b). In addition to these long regular hours, about 70 per cent of mining industry workers also work overtime either regularly or occasionally. In those sites with 12-hour shifts (which, as a rule, should not be extended by overtime), this typically means working not a few extra hours on a shift, but rather working additional shifts (Heiler and Pickersgill 2001). All these factors have combined to create very long hours of work in this industry, which has raised concerns about possible ill effects on workers’ health and safety, as well as high labour turnover among experienced workers in some sites.

160

Jon C. Messenger

50 hours or more per week; in the case of Danish engineers, for example, only 1 in 10 works the standard 37-hour week, and 1 in 4 works over 44 hours per week (EIROnline 2001d). In their paper, entitled ‘Work without limits’, Christensen and Pleman (2001) describe the case of two individuals who worked for a Danish firm constructing Internet web pages, which was based on interviews conducted by Denmark’s national radio service. The two individuals, called ‘M’ and ‘P’ in the paper, described the pressures of work in this sector; for example, there is a constant push to minimize the time allocated to complete work tasks, which often results in workers having insufficient time to complete their assigned tasks. This situation, in turn, means that there is a constant need to work overtime in order to complete tasks within the given timeframes. For ‘M’, the situation came to a head when he was injured (which was not work-related), but, due to his workload, he tried to continue working and ultimately collapsed physically. After his collapse, ‘M’ decided that he had to cut back his hours of work, and so he started going home at 5 p.m. However, this often meant that he was unable to complete coding his assigned part of the software, which produced tensions with his colleagues who needed his parts to complete their own work. As a result, his colleagues nicknamed him the ‘5 o’clock man’ and isolated him socially from the rest of the web page developers. According to the paper, ‘M’ was told to increase his output (i.e. to generate more ‘billable hours’), and when he refused to do so, he was fired. His colleague (‘P’) continued to work for the same firm, but suffered physical and psychological health problems, and eventually was diagnosed with chronic stress disorder – a problem frequently associated with long hours of work (see, for example, Spurgeon 2003). The Danish case provides a vivid, though perhaps extreme, illustration of the factors which appear to cause long hours of work in the IT industry. Here we see many of the factors described in Voss-Dahm (2001) at play. For example, this case clearly illustrates the tight time and fiscal constraints involved in project work in the IT sector and how these constraints can result in pressure for longer hours of work. It appears that these constraints are exacerbated by the uncertainty regarding the time actually needed to complete various tasks, which makes it easier to set unrealistic timeframes, and which in turn, results in even greater work pressures. In addition, the interdependence of the work and the high level of IT workers’ motivation show through in ‘M’s’ interaction with his colleagues, in which they ended up socially ostracizing him for failing to produce his part of the software on time because they could not finish their own work without it. Of course, meeting these rather unrealistic performance expectations would require additional hours of work, and one could easily see how meeting such expectations could result in more unrealistic task demands in the future, causing an upward spiral in working time. The problem of long hours is also particularly acute in certain occupations. Managerial and professional staff across countries tend to work

Working time in the enterprise 161 particularly long hours. In the United States, for example, in 1999 nearly 30 per cent of all managers worked 49 hours a week or more, and among male managers, this number increases to about 40 per cent (USDOL 2000). Even in the European Union, long hours of work are relatively common among managers and many categories of professional staff. Though this situation would appear to be contrary to the EU Working Time Directive, which specifies maximum hours of work in the EU member states, in fact the Directive specifically provides for a derogation (an exception) which allows member states to exclude managerial staff, and most EU states have adopted this exception. For example, one study found that male professionals across the EU work 13 per cent longer than other males; the gap is even larger for females, although in this case it is primarily due to the relatively low rate of part-time work among female managers, compared with high rates of part-time work among women in general (EIROnline 1997). Within the EU, the long hours phenomenon is particularly prevalent in the UK, which is not particularly surprising since it is the only country in the EU which exhibits a working time regime that is primarily market-based. One recent study found that managerial and professional occupations combined accounted for well over half (58 per cent) of all employees working more than 48 hours per week (TUC 2002). Box 5.2 provides a further discussion of the phenomenon of long hours working among managers and professionals in one firm in the UK. Thus, it appears that there are a number of circumstances that are leading to longer hours of work. However, the driving force overall appears to be firms’ search for competitive advantage by, for example, implementing working time structures that provide for rapid response to changes in customer demands, as in the case of Frufood, or which focus on extending operating hours to reduce unit costs of production, as appears to be the situation that exists in the Australian mining industry. In both of these cases, the use of paid overtime as and when needed (even on short notice) is relatively straightforward in the market-based working time regimes which exist in Australia and the UK. In addition, while national working time regulations may have an effect on restricting hours of work for employees in general, it appears that they have less of an impact in restricting excessively long hours in certain sectors (e.g. the IT industry), and certain occupations, such as managerial and professional staff, which tend to be exempted from these regulations. The nature of professional and managerial work – which is focused on task completion rather than the number of hours worked and remunerated as a salary rather than an hourly wage – is also an important factor in explaining why overtime (albeit unpaid for salaried employees) is the most typical way that firms employing these types of workers respond to fluctuations in demand. At the same time, it is important to recall that such long hours of work are clearly out of sync with the working time preferences of most

162

Jon C. Messenger

Box 5.2 Long hours of work for professionals and managers Boots is a major chain of department stores located in the UK. Contractual weekly working hours at Boots are only 37.5 hours per week, but the actual working time for several groups of professional and managerial staff is much longer. For example, the engineering support staff at the store have contracts that include overtime, resulting in a workweek that typically exceeds 48 hours – the maximum under the UK working time regulations, which are based on the EU Working Time Directive. Because these workers have an annualized hours contract that averages working time over a long period (52 weeks), however, it is still possible to hold weekly hours of work to less than 48 hours on average over the entire period. Managers at Boots also reflect the general pattern of long hours of work for managerial staff. Boots is just one of many UK companies which exempts higher-level middle managers from coverage under the working time law by using an ‘opt-out’ provision in the EU Working Time Directive. This provision allows firms to exclude managerial staff from the Directive’s requirements. Likewise, management trainees at Boots are required, as a condition of employment, to sign contracts that contain ‘opt-out’ clauses whereby they voluntarily agree to be exempted from the 48-hour working time limit (EIRR 2000).

workers, and it is those workers that are working the longest hours who are the ones most likely to prefer a substantial reduction in their weekly working time. Short hours: part-time work Short hours, more commonly known as ‘part-time work’, refer to hours of work that are typically shorter than the established ‘normal’ hours of work in a country. The definition of part-time work varies from country to country, but it is most commonly less than 30 hours per week (see endnote three). Using this definition, 16.3 per cent of all workers across the EU work part-time, although the figures vary considerably across member states – from a high of 32.1 per cent in the Netherlands to a low of 5.4 per cent in Greece (OECD 2002: 18–19). The comparably defined rate of part-time work is somewhat lower in the United States, 12.8 per cent, but considerably higher in Australia and Japan at 26.2 per cent and 23.1 per cent, respectively (OECD 2002: 18–19).5 It is interesting to note that, for Japan, comparable figures are particularly difficult to derive because (depending on the data source) workers may be classified as

Working time in the enterprise 163 having a ‘part-time’ employment status regardless of the number of hours worked.6 There are several important aspects of part-time work that should be emphasized in order to provide some context for understanding the enterprise cases presented in this section. First and foremost, it is impossible to discuss part-time work without considering its gender dimension. The vast majority of part-time workers are female: in the EU, for example, 77.1 per cent of all part-time workers are women (Eurostat 2001). As noted previously, women generally have a disproportionate share of the responsibilities for household and care-related tasks, and therefore the traditional organization of work into units called ‘full-time jobs’ was not well suited to them. The dramatic increase in part-time work in many industrialized countries, particularly in the EU and Japan, testifies to its potential usefulness as a vehicle for assisting women to facilitate a balance between paid work and family responsiblities, as well as to employers’ interest in using part-time work as a tool for tapping into the female workforce (see Fagan, Chapter 4, for a comprehensive discussion of the gender dimension of working time). Second, there are important trade-offs to be considered regarding parttime work, and the primary issue here concerns the quality of these jobs. Available evidence suggests that the quality of part-time jobs varies dramatically, both across countries and among different firms in each country. In general, though, part-time workers tend to earn less, even on an hourly basis, than comparable full-time workers and have lower levels of nonwage benefits (e.g. access to pension schemes and social security provisions, or in the USA, to company-provided health insurance). This situation holds even after controlling for a variety of other factors such as industry, occupation and differences in individual characteristics (see, for example, Polivka et al. 2000; Rubery 1998). Third, there are important differences within the sphere of part-time work. In particular, there appear to be two particularly important factors that influence the quality of work in part-time jobs. The first is the number of hours worked in the part-time job. This is the concept of ‘marginal’ part-time work, which is typically defined as hours of work less than 20 per week. It is those workers who are working in ‘marginal’ part-time jobs who have the largest ‘working time deficit’ – the ‘gap’ between their actual hours of work and their preferred hours (an issue which is extensively discussed by Lee, Chapter 2) and are also the most likely to prefer an increase in their working time (an issue which is extensively discussed by Fagan, Chapter 4). The other important difference among part-timers is whether or not the individual is working part-time on a voluntary basis, which is actually a fairly direct measure of workers’ preferences regarding part-time work. In this case, the key issue is not the number of hours that are worked, but whether or not the worker is working in a part-time job by choice – that is, they prefer to work part-time – or, rather, because they

164

Jon C. Messenger

were unable to find a full-time position – i.e. they would prefer not to work part-time.7 There is evidence that a substantial proportion of the part-time work in many industrialized countries is involuntary (see Lee, Chapter 2, for a discussion of this issue). Finally, it is important to consider why certain employers may prefer to use part-time workers rather than full-timers. According to Rubery (1998), these factors include cost and price pressures, the need for flexible working time schedules including covering unsocial hours, and the individual characteristics of workers (e.g. gender). Similarly, O’Reilly (1996) emphasizes that using part-time workers is a cost-effective means of extending operating hours or the availability of services because part-time workers, unlike full-timers, are not likely to receive wage premiums for working additional hours above their normal hours or for working at ‘off hours’, such as evenings, nights and weekends (often referred to in Europe as ‘unsocial hours’). For example, part-timers in many countries can be hired to work weekend-only shift schedules without being paid premium wages; such schedules may even be considered ‘attractive’ by some workers with family responsibilities (primarily women). Moreover, firms in countries that do not require the equal treatment of full-time and parttime workers, such as the USA, can save money on staffing costs, particularly for non-wage benefits, by hiring part-timers instead of full-timers (Houseman 2001). While part-time work is widely used by employers across the industrialized world, however, firms in certain sectors are more likely to make use of part-time work. This is particularly true for enterprises in the service industries, in which (as discussed earlier in this chapter) labour costs are the single largest category of firms’ operating costs. In the EU, for example, part-time work is especially prevalent in three sectors – wholesale and retail trade; hotels and restaurants; and other services (Bodin 2001). Keeping these important aspects of part-time work in mind, the case studies in this section will illustrate how firms, particularly those in the service industries, are reducing costs and extending shop opening hours through an ever-greater use of part-time work. Both the positive and negative features of part-time work will be explored, particularly those two important factors that appear to be directly linked to workers’ preferences for part-time work: •

•

Regular part-time work, which will be defined broadly, to include a ‘substantial’ number of hours of work ranging from 20 up to a maximum of 34 hours per week,8 compared with more ‘marginal’ forms of part-time work (which, for purposes of consistency with earlier chapters, is defined as less than 20 hours a week). Part-time work which is voluntarily sought vs. ‘involuntary’ part-time work, i.e. workers who are working part-time because they were unable to obtain a full-time job.

Working time in the enterprise 165 It is useful to begin by looking at a fairly typical example of regular parttime work – i.e. part-time with a ‘substantial’ number of hours. The case of the Swedish retail trade sector (Anxo and Nyman 2001) provides an illustrative example of how part-time work with ‘substantial’ hours is structured in a country where there is a relatively high proportion of part-timers – 14.5 per cent of total employment (OECD 2001). In Sweden, the vast majority of part-time jobs average between 20 and 34 hours per week (see Lee, Chapter 2). At the same time, shop opening hours in Sweden have been unregulated for many years, and as a result, the hours of operation in the Swedish retail trade are relatively long. Working time in the retail trade in Sweden, as in most sectors, is largely determined via the negotiation of collective agreements between trade unions and employer organizations, which result in national-level industry agreements with some flexibility for further negotiations at the sectoral and enterprise levels. Normal hours of work for full-time employees in the industry are 40 hours per week, and actual hours of work average 38.5 per week; however, a much higher proportion of employees (45 per cent) are part-time workers than in the country as a whole, and 90 per cent of those part-timers are female (Anxo and Nyman 2001). The wages and working conditions of part-time employees, like full-timers, are covered under the collective agreement for the retail sector, providing a strong and sectorspecific regulatory framework for the retail industry. According to Anxo and Nyman (2001), this high level of reliance on part-timers in the Swedish retail sector is linked to the relatively long shop opening hours in this sector, which may extend until 22:00 hours on weekdays, and which often include Sunday opening hours as well. For example, in the two specific enterprises that they use in their discussion of the Swedish retail sector, a large grocery store chain (ICA Maxi) and a fashion store chain (Kapp Ahl), the shop opening hours are 72 and 55 hours per week, respectively. The need for workers to cover these relatively long opening hours – far longer than the weekly working time of individual workers, but not necessarily long enough to justify multiple full-time shifts – makes the use of part-time shifts attractive. However, the authors also cite other employer objectives, including the reduction of overtime costs and covering for what they term ‘legal absenteeism,’ including not only sick leave and annual leave, but a variety of other types of leave (e.g. parental leave, educational leave) which are mandated under Swedish law. Contrast these ‘rather traditional work organisation and working time patterns’, as Anxo and Nyman (2001: 145) call them, with the situation of ‘marginal’ part-time workers in the UK, a country where over half of all part-timers (see Table 2.2) work very short hours. For example, the case of Call Tec (fictitious name) – see Box 5.3 – a UK company operating five call centres, illustrates how the use of part-time shifts combined with other types of ‘flexibilization’, in this case, so-called ‘zero hours contracts’, can result in very short, as well as highly uncertain, hours of work (Purcell et

166

Jon C. Messenger

Box 5.3 Very short hours of work: ‘zero hours contracts’ Call Tec (fictitious name) provides telephone help lines, more commonly known as ‘call centres’. The company is a call centre operator providing market research and other information services for a variety of firms (e.g. telephone ‘hotlines’ for publicity campaigns). The highly uncertain nature of demand in this emerging service ‘industry’, combined with the firm’s business decision to focus initially on winning contracts related to the sales campaigns of other firms, caused much of this company’s work to be based upon shortterm contracts varying from a few weeks to a few months. As a result, the company’s demand for workers was highly variable. To respond to this high level of unpredictability, the company relies on a combination of so-called ‘zero hours contracts’ with parttime shifts. ‘Zero hours contracts’ are a type of working time arrangement in which employers, instead of specifying the number of hours to be worked over a particular period, contract with employees to work only as and when they are needed. This type of arrangement is also referred to as ‘on call work’, since workers work only when they are ‘called in’ by the employer. In this case, the company in question used three weekday shifts (day, evening and night), plus weekend shifts – though a much larger number of the employees worked on the evening shift than on the other shifts. The result of this ‘on call’ working time arrangement was an extraordinarily high level of flexibility to respond to changes in customer demands, but also a very low level of employee commitment and very high turnover among the company’s staff, most of whom were students. More recently, the company has sought to develop longer-term contracts for its services, and is also developing certain human resource management policies (e.g. multiple job categories for telephone operator positions based on skill level) in order to recruit and retain better-qualified employees, such as those with foreign language skills (Purcell et al. 1999).

al. 1999). This type of short hours working, termed ‘on call’ work, involves no fixed hours of work at all: workers only work as and when they are needed by their employer. In market-based working time regimes, there are few if any restrictions on this type of working time arrangement. Although ‘on call’ work only accounts for a very small percentage of employment in any industrialized country, this type of working time arrangement is most common in market-based working time regimes, particularly the USA and the UK, because of the lack of institutional constraints on working time in those countries. While such an arrangement

Working time in the enterprise 167 may suit some types of workers (e.g. students), ‘on call’ working is often unpopular with workers: for example, one recent US study found that nearly half of all ‘on call’ workers in that country would have preferred a job with fixed hours of work (DiNatale 2001). Another country with a relatively high proportion of part-timers who are working very short hours is Germany (52 per cent; see Table 2.2). However, in contrast with the UK, where students make up a substantial proportion of the ‘marginal’ part-time workforce, in Germany, it is primarily female second earners who have ‘marginal’ part-time jobs. One recent study of the food retail sector in Germany (Deiß 1999) found that over three quarters of the workers in German supermarkets are part-time workers, and of these workers, half of them are ‘marginal’ part-timers. Part-time work allows enterprises to better adapt staffing levels in response to customer traffic in stores, since the shorter part-time shifts can be more easily combined to cover periods of peak demand than full-time shifts. These workers also provide the supermarkets with an important numerical ‘flexibility buffer’; they can be brought in quickly when additional workers are needed to respond to unanticipated fluctuations in customer traffic. This means that these workers are often required to work at short notice and to work ‘unsocial hours,’ particularly evenings and weekends (ibid.). Short hours of work are often linked with the phenomenon of ‘involuntary’ part-time work, in which workers who would prefer to work full-time end up working in part-time jobs because full-time jobs are not available.9 Involuntary part-time work has been a particular issue of contention in France, a state-regulated working time regime in which women have a relatively long tradition of participation in the labour market. This tradition has been primarily based on full-time work, with government support in areas such as education and child-care (Daune-Richard 1998). However, in the past 20 years, part-time work in France has more than doubled from 6 per cent to 14 per cent of total employment. This dramatic increase in part-time has emerged in large part as a result of a variety of financial incentives provided by the French government to promote the creation of part-time jobs with the objective of reducing unemployment, which – given the traditionally strong role of the state in regulating both working time and employment – has proven to be an effective approach (EIROnline 2001a). At the same time, however, there has been an increasing proportion of part-time workers with limited skills for whom part-time jobs were the only available alternative to unemployment (Daune-Richard 1998). The result is a high level of ‘involuntary’ part-time work – as much as 44 per cent by some measures – although there has been a modest decline recently (EIROnline 2001a). A case study of a clothing retailer in France (see Box 5.4) provides an illustration of involuntary part-time work in an industry in which part-timers account for over a quarter of the workforce (Jany-Catrice 2001).

168

Jon C. Messenger

Box 5.4 ‘Involuntary’ part-time work in the French retailing industry A few large firms, primarily food retailers, who are also increasingly expanding into other areas of retailing as well, dominate the retail trade sector in France. This sector is also characterized by intense price competition, which as discussed earlier, is an important factor in promoting the use of part-timers. One recent study (Jany-Catrice 2001) examined the case of a large fashion retailer in France that is part of a multinational chain that markets ‘trendy’ clothing at low prices. This particular company also has a strong focus on selfservice. This fashion chain has a largely full-time workforce, but in the case study store, the workforce is largely part-time. 80 per cent of the store’s workforce is female and most employees are also quite young. Very flexible hours of work are a critical component of the firm’s strategy to respond to changes in the workload over time. The typical working time for the sales force in this store is 18.5 hours per week and, with only a few exceptions, workers have ‘permanently changing schedules without a fixed day off’ (ibid: 110–11). This French fashion store also exhibits a very high level of ‘involuntary’ part-time work – even higher than the rate in France as a whole. Interviews found that ‘90 per cent of all part-timers would prefer to work longer hours or even to be hired on a full-time contract’ (ibid.: 112). This result may be an indication that workers are unable to work enough hours to earn an adequate income. Coupled with the low hourly wages that these workers receive, the end result is that the French retail trade sector has one of the highest proportions of working poor in the country.

By contrast, in the USA, it is not short hours per se, nor the involuntary nature of the work (the overwhelming majority of US part-timers voluntarily work part-time), but rather the lower wages and especially the lack of benefits for part-time workers that is the key issue surrounding the quality of part-time work in that country. This is not to say that the issue of the quality of part-time jobs is not important in other industrialized countries; it remains an issue in most of these countries. However, there are certain unique circumstances in the USA that make this problem particularly acute. Specifically, in the USA – unlike most of the industrialized world – health insurance and pension coverage are provided not via the government, but directly by employers. In this situation, employerprovided ‘fringe benefits’ become absolutely essential to workers’ wellbeing. Yet, here is where the difference between part-time and full-time

Working time in the enterprise 169 workers in the USA is the greatest: for example, while 86 per cent of all full-time employees were offered health insurance through their employers, only 35 per cent of part-time workers had an option to obtain employer-provided health insurance (Polivka et al. 2000: 79).10 The expectation that part-timers in the USA are simply not entitled to the same benefits as full-timers is so pervasive that those US firms which do offer benefits such as health insurance to their part-time employees are typically viewed as ‘model’ employers. For example, CIGNA Corp. provides full benefits, including health insurance, to employees working 24 hours a week or more; the company’s justification for this approach is that it reduces staff turnover, thereby lowering their training and recruitment costs (Brown 1998). Nevertheless, part-time work can play an important role in helping some workers to adjust their work schedules to meet their working time preferences. In such cases, part-time work is likely not only to be voluntarily chosen, but actively sought. For example, those workers who have family responsibilities, particularly women with young children, often prefer to work part-time shifts with working hours during the school. According to one recent study in the EU, women with children living at home were 26 per cent more likely to work part-time than other women; this figure is even higher – nearly 30 per cent – if they have a child less than six years old (Lilja and Hämäläinen 2001: 39). While this does not necessarily mean that part-time work reflects the working time preferences of these workers, there is substantial evidence that women in particular would prefer a workweek that corresponds to ‘substantial’ parttime hours (see the evidence presented by Fagan, Chapter 4). Overall, short hours or part-time work is an area of working time that is growing rapidly and promises to continue to do so for the foreseeable future. For the most part, it appears that part-time work is designed primarily to meet enterprises’ business objectives, particularly in service industries such as retail trade, where competitive pressures are among the highest in any industry and keeping staffing costs low is the key to price competitiveness. Short hours of work allow firms to reduce their staffing costs by not paying premium wages to cover additional hours over normal hours of work (because such additional hours typically do not exceed the hour thresholds for receiving overtime pay) and ‘unsocial’ hours of work (particularly weekend work), and by providing lower wages and fewer benefits to part-timers in some countries. Part-time work also allows enterprises to better adapt staffing levels in response to customer traffic in stores, since the shorter part-time shifts can be more easily combined to cover periods of peak demand, and to rapidly adjust work schedules in response to customer demands – in effect, using workers as a ‘flexibility buffer’. There is also an indisputable gender dimension to part-time work, as the vast majority of part-time workers in all of the industrialized countries are women; thus, part-time work provides firms with a tool to tap into

170

Jon C. Messenger

the female workforce, especially in countries where there is a demand among mothers for part-time positions. From the perspective of workers, when part-time work is ‘involuntary’, it is obviously contrary to the preferences of the workers concerned, and may also be an indication that workers are unable to obtain sufficient earnings.Very short hours of work (less than 20 hours per work) are also generally contrary to the preferences of most workers, and such ‘marginal’ part-time jobs are often associated with a lower quality of work. Nevertheless, part-time work can provide a means of meeting workers’ needs and preferences in some cases – such as when there is a need to balance work and family responsibilities – if part-time workers are treated on an equal basis with full-timers, if the hours of work are ‘substantial’ (i.e. 20 hours per week or more), and if work schedules are fairly predictable and structured with those specific workers’ needs in mind. ‘Flexible’ working time arrangements: new approaches to the standard workweek Over the past two decades, there has been a move towards working time arrangements which allow hours of work to vary over the day, week, month, or even an entire year. Although such variation or ‘flexibility’ in hours of work is hardly a new phenomenon – overtime and shiftwork have existed for nearly a century – there has nonetheless been a marked trend away from the traditional ‘standard workweek’ of daytime, weekday shifts and towards a more diverse array of possible options for arranging any given number of hours of work. These ‘flexible’ working time arrangements include everything from more diversified shift patterns and increased weekend working to flexible starting and finishing times, ‘compressed’ workweeks (where a set number of hours is worked in fewer days), and the averaging of working hours over extended periods of time up to a year.11 Because of their potential importance to workers, employers, or both, this section will focus on the following types of ‘flexible working time’: •

Formal flexi-time programmes allow workers to vary their starting and ending times, and, in some cases, even the number of hours that they work in a particular week(s).12 Generally, flexi-time programmes involve establishing a period of ‘core’ hours when all employees are required to be at work (e.g. 10:00–16:00); however, some programmes operate with no core hours at all. These ‘core’ hours are bracketed on either side by periods of ‘flexible’ hours (e.g. 7:00–10:00 and 16:00–19:00) when employees can choose which hours to work, as long as the contractually required hours are worked for a specified work period. Flexi-time programmes may also permit workers to build up ‘credits’ or accumulate ‘deficits’ in hours worked, up to a maximum;

Working time in the enterprise 171

•

•

any ‘deficits’ can be made up in future work periods, while accumulated ‘credits’ can be exchanged for paid time off. The concept of ‘time banking’ or working time accounts involves keeping track of hours worked in ‘accounts’ for individual workers. Like some flexi-time programmes, time banking permits workers to build up ‘credits’ or accumulate ‘deficits’ in hours worked, up to a maximum amount; however, the periods involved are much longer, ranging from several months to a year or more. The rules of the specific time banking arrangement determine how and when the excess hours accumulated in the time banking account can be ‘spent’; in some cases, there are significant restrictions based on enterprises’ operational needs (e.g. the amount of notice required to take time off). In the case of shorter-term accounts, which are the most typical, hours worked that are above contractually-agreed hours can be taken as paid time off; in the case of long-term accounts, such as those in some sectors in Germany, accumulated credit hours can also be used for longer paid holidays, sabbaticals or even converted to cash for early retirement (IDS 2001c). Annualized hours or ‘hours averaging’ schemes allow for variations in working time over a longer period. Typically, annualized hours schemes set an average number of hours to be worked per week (e.g. 40 hours per week) over a specified period of time, called the ‘reference period’, which may vary from several weeks to as long as an entire year. This approach allows firms to increase hours of work during, for example, periods with a high volume of work, which can then be offset by shorter hours of work at other times when the company’s workload is less. The number of hours to be worked in any given week is determined by employers based on the needs of the enterprise, typically within a certain band of maximum and minimum weekly hours, subject to any additional conditions that may exist in collective agreements such as minimum notice periods. Annualized hours schemes can also be combined with elements of flexi-time or time banking in a myriad array of arrangements that are often unique to a specific enterprise.

Reliable information on the incidence of these flexible working time arrangements is scarce, and the available figures are not directly comparable across countries. However, the limited data available are nonetheless useful in providing some sense of how widespread these arrangements are among workers or among companies in a handful of industrialized countries. For example, formal flexi-time programmes,13 which have existed in many countries (in one form or another) since the 1970s are fairly widespread, covering roughly 1 in 10 workers in the USA, the UK and Japan; 14 per cent of workers in France; and a much higher proportion of workers in Germany, about 35 per cent (EIROnline 2001b: 15;

172

Jon C. Messenger

USDOL, BLS 2001: 1 (from Table 3); Japan Institute of Labour, 2002 Table 50: 63). In contrast, annualized hours schemes are relatively rare or nonexistent in some of these countries, but widespread in others. For example, such schemes cover only 3 per cent of workers in France14 and 4 per cent of workers in the UK (EIROnline 2001b: 15); in the USA, this type of working time arrangement is not allowed at all.15 Annualized hours schemes are the most prevalent in Germany, where they are used in 28 per cent of companies, and in Japan where they are found in 36 per cent of companies covering 23 per cent of all workers in the country (German Association of Chambers of Industry and Commerce (DIHT) 2000, reported in IDS 2001c: 125; Japan Institute of Labour 2002, Table 50: 63). As for ‘time banking’ accounts, although reliable data is unavailable for most industrialized countries, it is well established that such accounts are particularly widespread in Germany: 37 per cent of German workers have access to such individualized time accounts (IDS 2001c: 127).16 The only other industrialized country for which usable data regarding the incidence of time banking was determined to be available is France, where approximately 12 per cent of all employees had access to some type of a ‘time account’ (EIROnline 2001b: 15). The move towards more ‘flexible’ hours of work has largely been spearheaded by enterprises, often based on changes in work organization in response to the driving forces discussed earlier in this chapter. Bosch (in OECD 1995) has identified three main reasons why enterprises are seeking this increased flexibility in working time. First, firms are interested in using flexible working time to extend their hours of operation. In the case of manufacturing firms, as noted earlier, extended operating hours is a method of improving the utilization of their existing capital stock, and thus reducing their unit costs of production. In the case of service sector firms, extending shop opening hours has no immediate cost advantages, but rather is a tool to broaden the company’s appeal to different customer groups (who may prefer to shop at different times) and hence increase their competitive advantage. Second, firms are interested in using flexible working time to better adapt their hours of operation to fluctuations in market demand, and in particular, annualized hours schemes are effective in allowing enterprises to readily increase working time in periods of peak demand and reduce it at times when demand is slack. Finally, firms are also interested in using some flexible working time arrangements as a cost-cutting measure – although this particular aspect of temporal flexibility is sometimes limited by the provisions of national law. In particular, flexible hours schemes of various kinds can serve as a costeffective alternative to overtime. The introduction of either annualized hours schemes or time banking accounts – which permit hours to be averaged or accumulated over periods of time ranging from several weeks to a year – eliminates the need for employers to pay overtime premia to

Working time in the enterprise 173 workers for working hours in excess of the statutory ‘normal’ hours in any given week.17 In fact, the results of an enterprise survey by the German Association of Chambers of Industry and Commerce (DIHT) show that the primary objective cited by most German firms (60 per cent of respondents) in introducing flexible working hours was to ‘avoid overtime’; the second most frequently cited objective (28 per cent of respondents) was to ‘extend operating or service hours’ (DIHT 2000, reported in IDS 2001c: 125). In terms of workers’ needs and preferences regarding their work schedules, ‘flexible’ working time arrangements would appear to be a ‘mixed bag’. On the one hand, workers generally welcome the increased autonomy or ‘time sovereignty’ that is embodied in many of these arrangements, especially flexi-time and many forms of ‘time banking’. In fact, a growing body of evidence (e.g. Golden 2000; Fagan 2001b; see also Fagan, Chapter 4) points to workers’ increasing desire for influencing their work schedules. On the other hand, as we have seen in the discussion of parttime work, workers generally prefer to have some predictability in their weekly hours of work, and express particularly strong preferences not to work nights and weekends, especially not Sundays (see Fagan, Chapter 4, for a discussion of workers’ preferences regarding ‘unsocial’ work schedules). For example, one study (OECD 1995: 25) reported that, based on a representative survey of EU employees, only 52 per cent of them were willing to work variable hours, only 22 per cent were willing to work nights, and only 21 per cent were willing to work on Sundays – even in exchange for higher wages or additional paid leave.18 Since annualized hours schemes (by definition) require variable hours of work from one week to the next, and often require at least some work during ‘unsocial hours’, typically in the evenings or on weekends, these types of schemes would appear unlikely to be in line with the working time preferences of most workers. Of course, the realities here may not be so clear-cut: annualized hours schemes can also be combined with elements of flexi-time or time banking in ways which make it difficult to predict a priori how well workers would receive any particular model. Keeping these potential trade-offs in mind, the case studies in this section will illustrate how working time is evolving with the increasing use of flexi-time schemes, annualized hours, time banking, and other innovative methods of organizing a given amount of working time by firms pursuing specific business objectives. For example, as discussed earlier in this section, the extension of shop opening hours in response to diversifying customer demands has been an important factor shaping the rearrangement of working time. Once shop opening hours are extended beyond the range of an individual worker’s normal workday, it becomes necessary to develop alternative working time arrangements which can bridge the gap between the total number of hours that the shop is open and the total number of hours that any individual worker works. In the Netherlands, for

174

Jon C. Messenger

example, the implementation of a new Trading Act in 1996 allowed firms to expand their shop opening hours by 11 hours per week. As a result, many retail trading firms in that country reorganized their employees’ working time in ways that were specifically designed to exploit this opportunity. The case of one Dutch department store illustrates how this extension of shop opening hours was achieved via the reorganization of working time (Jansen and de Jong 1999). This particular company operates a chain of department stores selling ‘lifestyle-oriented’ products, and it includes six stores that are among the largest in the Netherlands. Since the early 1980s, the firm has developed and strengthened its working time flexibility with the support of its trade unions and works councils. In 1995, in preparation for the implementation of the new Trading Act in the Netherlands, the store’s management and the trade unions reached an agreement on a new working time scheme combining a compressed workweek with annualized hours. A ‘compressed workweek’ involves placing a set number of hours of work into a smaller number of days: for example, instead of working 40 hours in 5 ⫻ 8-hour days, a ‘compressed workweek’ might involve working the same number of hours in only 4 ⫻ 10-hour days. In this case, the agreement was for an average of 35 hours per week over four working days. These average weekly hours of work were derived based on total annual hours equal to 1,826 – i.e. 35 hours per week times 52 weeks (plus 6 additional ‘attendance hours’). In fact, what this means is that employees can be called upon to work whatever hours are necessary each week (up to the legal maximum, which is 48 hours per week) as long as the average hours of work do not exceed 35 per week over the course of the year. In return for this flexibility, workers were guaranteed a fixed salary, regardless of the number of hours actually worked during a particular period. In addition, this working time model includes a variety of provisions designed to provide workers with additional compensation (typically in the form of paid time off) for working overtime (e.g. over 9 hours per day, 520 hours per quarter, or 1,825 hours per year) and for hours worked in the evenings and on weekends. Finally, there are specified limits on certain types of ‘unsocial’ hours, which is of particular importance given the general unpopularity of these hours among workers. For example, a worker is expected to work a maximum of two evenings per week and a maximum of 13 Sundays per year; any work above these thresholds is strictly voluntary. In addition to extending operating and shop-opening hours, firms are interested in using flexible working time to better adapt their hours of operation to fluctuations in market demand. As discussed earlier in this section, flexible hours of work, and in particular, annualized hours schemes, are an effective approach for allowing enterprises to more readily increase working time in periods of peak demand and reduce it in times when demand is slack. Box 5.5 provides an example of how an annu-

Working time in the enterprise 175 Box 5.5 The use of annualized hours in seasonal industries VSFL, is one of the largest photo finishing laboratories and professional portrait developers in Sweden (it also serves parts of Denmark and Norway). One of the characteristics of this industry is its highly seasonal nature, with 50 per cent of its sales occurring in the summer months. Traditionally, VSFL employees worked a set number of hours per week throughout the entire year, but that system proved to be too inflexible to meet customer demands – particularly demands for quick turnaround of photo orders. In response, the company adopted a new system of work organization based on self-directed work teams with job rotation and training to expand employee skills. In addition, the company adopted a working time system based upon annualized hours. This system allows the teams to increase or decrease hours of work in response to fluctuations in customer demand. In the summer, employees in the professional portrait business work 45 hours a week, but they work only 30 to 40 hours per week during other times of the year. In the photo finishing business, the company operates three different shift patterns. During the peak summer months, the company operates 24 hours a day with three 8-hour shifts, but it generally uses only one 8-hour shift or two 6-hour shifts during the remainder of the year. The self-directed work teams also play an important role in setting working time patterns, by determining when to change shift patterns in response to changes in customer demand (The European Commission, Directorate-General for Employment, Industrial Relations and Social Affairs 1998).

alized hour scheme is being used in one firm in a highly seasonal industry in Sweden. Although this type of approach is not widely used in most of the industrialized countries (and, as noted earlier, is prohibited for hourly workers in the USA), it can serve an important function in helping to ‘smooth out’ peaks and troughs in anticipated demand for products and services, such as those in industries that are highly seasonal in nature. Two interesting enterprise examples of the increasing ‘flexibilization’ of working time come from Germany: the case of Quelle, the largest mail order house in Germany (see Box 5.6), and the case of Bremer Landesbank (BL), the largest bank in the north-west coastal region of Germany and one of the largest publicly-owned banks in the country. Both of these cases provide excellent examples of flexible working time models that arrange a ‘standard’ number of hours of work in a distinctly nontraditional way. Both of these cases also demonstrate how flexible hours of

176

Jon C. Messenger

Box 5.6 ‘Net annual working time’: a new model for standard hours Quelle, the largest mail order house in Germany, has introduced working time flexibility to meet the objective of fast delivery of customer orders. Because of customer demands for faster delivery of their orders, plus dramatic shifts in workload levels during the course of the year that typically required the extensive use of overtime, the company decided to introduce a new working time model. This new model is based on the concept of ‘Net Annual Working Time’, which is the total number of hours to be worked during the following calendar year. The annual figure for hours of work amounts to an average of 31 hours per week over the entire year (this average includes periods of paid annual leave). This average is agreed upon with employees by the end of the prior calendar year. Based on this annual figure, a ‘Monthly Time Plan’ is developed each month which specifies the shifts and hours of work for the next month, and hours of work can vary between 6:00 and 23:00, 5 days a week. The ‘Net Annual Working Time’ must be reached within the calendar year, and salaries are based on annual hours, paid in equal monthly payments regardless of actual hours of work. Thus, no overtime is paid for hours exceeding the average each week. However, time worked in excess of ‘Net Annual Working Time’ is paid at normal rates up to 110 per cent of this figure, and overtime is paid for those hours in excess of 110 per cent of ‘Net Annual Working Time’ (European Commission, Directorate-General for Employment Industrial Relations and Social Affairs 1998).

work can provide a cost-effective alternative to the use of overtime (The European Commission, Directorate-General for Employment Industrial Relations and Social Affairs 1998). With an increased level of competition and growing demands from its customers, BL had to focus on controlling costs and providing higher levels of customer service. In response to this competitive situation, the BL management organized a workshop involving employees for the purpose of discussing alternative working time arrangements. Based on brainstorming sessions during this workshop, the management developed (with the assistance of working time consultants) a new working time model, which was presented to the Bank’s works council. It should be noted that, in Germany, the works council has a statutory right of codetermination regarding the beginning and ending of daily working hours and regarding the distribution of working hours over the days of the week.19 Management and the works council then negotiated a final agreement on the new working time system to be implemented at the Bank.

Working time in the enterprise 177 The result of these negotiations was a working time model that combines annualized hours with time banking. This model is based on the concept of ‘function time’, with different working times for each function in the bank. The normal daily working time is 7.8 hours per day (or 7 hours and 48 minutes), and the normal ‘bandwidth’ (range) of daily working time is from 7:30 to 18:30. Within these broad parameters each work group is able to propose customized hours of work for each of the functions for which they are responsible; these proposed hours of work are subject to approval by management and the staff works council. Once approved, the teams can then determine their own staffing arrangements, as long as they ensure that required tasks are properly performed and appropriate services are provided to customers. The key to the Bank’s working time system is a new model of time accounting based on a so-called ‘traffic light’ approach. Instead of having a set period of time (called the ‘accounting period’) over which working time must be calculated, the Bank’s new model allows for an ongoing calculation of working time based upon deviations from the established contractual normal hours of work. Like a traffic light, there are three possible options: ‘green zone’, ‘amber (yellow) zone’ and ‘red zone’. On any given day, an employee’s actual hours of work are considered to be ‘normal’ (green zone) if they fall within +/– 20 hours of the contractually agreed hours of work (i.e. 7.8 hours per day times the number of days worked). However, if the hours of work exceed (or are less than) contractual hours by more than 20, but less than 30, then the employee is in the ‘amber zone’, which means that action must be taken to get them back within the ‘normal’ range, i.e. the employee must work fewer (or more) hours per day until the actual hours worked are back in the ‘green zone’. Finally, a deviation of more than 30 hours above or below contractually agreedupon hours is considered to be in the ‘red zone’; such a major deviation from ‘normal’ hours is prohibited unless it has been explicitly authorized by the management. It is reported that this new working time system has resulted in the following benefits: greater customer access to banking services beyond normal bank opening hours, thus improving the level of service to customers, as well as increased satisfaction among BL staff due to increased flexibility and autonomy in their hours of work (The European Commission, Directorate-General for Employment Industrial Relations and Social Affairs 1998). In summary, it seems clear that the ‘standard workweek’ is changing. Firms are looking to working time arrangements such as flexi-time, annualized hours, time banking, compressed workweeks, and other variants of these approaches, which allow them to extend operating hours; vary both the hours and the timing of work in ways that better respond to the demands of particular markets; and reduce their staffing costs (e.g. overtime). These flexible working time arrangements may have both positive and negative implications for workers: offering them the potential to

178

Jon C. Messenger

influence their work schedules, but also reducing the predictability of those schedules. Some of these arrangements, particularly annualized hours schemes, may also require an increase in working during ‘unsocial hours’ – evenings, nights, and weekends. While the impacts of any particular scheme are difficult to predict in advance, it seems likely that those flexible working time arrangements that allow workers a substantial degree of control or influence over working (or ‘time sovereignty’) – such as flexi-time programmes and ‘time banking’ accounts that operate like an extension of flexi-time over longer periods – offer the best opportunity of adapting ‘flexible’ working time to workers’ needs and preferences.

Balancing business’ objectives with workers’ preferences: some innovative cases The preceding sections have attempted to illustrate those actual working time patterns which are emerging at the enterprise level. While these case studies are not based on a representative sample of enterprises, they nonetheless provide some useful illustrations of certain broad trends in working time (e.g. the increasing use of part-time work, the move towards more flexible hours of work) which are occurring at the enterprise level across the industrialized world. These cases reflect the reality that, for the most part, changes in working time patterns are being driven predominantly by the business objectives and strategies of firms in an intensely competitive globalized marketplace, and only marginally by the needs and preferences of workers. Nevertheless, a growing number of enterprises have begun to pay attention to the needs and preferences of workers regarding various aspects of their working lives, including working time. While employers’ business objectives and workers’ preferences regarding working time – both their hours of work and how those hours are arranged – would often appear to be in conflict, some pioneering enterprises are creating working time arrangements and work schedules that intentionally seek to balance business objectives with the working time needs and preferences of various types of workers. There are a wealth of research studies which provide evidence regarding the benefits to businesses of adopting working time patterns which are better adapted to workers’ needs and preferences. For example, there is a substantial body of evidence that has developed over the years showing that reductions in ‘excessively’ long hours of work – which are clearly contrary to workers’ preferences – have resulted in substantial productivity gains over the years (see, for example, Bosch and Lehndorff 2001, and White 1987, for a review of the relevant literature). The productivity gains that can be achieved through a reduction in ‘excessively’ long hours are derived through two main avenues: an increased pace of work (resulting from physiological adaption by workers to shorter hours) and the elimina-

Working time in the enterprise 179 tion of inefficient work practices, such as unscheduled breaks. These productivity gains are often related to changes in work organization, new methods of production, more efficient working practices and similar factors which typically accompany reductions in working time. While, in theory, these types of changes could be implemented with longer hours, in practice the changes are often so closely linked (e.g. with collective agreements that simultaneously implement reduced working time and new patterns of working time, such as flexible hours), that they are nearly impossible to distinguish (White 1987).20 Reducing long hours of work can also provide benefits to businesses due to reductions in absenteeism and staff turnover. For example, it is well established that there is a positive relationship between longer hours of work and absenteeism (see, for example, Barmby et al. 2002). In addition, a recent UK study based on data from the 1998 Workplace Employee Relations Survey provides empirical evidence demonstrating a significant positive association between long hours of work and higher staff turnover (Meager et al. 2002: 2). Another related body of research indicates that working time arrangements that take workers’ needs and preferences into account can have positive effects on employees’ morale and motivation, which in turn can lead to improved company performance. For example, flexible working time arrangements such as flexi-time and compressed workweeks have been shown to have positive effects on employee attitudes and morale (Hogarth et al. 2001; Gottlieb et. al. 1998). Logically, it makes intuitive sense that workers who have better morale would be more productive, but there is also empirical evidence to support this conclusion. For example, a recent empirical study of employees’ emotions on the job, Working Today: Exploring Employees’ Emotional Connection to their Jobs, shows a statistically valid, positive relationship between employees’ emotions about work and company performance (Towers Perrin and Gang & Gang 2003). This study concludes that: Finally, there is a statistically significant relationship between strong positive employee emotion and superior financial results for a company . . . Gaining this discretionary effort from employees may be the last remaining source of increased productivity, now that so many companies have already captured the efficiencies of technology and streamlined work processes. (ibid.: 2) A number of enterprise-level studies also point to business benefits resulting from the adoption of various types of flexible working time arrangments, particularly flexi-time. For example, one study of 6 large US companies – Lucent Technologies, Honeywell International Inc., Motorola Inc., Kraft Foods Inc., Amway Corp. and Bristol Myers Squibb Co. – found

180

Jon C. Messenger

that 70 per cent of managers and 87 per cent of workers reported positive impacts on productivity resulting from the use of flexi-time (Boston College Center for Work and Family 2000: 2). Other reported benefits of flexible hours include reduced tardiness and absenteeism, reduced staff turnover (improved employee retention), and more effective recruitment of a diverse pool of new hires (Avery and Zabel 2001; Catalyst 1997; Hogarth et al. 2001). Broader firm policies designed to assist workers in reconciling their work and family responsibilities – such as child-care, parental leave, and the right to return to work on a part-time basis – also show positive effects, including better retention rates, reduced absenteeism, and productivity gains (OECD 2001). Ultimately, however, it may not be any of these working time arrangements per se that are the key to achieving a positive impact on productivity. Although, as we have seen, the type of working time arrangement can certainly produce important benefits for businesses, there is evidence that the most important factor may be workers’ ability to choose their working time arrangement – it is this exercise of ‘choice’ that shows the strongest impact on employees’ job performance, and hence on firms’ productivity. For example, one comprehensive study focusing on the effects of flexible work arrangements (Gottlieb et al. 1998) concluded that workers’ ability to choose their working time arrangement was strongly related to their job performance; based on their managers’ ratings, workers who had chosen their work arrangement exhibited higher levels of job performance, better relationships with customers and co-workers, and slightly better relationships with their managers than those workers who were assigned to their work arrangement (or hired into a given work arrangement). This study concluded that: [T]here is a compelling case for having employees choose flexible work arrangements. Based on the data from the Workplace Flexibility Study it appears that the benefits of flexible work arrangements arise from and are contingent upon the subjective sense of enhanced control that employees gain from choosing the arrangement that is right for them. (ibid.: 123) Thus, taking workers’ needs and preferences into account can be expected to provide benefits not only to workers, but to enterprises as well. These initiatives include several of the working time arrangements discussed earlier in this chapter, such as a variety of flexi-time arrangements and similar schemes to give workers greater control over their work schedules; high-quality part-time working in jobs with a substantial number of hours and good wages and benefits; and also some innovative approaches to more ‘standard’ work patterns, in which firms actively seek to obtain input on workers’ preferences when developing work schedules. The remainder

Working time in the enterprise 181 of this section will explore a few enterprise cases that illustrate innovative attempts to create working time arrangements that balance business objectives with the needs and preferences of workers regarding working time, in order to answer the following questions: • • •

What working time patterns/arrangements were implemented? How were these patterns/arrangements implemented? What results were achieved?

Flexi-time programmes and flexible hours schemes As discussed earlier in this chapter, flexi-time arrangements can offer workers the opportunity to structure their hours of work to better fit their individual situations, such as their family responsibilities. The use of flexitime arrangements has been expanding in Japan. One survey of Japanese companies conducted by the RENGO, the Japanese Trade Union Confederation, found that most companies in the country had instituted some type of a flexible working hours system (Rengo 2001b). While the vast majority of the companies surveyed used a standard flexi-time system with fixed core hours, a small but increasing proportion of firms are adopting a new approach known as ‘discretionary working hours systems’; these innovative work systems are described in Box 5.7. The Ford Motor Company of Australia provides an illustration of a company-led effort at cultural change based on an increase in working time flexibility aimed at increasing workers’ control of their work schedules (Australian Department of Employment, Workplace Relations, and Small Business, Work and Family Unit 2000). Ford Australia is attempting to change from having a male-dominated ‘long hours culture’ to one which is more conducive to a healthy work–life balance. The key to Ford Australia’s change effort is its Core Meeting Hours Policy, which states that meetings must only be held during ‘core hours’ between 9:30 a.m. and 3:00 p.m. This policy is combined with flexible working hours guidelines, including flexible starting and ending times, compensatory time off, and a variety of ad-hoc arrangements for flexible hours, days off and telecommuting, plus an option for employees to negotiate their hours of work. These flexi-time components are part of the firm’s broader ‘Life/Work Initiative’, which also includes a variety of ‘family-friendly’ benefits such as paid maternity, paternity, and adoption leave, workplace child-care, and a dependent care scheme. Ford Australia began the change process with a pilot scheme in one part of the company, the Product Development Division. This process began with a discussion of the firm’s business plan and workplace operations, which generated the ideas that resulted in the ‘Life/Work Initiative’. This programme was later extended to the company as a whole. Strong leadership from senior company executives has been a key factor in the

182

Jon C. Messenger

Box 5.7 Maximum flexi-time: discretionary working hours systems in Japan Some Japanese companies have been experimenting with an innovative concept called ‘discretionary working hours systems’, based on the idea of allowing individual workers to decide their own hours of work, and often the location where the work is performed as well. In other words, workers perform their work at their own ‘discretion’. In this type of a system, work corresponds not to the number of hours actually worked, but instead to a ‘notional’ (implicit) average number of hours that is regarded to be adequate to perform the requirements of the job. Under such an approach, performance evaluation and compensation systems become particularly critical, since a significant portion of the workers’ pay is based on performance rather than on the number of hours worked. Discretionary working hours systems are being tested with a small number of workers in companies such as Matsushita Electric Industries (Sato 2001). This company introduced a discretionary hours system covering 243 research and development staff and designers in several different establishments within the company. The ‘notional’ working hours for employees covered by this system are 7 hours and 45 minutes per day (plus an extra 18 hours per month divided by the number of scheduled work days). The actual number of hours worked depends on the individual worker. Performance bonuses are an integral part of workers’ pay, comprising 10 per cent of their total earnings. Key firm decisions in implementing this system include which workers to include and the lowest grade level to be covered, as well as how to structure the performance evaluation and pay schemes. While it is too early to know the results, this new system clearly represents a radical change from normal practices, particularly in a nation where salaries are based largely on time worked, including both current working hours and the number of years of seniority.

success of the programme to date. There has also been a focus on data collection and evaluation, with the development of key indicators for the purpose of measuring the company’s progress in improving the working environment; for example, one indicator measures workload stress based on an annual survey. The results of Ford Australia’s ‘Life/Work Initiative’ are both tangible and intangible. From the company’s perspective, there has been an increase in productivity, employee morale, work quality, and also an increase in the breadth and diversity of the talent pool interested in the

Working time in the enterprise 183 company, which in turn has increased the quality of its new hires. From the point of view of employees, there has been an improvement not just in the flexibility of working hours, but also ‘employees report a greater ability to access flexible hours without feeling guilty’ (ibid.: 7). The end result for Ford Australia’s employees was that they reported being better able to balance their work with other aspects of their lives. High-quality part-time working As discussed in the section on short hours earlier, part-time work encompasses a broad range of situations, which vary dramatically in terms of their responsiveness to workers’ needs and preferences. Part-time work that is responsive to workers’ working time preferences is typically regular part-time work with a ‘substantial’ number of hours (i.e. 20 to 34 hours per week) and that is performed on a voluntary basis. In addition, an element of predictability also appears to be important: part-time workers need to know their work schedules far enough in advance to be able to adequately plan their paid working time around their personal responsibilities and other interests (see also Fagan, Chapter 4). The Netherlands has been a leader in attempts to promote a better balance between work and non-working life. The Dutch have the highest rate of part-time work in the industrialized world, and, as discussed by McCann in Chapter 1, the Netherlands has implemented a number of pieces of national legislation in recent years designed to promote work–life balance. The relevant legislation includes the 1996 equality law giving part-timers a legal right to equal treatment, and the Right to Change Working Hours Law in 2000, giving workers the legal right to change from full-time to part-time hours or the reverse (IDS 2001a). Moreover, employers and employees in many Dutch companies are collaborating to create working time arrangements that are responsive to the different working time needs and preferences of workers at various stages of their lives – see Box 5.8 for a brief review of such a scheme at one Dutch firm. The case of Northern Trust, which is part of the British National Health Service (NHS), provides a good illustration of how ‘regular’ part-time work with substantial hours can be structured to simultaneously meet business objectives and workers’ preferences, even in a country where (as discussed previously) more than half of all part-time workers work in ‘marginal’ part-time jobs (Arrowsmith and Mossé 2000). First, some background is necessary. The NHS was operated on a centralized basis until 1990, when the NHS Community Care Act was enacted. That legislation, part of a push by the then-Conservative government to decentralize public agencies in the UK with the objective of increasing efficiency, encouraged local components of the NHS to establish themselves as separate entities – ‘self-governing hospital, community care or ambulance trusts’ (ibid.: 289). These ‘trusts’, though still subject to certain national-level controls

184

Jon C. Messenger

Box 5.8 Meeting workers’ preferences over the life cycle: flexible options for older workers in the Netherlands Wavin, a large Dutch firm manufacturing plastic pipe systems, has created two working time schemes designed specifically to meet the working time needs of one group of workers: older workers. The two options include a flexible working time and retirement scheme, and also a so-called ‘career bending’ scheme giving workers the option to scale back their work commitments beginning from the age of 45. The flexible retirement scheme was actually a creative response – agreed upon with the company’s three trade unions – to cost problems with the firm’s pension scheme. The pension scheme was predicted to become unsustainable as large numbers of workers began to retire (pension fund shortfalls are now a serious problem in many European countries). The company responded by developing a flexible retirement scheme that permits workers to adopt nearly any type of working time pattern they choose – either full-time or part-time, ranging from one to five days per week – and retire between the ages of 55 and 70, rather than the previous ages of 65 (with a full pension) or 62 (with 85 per cent of their full pension). While workers gained increased flexibility to work their preferred hours, the firm gained financially from a restructuring of the pension system to ensure its long-term solvency. Under the new scheme, retirement benefits are based on age, with benefits ranging from 30 per cent of former pay at age 55 to 100 per cent of pay at age 70. This scheme also includes financial protections for workers in their sixties who were already eligible for early retirement under the previous scheme (EIRR 1999).

regarding the terms and conditions of employment of their employees, have the authority to hire their own staff and determine their own human resource and industrial relations policies. As in many sectors of the UK, part-time work is a relatively common practice in the NHS Trusts, and Northern Trust’s use of part-time work, in and of itself, is not especially noteworthy. What is interesting, however, is how Northern Trust specifically focused on its workers’ preferences as a strategy to meet a business objective. In this case, Northern Trust’s business objective was to reconcile tensions between cost and quality in the delivery of its breast-screening services, part of a national programme of breast-screening for all women between 50 and 64 years old. Essentially, the problem was that, for cost reasons, the time allotted for each breastscreening X-ray was only five minutes, but yet (for obvious reasons) the results had to be highly accurate. Also, the X-rays were actually developed

Working time in the enterprise 185 off-site, so if there was any type of error, the patients would have to be called back for another X-ray, both increasing costs and possibly upsetting the patients as well. The management at Northern Trust was concerned that staff working full-time hours would be unable to handle the combined pressures of time and accuracy while delivering such a high volume of breast-screening examinations. At the same time, they realized that they had a workforce comprised largely of women, many of whom also had family responsibilities. Their solution was to offer positions in breast-screening on a halftime basis, with 2-day and 3-day weeks; that is, one week workers work 2 days, and the following week they work 3 days, averaging out to half-time (or roughly 20 hours per week). The breast-screening function was housed in four mobile units, with four staff assigned to each unit. Within each unit, 2 of the 4 half-time staff in the unit work at any given time. Days of work are fixed for each employee (the exact days of work depend on whether they are working a 2-day or a 3-day week), although they have the option of exchanging work days among themselves at their own discretion. The results of this part-time arrangement proved to be very positive, both for the organization and the workers involved. Northern Trust was able to achieve its objective of providing highly accurate breast-screening services – while simultaneously holding down costs – by limiting the time per examination. This was critical, given the fact that the organization is responsible under the UK’s national breast-screening programme for providing breast-screening examinations for all 128,000 of the women in its region at least once every three years. At the same time, the positions were highly valued by the workers because ‘part-time positions in radiography were few and far between, particularly at the higher-graded Senior 1 level’ (ibid.: 299). In addition, the predictability of the fixed days and hours of work, and the fact that workers were not required to work ‘unsocial hours’ made the positions ideally suited to the needs of the women in these positions, most of whom had school-age children. As a result, workers reported that they were strongly motivated to do a good job, which is probably the main reason why they were able to meet the quite demanding business objective of providing high-quality breast-screening services in a very short period of time. In fact, this particular effort with part-time workers was so successful that it resulted in an expansion of part-time opportunities throughout the organization, including opportunities for part-time work in more senior grades (ibid.) Using workers’ working time preferences to design work schedules The enterprise cases discussed above have provided some examples of how working time arrangements can be restructured to achieve a better balance between enterprises’ business objectives and workers’ needs and

186

Jon C. Messenger

preferences. However, even the most traditional types of working time patterns, such as simple shiftwork systems, can be designed in ways that take workers’ needs and preferences into account. Often, the approach involved can be as simple as asking workers which hours they would prefer to work, and using the resulting answers as an input into the process of structuring the company’s work schedules. For example, the case of a hypermarket in Germany illustrates how work schedules – in this case, shiftwork schedules – can be designed to incorporate workers’ preferences (Haipeter 2001). The store in question (known simply as ‘Store A’ in the case study) is a hypermarket that sells a wide range of products, in competition with a wide variety of grocery stores and other retailers. The company’s overall business objective is to compete successfully both on price and also on service quality, based upon a broad range of products; a quick and easy return policy; and high-quality customer service. Thus, employees are an important resource for implementing the firm’s quality-focused service strategy. This German hypermarket, like many retailers, has moved increasingly towards the use of part-time workers as a means of increasing its flexibility to respond to changes in market demands. In particular, the use of parttimers is considered to facilitate flexible scheduling by making it easier to vary the total number of hours that are worked and when those hours are worked. Unlike many retailers, however, this store does not use ‘marginal’ part-timers. The hypermarket employs only regular part-time workers on permanent contracts, largely because marginal part-timers are typically unskilled workers, and the store needs skilled workers to implement its quality service strategy. The use of part-time shifts with flexible shift schedules in the retail sector is a very typical working time approach. But how those shifts are determined in Store A is far from typical. The key element is the use of workers’ working time preferences in the development of the shift schedules. In fact, two types of data are used as input into the shift schedules: historical data on customer flows on different days and at different times over the course of the year, and data gathered from employees regarding their preferred hours of work. Both types of data are input into work scheduling software packages on the firm’s computer system, which then designs work schedules at the store for the following 3-week period. There are three constraints on the scheduling process, which were agreed with the company’s works council: (1) shifts must be at least 4 hours; (2) shifts must start within a range of established ‘core’ hours; and (3) overtime (if any) must be balanced within the same year. In addition, it should also be noted that, whatever their work schedules, workers also have the option of ‘swapping shifts’ with co-workers (subject to supervisory approval). The positive outcomes that resulted from such a seemingly simple process are difficult to overstate. One key statistic from an internal employee survey seems to say it all: 95 per cent of the hypermarket’s

Working time in the enterprise 187 employees are satisfied with the firm’s working time system (Haipeter 2001: 119). In addition, Haipeter (ibid.: 119) reports that the company’s active efforts to seek out workers’ inputs regarding their working time preferences, combined with workers’ ability to ‘swap’ shifts, has resulted in ‘a culture of negotiation based on mutual give and take . . . between employees and supervisors, a culture that has a high degree of normative and cohesive force’. Although the case study does not clearly state the specific business outcomes that were achieved, such a culture is bound to translate into improvements in whatever areas the store wants to emphasize – in this case, for example, better customer service. Lessons learned from these innovative cases As we have seen, some enterprises are implementing innovative working time patterns and arrangements that are designed both to achieve business objectives and to meet workers’ needs and preferences at the same time. These types of arrangements include various approaches to flexi-time and other types of flexible hours schemes; regular part-time working in highquality jobs; and the use workers’ preferences as an input in designing work schedules. The key factors that contributed to the success of these efforts included: some type of a mechanism to obtain information on the needs and preferences of the firms’ employees; a conscious attempt to align business objectives and strategies with workers’ needs and preferences in ways that are mutually reinforcing; a clear plan for how the working time arrangements could be used to implement the firms’ business strategies; strong management support, especially at the executive level; and some type of feedback loop that assists firms in evaluating progress and identifying potential problems. The results of implementing these working time arrangements included positive outcomes for both firms and workers. From the perspective of employers, these outcomes include improved employee morale, job performance and productivity, as well as reduced absenteeism, increased retention (reduced staff turnover) and improved recruitment of new employees. From the viewpoint of workers, benefits include increased job satisfaction and an improved ability to balance paid work with their personal lives, including their family responsibilities. Moreover, the most important factor in achieving positive results for both enterprises and workers appears to be workers’ ability to choose their working time arrangement – it is this exercise of ‘choice’ that appears to have the strongest impact on employees’ job performance, and hence on firms’ productivity.

Conclusion Working time arrangements at the enterprise level are shaped by the legal and regulatory frameworks governing working time – that is, the working

188

Jon C. Messenger

time regimes – that exist in each country, as well as both the broader socioeconomic context and the competitive realities in different economic sectors. However, more and more, transnational factors – factors such as the intensification in competitive pressures in many industries as a result of the increasingly global nature of competition and changes in market demand – enabled by advances in information and communications technologies, are providing the context within which hours of work and working time arrangements are being developed. Women’s increasing participation in the labour market has also had an effect, particularly regarding the extensive development of part-time work in many countries. Finally, enterprises themselves are, in a more decentralized environment, finding themselves with more room to manoeuvre, and thus are exercising a greater degree of control over their industrial relations and human resource management policies, including working time. As noted earlier in this chapter, this factor helps to account for the enormous degree of variation in these policies across firms – even among firms in the same country and the same industry. In general, however, firms have been exercising this increasing discretion in the direction of seeking greater flexibility to respond to the particular mix of competitive pressures they face – including pushing for greater flexibility in working time. The hours of work and working time arrangements that are emerging vary dramatically, not only across countries, but across industrial sectors within the same country, and even among firms in the same industry depending on their particular business objectives and strategies. First, it appears that there are a number of circumstances that are leading to longer hours of work in certain industrial sectors and occupations. The driving force overall appears to be firms’ search for competitive advantage by, for example, implementing working time structures that provide for rapid response to variations in customer demands or which focus on extending operating hours to increase output while reducing unit costs of production. While national working time laws and regulations have an effect on restricting the hours of work for employees generally, it appears that they have less of an impact in restricting excessively long hours in certain sectors, such as IT, and certain occupations, such as professional and managerial staff, which tend to be exempted from working time regulations (although some companies may still seek to adapt their working time patterns to different regulatory frameworks). It is also important to note that, while working time patterns which induce excessively long hours are likely to have negative effects on workers, such as increased health problems and burnout – and are clearly contrary to the working time preferences of most workers – they may also have negative effects on firms, including low staff commitment to the business, increased absenteeism and staff turnover, and ultimately, reduced productivity, and of course negative externalities as well (such as increasing health care costs). Short hours, i.e. part-time work is an area of working time that is

Working time in the enterprise 189 growing rapidly and promises to continue to do so for the foreseeable future. Part-time work has an indisputable gender dimension, as the vast majority of part-time workers in all of the industrialized countries are women – and typically women who are attempting to balance paid employment with family responsibilities. For the most part, it appears that part-time working is designed to meet enterprises’ business objectives, particularly in service industries where competitive pressures, especially cost competition, is intense. Short hours of work allow firms to reduce staffing costs (especially for low-skilled jobs) by, for example, covering ‘unsocial’ hours of work without having to pay premium wages or (in some countries) offering part-timers lower wages and fewer benefits than fulltime workers. Part-time work also allows enterprises to better adapt staffing levels in response to customer traffic in stores, since the shorter part-time shifts can be more easily combined to cover periods of peak demand and also to rapidly adjust work schedules in response to demand – in effect, using workers as a ‘flexibility buffer’. Part-time work also offers firms an important tool for tapping into the female workforce, allowing companies to recruit and retain key staff, especially in countries where there is a demand among mothers for part-time positions. Unfortunately, very short hours of work (i.e. ‘marginal’ part-time) are often associated with a lower job quality and, particularly when part-time work is ‘involuntary’, it may be an indication that workers are unable to obtain sufficient earnings. Nevertheless, part-time work can provide a means of meeting workers’ needs and preferences – if part-time workers are treated on an equal basis with full-timers, if the hours of work are ‘substantial’ (i.e. 20 hours per week or more), and if work schedules are fairly predictable and are structured with specific workers’ needs in mind. It is also clear that the ‘standard workweek’ is changing. While the 8hour day and the 40-hour week (or their equivalents in countries or sectors with shorter hours of work) have not disappeared from the industrialized world, their form is clearly undergoing a transformation. Increasingly, firms are looking to working time arrangements which allow hours of work to vary over the day, week, month, or even an entire year. These ‘flexible’ working time arrangements – flexi-time programmes, annualized hours, time banking, compressed workweeks, and variants of these schemes – allow variations in both the hours and timing of work in ways that better respond to the idiosyncratic demands of particular markets, as well reducing staffing costs, particularly overtime. Various types of these arrangements may also be combined, such as, for example, annualized hours schemes and time banking accounts. While the diversity of these working time arrangements and the lack of reliable data on them make it difficult to draw definitive conclusions about the extent to which they are in line with workers’ working time preferences, it seems likely that those arrangements that allow workers to exercise a substantial influence over their work schedules offer the best chance of adapting ‘flexible working time’ to

190

Jon C. Messenger

workers’ needs. For example, in a British survey, when employees were asked to identify their priorities for rearranging their working hours, the most popular reform was more flexibility and autonomy over their hours (Fagan 2001b). Some firms are implementing innovative working time arrangements designed to simultaneously meet business objectives and workers’ needs and preferences. These types of arrangements include various approaches to flexi-time and other types of flexible hours schemes, high-quality parttime working, and the use of workers’ preferences as an input in designing work schedules. These efforts can result in positive outcomes for both employers and workers, including improved productivity and reduced staff turnover and absenteeism for employers, and for workers, increased job satisfaction and an improved ability to balance paid work with their personal lives. In this context, the most important factor in achieving positive results for both enterprises and workers appears to be workers’ ability to choose their working time arrangement – it is this exercise of ‘choice’ that has the strongest impact on employees’ job performance, and hence on firms’ productivity. Implications for enterprise policies and practices Based upon the selected enterprise case studies that have been reviewed in this chapter, it is possible to extract some important lessons regarding what types of enterprise policies and practices offer the best possibility of simultaneously meeting both employers’ business objectives and the needs and preferences of different workers. It should be emphasized that, while these suggestions will focus on enterprise-level policies and practices, it is unlikely that workplace policies alone will be sufficient for achieving this balance. There is also a potentially useful role for government initiatives in certain areas, particularly in restricting excessively long hours of work for individual workers, in line with the ILO Reduction of Hours of Work Recommendation 1962 (R 116),21 and in promoting the equal treatment of part-time and full-time workers, in line with the ILO Part-Time Work Convention 1994 (No. 175). In addition, these enterprise-level suggestions will obviously need to be considered within the context of the specific country and industrial sector within which the firm is operating. Such an approach is recommended, for example, in the EU Working Time Directive, which includes an emphasis on encouraging the social partners to negotiate working time reforms, taking into account work–life balance issues. What policies and practices can firms adopt to take workers’ needs and preferences into account in ways that will also benefit the ‘bottom line’? First, firms would need to establish enterprise policies that clearly indicate their commitment to the objective of promoting work–life balance. A company policy statement of this type, such as that adopted by Ford Aus-

Working time in the enterprise 191 tralia, is an essential step in legitimating workers’ needs for a balance between their paid employment and their personal responsibilities and interests. However, statements of enterprise policy alone are not enough. Specific processes must be put in place to ensure that workers’ needs and preferences are systematically taken into account in determining both hours of work and how those hours are arranged. The first step in such a process would be to obtain information regarding the needs and preferences of the workforce through some type of vehicle such as a discussion group (e.g. the ‘brainstorming’ sessions at BL) or an employee survey. Then, employers would need to compare those needs and preferences with their current business objectives and business plan. The purpose of that comparison would be to determine the extent to which the company’s current business objectives and strategies fit, or don’t fit, with the workers’ needs and preferences. Although this type of a discussion needs to begin at the executive level, it is highly recommended that workers and their representatives be involved in these discussions from the beginning. At this stage, there may well be tradeoffs to be made, but the ultimate objective would be to try to align business objectives and strategies with workers’ needs and preferences to the greatest extent feasible given the enterprise’s competitive situation. The next step would be for managers and workers together to carefully examine the firm’s operating procedures to see how existing working time arrangements and work schedules might be restructured to both improve business outcomes and better reflect workers’ needs and preferences. Clearly, this is no simple task, one which is highly firm-specific, and not something for which a ‘cookbook-style’ recipe can be developed. Also, both the enterprise’s objectives and workers’ needs and preferences may change over time. Fortunately, the level of sophistication of currently available computer software makes it relatively simple to simultaneously solve for a variety of variables and constraints, thus facilitating the process of taking workers’ working time preferences into account in developing working time arrangements. For example, as was seen in the case of the German hypermarket, available software for planning work schedules allows for data on workers’ working time preferences to be used as an input, which can be combined with other inputs (e.g. production objectives) in order to generate optimal work schedules. Obviously, such an approach is best suited for large and medium-sized firms; less formal approaches along similar lines would be needed in small enterprises, where work scheduling is handled more informally. Finally, there needs to be some type of a feedback mechanism to measure the results of implementing the new working time arrangements. Ideally, such a feedback loop could be based on a system of key performance indicators, using data gathered at regular intervals, as in the case of Ford Australia’s indicator of work stress. Few firms in today’s world would run their business operations without some type of indicator(s) to measure

192

Jon C. Messenger

their business results, including key intermediary steps in the production/service delivery process. Thus, it may be possible to simply adapt the firm’s existing performance indicators for this purpose. Clearly, there are going to be some trade-offs that have to be made in developing working time patterns and arrangements that attempt to balance enterprises’ business objectives with the needs and preferences of different types of workers. As noted earlier in this section, government policy initiatives may also be necessary in certain areas, particularly in restricting excessively long hours of work and in promoting the equal treatment of part-time and full-time workers. Depending upon the firm’s competitive situation, certain kinds of worker needs and preferences will be able to be taken into account, while others may ultimately prove to be economically infeasible. Nonetheless, it is important to keep in mind that, in today’s intensely competitive business environment, it is the enterprise’s human resources – that is, its workers – that frequently hold the key to its competitive advantage.

Acknowledgements The author would like to thank Dominique Anxo, François Eyraud, Colette Fagan, Sangheon Lee, Deirdre McCann and Stephen Wandner for their helpful comments on various drafts of this chapter.

Notes 1 It should be noted that, the ‘flexible firm’ model is in many ways just a rearticulation of the theory of dual labour markets. In the ‘flexible firm’ model, it is the ‘core’ employees who work in internal labour markets, and the various groups of ‘peripheral’ employees (who may be part-time, temporary, self-employed consultants, etc.) who must continually sell their skills in the external labour market. It is also worth noting that, while Atkinson places part-time employees squarely in the ‘periphery’, some recent studies (e.g. O’Reilly and Fagan 1998) have emphasized that part-time work can be either primary or secondary employment, depending on how it is used by a particular firm. 2 Forty-eight hours of work per week has long been considered to be an important benchmark for determining ‘long’ hours of work, as two key ILO Conventions on hours of work (the Hours of Work (Industry) Convention 1919 (No. 1) and the Hours of Work Commerce and Offices Convention 1930 (No. 30)) and the EU Working Time Directive both use 48 hours as a key reference point. However, it should be noted that the 48-hour standard established in Conventions No. 1 and No. 30 refers to ‘normal’ hours of work, while the 48-hour standard in the EU Working Time Directive includes both normal hours of work and overtime. This threshold is in line with the findings by Lee in Chapter 2, which concluded that ‘working hours of no less than 50 hours [per week] are neither preferred by most workers, nor desirable from the perspective of health and safety’. 3 The definition of part-time work in most countries is based on the number of hours worked, although in some survey instruments (notably the EU Labour Force Survey) respondents are asked directly if they work part-time. The hours

Working time in the enterprise 193

4

5

6

7 8 9

10 11

12 13

14

15

threshold for defining part-time work varies across countries, but typically refers to work of less than 30 hours per week. However, in some industrialized countries (notably Sweden and the USA), a higher threshold of 35 hours is used to define the range of part-time work. It should be noted that this situation has since changed somewhat with the implementation of the Working Time Regulation 1998, which establishes a maximum workweek of 48 hours, including both normal hours of work and overtime. It should be noted that part-time work in the USA is officially defined as less than 35 hours per week. However, the figure presented here for the USA is for wage and salary employees working less than 30 hours per week. For Japan, the figure presented here is for all workers based on a workweek of 35 hours or less. In this case, part-time is not based upon hours worked, but rather refers to a lesser employment status in the firm. For example, according to Houseman and Osawa (2000), 30 per cent of all Japanese part-time employees worked the same hours per week as full-time workers. This ‘either-or’ categorization is undoubtedly a simplification of reality, as there are a variety of social and cultural factors which shape the degree to which part-time work is considered to be an ‘acceptable’ form of employment. As discussed in note 3, the definitions of part-time work vary across industrialized countries. The intention here is to define part-time work as broadly as possible, so as to consider a wide range of enterprise situations. It should be noted that the country and even the community context play an important role in shaping individuals’ perceptions regarding whether or not part-time work is ‘voluntary’ or ‘involuntary’. Workers are more likely to consider part-time work as ‘voluntary’ if part-tme work is typical in their working environment. However, part-time workers may have access to health insurance and pension coverage through other sources, often through their spouse’s employer. Many categorizations also include part-time work as a type of ‘flexible’ working time arrangement. In this section, however, the focus will be on reviewing alternative methods for organizing any given number of hours, regardless of whether the number of hours corresponds to full-time or part-time work. Many more enterprises which do not have formal flexi-time programmes permit employees to vary their daily hours of work on an informal basis. These statistics refer only to formal flexi-time programmes. In some countries, the proportion of workers who indicated that they had the ability to vary their daily work schedules informally, which is sometimes referred to as ‘informal flexi-time’, is much higher. In the USA, for example, 28.8 per cent of full-time wage and salary workers reported that they had ‘flexible work schedules that allowed them to vary the time that they began or ended work’, while only 11.1 per cent of the workers were in a formal flexi-time programme (USDOL, BLS 2001: 1). The ‘Aubry II’ Law, enacted in 1999, replaced several existing schemes allowing working time flexibility with a single model of annualized hours based on an average 35-hour workweek and a maximum of 1,600 hours of work per year (which takes into account periods of paid leave during the year). For those workers covered by the US Fair Labor Standards Act (FLSA), overtime premia must be paid for any hours worked in excess of the normal 40hour workweek; these ‘hourly’ workers cannot be compensated for extra hours worked by time off in lieu of additional wages. Workers who are not covered by FLSA, who are mostly salaried employees, are not entitled to overtime pay

194

16

17

18

19 20

21

Jon C. Messenger

in any case, and thus there is no financial incentive for firms to implement annualized hours schemes for these workers. There appears to be a significant overlap between flexi-time programmes and time banking accounts in the available data on flexible working time arrangements in Germany. This is perhaps not surprising given the similarities between more expansive forms of flexi-time and time banking accounts. Only if workers’ total hours worked exceed the total contractually agreed hours over the entire reference period would the firm be required to pay workers additional wages. Typically, where collective agreements implement annualized hours schemes, they contain specific provisions concerning additional remuneration to be paid if total hours of work exceed contractually agreed hours and also provisions regarding how to handle a situtation in which total hours worked are less than contractually agreed hours. The question in the European Union Survey asked, ‘Would you be willing to work different hours if you were offered higher wages or additional leisure time?’ regarding the following possible work schedules: early or midday shifts; night shifts; Saturday work; Sunday work; and ‘variable working time’. The term ‘co-determination’ denotes that workers, via the ‘works council’ established within the firm, have the right to make decisions regarding working hours on an equal basis with the firm’s management. More recent empirical studies of the effect of reductions in working time have focused on the reduction of hours of work from a lower baseline level (e.g. 40 hours per week or less). Since productivity gains are likely to be a decreasing function of working time, these studies show weak or no effects in countries in which average hours of work are already relatively short (see, for example, Anxo and Bigsten 1989). Article I, Paragraph 5 of this Recommendation states that: Where the duration of the working week exceeds forty-eight hours, immediate steps should be taken to bring it down to this level without any reduction in the wages of the workers as at the time hours of work are reduced.

6

Implications for working time policies Dominique Anxo, Colette Fagan, Sangheon Lee, Deirdre McCann and Jon C. Messenger

Introduction: the dimensions of ‘decent working time’ The preceding chapters have explored different aspects of individual workers’ needs and preferences with respect to their working time by addressing one kind of ‘decent work deficit’ (ILO 2001a): the gap between the working time schedules which individuals need or would prefer and the ones they are required to work. In adopting this approach, they provide a basis from which to consider how the goal of decent work can be advanced in the arena of working time, allowing a notion of ‘decent working time’ to take shape. Drawing on the discussions presented here, the most significant dimensions of decent working time can be identified: working time arrangements should be healthy; ‘family-friendly’; promote gender equality; advance productivity; and facilitate worker choice and influence over their hours of work. Each of these dimensions of decent working time is analysed in more detail in the remainder of this chapter together, along with the kinds of policy measures which can support them. In fact, as will become clear, a number of the same policy initiatives – quality part-time work, flexi-time and collective reductions in working time, for example – can be employed to advance different dimensions of decent working time. Of course working time-focused initiatives are not the only kind of policy interventions that can advance decent working time – a number of other policy arenas beyond the scope of this report are also relevant. Perhaps most significantly, working time preferences tend to be influenced by income distribution including across the individual’s life cycle. Women’s preferences, for example, can be seen to shift when more affordable child-care is made available, while the reduction of excessively long hours is inevitably as dependent on wage policies as working time measures. The implication for working time policies is that they must be affordable to individual workers if they are to be effective. Moreover, as this book has demonstrated, the goal of advancing gender equality – which has been identified as at the core of the ILO’s decent work agenda (ILO 2000a) – needs to be integrated into all working time

196

D. Anxo et al.

policies if they are to be prevented from entrenching existing gender inequities. However, although the goal of gender equality is shared across industrialized countries, the ways in which it is realized take different forms in different countries depending on the gender ‘contract’(explained by Fagan in Chapter 4) prevalent in each of them. In crafting working time policies, then, sensitivity to the national context is needed, not only with respect to gender equality but also to the relationship between different levels of intervention and the cost-sharing arrangements for various measures among the main actors.

Healthy working time Healthy working time represents a traditional but essential dimension of decent working time, in which workers’ well-being is advanced through addressing the health and safety implications of various working time arrangements, such as long hours of work and night work – especially when these working time patterns, notably night work, are sometimes socially or technically necessary (e.g. hospitals). As discussed in Chapters 2 and 4, long working hours and unsocial working hours (especially night work) are neither preferred nor healthy. It is also important to emphasize that the effects of long hours and ‘unsocial’ working hours such as night work are not confined to individual workers, but also concern their families and the society as a whole (Spurgeon 2003). In some cases, these negative effects, particularly when they are revealed over an extended period of time, are not properly considered by individual workers and employers in determining working hours. For example, young workers may want to work very long hours to earn higher incomes without realizing the longterm consequences for their future health. The presence of these negative externalities suggests the need for regulation against excessively long hours and protective measures concerning ‘unsocial’ working hours. The underlying principle here should be that unhealthy working time should not be used as a tool to improve company profitability, which is a principle that is referred to in the international standards on working time, most significantly the Hours of Work (Industry) Convention 1919 (No. 1) and the Hours of Work (Commerce and Offices) Convention 1930 (No. 30). More recently, this principle has been highlighted in the EU Directive on Working Time of 1993 which states that ‘the improvement of workers’ . . . health at work is an objective which should not be subordinated to purely economic considerations’ (European Industrial Relations Review 1994: 29). Of course, discouraging excessive hours of work and unsocial working hours is not a sufficient condition for healthy working time. In fact, the available evidence suggests that shorter working hours may involve higher work intensity, which can also negatively impact on the mental as well as the physical heath of workers. Such interaction between the length of

Implications for working time policies 197 working time and work intensity needs much more attention, particularly given the indications of work intensification in the 1990s in many industrialized countries (Boisard et al. 2002; Fagan and Burchell 2002; Green and McIntosh 2001). Obviously, however, appropriate policies to protect workers against long and ‘unsocial’ working hours are a necessary condition for healthier working time. It is worthwhile reiterating our finding that weak regulatory and institutional frameworks regarding working time are strongly associated with a higher incidence of long working hours (Chapters 2 and 3). In promoting healthy working time, however, the establishment of a sound regulatory framework is not a panacea. Strong economic incentives towards long hours among certain types of workers can reduce the effectiveness of such regulations. First, these incentives may be considerably higher among low-paid workers when they have to rely on overtime payments as an important source of additional income. If workers’ dependency on overtime work is utilized as part of company cost-saving strategies, longer working hours might be institutionalized while increasing downward pressures on wages. In this respect, the establishment of a minimum wage would be helpful in preventing such a vicious circle between long working hours and low wages, which can be considered a type of ‘race to the bottom’ among low-paid workers. Unhealthy working time concerns also some types of high-paid workers, particularly professionals. In this case the reasons are more complicated, however. For example, long working hours among IT workers appear to be associated with the shortage of skilled workers and the nature of their jobs, as is discussed in Chapter 5 (i.e. for these workers, for a number of reasons, there is no clear demarcation between work and non-work). In other cases, corporate culture, intense competition among colleagues, and management strategies, as well as different institutional frameworks, induce workers to work long hours (Meager et al. 2002). In addition, there are increasing concerns about health and safety among these workers. Both employees and employers across the European Union are reported to be becoming more aware of the dangers involved in stress brought about by unmanageable workloads (European Commission 2002a). In the UK, for example, work-related stress cases (many of which involve accusations against employers of unmanageable workloads brought by unions on behalf of their members) increased twelve-fold, with 6,428 new cases reported in 2000, compared with just 516 in 1999 (Trade Union Technical Bureau 2002). Admittedly, despite the strong concerns about long working hours among professionals and managers and their effects on the personal lives of workers, it is often not easy to address unhealthy working time among these individuals, in part because of insufficient legal support and the strength of an incentive structure which rewards long hours. Nonetheless, it is important to keep their working-hour norm at a reasonable level, and

198

D. Anxo et al.

as close as possible to the national standard. One notable experience in this direction is found in France where the legal regulation of working hours was to a large extent extended to cover managerial and professional staff (cadres), through regulation of the number of days worked (via extended paid holidays) rather than the total number of hours used for other workers (IDS 2002). In some circumstances it may be feasible to incorporate some of these workers into workplace collective agreements, but additional new time ‘accounting currencies’ (Bosch 1999b) and new mechanisms of regulation that centre around work organisation and the volume of hours may be needed. As Fagan argues in Chapter 4, this might include developing better time-accounting and budgeting methods to measure the required input for particular tasks and hence to monitor workloads, combined with incentives and targets for line managers to reorganize work methods and to replace the ‘long hours culture’ with ‘smart working’. Collective negotiations are a very effective method of promoting healthy working time, while taking into account company – or industryspecific – circumstances and the needs of individual workers. As discussed in Chapter 2, higher union density is itself associated with a lower incidence of excessively long working hours. The presence of trade unions can increase the effectiveness of monitoring working time developments in relation to the agreed or legislated norm regarding working time. However, they cannot be a replacement for legislation, especially with respect to unorganized workers. Thus, it is essential to have a legal framework that specifies minimum standards while allowing for flexibility through collective negotiations. Finally, long working hours are largely a male phenomenon, which implies that men have insufficient time for family life. This phenomenon and the resulting gender inequality in the sharing of family responsibilities between men and women are discussed in the following section.

‘Family-friendly’ working time The available empirical evidence (see, for example, Anxo et al. 2002; Gershuny 2000) indicates that the allocation of an individual’s time among various activities (paid work, housework, caring, leisure and other social activities) is very sensitive to the composition of their household (e.g. the number and age of individuals in the household) and also to different phases in the life cycle. In particular, the presence and age of children dramatically affect households’ time allocation and the gender division of labour. Time is a scarce resource and the presence of children – by increasing the time devoted to domestic and caring activities – must therefore affect the time spent on other activities (paid work, leisure, etc.). In Chapter 3, we saw that the gendered impact of children on time allocation varies dramatically across countries and societal contexts. In other

Implications for working time policies 199 words, the ability to combine family commitments and work exhibits a large variation among industrialized countries. A part of this time-consuming activity can be and is to a varying extent already outsourced (e.g. childcare facilities, private and public schooling, domestic help and other market-based household services), yet there is obviously a substantial portion of this activity which remains the responsibility of the family/ household unit. Not all care activities can reasonably be outsourced. The increasingly accepted idea that children have a ‘right’ to be with their parents and also parents’ preferences to spend time taking care of their children themselves are two factors that limit the extent to which childcare activities can be outsourced. Yet, rigidity in working time arrangements, combined with a low provision of child-care and other family-related support facilities, tends to reinforce the traditional ‘male breadwinner–female homemaker’ division of labour and create difficulties in combining paid employment with family responsibilities. In fact, the empirical evidence reported by Anxo in Chapter 3 clearly shows that males’ labour supply, be it in terms of labour market participation or working time, is hardly affected by children, and that the burden of caring and domestic work remains disproportionably the responsibility of mothers – even when mothers are also in paid employment. Of course, being able to fulfil care responsibilities and social activities is not just about how much time is available, it is also about when people are available, and the energy that they have left for the practical and emotional inputs to social relationships and activities outside of the workplace. Intense workloads, non-standard schedules and unpredictable variations in working hours seem to be spreading. Each of these factors increase the probability that people will evaluate their employment as incompatible with family and other activities (as discussed by Fagan, Chapter 4), as well as increasing their reports of work-related illnesses. Where regulations exist to address these issues they usually come under health and safety policy, yet the links with work-family reconciliation should also be drawn into policy formulation. Developments in EU legislation and policy are instructive here. The 1993 EU Working Time Directive draws together standards on health and safety and work–family reconciliation and encourages the social partners to take both into account in their negotiations, although there is no requirement for them to do so. Hence a second important dimension of ‘decent working time’ is providing parents who are engaged in paid employment – both fathers and mothers – with the time that they need to handle their family responsibilities, in line with the ILO’s Workers with Family Responsibilities Convention 1981 (No. 156). Such responsibilities include both caring for family members (not only their children, but also elderly relatives, etc.) and also performing other necessary household tasks (cooking, cleaning, shopping, laundry, etc.). ‘Family-friendly’ working time allows individuals sufficient time to meet their domestic obligations, and is therefore of benefit not

200

D. Anxo et al.

only to the individuals in question, but also to society as a whole. In particular, working time-focused ‘family-friendly’ measures1 (in contrast to other types of work–family initiatives such as the provision of onsite childcare services) address the need for parents to have sufficient time to spend with their families on a daily basis, and thus promote the value of parenting for those workers who want to provide care themselves. The possibility of using working time arrangements to facilitate the combination of work and family commitments may be achieved by different but complementary means. These ‘family-friendly’ working time measures include a variety of different policy options. For example, a collective reduction in full-time hours (which was discussed in the previous section with regard to ‘healthy working time’) makes full-time employment more compatible with raising children and other time-consuming commitments. Although such reductions can be introduced through annualization of hours or an extension of annual leave entitlements, the reference period of weekly hours is also important to provide time for the regular activities and timetables of domestic life. Alternatively, an individual right to reduce working time for family reasons could be established, similar to laws in several countries regarding the right to request changes in working time (discussed below). However, for most parents, the right to reduce working time or take leave to take care of children or other dependants is in practice limited by financial constraints. Therefore, such a right to adapt working time for family reasons would need to be complemented by some form of financial support to help replace at least part of the income foregone. As we have seen throughout this book, part-time work represents a particularly important vehicle for balancing paid employment with personal life, including family responsibilities. However, part-time jobs are, on average, of lesser quality than comparable full-time jobs (wages, nonwage benefits such as pensions – and health insurance in the USA – social insurance coverage, training, career development, etc.). We have also seen that ‘substantial’ part-time hours (20–34 hours) are much more popular than ‘marginal’ part-time hours, and that many full-timers would prefer to switch to part-time work but for the workplace obstacles and career penalties that part-time work currently incurs, as well as the financial constraints which often limit individual workers’ ability to reduce their hours of work. Thus, it is important to improve the quality of available part-time work. Equal treatment regulations in employment, non-wage benefits and social protection systems help to improve the conditions of part-time work. Promoting the equal treatment of part-time and full-time workers is a principle that is enshrined in the ILO’s Part-Time Work Convention 1994 (No. 175), and constitutes a crucial step towards improving the quality of parttime work. In the EU, this principle has been written into law through the adoption of the 1997 Part-Time Work Directive. A complementary mechanism is the introduction of employee entitlements to reduce their

Implications for working time policies 201 hours in their current jobs as part of work-family reconciliation measures, with a right to resume full-time hours in a later period. Entitlements to work reduced full-time or part-time hours, perhaps as an option within extended parental leave schemes, enhance the autonomous flexibility of parents. This option has been available for some time under the Swedish parental leave system and more recently has been introduced with varying degrees of regulated entitlement in a number of other countries, notably in the Netherlands (see the discussion later in this chapter regarding choice and influence regarding working time). Other types of flexible working time arrangements can contribute to the reconciliation of employment and family responsibilities, particularly various forms of flexi-time programmes and ‘time-banking’ accounts, which allow workers to vary their hours of work on a daily basis based on their individual needs. Opportunities to ‘telecommute’ to their jobs by working from home can also provide workers with the ability to tailor the hours that they work on a daily basis. Maternity, paternity, parental and other care-related leave systems provide longer-term working time adjustments for care responsibilities (and physical recovery in the case of maternity leave). Statutory entitlements to maternity, paternity and parental leave have been increased in almost all OECD countries in recent years, and some countries have also introduced measures to reserve periods of leave specifically for fathers (OECD 2001). It should be noted that the compatibility of different working time arrangements with family life are mediated, particularly for mothers, by the availability of pre-school child-care services, school opening hours and after-school care. Similarly, leave arrangements, working time autonomy and care services are important for those with elder care responsibilities. Another important consideration in relation to work–family reconciliation measures is that take-up of extended leave or working time adjustments is often low, and hence there is a large implementation gap between policy and practice. Take-up rates tend to be greatest where there are statutory entitlements rather than voluntary workplace agreements, and in the case of extended leave, where the leave is paid at a reasonably high earnings-replacement rate (Moss and Deven 1999; OECD 2001). In some countries, such as the USA, trade unions have played an important role in the development of workplace work–family policies in the absence of statutory entitlements, but elsewhere collective bargaining has been more limited. Nevertheless, the low take-up rates for work–family reconciliation measures cannot simply be interpreted as indicating that employees do not want to make use of these benefits, as part of the reason may be that there are penalties or obstacles in their way at the workplace. Lewis (1997) has argued that workplace policies on reconciliation measures such as rights to working time adjustments, extended leave or working from home are

202

D. Anxo et al.

often in practice ‘symbolic statements’ that are undermined by wider organizational cultural norms of full-time and long or variable hours of work. For example, employees may lack a sense of genuine entitlement to take up the opportunity to work part-time or take extended leave, perhaps because of the reluctance of their line manager to sanction this arrangement, or an awareness that this will rebound on the workloads of colleagues. Alternatively, employees may be aware that by making such working time adjustments this action will be interpreted by their managers as a lack of commitment to the company and hence their future promotion potential and perhaps even their job security will be compromised. Hence, the development and implementation of ‘family-friendly’ working time measures will require a number of supporting measures at the workplace to ensure that they become embedded, legitimated and used in organizational practices. These supporting measures are also pertinent to the particular problem that take-up rates for these measures are systematically much lower for men than women, and so are discussed in the next section, which focuses on promoting gender equality through working time.

Gender equality through working time Yet another important dimension of ‘decent working time’ concerns the use of working time as a tool for promoting gender equality. The fundamental principle of achieving equality of opportunity and treatment between women and men in the world of work is enshrined in a number of ILO conventions, in particular the ILO’s Discrimination in Employment and Occupation Convention 1958 (No. 111) which establishes as a fundamental principle the elimination of discrimination in respect of employment and occupation (ILO 2000a). The gender equality dimension has two important implications for working time policies. First, there is a need for working time policies that advance the objective of gender equality through gender-neutral measures. Second, there is also a need to ensure that measures that are effective in advancing the other dimensions of decent working time do not have detrimental effects for gender equality. In this sense, the gender equality dimension of ‘decent working time’ is ‘cross-cutting’ because it affects all the other dimensions as well. The promotion of gender equality through working time measures necessitates a combination or ‘portfolio’ of instruments. Many, but not all of these measures, have been discussed above in relation to expanding working time arrangements that better accommodate the rhythms and demands of family life (see above). However, it is important to differentiate between working time measures that are ‘family-friendly’ per se and those that promote gender equality. Some measures that are familyfriendly may enable one parent to allocate more time to child-care care.

Implications for working time policies 203 However, if this parent is always the mother, then this measure may facilitate a balance between work and family responsibilities, but may also simultaneously reinforce gender inequality (e.g. many years of working part-time, long periods of parental leave, etc.) by modifying the traditional ‘male breadwinner’ arrangement only to allow women some marginal forms of labour market participation, rather than paving the way for a more significant transformation. To promote gender equality, measures must enable women to be on an equal footing with men in employment (the level or status of jobs, promotion prospects etc.) and encourage a more equal sharing of the time-consuming aspects of family responsibilities between men and women. This objective can be facilitated by encouraging more men to adjust their working time at different stages in the life course, such as when they have young children. As discussed by Fagan in Chapter 4, a coordinated combination or ‘portfolio’ of policies is required to promote gender equality because the efficacy of one particular instrument is usually contingent upon other measures. For example, policies designed to promote equal treatment between part-timers and full-timers may be undermined if part-time workers are concentrated into a narrow range of occupations where there are few, if any, similar full-time positions with whom comparisons regarding the ‘equal treatment’ of part-time workers could be made. So what are the policy elements that are specific to promoting gender equality over and above that of making employment more compatible with the time-consuming activities of raising children, caring for elders and other family responsibilities? First, policies are needed to close the ‘gender gap’ in hours of work by restricting excessive hours and promoting longer hours for part-time workers. A particularly difficult issue is how to change the ‘long hours culture’ that is often associated with the ‘self-determined’ working time patterns of a growing number of managers and ‘new economy’ professionals who largely fall beyond the scope of traditional working time regulations, as discussed in the first section of this chapter. The long hours in these types of occupations make it difficult for women with care responsibilities to advance, and hence reinforce gender segregation. As was discussed in the previous section on ‘family-friendly’ working time, part-time work is an important means of reconciling employment with other activities. The key issue here is that the quality of part-time work must be enhanced if it is to be used to promote gender equality (OECD 1994; OECD 2001; O’Reilly and Fagan 1998). As discussed earlier, this can be accomplished primarily through measures to promote the equal treatment of part-time workers and full-time workers and also to facilitate movement from full-time to part-time work and vice versa. In addition, however, it is also important to extend the availability of part-time work across all occupations and position levels, so that part-time work does not remain segregated in lower-paid occupations and lower-level jobs, but

204

D. Anxo et al.

instead becomes a viable option across a broad array of occupations and at all levels of the organizational hierarchy. To achieve this objective, measures are needed to increase opportunities for part-time work in a wider range of jobs and to encourage employers to arrange part-time work around ‘substantial’ rather than ‘marginal’ part-time hours. Under these conditions under-employment plus the loss of earnings and career advancement associated with part-time work is reduced, and part-time workers become more integrated alongside fulltimers in the workplace and the wider employment hierarchy (O’Reilly and Fagan 1998). Useful policy measures here may include fiscal reform to remove any cost incentives for creating ‘marginal’ over ‘substantial’ parttime jobs, perhaps in conjunction with regulations to establish minimum hour thresholds. Dissemination of good practice examples can also contribute to increasing the opportunities for part-time work in a wider range of jobs by reducing information and design costs. Another important issue for promoting gender equality involves using working time adjustments to involve men in care work. Currently it is women who make most of the adjustments to their working lives to accommodate care responsibilities. Very few men switch to part-time work or take advantage of other reconciliation measures to adjust their hours, and they use only a small fraction of the available parental and other family-related leave entitlements. This situation creates a ‘Catch-22’ dilemma of work–family reconciliation measures, for these policies are designed to promote women’s employment integration and gender equality, but they may simultaneously serve to reinforce gender inequality in the division of care responsibilities and so undermine the achievement of gender equity in employment (Moss and Deven 1999). Hence, it is important that work–family reconciliation measures incorporate elements that are designed to increase male take-up and hence promote gender equality in the use of these measures. The empirical evidence presented in Chapters 3 and 4 of this book suggests the need for a pro-active policy in order to favour a larger involvement of fathers in both domestic and caring activities. One way to tackle this is to empower fathers by providing them with the same rights as mothers to take leaves of absence for family reasons or reduce their working time when they have young children. Research has shown that male take-up of paternity and parental leave has increased, particularly in the Nordic countries, and that there are a number of factors that contribute to these changes in men’s behaviour (Moss and Deven 1999). The Swedish model of parental leave is one of the most successful in terms of being extensively used by mothers and with a growing take-up by fathers – half of all Swedish fathers now take parental leave for an average of two months, which though still much less than the average of 11 months taken by Swedish mothers, is greater than that found in most other industrialized countries (Haas and Hwang 1999). The policy elements that have con-

Implications for working time policies 205 tributed to raising men’s use of parental leave in Sweden include a high earnings replacement ratio, a non-transferable ‘daddy quota’ of leave, flexible options as to when and how the leave is taken, plus promotional campaigns and target setting which have been underpinned by cross-party support (Moss and Deven 1999). The fact that fathers as well as mothers can face discriminatory treatment when they want to reduce their working time or take leave to accommodate care responsibilities also needs to be addressed. Measures to remove organizational resistance to men adapting their working time would have wider benefits for female colleagues as well by ‘normalizing’ and legitimizing a range of working time arrangements. Managerial resistance and the negative attitudes of colleagues create obstacles to men’s use of reconciliation measures, particularly in male-dominated workplaces where few examples have been set by women’s use of leave. Thus, enterprise level action is needed, first, in the form of information campaigns to raise employees’ awareness of their entitlements. Second, managers need training and information on good practice to skill them in how to plan and rearrange workloads to accommodate these adjustments. Third, targets for take-up rates, perhaps set in conjunction with incentives for the relevant line managers are needed to monitor the situation. Fourth, information on employees’ evaluations of the existing policies should be collected. In other words, a new standard has to be developed, whereby the use of family reconciliation measures is positively valued and encouraged for both men and women. A number of social actors have a role to play in developing this standard, including governments, employers’ organizations and workers’ organizations.

Decent working time as productive working time ’Decent working time’ also has important advantages for enterprises in today’s highly competitive global economy. More and more enterprises are recognizing that listening to the needs and preferences of their employees for a healthy work–life balance isn’t just the right thing to do, but it also makes good business sense. Why? Because promoting work–life balance can serve as an effective competitiveness strategy. Specifically, enterprise policies and practices that improve work–life balance by taking workers’ needs and preferences into account can increase firm productivity in a number of different ways. For example, as was discussed in Chapter 5, there is a substantial body of empirical evidence demonstrating that reductions in ‘excessively’ long hours of work – typically linked with changes in work organization, methods of production and similar factors – have resulted in substantial productivity gains over the years (see, for example, Bosch and Lehndorff 2001). Because the productivity gains connected with reductions in working time tend to decrease as the length of working time decreases, the largest potential gains can be expected from

206

D. Anxo et al.

reductions in very long hours of work (over 48 hours per week or more) – exactly the ‘excessive’ hours of work discussed in this book. In addition, because long hours of work are positively related to absenteeism and staff turnover, shortening these hours can also provide benefits to firms in terms of reduced absenteeism and lower staff turnover (Barmby et al. 2002; Meager et al. 2002). Other empirical research indicates that productivity gains result not only from physiological factors such as reduced fatigue (as in the case of workers who are working long hours on a regular basis), but also from an improvement in employee attitudes and morale, leading to increased productivity and ultimately, higher profits for the firm and its shareholders. As discussed in Chapter 5, flexible working time arrangements such as flexi-time and compressed workweeks have positive effects on employee morale (see, for example, Hogarth et al. 2001). Logically, it makes intuitive sense that workers with better morale would be more productive, and in fact, there is some recent empirical evidence which shows a statistically valid, positive relationship between employee emotions about work and company financial performance (Towers Perrin and Gang & Gang 2003). There are also numerous enterprise-level studies that show the business benefits of adopting policies that promote work–life balance, as was discussed by Messenger in Chapter 5. For example, one study of six large American companies found that 70 per cent of managers and 87 per cent of workers reported that working time arrangements which allowed workers to adjust their work schedules in response to their individual needs, such as flexi-time and telecommuting, had positive effects on productivity (Boston College Center for Work and Family 2000). Flexible working time arrangements, like reductions in very long hours of work, can also result in benefits to firms due to decreased tardiness and absenteeism, as well as serving as an important tool for recruiting (and retaining) new employees, particularly when labour markets are tight (see Avery and Zabel 2001, for a review of the relevant literature). For example, the UK government’s current initiative to promote the work–life balance has identified a large number of British companies that have introduced ‘flexible ways of working which will both help the business and enable staff to balance work and the rest of their lives’ (United Kingdom Department for Education and Employment et al. 2000: 2). Many firms involved with this Work–Life Balance Campaign offer company packages that include a menu of options for flexible working time arrangements (e.g. flexi-time, part-time, telecommuting, etc.) as a recruitment and retention strategy, particularly given a more diverse workforce, and 52 per cent of them reported that these measures helped lower labour turnover, reduce absenteeism, and retain employees, particularly female employees (Hogarth et al. 2001: 233). In addition, the available evidence indicates that, overall, broader work–family reconciliation policies – including childcare, leave and rights to return to work on a part-time basis – increase

Implications for working time policies 207 women’s labour market attachment and, at the enterprise level, the positive effects include better retention rates, reduced absenteeism and productivity gains (OECD 2001). While reducing long hours and adopting various types of flexible working time arrangements can indeed produce important business benefits, it appears that the most important factor is not the working time arrangement per se, but rather it is workers’ ability to choose their arrangement – the exercise of ‘choice’ – that shows the strongest impact on employees’ job performance, and hence on firms’ productivity. For example, as discussed in Chapter 5, a comprehensive study of the effects of flexible work arrangements found that (based on their managers’ ratings) workers who had requested their current work arrangement had higher levels of job performance, and were rated higher in customer, co-worker and manager relations, than workers who were assigned or hired into their work arrangement (Gottlieb et al. 1998). Thus, by adopting enterprise policies that offer workers a range of working time options and allow them the opportunity to choose the arrangement that best meets their individual needs, firms can also reap important business benefits. Finally, the existence of work–life balance policies in an enterprise can also serve as an important indication of harmonious labour relations and corporate social responsibility. This recognition can, in turn, burnish a firm’s image and make them that much more attractive to both potential customers and potential recruits. For example, US firms compete with each other to be recognized as one of the nation’s best companies for the work–family balance by publications such as Working Mother magazine, and actively use this recognition in their public relations efforts in order to promote a positive corporate image. Nevertheless, while policy measures to meet workers’ needs and preferences and hence promote work–life balance have important benefits, there are also costs associated with most of them. These costs may be fairly minimal, as in the case of flexi-time, or quite substantial, as in the case of extended periods of paid leave. The costs may be financial, such as those associated with providing financial support for parents during periods of parental leave, but they may also be non-financial, such as the cost of foregoing the services of a particularly valuable employee for a period of time. Estimates of the costs vary, depending on whether the policy measures are considered as a short-term form of consumption or a longer-term form of investment, and also on whether the costs are calculated from a macroeconomic societal level or from an enterprise perspective (Rubery et al. 2003). However, it must be recognized that the benefits of these measures are not reaped by enterprises alone, and thus the costs cannot be borne by enterprises alone; the costs need to be shared among the different actors, including governments, employers and workers.

208

D. Anxo et al.

Choice and influence regarding working time The focus of this chapter has been on the ‘decent work deficit’ between workers’ preferred and actual working time arrangements. If this deficit is thought of as ‘a gap between the world that we work in and the hopes that people have for a better life’ (ILO 2001a: 8), it suggests that ‘decent working time’ can be promoted by increasing the choice and influence of individuals over their working hours and schedules. This kind of choice can be enhanced and realized in two related ways. First, increasing the number of available working time options is in itself a way of allowing workers a choice over how they arrange their working lives including how they balance paid work with their caring and other commitments. In addition, working time policies can ensure that workers’ preferences are realized by allowing workers the opportunity to exercise a direct influence over the length and arrangement of their working hours. This ‘individual influence’ approach recognizes that decent working time has a procedural as well as a substantive dimension and cannot therefore be determined solely at the macro level, but instead must involve identifying and promoting the substantive outcomes which workers themselves need and prefer. Policy measures that advance worker influence can be adopted at all levels. As discussed by Messenger in Chapter 5, for example, organizations that operate ‘time banking’ accounts can integrate them into annualized hours schemes. Indeed, any firm-level practices that facilitate worker influence over scheduling would be of some benefit to individual workers and can be encouraged through a range of government incentives. At the national level, legislation in Europe that allows collective agreements to implement or adapt working time standards is increasingly being introduced and can enhance worker influence by facilitating the involvement of the social partners in determining the ways in which working time flexibility is introduced (EIROnline 1998). More directly, legislation has been introduced in a number of European countries (e.g. the Netherlands and Germany) throughout Europe that provides individual workers with a right to request working time changes. The strongest version of this kind of legislation permits workers to request both increased or decreased hours and changes in their scheduling; carefully limits the grounds on which working time changes can be refused; and enables individuals to return to their original working time arrangement at a later stage without any career penalties. These kinds of laws are most effective when enacted in conjunction with legislation ensuring equal treatment of part-time workers, which is discussed in the section on gender equality above. Indeed, enacting both kinds of measures may have the potential to avoid the risk of further entrenching the dichotomy between full-time and parttime work which has served to discriminate against women workers, favouring instead a notion of the ‘modulation’ of working hours by facili-

Implications for working time policies 209 tating smoother transitions between different working time arrangements, including over the life-cycle of a worker. As McCann highlighted in Chapter 1, however, there is a need to strike a balance between legislation and collective bargaining in a way that ensures that individual influence over working hours does not detract from strong legal standards. Most significantly, fundamental legal rights can be enacted as a basis for collective bargaining that are not subject to derogation. It is also apparent that individual rights are best realized from a position of collective strength. There is therefore cause for concern over ‘bargaining’ by unrepresentative bodies and by individuals, in line with traditional labour law concerns about the protection of groups with limited bargaining power. As a consequence, in addition to ensuring that nonunionized workers are covered by legislative measures, laws that strengthen unions such as those on independence, recognition and the right to strike, would also strengthen worker influence. This approach, then, would both draw on and advance the ILO’s fundamental principle of freedom of association and the effective recognition of the right to collective bargaining (see ILO 2000b). There is also a need to ensure that collective bargaining is sensitive to the needs and preferences of individual workers, particularly at the enterprise level. In some areas of employment the ‘insiders’ may be predominately male, or young and childless and hence may agree to working time arrangements that contribute to the exclusion of those with care responsibilities from the workplace. The interests of women and of workers with care responsibilities, then, need to be identified and represented. Requiring that the representativity of worker bodies be enhanced and, for example, requiring that both of the above groups are represented on negotiating bodies can help to address this concern. The principle of representativity also demands that different kinds of employers be represented on employer bodies, particularly small employers. Attention should also be directed at ensuring that broader goals can be realized in conjunction with the facilitation of individual choice, including taking account of the need to pursue community values in drafting legislation. The need for preserving time traditionally dedicated to societal and family needs, for example, can be advanced by working time policies that take into account the opening hours of public and private services. Finally, if workers are to be allowed a meaningful choice with respect to their working time arrangements, the measures just outlined must be underpinned by a range of other initiatives. Most obviously, a real choice over working time arrangements cannot be realized without the widespread provision of high quality child-care, as noted above, and also depends on ensuring that individual workers are aware of the options available to them, for example, through policies on information dissemination and the transparency of legal rights.

210

D. Anxo et al.

Conclusion This chapter has shown that there are substantial gaps between, on the one hand, the actual hours of work and working time arrangements that currently exist, and on the other, workers’ needs and preferences regarding their working time. In particular, there is a substantial group of workers with ‘excessively’ long hours in comparison with their preferred hours, and simultaneously, a substantial group of workers whose working hours are substantially shorter than what they would prefer. The chapter also reviewed the implications of these imbalances or ‘gaps’ between workers’ actual and preferred working times, concluding that they have a number of negative impacts on workers’ ability to balance their work responsibilities with other aspects of their lives, such as family commitments, other social activities, and the increasing need for them to enhance their skills via lifelong learning. The result is a ‘decent work deficit’ in the area of working time resulting from these important gaps between workers’ needs and preferences regarding their working time and their actual working time arrangements. Based on these findings, this chapter has proposed a range of possible policies and practices that are designed to close these working time gaps and thus advance the goal of ‘decent working time.’ These measures include suggestions for both national and enterprise-level policies that are designed to promote an improved work–life balance by better reflecting workers’ needs and preferences regarding working time, while meeting enterprises’ business requirements. These policy measures are arrayed along five dimensions: healthy working time; ‘family-friendly’ working time; gender equality through working time; productive working time; and choice and influence regarding working time. Achieving ‘decent working time’ requires a mix of policy measures that simultaneously address all five of these dimensions. Working time arrangements are not set in stone, but reflect the legal and regulatory framework within a given country, the competitive realities in different industrial sectors, and the objectives and choices of the various actors involved, particularly enterprises. In establishing policies and practices that promote ‘decent working time’, it is important to recognize that different types of policies are likely to be required in different environments, and that solutions will need to be customized. Nonetheless, the evidence provided in this report demonstrates that the ‘status quo’ on working time in the industrialized world is out of sync with the needs and preferences of many workers, and that action is needed to find the balance between workers’ needs and preferences and enterprises’ business requirements – and thus make ‘decent working time’ a reality.

Implications for working time policies 211

Note 1 ‘Family-friendly’ measures include a variety of different policies and practices, including not only measures relating to hours of work but also other measures such as maternity protection and affordable, accessible and high-quality childcare services.

Bibliography

ACIRRT (Australian Centre for Industrial Relations Research and Training) (1999) Australia at Work: Just Managing?, Sydney: ACIRRT. Anker, R. (1998) Gender and Jobs: Sex Segregation of Occupations in the World, Geneva: International Labour Organization. Anxo, D. (1999) ‘Working time: research and development’, in Employment and Social Affairs, Industrial Relations and Industrial Change, Brussels: European Commission. Anxo, D. and Bigsten, A. (1989) ‘Working hours and productivity in Swedish manufacturing’, Scandanavian Journal of Economics, 91(3): 613–19. Anxo, D. and Nyman, H. (2001) ‘The retail trade sector in Sweden’, in E. Mermet and S. Lehndorff (eds) New Forms of Employment and Working Time in the Service Economy (NESY): Country Case Studies Conducted in Five Service Sectors, Brussels: European Trade Union Institute. Anxo, D. and O’Reilly, J. (2000) ‘Working time regimes and transitions in comparative perspective’, in J. O’Reilly, I. Cebrián and M. Lallement (eds) Working Time Changes: Social Integration through Transitional Labour Markets, Cheltenham: Edward Elgar Publishing. Anxo, D. and O’Reilly, J. (2002) ‘Working time transitions and transitional labour markets’, in G. Schmid and B. Gazier (eds) The Dynamics of Full Employment, Cheltenham: Edward Elgar Publishing. Anxo, D., Bosch, G., Bosworth, D., Cette, G., Sterner, T. and Taddei, D. (eds) (1995) Work Patterns and Capital Utilization: An International Comparative Study, Dordrecht: Kluwer Academic Publishers. Anxo, D., Flood, L. and Kocoglu, Y. (2002) ‘Offre de travail et répartition des activités doméstique et parentale au sein du couple: une comparaison entre la France et la Suède’, Économie et Statistique, 2: 352–3. Anxo, D., Flood, L. and Rubery, J. (1999) Household Income Distribution and Working Time Patterns: An International Comparison, Gelsenkirchen: Graue Reihe, Des Instituts Arbeit und Technik. Appelbaum, E. and Batt, R. (1995) ‘Alternative models of production’, in The New American Workplace, Ithaca, NY: Industrial Labor Relations Press. Arrowsmith, J. and Mossé, P. (2000) ‘Health care reform and the working time of hospital nurses in England and France’, European Journal of Industrial Relations, 6(3): 283–306. Arrowsmith, J. and Sisson, K. (2001) ‘Working time developments and the quality

Bibliography

213

of work’, EIROnline. Online. Available at HTTP: (accessed 5 December 2001). Atkinson, J. (1984) ‘Manpower strategies for flexible organizations’, Personnel Management, 16 (2): 28–31. Atkinson, J. and Meager, N. (1986) Changing Working Patterns: How Companies Achieve Flexibility to Meet New Needs, London: National Economic Development Office. Australian Bureau of Statistics (1999) ‘Work – paid work: decline of the standard working week’, Australian Social Trends 1999. Online. Available at HTTP: (accessed 13 February 2003). Australian Bureau of Statistics (2002) ‘Work – paid work: employment arrangements’, Australian Social Trends 2002. Online. Available at HTTP: (accessed 13 February 2003). Australian Council of Trade Unions (2003) ‘80% of workers want family friendly laws and a cap on hours’, ACTU National Workplace Survey. Online. Available at HTTP: (accessed 13 February 2003). Australian Department of Employment, Workplace Relations, and Small Business, Work and Family Unit (2000) ACCI Corporate Work and Family Awards 2000 Finalists Booklet, Canberra ACT: Department of Employment, Workplace Relations, and Small Business, Work and Family Unit. Australian Department of Employment, Workplace Relations, and Small Business, Work and Family Unit (2001) ACCI Corporate Work and Family Awards 2001 Finalists Booklet, Canberra ACT: Department of Employment, Workplace Relations, and Small Business, Work and Family Unit. Avery, C. and Zabel, D. (2001) The Flexible Workplace: A Sourcebook of Information and Research, Westport, CT: Quorum Books. Barmby, T., Ercolani, M., and Treble, J. (2002) ‘Sickness absence: an international comparison’, The Economic Journal, 112, June: F315–F331. Barrett, G. and Doiron, D. (2001) ‘Working part time: by choice or by constraint’, Canadian Journal of Economics, 34(4): 1042–65. Baxter, J. (1998) ‘Will the employment conditions of part-timers in Australia and New Zealand worsen?’, in J. O’Reilly and C. Fagan (eds) Part-time Prospects: An International Comparison of Part-Time Work in Europe, North America and the Pacific Rim, London: Routledge. Beechey, V. and Perkins, T. (1987) A Matter of Hours: Women, Part-time Work and Labour Markets, Cambridge: Polity Press. Bell, D. and Hart, R. (1998) ‘Working time in Great Britain, 1975–1994: evidence from the New Earnings Survey panel data’, Journal of the Royal Statistical Society, Series A, 161(3): 327–48. Bell, L. and Freeman, R. (1995) ‘Why do Americans and Germans work different hours?’, in F. Butler, W. Franz, R. Schettkat and D. Soskice (eds) Institutional Frameworks and Labor Market Performance, London: Routledge. Bell, L. and Freeman, R. (2001) ‘The incentive for working hard: explaining hours worked differences in the US and Germany’, Labour Economics, 8: 181–202. Bettio, F. and Villa, P. (1998) ‘A Mediterranean perspective on the break-down of the relationship between participation and fertility’, Cambridge Journal of Economics, 22(2), March: 137–71. Bielenski, H., Bosch, G. and Wagner, A. (2002) Working Time Preferences in

214

Bibliography

Sixteen European Countries, Dublin: European Foundation for the Improvement of Living and Working Conditions. Blossfeld, H.P. and Hakim, C. (eds) (1997) Between Equalisation and Marginalisation: Women Working Part-Time in Europe and the USA, Oxford: Oxford University Press. Bluestone, B. and Rose, S. (2000) ‘The enigma of working time trends’, in L. Golden and D. Figart (eds) Working Time: International Trends, Theory and Policy Perspectives, London: Routledge. Bodin, R.-P. (2001) ‘Wide-ranging forms of work and employment in Europe: review and challenges for the players’, paper presented at the ILO Conference on the Future Work, Employment and Social Protection, Annecy, January. Boisard, P., Carton, D., Gollac, M. and Valeyre, A. (2002) Temps et travail: la durée du travail, Dublin: European Foundation for the Improvement of Living and Working Conditions. Boreham, P. (1992) ‘The myth of Post-Fordist management: work organization and employee discretion in seven countries’, Employee Relations, 14(2): 13–24. Bosch, G. (1995) ‘Synthesis report’, Flexible Working Time: Collective Bargaining and Government Intervention, Paris: Organization for Economic Co-operation and Development. Bosch, G. (1997) ‘Annual working hours: an international comparison’ in G. Bosch, D. Meulders, and F. Michon (eds) Working Time: New Issues, New Measures, New Norms, Brussels: Editions du Dublea. Bosch, G. (1999a) ‘Working time and work organisation: recent trends in working time’, Journal of Human Resource Costing and Accounting, 4(2): 11–26. Bosch, G. (1999b) ‘Working time: tendencies and emerging issues’, International Labour Review, 138(2): 131–49. Bosch, G. and Lehndorff, S. (2001) ‘Working-time reduction and employment: experiences in Europe and economic policy recommendations’, Cambridge Journal of Economics, 25: 209–43. Bosch, G. and Lehndorff, S. (eds) (forthcoming) Working in the Service Sector: A Tale of Different Worlds, London: Routledge. Bosch, G., Dawkins, P. and Michon, F. (1994) Times are Changing: Working Time in 14 Industrialised Countries, Geneva: Institute for International Labour Studies. Bosch, G., Meulders, D. and Michon, F. (eds) (1997) Working Time: New Issues, New Measures, New Norms, Brussels: Editions du Dublea. Boston College Center for Work and Family (2000) Measuring the Impact of Workplace Flexibility: Findings from the National Work/Life Measurement Project. Executive Summary, Boston: Boston College Center for Work and Family, Wallace E. Carroll School of Management. Online. Available at HTTP: http://www.bc.edu/centers/cwf/research/highlights/meta-elements/pdf/flexexecsumm.pdf (accessed 22 March 2002). Boulin, J.-Y. and Hoffman, R. (eds) (1999) Lifetime Working Hours and New Work Organisation, Brussels: European Trade Union Institute. Brown, B.L. (1998) ‘Part-time work and other flexible options’, ERIC Clearinghouse on Adult Career, and Vocational Education, ERIC Digest No. 152. Online. Available at HTTP: (accessed 27 January 2004). Bruegel, I., Figart, D.M. and Mutari, E. (1998) ‘A feminist perspective on full

Bibliography

215

employment and work redistribution’, in J. Wheelock and J. Vail (eds) Work and Idleness: The Political Economy of Full Employment, London: Kluwer. Bruyn-Hundt, M. (1996) ‘Scenarios for a redistribution of unpaid work in the Netherlands’, Feminist Economics, 2(3): 129–33. Burchell, B., Ladipo, D. and Wilkinson, F. (2002) Job Insecurity and Work Intensification, London: Routledge. Campbell, I. (1997) ‘Working-time: comparing Australia and Germany’, in P. James, W. Veit and S. Wright (eds) Work of the Future: Global Perspectives, Sydney: Allen and Unwin. Campbell, I. (1999) ‘Full-time employees and the expansion of unpaid overtime in Australia’, paper presented at the International Symposium on Working Time, Amsterdam, February. Campbell, I. and Brosnan, P. (1999) ‘Labour market deregulation in Australia: the slow combustion approach to workplace change’, International Review of Applied Economics, 13(3): 353–94. Caputo, R. and Cianni, M. (2001) ‘Correlates of voluntary vs. involuntary parttime employment among US women’, Gender, Work and Organization, 8(3): 311–25. Catalyst (1997) A New Approach to Flexibility: Managing the Work/Time Equation, New York: Catalyst. Christensen, E. and Pleman, A. (2001) ‘Work without limits’, paper presented at the Vienna Conference Workshop of PROSA (the Danish trade union for IT workers), Vienna, September. Commission of the European Communities (2003) Communication from the Commission to the Council, the European Parliament, the European Economic and Social Committee and the Committee of the Regions Concerning the Re-examination of Directive 93/104/EC Concerning Certain Aspects of The Organization of Working Time, Brussels: Commission of the European Communities. Online. Available at HTTP (accessed 27 January 2004). Conaghan, J. (2002) ‘Women, work, and family: a British revolution?’, in J., Conaghan, R.M. Fischl and K. Klare (eds) Labour Law in an Era of Globalization: Transformative Practice and Possibilities, Oxford: Oxford University Press. Costa, D. (2000) ‘Hours of work and the Fair Labor Standards Act: a study of retail and wholesale trade, 1938–1950’, American Economic Review, 53(4): 648–64. Cranfield School of Management (1997) Working Time and Contract Flexibility in the EU, Cranfield: Cranfield University. Daune-Richard, A.-M. (1998) ‘How does the “societal effect” shape the use of part-time work in France, the UK, and Sweden?’, in C. Fagan and J. O’Reilly (eds) Part-Time Prospects: An International Comparison of Part-Time Work in Europe, North America and the Pacific Rim, London and New York: Routledge. Dawkin, P. and Baker, M. (1993) ‘Australia’, in G. Bosch, P. Dawkins and F. Michon (eds) Times Are Changing: Working Time in 14 Industrialized Countries, Geneva: International Institute for Labour Studies. Deakin, S. (2002) ‘The Many Futures of the Contract of Employment’, in J. Conaghan, R.M. Fischl and K. Klare (eds) Labour Law in an Era of Globalization: Transformative Practice and Possibilities, Oxford: Oxford University Press. Deiß, M. (1999) Flexible Employment Policies and Working Conditions: Final

216

Bibliography

Report for Germany. Report prepared for the European Foundation for the Improvement of Living and Working Conditions, Munich: Institut für Sozialwissenschaftliche Forschung e.V. Delsen, L. (1998) ‘When do men work part-time?’, in J. O’Reilly and C. Fagan (eds) Part-time Prospects: An International Comparison of Part-Time Work in Europe, North America and the Pacific Rim, London and New York: Routledge. Den Dulk, L. (2001) Work-Family Arrangements in Organisations: A CrossNational Study in the Netherlands, Italy, the United Kingdom and Sweden, Amsterdam: Rozenberg Publishers. Dhondt, S., for the European Foundation (1998) Time Constraints and Autonomy at Work in the European Union, Luxembourg: Office for Official Publications of the European Communities. Dickens, L. (2000) ‘Collective bargaining and the promotion of gender equality at work: opportunities and challenges for trade unions’, in Transfer, 6(2): 193–208. DiNatale, M. (2001) ‘Characteristics of and preference for alternative work arrangements, 1999’, Monthly Labour Review, March: 28–49. Dooley, D., Prause, J. and Ham-Rowbottom, K. (2000) ‘Underemployment and depression: longitudinal relationships’, Journal of Health and Social Behavior, 41: 421–36. Drago, R. (2000) ‘Trends in working time in the US: a policy perspective’, Labor Law Journal, winter: 212–18. Drolet, M. and Morissette, R. (1997) ‘Working more? Working less? What do Canadian workers prefer?’, Research paper series of the Analytical Studies Branch, No. 104, Ottawa: Statistics Canada. Eastman, W. (1998) ‘Working for position: women, men, and managerial work hours’, Industrial Relations, 37(1): 51–66. Edwards, L. and Burkitt, N. (2001) Wanting More from Work? Expectations and Aspirations of People in Low-Paid Jobs, London: UK Department of Education and Employment Research Brief. EIROnline (European Industrial Relations Observatory Online) (1997) Lightening the Load of Long Working Hours among Europe’s Managerial Staff, Dublin: European Foundation for the Improvement of Living and Working Conditions. Online. Available at HTTP: (accessed 6 December 2001). EIROnline (1998) Flexibility of Working Time in Europe, Dublin: European Foundation for the Improvement of Living and Working Conditions. Online. Available at HTTP: (accessed 16 December 2003). EIROnline (2001a) Involuntary Part-time Work Declines, Dublin: European Foundation for the Improvement of Living and Working Conditions. Online. Available at HTTP: (accessed 1 February 2002). EIROnline (2001b) Working Time Developments and the Quality of Work, Dublin: European Foundation for the Improvement of Living and Working Conditions. Online. Available at HTTP: (accessed 27 January 2003). EIROnline (2001c) ‘Progressive Retirement in Europe’. Online. Available HTTP: (accessed 27 January 2004).

Bibliography

217

EIROnline (2001d) Professional Staff Work Substantial Additional Hours, Dublin: European Foundation for the Improvement of Living and Working Conditions. Online. Available at HTTP: (accessed 22 October 2001). EIRR (European Industrial Relations Review) (1999) ‘Flexibility for older workers at Wavin’, European Industrial Relations Review, 306, July: 28–30. EIRR (2000) ‘Working time: an Anglo-Dutch comparison’, European Industrial Relations Review, 312, January: 17–20. Esping-Andersen G. (1990) The Three Worlds of Welfare Capitalism, Cambridge: Polity Press. Esping-Andersen, G. (1999) The Social Foundations of Postindustrial Economies, Oxford: Oxford University Press. European Commission (1999) ‘Tax, working time arrangements and women’s employment in the EU’, Supplement to Equal Opportunities Magazine, No. 7. European Commission (2000a) European Social Statistics: Labour Force Survey Results 1999, Luxembourg: Office for Official Publications of the European Communities. European Commission (2000b) Industrial Relations in Europe, Brussels: Office for the Official Publications of the European Communities. European Commission (2002a) Adapting to Change in Work and Society: A New Community Strategy on Health and Safety at Work 2002–2006, Communication from the Commission, Brussels: COM (2002) 118 final. European Commission (2002b) Employment in Europe 2002, Brussels: European Commission. European Commission, Directorate-General for Employment Industrial Relations and Social Affairs (1998) New Forms of Work Organisation: Case Studies. Final Report, Brussels: European Commission. European Commission’s Group of Experts on ‘Gender and Employment’ (1998) Working Time Patterns in the European Union: Policies and Innovations from a Gender Perspective, Luxembourg: European Commission. European Foundation for the Improvement of Living and Working Conditions (2002) Quality of Work and Employment in Europe: Issues and Challenges, Foundation paper No. 1, Dublin: European Foundation for the Improvement of Living and Working Conditions. European Industrial Relations Observatory (2001) ‘Report evaluates 35-hour week law at half-way stage’. Online. Available at HTTP: (accessed 1 March 2003). European Industrial Relations review (1994) No. 242, March. European Trade Union Institute (2000) Working Time in Europe: A European Working Time Policy in the Public Services, Brussels: European Trade Union Institute. Eurostat (2001) European Social Statistics: Labour Force Survey Results 2000, Luxemburg: Office for Official Publications of the European Communities. Euwals, R. (2001) ‘Female labour supply, flexibility of working hours, and job mobility’, The Economic Journal, 111, C-120–34. Evans, A. (1975) Hours of Work in Industrialised Countries, Geneva: International Labour Office. Evans, J. (2001) ‘Firms’ contribution to the reconciliation between work and

218

Bibliography

family life’, in Labour Market and Social Policy Occasional Papers, No. 48, ref. DEELSA/ELSA/WD(2001)1, Paris: Organisation for Economic Co-operation and Development. Evans, J., Lippoldt, D.C. and Marianna, P. (2001) ‘Trends in working hours in OECD countries’, Labour Market and Social Policy Occasional Papers, No. 45, ref. DEELSA/ELSA/WD(2000)6, Paris: Organisation for Economic Co-operation and Development. Fagan, C. (1996) ‘Gendered time schedules: paid work in Great Britain’, Social Politics, 3(1): 72–106. Fagan, C. (2001a) ‘European Foundation for the Improvement of Living and Working Conditions’, Gender, Employment and Working-Time Preferences in Europe, Luxembourg: Office for Official Publications of the European Communities. Fagan, C. (2001b) ‘Time, money and the gender order: work orientations and working time preferences in Britain’, Gender, Work and Organization, 8(3), July: 239–66. Fagan, C. (2001c) ‘The temporal re-organisation of employment and household rhythm of work schedules: the implications for gender and class relations’, The American Behavioural Scientist, 44(7), March: 1199–212. Fagan, C. and Burchell, B.J. (2002) Gender, Jobs, and Working Conditions in the European Union, Dublin: European Foundation for the Improvement of Living and Working Conditions. Fagan, C. and Lallement, M. (2000) ‘Working time, social integration and transitional labour markets’, in J. O’Reilly, I. Cebrián and M. Lallement (eds) Working-Time Changes: Social Integration through Transitional Labour Markets, Cheltenham: Edward Elgar Publishing. Fagan, C. and O’Reilly, J. (1998) ‘Conceptualizing part-time work: the value of an integrated comparative perspective’, in C. Fagan and J. O’Reilly (eds) Part-time Prospects: An International Comparison of Part-Time Work in Europe, North America and the Pacific Rim, London and New York: Routledge. Fagan, C., Plantenga, J. and Rubery, J. (1995) ‘Part-time work and inequality; lessons from the Netherlands and the UK’, in J. Lapeyre and R. Hoffman (eds) A Time for Work, A Time for Living, Brussels and London: ETUI and Labour Research Department. Fajertag, G. (1999) ‘New paths in working-time policies in Europe: the difficult challenge of reconciling employers’ and employees’ need for flexibility’, Labour and Industry, 9(3): 145–60. Farber, H. (1999) ‘Alternative and part-time employment arrangements as a response to job loss’, Journal of Labor Economics, 17(4): S142–S169. Figart, D.M. and Mutari, E. (2000) ‘Working time regimes in Europe: can flexibility and gender equity coexist?’, Journal of Economic Issues 34(4): 847–71. Fletcher, J. and Rapoport, R. (1996) ‘Work-family issues as a catalyst for organizational change’, in S. Lewis and J. Lewis (eds) The Work-Family Challenge: Rethinking Employment, London: Sage. Fraser, N. (1997) Justice Interruptus: Critical Reflections on the ‘Postsocialist’ Condition, New York and London: Routledge. Fredman, S. (1998) Women and the Law, Oxford: Oxford University Press. Gallie, D., White, M., Cheng, Y. and Tomlinson, M. (1998) Restructuring the Employment Relationship, Oxford: Oxford University Press.

Bibliography

219

Gershuny, J. (2000) Changing Times: Work and Leisure in Post-Industrial Society, Oxford: Oxford University Press. Gershuny, J., Godwin, M. and Jones, S. (1994) ‘The domestic labour revolution: a process of lagged adaptation’, in M. Anderson, M. Bechhofer and J Gershuny (eds) The Social and Political Economy of the Household, Oxford: Oxford University Press. Ghez, G. and Becker, G. (1975) The Allocation of Time and Goods over the Life Cycle, New York: NBER. Ginn, J. and Arber, S. (1998) ‘How does part-time work lead to low pension income?’, in J. O’Reilly and C. Fagan (eds) Part-Time Prospects: An International Comparison of Part-Time Work in Europe, North America and the Pacific Rim, London and New York: Routledge. Golden, L. (2000) ‘Better timing? Work schedule flexibility among US workers and policy directions’, in L. Golden and D. Figart (eds) Working Time: International Trends, Theory and Policy Perspectives, London: Routledge. Golden, L. and Figart, D. (2000) Working Time: International Trends, Theory and Policy Perspectives, London: Routledge. Goss, D. and Adam-Smith, D. (2001) ‘Pragmatism and compliance: employer responses to the working time regulations’, Industrial Relations Journal, 32(3): 195–208. Gottlieb, B., Kelloway, E.K. and Barham, E. (1998) Flexible Work Arrangements: Managing the Work-Family Boundary, Chichester: John Wiley & Sons, Ltd. Goudswaard, A. and de Nanteuil, M. (2000) Flexibility and Working Conditions: A Qualitative and Comparative Study in Seven EU Member States, Dublin: European Foundation for the Improvement of Living and Working Conditions. Green, F. (2001) ‘It’s been a hard day’s night: the concentration and intensification of work in late twentieth-century Britain’, British Journal of Industrial Relations, 39(1): 53–80. Green, F. and McIntosh, S. (2001) ‘The intensification of work in Europe’, Labour Economics, 8: 291–308. Guerrina, R. (2002) ‘Feminist critique of European policies on motherhood and employment’, The European Journal of Women’s Studies, 9(1): 49–68. Haas and Hwang (1999) ‘Parental leave in Sweden’, in P. Moss and F. Deven (eds) Parental Leave: Progress or Pitfall?, Brussels: CVBG-Centrum voor Bevolkingsen Gezinsstudie. Haipeter, T. (2001) ‘The retail trade sector in Germany’, in E. Mermet and S. Lehndorff (eds) New Forms of Employment and Working Time in the Service Economy (NESY): Country Case Studies Conducted in Five Service Sectors, Brussels: European Trade Union Institute. Hakim, C. (1997) ‘A sociological perspective on part-time work’, in H.-P. Blossfeld and C. Hakim (eds) Between Equalization and Marginalization: Women Working Part-Time in Europe and the United States of America, Oxford: Oxford University Press. Hakim, C. (1998) Social Change and Innovation in the Labour Market, Oxford: Oxford University Press. Hakim, C. (2000) Work-Lifestyle Choices in the 21st Century: Preference Theory, Oxford: Oxford University Press. Harkness, S. (1999) ‘Working 9 to 5?’ in P. Gregg and J. Wadsworth (eds) The State of Working Britain, Manchester: Manchester University Press.

220

Bibliography

Hart, R. (1987) Working Time and Employment, London: Allen & Unwin. Heiler, K. (1998) ‘The “petty pilfering of minutes” or what has happened to the length of the working day in Australia?’, International Journal of Manpower, 19(4): 266–80. Heiler, K. and Pickersgill, R. (2001) ‘Shiftwork and rostering arrangements in the Australian mining industry: an overview of key trends’, Australian Bulletin of Labour, 27(1): 20–41. Heiler, K., Pickersgill, R and Briggs, C. (2000) Working Time Arrangements in the Australian Mining Industry: Trends and Implications with Particular Reference to Occupational Health and Safety, International Labour Organization Sectoral Activities Programme Working Paper 162, Geneva: ILO. Hetrick, R. (2000) ‘Analyzing the recent upward surge in overtime hours’, Monthly Labor Review, February: 30–3. Hippo, Y. (1993) ‘Japan: the reduction in working time – an unfinished cultural revolution’, Futures, 25(5): 537–50. Hogarth. T., Hasluck, C., Pierre, G., Winterbotham, M. and Vivian, D. (2001) Work-life balance 2000: Results from the Baseline Study, Norwich: United Kingdom Department for Education and Employment. Horrell, S. and Rubery, J. (1991) Employers’ Hours of Work Survey, London: HMSO. Houseman, S. (2001) ‘Why employers use flexible staffing arrangements: evidence from an establishment survey’, Industrial and Labor Relations Review, 55(1): 149–70. Houseman, S. and Osawa, M. (1998) ‘What is the nature of part-time work in the United States and Japan?’, in J. O’Reilly and C. Fagan (eds) Part-Time Prospects: An International Comparison of Part-Time Work in Europe, North America and the Pacific Rim, London and New York: Routledge. Houseman, S. and Osawa, M. (2000) ‘The growth of nonstandard employment in Japan and the United States: a comparison of causes and consequences’, paper presented at W.E. Upjohn Institute, the Japan Foundation and Japan Women’s University Kalamazoo Conference on Non-standard Employment in Japan, Europe and the United States, Kalamazoo, Michigan, August. Incomes Data Services (IDS) International Department (2001a) ‘Balancing work and family life the Dutch way’, IDS Employment Europe, 477, September: 20–1. IDS International Department (2001b) Pay and Conditions in France, London: Incomes Data Services, Ltd. IDS International Department (2001c) Pay and Conditions in Germany, London: Incomes Data Services, Ltd. Income Data Services (IDS) (2002) ‘France: managers’ hours and the European Social Charter’, Employment Europe, 483, March: 5. International Labour Organization (ILO) (1995) Conditions of Work Digest: Working Time Around the World, 14, Geneva: ILO. ILO, Bureau for Gender Equality (2000a) Decent Work for Women: An ILO Proposal to Accelerate the Implementation of the Beijing Platform for Action, Geneva: ILO. ILO (2000b) Your Voice at Work: Global Report under the Follow-Up to the ILO Declaration on Fundamental Principles and Rights at Work, International Labour Conference, 88th Session 2000, Geneva: ILO.

Bibliography

221

ILO (2001a) Reducing the Decent Work Deficit: A Global Challenge, International Labour Conference, 89th Session 2001, Geneva: ILO. ILO (2001b) World Employment Report 2001, Geneva: ILO. ILO (2002a) Key Indicators of the Labour Market 2001–2, Geneva: ILO. ILO, Sectoral Activities Programme (2002b) The Evolution of Employment, Working Time, and Training in the Mining Industry, Geneva: ILO. Intratest Burke Sozialforschung (1998) ‘Employment Options of the Future – Field Report’, prepared for the European Foundation for the Improvement of Living and Working Conditions. Jacobs, J.A. and Gerson, K. (1998) ‘Who are the overworked Americans?’, Review of Social Economy, 56 (December): 442–59. Jacobs, J.A. and Gerson, K. (2000) ‘Who are the overworked Americans?’, in L. Golden and D. Figart (eds) Working Time: International Trends, Theory and Policy Perspectives, London: Routledge. Jacobs, J.A. and Schmidt, M. (2001) ‘The right to part-time work: the Netherlands and Germany compared’, International Journal of Comparative Labour Law and Industrial Relations, 17(3): 371–84. Jansen, B. and de Jong, M. (1999) ‘Working time organisation in a chain of department stores in the Netherlands’, paper presented at the 7th International Symposium on Working Time, Gelsenkirchen, Germany, February. Jany-Catrice, F. (2001) ‘The retail trade sector in France’, in E. Mermet and S. Lehndorff (eds) New Forms of Employment and Working Time in the Service Economy (NESY): Country Case Studies Conducted in Five Service Sectors, Brussels: European Trade Union Institute. Japan Institute of Labour (2001) The Labour Situation in Japan 2000/2001, Tokyo: Japan Institute of Labour. Japan Institute of Labour (2002) Working Life Profile 2002, Tokyo: Japan Institute of Labour. Japan Ministry of Labour (2000) General Survey on Wages and Working Hours System, Tokyo: Policy Planning and Research Department. Japan Statistics Bureau (2002), Report on the Special Survey of the Labour Force Survey 2001, Tokyo: Japan Statistics Bureau. Jefferys, S. (2000) ‘A “Copernican revolution” in French industrial relations: are the times a’ changing?’, British Journal of Industrial Relations, 38(2): 241–60. Kahn, S. and Lang, K. (1992) ‘Constraints on the choice of work hours: agency versus specific-capital’, Journal of Human Resources, 27(4): 661–78. Kezuka, K. (2000) ‘Legal problems concerning part-time work in Japan’, Japan Labor Bulletin, 39(9): 6–12. Landers, R., Rebitzer, J. and Taylor, L. (1996) ‘Rat race redux: adverse selection in the determination of work hours in law firms’, American Economic Review, 86(3): 329–48. Lehndorff, S. (2000) ‘Working time reduction in the European Union: a diversity of trends and approaches’, in L. Golden and D. Figart Working Time: International Trends, Theory and Policy Perspectives, London: Routledge. Lehndorff, S. (2001) New Forms of Employment and Working Time in the Service Economy NESY: The Case of the Retail Trade Sector, Brussels: European Trade Union Institute. Lewis, S. (1997) ‘ “Family friendly” employment policies: a route to changing

222

Bibliography

organizational culture or playing about at the margins?’, Gender, Work, and Organization, 4(1): 13–23. Lilja, R. and Hämäläinen, U. (2001) Working Time Preferences at Different Phases of Life, Dublin: European Foundation for the Improvement of Living and Working Conditions. Lo, V.I. (1996) ‘Atypical employment: a comparison of Japan and the United States’, Comparative Labor Law Journal, 17: 492–525. Locke, R., Kochan, T. and Piore, M. (1995) ‘Reconceptualizing comparative industrial relations: lessons from international research’, International Labour Review, July, Geneva: ILO. Low Pay Commission (2000) The National Minimum Wage: The Story So Far, London: Stationery Office. Macintyre, S. (1989) ‘Neither capital nor labour: the politics of the establishment of Arbitration’, in S. Macintyre and R. Mitchell (eds) Foundations of Arbitration: The Origins and Effects of State Compulsory Arbitration, 1890–1914, Melbourne: Oxford University Press. Maddison, A. (1995) Monitoring the World Economy, 1820–1992, Paris: OECD. Maric, D. (1977) Adapting Working Hours to Modern Needs, Geneva: International Labour Office. Mata-Greenwood, A. (2001) ‘The hours that we work: the data we need, the data we get’, Bulletin of Labour Statistics, 2001–1, Geneva: ILO. Meager, N., Kodz, J., Davis, S., Lain, D., Sheppard, E., Rick, J., Strebler, M., Bates, P., Cummings, J., Anxo, D., Gineste, S. and Trinczek, R. (2002) Working Long Hours in the UK Executive Summary: A Review of the Research Literature, Analysis of Survey Data and Cross-National Organizational Case Studies, Employment Relations Research Series No. 16. United Kingdom: Department of Trade and Industry. Online. Available at HTTP: (accessed 11 December 2003). Merllié, D. and Paoli, P. for the European Foundation (2001) Third European Survey on Working Conditions, 2000, Luxembourg: Office for Official Publications of the European Communities. Milkman, R. (1998) ‘The new American workplace: high road or low road?’, in P. Thompson and C. Warhurst (eds) Workplaces of the Future, Basingstoke: Macmillan Press Ltd. Moss, P. (1996) ‘Reconciling employment and family responsibilities: a European perspective’, in S. Lewis and J. Lewis (eds) The Work-Family Challenge: Rethinking Employment, London: Sage. Moss, P. and Deven, F. (eds) (1999) Parental Leave: Progress or Pitfall?, Brussels: CVBG-Centrum voor Bevolkings- en Gezinsstudie. Munck, B. (2001) ‘Changing a culture of face time’, Harvard Business Review, November: 125–31. Murray, J. (1999) ‘Social justice for women?: the ILO’s convention on part-time work’, International Journal of Comparative Labour Law and Industrial Relations, 15(1): 3–19. Murray, J. (2001) Transnational Labour Regulation: The ILO and EC Compared, The Hague: Kluwer Law International. National Economic Planning Agency (2001) Reduction of Working Time: Lessons from its Analysis, Paris: National Economic Planning Agency. Neathey, F. and Arrowsmith, J. (2001) Implementation of the Working Time Regu-

Bibliography

223

lations, Employment Relations Research Series No. 11, London: Department of Trade and Industry. Noreau, N. (2000) Longitudinal Aspect of Involuntary Part-Time Employment, Income Statistics Division, Statistics Canada. Oeij, P. and Wiezer, N.M. (2002) New Work Organisation, Working Conditions, and Quality of Work: Towards the Flexible Firm, Dublin: European Foundation for the Improvement of Living and Working Conditions. O’Reilly, J. (1996) ‘Labour market adjustments through part-time work’, in G. Schmid, J. O’Reilly and K. Schömann (eds) International Handbook of Labour Market Policy and Evaluation, Cheltenham: Edward Edgar Publishing. O’Reilly, J. and Bothfeld, S. (2002) ‘What happens after working part-time?: integration, maintenance or exclusionary transitions in Britain and Western Germany’, Cambridge Journal of Economics, 26: 409–39. O’Reilly, J. and Fagan, C. (eds) (1998) Part-Time Prospects: An International Comparison of Part-Time Work in Europe, North America, and the Pacific Rim, London and New York, Routledge. O’Reilly, J., Cebrián, I. and Lallement, M. (eds) (2000) Working Time Changes: Social Integration through Transitional Labour Markets, Cheltenham: Edward Elgar Publishing. OECD (Organisation for Economic Co-operation and Development) (1994) Women and Structural Change: New Perspectives, Paris: OECD. OECD (1995) Flexible Working Time: Collective Bargaining and Government Intervention, Paris: OECD. OECD (1997) ‘Economic performance and the structure of collective bargaining’, OECD Employment Outlook 1997, Paris: OECD. OECD (1998) ‘Working hours: latest trends and policy initiatives’, OECD Employment Outlook 1998, Paris: OECD. OECD (1999) ‘Recent labour market developments and prospects: special focus on the quality of part-time jobs’, OECD Employment Outlook 1999, Paris: OECD. OECD (2001) ‘Work and family life: How do they balance out?’, OECD Employment Outlook 2001, Paris: OECD. OECD (2002) OECD in Figures: Statistics on the Member Countries, Paris: OECD. Ozaki, M. (ed.) (1999) Negotiating Flexibility: The Role of the Social Partners and the State, Geneva: ILO. Paoli, P. and Merllié, D. (2001) Third European Survey on Working Conditions 2000, Luxembourg: Office for Official Publications of the European Communities. Peetz, D. (2001) Individual Contracts, Collective Bargaining, Wages and Power, Discussion Paper No. 437, London: Centre for Economic Policy Research. Perrons, D. (1999) ‘Flexible working patterns and equal opportunities in the European Union: conflict or compatibility?’, The European Journal of Women’s Studies, 6: 391–418. Pfau-Effinger, B. (1998) ‘Culture or structure as explanations for differences in part-time work in Germany, Finland and the Netherlands?’, in J. O’Reilly and C. Fagan (1998) Part-Time Prospects: An International Comparison of Part-Time Work in Europe, North America, and the Pacific Rim, London and New York, Routledge. Phillimore, A.J. (1989) ‘Flexible specialization, work organisation and skills:

224

Bibliography

approaching the “second industrial divide” ’, New Technology, Work and Employment, Oxford: Basil Blackwell Ltd. Piore, M. and Sable, C. (1984) The Second Industrial Divide: Possibilities for Prosperity, New York: Basic Books. Plantenga, J. (1992) ‘Women and work in the Netherlands: some notes about female labour force participation and the nature of the welfare state’, in G. Buttler, H.-J. Hoffmann-Nowotny and G. Schmitt-Rink (eds) Acta Demographica, Heidelberg: Physika. Plantenga, J. (1997) ‘European constants and national particularities: the position of women in the EU labour market’, in A.G. Dijkstra and J. Plantenga (eds) Gender and Economics: A European Perspective, London: Routledge. Plantenga, J. (2002) ‘Combining work and care in the Polder model: an assessment of the Dutch part-time strategy’, Critical Social Policy, 22(1): 53–71. Plantenga, J. and Remery, C. (2001) ‘Information technologies in the Netherlands’, in E. Mermet and S. Lehndorff (eds) New Forms of Employment and Working Time in the Service Economy (NESY): Country Case Studies Conducted in Five Service Sectors, Brussels: European Trade Union Institute. Plantenga, J., Schippers, J. and Siegers, J. (1999) ‘Towards an equal division of paid and unpaid work: the case of the Netherlands’, Journal of European Social Policy, 9(2): 99–110. Polivka, A., Cohany, S. and Hipple, S. (2000) ‘Definition, composition, and economic consequences of the nonstandard workforce’, in F. Carré, M. Ferber, L. Golden and S. Herzenberg (eds) Nonstandard Work: The Nature and Challenges of Changing Employment Arrangements, Champaign, IL: Industrial Relations Research Association. Presser, Harriet B. (1999) ‘Towards a 24-hour economy’, Science, 284: 177–8. Purcell, K., Hogarth, T. and Simm, C. (1999) Whose Flexibility? The Costs and Benefits of ‘Non-Standard’ Working Arrangements and Contractual Relations, York: York Publishing Services/Joseph Rowntree Foundation. Rebitzer, J. and Taylor, L. (1995) ‘Do labor markets provide enough short-hour jobs?: An analysis of work hours and work incentives’, Economic Inquiry, 33: 257–73. Reich, R. (2000) The Secret of Success, New York: Alfred A. Knopf Books. Rengo (the Japanese Trade Union Confederation) (2001a) The Spring Struggle for a Better Life 2001, Rengo White Paper, Tokyo: Rengo. Rengo (2001b) Survey on Flexibility of Working Hours and Changes in Working Styles. Online. Available at HTTP: <www.rengo-soken.or.jp/english/report/ no7/r7.htm> (accessed 27 January 2004). Rubery, J. (1998) ‘Part-time work: a threat to labour standards?’, in C. Fagan and J. O’Reilly (eds) Part-Time Prospects: An International Comparison of PartTime Work in Europe, North America and the Pacific Rim, London and New York: Routledge. Rubery, J. (2002) ‘Gender mainstreaming and gender equality in the EU: the impact of the EU employment strategy’, Industrial Relations Journal, 33(5): 500–22. Rubery, J., Humphries, J., Fagan, C., Grimshaw, D. and Smith. M. (2003) ‘Equal opportunities as a productive factor’, in B. Burchell, S. Deakin, J. Michie and J. Rubery (eds) Systems of Production: Markets, Organization, and Performance, London and New York: Routledge.

Bibliography

225

Rubery, J., Smith, M., Anxo, D. and Flood, L. (2001) ‘The future European labor supply: the critical role of the family’, Feminist Economics, 7(3): 33–69. Rubery, J., Smith, M. and Fagan, C. (1999) Women’s Employment in Europe: Trends and Prospects, London: Routledge. Rubery, J., Smith, M. and Fagan, C. (1998a) ‘National working-time regimes and equal opportunities’, Feminist Economics, 4(1): 71–102. Rubery, J., Smith, M., Fagan, C. and Grimshaw, D. (1998b) Women and European Employment, London: Routledge. Ruivo, M., Gonzalez, M. and Varejao, J. (1998) ‘Why is part-time work so low in Portugal and Spain?’, in J. O’Reilly and C. Fagan (eds) Part-Time Prospects: An International Comparison of Part-Time Work in Europe, North America and the Pacific Rim, London and New York: Routledge. Sato, H. (2001) ‘The Japanese system of working hours: current status and major issues’, ILO Conditions of Work Branch, unpublished paper. Scheuer, S. (1999) ‘The impact of collective agreements on working time in Denmark’, British Journal of Industrial Relations, 37(3): 465–81. Schmidt, M. (2001) ‘The right to part-time work under German Law: progress in or a boomerang for equal employment opportunities?’, Industrial Law Journal, 30(4): 335–51. Schor, J. (1992) The Overworked American: The Unexpected Decline of Leisure, New York: Basic Books. Seifert, A. (1999) ‘Employment protection and employment promotion as goals of collective bargaining in the Federal Republic of Germany’, International Journal of Comparative Labour Law and Industrial Relations, 15(4): 343–63. Seifert, H. (1994) ‘Working-time policy in Japan’, in J. Lapeyre and R. Hoffman (European Trade Union Institute) (eds) A Time for Working, a Time for Living, Brussels: joint publication of the European Trade Union Institute and the Labour Research Department. Sisson, K. and Marginson, P. (2001) ‘ “Soft regulation” – travesty of the real thing or new dimension?’ ESRC ‘One Europe or Several?’ Programme Working Paper 32/01, Brighton: Sussex European Institute. Online. Available at HTTP: (accessed 27 January 2004). Spurgeon, A. (2003) Working Time: Its Impact on Safety and Health, Seoul: ILO and Korean Occupational Safety and Health Research Institute. Supiot, A. (1996) ‘On-the-job time: time for agreement’, International Journal of Comparative Labour Law and Industrial Relations, 2(3): 195–211. Supiot, A. (2001) Beyond Employment: Changes in Work and the Future of Labour Law in Europe, Oxford: Oxford University Press. Supiot, A. (2002) ‘Towards an International Social Order? Preliminary observations on the “new regulations” in work, employment and social protection’, in P. Auer and C. Daniel (eds) The Future of Work, Employment and Social Protection: The Search for New Securities in a World of Growing Uncertainties, Geneva: ILO. Statistics Bureau and Statistics Center, Japan. Online. Available at HTTP: (accessed 15 January 2003). Stewart, M. and Swaffield, J. (1997) ‘Constraints on the desired hours of work of British men’, Economic Journal, 107: 520–35.

226

Bibliography

Stratton, L. (1996) ‘Are “involuntary” part-time workers indeed involuntary?’, Industrial and Labor Relations Review, 49(3): 522–36. Takagi, I. (1993) ‘Japan’, in G. Bosch, P. Dawkins and F. Michon (eds) Times are Changing: Working Time in 14 Industrialised Countries, Geneva: Institute for International Labour Studies. Tijdens, K. (2002) ‘Gender roles and labor use strategies: women’s part-time work in the European Union’, Feminist Economics, 8(1): 71–99. Tilly, C. (1991) ‘Reasons for the continuous growth of part-time employment’, Monthly Labor Review, March: 10–18. Tilly, C. (1996) Half a Job: Bad and Good Part-Time Jobs in a Changing Labor Market, Philadelphia, PA: Temple University Press. Towers Perrin and Gang & Gang (2003) Working Today: Exploring Employees’ Emotional Connections to their Jobs, New York: Towers Perrin. Trade Union Congress (2000) Work Smarter: An End to Burnout Britain, London: TUC dossier. Trade Union Congress (2002) About Time: A New Agenda for Shaping Working Hours, London: TUC. Trade Union Technical Bureau (2002) Stress at Work. Trade Union Technical Bureau Newsletter Special Issue No. 19–20. Online. Available at HTTP: (accessed 11 December 2003). Trejo, S. (1993) ‘Overtime pay, overtime hours, and labor unions’, Journal of Labor Economics, 11(2): 253–78. United Kingdom Department for Education and Employment; United Kingdom Minister for Women; the National Assembly for Wales; and the Scotland Office (2000) Creating a Work-Life Balance: A Good Practice Guide for Employers, Sudbury: United Kingdom Department for Education and Employment, September. USDOL (US Department of Labor) (1999a) Report on the American Workforce, Washington, DC: USDOL. USDOL (1999b) The State of the American Workforce, Washington, DC: USDOL. USDOL, Bureau of Labor Statistics (BLS) (2000) ‘Are managers and professionals really working more?’, Issues in Labor Statistics, May, Washington, DC: USDOL, BLS. USDOL Bureau of Labor Statistics (BLS) (2001a) May supplement, USDL 02–225 Workers on Flexible and Shift Schedules in 2001. Online. Available at HTTP: (accessed 15 January 2003). USDOL, Bureau of Labour Statistics (BLS) (2001b) Workers on Flexible and Shift Schedules in 2001. Summary. USDL 02–225, Washington, DC: USDOL, BLS. Online. Available at HTTP: (accessed 20 February 2003). Visser, J. (2002) ‘The first part-time economy in the world: a model to be followed?’, Journal of Europe Social Policy, 12(1): 23–42. Visser, J. (2003) ‘Negotiated flexibility, working-time and transitions in the Netherlands’, in J. O’Reilly (ed.) Regulating Working-Time Transitions in Europe, London: Edward Elgar Publishing. Voss-Dahm, D. (2001) ‘Information technologies in Germany’, in E. Mermet and S. Lehndorff (eds) New Forms of Employment and Working Time in the Service Economy (NESY): Country Case Studies Conducted in Five Service Sectors, Brussels: European Trade Union Institute.

Bibliography

227

Wailes, N. and Lansbury, R.D. (1999) ‘Collective bargaining and flexibility: Australia’, paper prepared for the ILO Infocus Programme on Social Dialogue, 1999. Online. Available at HTTP: (accessed 27 January 2004). Wajcman, J. (1998) Managing Like a Man, Cambridge: Polity Press. Walsh, J. (1999) ‘Myths and counter-myths: an analysis of part-time female employees and their orientations to work and working hours’, Work, Employment and Society, 13(2): 179–203. Warhurst, C. and Thompson, P. (1998) ‘Hands, hearts, and minds: changing work and workers at the end of the century’, in P. Thompson and C. Warhurst (eds) Workplaces of the Future, Basingstoke: Macmillan Press Ltd. White, M. (1987), Working Hours: Assessing the Potential for Reduction, Geneva: ILO. Wielers, R. and van de Meer, P. (2001) ‘Low educated workers and part-time work: the Netherlands 1973–1991’, unpublished paper. Wooden, M. and Loundes, J. (2001) How Unreasonable are Long Working Hours?, Melbourne Institute Working Paper No. 1/02, Melbourne: Melbourne Institute of Applied Economic and Social Research. Wong, G. and Picot, G. (2001) Working Time in Comparative Perspective, vol. I, Kalamazoo, Michigan: W.E. Upjohn Institute for Employment Research. Yokoyama, K. (2000) ‘Flux time: new work schedules and wage structures are allowing workers flexibility and helping companies motivate – and keep – valuable employees’, The Nikkei Weekly, 26 June: 3.

Index

absenteeism 165, 179–80, 187–90, 206–7 age-related variations in labour market participation 92 Amway Corp. 179 annualized hours 30–1, 171–8, 189, 200, 208 Anxo, D. 165 passim Atkinson, J. 151–2 Aubry laws 19, 21, 24 Australia 11–12, 24, 30, 37–8, 41–4, 48–9, 62–3, 66–7, 72–84, 88–9, 111, 114–18, 124, 128–30, 139, 141, 154–62 Austria 13, 22, 42, 48 Barmby, T. 179 Barrett, G. 51 Belgium 34, 116 Bell, L. 129 Bielenski, H. 6 Bluestone, B. 45, 129 Boots (company) 162 Bosch, G. 128, 172 breast-screening 184–5 Bremer Landesbank (BL) 175–7, 191 Bristol Myers Squibb Co. 179 Canada 40, 51, 111 caring responsibilities 21–2, 110, 117, 138–44, 149, 163, 181, 201, 204–5, 209 child-care arrangements 6, 13–16, 22, 26, 67–8, 72–3, 82–3, 87, 90–1, 95–8, 110–14, 117, 130, 133, 135, 140, 144, 149, 167, 180–1, 195, 199–202, 209 children’s effect on parents’ labour market participation 94–8, 130, 198–9 choice of working time arrangements available to employees 187, 190, 207–9 Christensen, E. 160 CIGNA Corp. 169

co-determination rights 176 collective bargaining 21, 24–7, 34–5, 64, 138, 153–4, 165, 198, 209 community life 26–7 company-level agreements 25–6 competitiveness and competitive advantage 32, 148, 169, 172, 188, 192, 205 ‘compressed’ working week 170, 174, 177, 179, 189, 206 computer software for work scheduling 191 Conaghan, J. 15 core hours 170, 181 core workers 152 Czech Republic 30 ‘decent working time’ concept 7, 9, 139, 195, 199, 202–5, 208–10 decentralization of industrial relations, 24–5, 150 ‘deficits’ in working hours 46–7, 57, 163, 208 Deiß, M. 167 Denmark 12, 14, 22, 24, 32, 35–6, 49, 51, 114–17, 142, 154, 158–60 derogations from legislative provisions 21, 25, 161, 209 developing countries 9 devolved regulatory regimes 25–6 discretionary effort by employees 179–82 discrimination in employment 202, 208 Doiron, D. 51 domestic work 111, 142, 144, 163, 199 econometric analysis 91 economic theory of the labour market 5, 39 educational attainment 93–4

Index 229 employment creation 1, 19, 34–5 Employment Options for the Future Survey 39–40 employment rates 69–74, 111, 142 equality of opportunity and treatment between workers 18–19, 25–6, 92, 98, 138–9, 183, 195–6, 200–3 Esping-Andersen, G. 62 European Court of Justice 11 European Foundation for Living and Working Conditions 152 European Union (EU) 5–6, 15; PartTime Working Directive 17, 139, 200; Working Conditions Survey 47; Working Time Directive 11, 21, 25, 39, 41, 114–15, 162, 171, 191, 196, 199 Euwals, R. 6 evening work 110, 118, 135, 164, 167, 173–4, 178 ‘excessive’working hours 41–5, 52–3, 56, 87, 97, 210 extended leave 201–2

Germany 1, 5, 13, 19–24, 30–1, 35–7, 49, 51, 62–4, 67–8, 71–6, 79–82, 85–6, 89–97, 114, 116, 125, 154, 167, 171–6, 186, 208 Gerson, K. 129 globalization 2, 108, 149, 178 Golden, L. 126, 129 Gottlieb, B. 180 Greece 48, 51, 115–17, 162

Fagan, C. 218 passim family responsibilities 3, 5, 11, 15–16, 19, 23, 27, 76, 96, 98, 117, 125, 135, 144, 155, 163–4, 169, 180, 187, 189, 198–200, 203 Finland 12, 22, 51, 116–17 flexibility in organization of working time 1–3, 11–15, 19–20, 27, 97–8, 108, 148–56, 170–6, 180–3, 187–9; autonomous 110–11, 125–6, 136, 140–1 ‘flexible firm’ model 151–2 flexible specialization 151–2 flexi-time 5, 111, 125–9, 141, 170–3, 177–82, 189–90, 201, 206–7 Ford Australia 181–3, 191–2 Fordism 151–2 framework agreements 21 France 1, 5, 12–13, 19–21, 24, 30–4, 62–4, 67–74, 77–80, 84–5, 89–90, 94, 96, 114–17, 139, 154, 167–8, 171–2, 198 Freeman, R. 129 fringe benefits 168–9

ICA Maxi (retail chain) 165 Iceland 30 image, corporate 207 individuals’ influence over their own working time 21–7, 208 information technology sector 158–60, 197 intensity of work 141, 196–7 International Labour Office (ILO) 4, 7, 9, 10, 195, 209; Discrimination in Employment and Occupations Convention 202; Hours of Work Conventions 196; Part-Time Work Convention 17, 57, 190, 200; PartTime Work Recommendation 22; Reduction of Hours of Work Recommendation 190; Workers with Family Responsibilities Convention 199 Ireland 13, 51, 115–17 Italy 24, 34, 62–4, 67, 72–7, 80–5, 90–6, 116, 125

gap between actual and preferred working time 7–8, 10, 20, 29, 39–57, 61, 163, 195, 208–10 gender equality, promotion of 202–4; see also equality of opportunity and treatment between workers Chapter 4

Haipeter, T. 187 Hakim, C. 47 Harkness, S. 43 health insurance 14, 169 health and safety concerns 4, 10–11, 27, 31, 41, 138, 141, 155, 196, 199 home-working 141, 201 Honeywell International 179 household data on working time 60–2, 78–83 Houseman, S. 14 passim human resource management 150, 188

Jacobs, J. 129 Jany-Catrice, F. 168 Japan 5, 11, 14, 18, 30–4, 41–5, 49, 115–16, 129–30, 139–40, 149, 154–7, 162–3, 171–2, 181–2 job satisfaction 187, 190 job-seekers 131–2 Kapp Ahl (retail chain) 165

230

Index

key performance indicators 192 Korea 30 Kraft Foods 179 legislation on working time 1–5, 12, 17–27, 31–2, 138, 153, 183, 198, 208–9 leisure time 4, 10–13, 135 Lewis, S. 201–2 life-cycle, variations over the 98, 184, 203, 208–9 limitation of working hours 11, 27, 31, 114–15, 139 Locke, R. 150 lone mothers 72–7, 82–3, 94, 108 long hours 79–82, 88–90, 96, 108–9, 113–18, 123–4, 130, 136, 154–62, 179, 188, 195–7, 206–7, 210; culture of 138–41, 181, 198, 203 Lucent Technologies 179 Luxembourg Income Study 62 Maddison, A. 31 managerial jobs 160–2, 198, 203 Maric, D. 4 married couples 14, 84–6, 89, 92–5 Marriott Hotels 139 mass production 151 Matsushita Electric Industries 182 Mexico 30 minimum hours provisions 204 minimum wage provisions 197 mining industry 158–61 ‘modulation’ of working time over life-cycle 98, 208–9 morale 179, 187, 206 Moss, P. 144 motivation of employees 179 Motorola Inc. 179 National Health Service: Northern Trust 183–5 Netherlands 5, 12, 17–19, 22–4, 35–7, 41–2, 48–51, 54–6, 62–4, 67, 69, 72–82, 85–6, 89, 92–8, 114–17, 140, 154, 158, 162, 173–4, 183–4, 201, 208 New Zealand 30, 43, 48–9, 115 night work 11, 108, 118, 123, 135, 156, 164, 173,178, 196 ‘normal’ hours 31–7, 43, 154, 158, 177 Norway 30, 51, 117, 129, 133 Nyman, H. 165 ‘on-call’ commitments 13, 166–7 ‘Opportunity Now’ initiative 142

O’Reilly, J. 164 Organization for Economic Cooperation and Development (OECD) 126–8 Osawa, M. 14 overtime 12–13, 43–5, 155–7, 165, 170–6, 186, 189; unpaid 43–4, 161 over-working 108–9, 115–18 parental leave 5, 22, 66–8, 72, 83, 87, 96, 98, 140–2, 165, 180–1, 201, 204–7 participation in the labour force, determinants of 91–6 part-time work 3–5, 13–18, 22–4, 27, 109, 116–17, 139–40, 162–70, 200–4; high-quality 183–5, 190; involuntary 46–56, 117, 163–4, 167, 170, 189; marginal 46–9, 78–9, 87, 116, 118, 139, 155–6, 163–7, 170, 183, 189, 204 pay differentials 17–18, 87–8, 97 pension funds 184 ‘peripheral’ workers 152 Perrons, D. 124 Phillimore, A.J. 151 Plantenga, J. 158 Pleman, A. 160 policies on working time 136–42, 195; at enterprise level 190–1 Portugal 32–4, 48, 51, 114–17 preferences of workers regarding working time 5–8, 10, 15, 20–1, 25–6, 29, 39–41, 51–2, 129–33, 156, 161–2, 169–70, 173, 178, 181–7, 190–2, 208–10; benefits to businesses of adapting to 178–80, 187 productivity gains 178–80, 188, 205–7 professional jobs 160–2, 198, 203 public relations 207 Purcell, K. 110 quality of life 11, 20 quality of work 52–4, 57, 116, 163, 168, 170, 189, 200, 203 Quelle (company) 175–6 recruitment strategies 206 reductions in working time 19–21, 27, 30, 139 reference groups 54 reference periods 12–13, 27, 171, 200 regulation of working time; changes in techniques for 37–9; levels of 63–4 Remery, C. 158

Index 231 representativeness of negotiators 26, 209 retailing 154, 165, 168–9, 173–4, 186–7 retention of workers see turnover retirement schemes 22, 184 revealed preference 5 Rose, S. 45, 129 Rubery, J. 164 salaried employees 161 scheduling of working hours 118–24, 135–6, 156, 173, 177–8, 183, 186–7, 191 seasonal industries 175 segmentation/segregation in the labour market 13, 57, 109, 117–18, 126, 136–40, 203 self-employment 118, 126 shiftworking 136, 157, 164, 170, 175, 186–7 ‘smart working’ 139, 198 social responsibility 207 social security contributions 13–14, 67, 113 Spain 22, 24, 43, 48, 51, 115–17 Spurgeon, A. 41 ‘standard’ hours 10, 15, 80–5, 109, 123, 136, 170, 176–7, 180 stress at work 160, 182, 192, 197 students 47–9, 116–17, 166–7 Supiot, A. 3 ‘surpluses’ of working hours 41–6, 56–7 Sweden 5, 16, 22, 24, 30, 32, 42, 51, 62–6, 71–83, 86–7, 92–8, 116–17, 125, 133, 140, 142, 154, 165, 175, 201, 204–5 Switzerland 30 taxation 14, 63, 67–8, 87, 144, 204 Taylorism 151 temporary employment 2 Thompson, P. 153 Tilly, C. 46 time-accounting 198 time-banking systems 5, 111, 125, 141, 171–3, 177–8, 189, 201, 208 time-budgeting 138 time-related under-employment 46 time transfer 57

trade unions 1, 25–6, 44, 197–8, 201 trends in working hours 29–30 turnover of staff 179–80, 187–90, 206 ‘twilight’ shifts 123 under-employment 5–6, 46 unemployment 1, 51 United Kingdom 5, 11–13, 16, 21, 24, 37–44, 48–5, 54–6, 62–3, 66–7, 71–84, 88–9, 92–7, 109, 115–17, 140–2, 154, 156–7, 161–2, 165–6, 171, 179, 183–4, 190, 197 United States 5, 11, 14, 30–1, 40–6, 51, 54, 62–3, 66–7, 71–4, 79–84, 88–9, 92–7, 111, 114–18, 123–30, 139–41, 154–7, 161–72, 175, 201, 207 unsocial hours 2–3, 10–13, 108–9, 136, 164, 167, 169, 173–5, 185, 196–7 ‘usual’ working hours 111–13 Voss-Dahm, D. 158, 160 VSFL (company) 175 wage differentials see pay differentials Warhurst, C. 153 weekend working 108, 110, 118, 123, 135, 164, 167, 169, 173–4, 178 welfare state regimes 62–9, 84, 87 women’s employment 3, 5, 13, 16, 49, 138, 149–50, 163 work–life balance 4–7, 16, 19–20, 23, 31, 111, 125, 129–30, 133–5, 138–42, 155, 163, 170, 183, 187, 190–1, 201, 205–7 work organization, changes in 148–54, 172, 179–80 working conditions 47, 165 Working Mother magazine 207 working time accounts 171; see also time-banking Workplace Employee Relations Survey 179 works councils 176–7 Xerox Corporation 139 young workers 49 zero-hours contracts 110, 165–6