Criminality at Work 0198836996, 9780198836995

Table of contents :
cover
half title
Criminality at Work
Copyright
Table of Contents
List of Contributors
1. Criminality at Work: A Framework for Discussion
part 1
2. Workplace Welfare and State Coercion
A. Introduction
1. Social Democracy
2. Why a Social Democratic Perspective?
3. Social Democratic Criminal Law
B. Workers with no Right to Work
C. Physical Safety
1. The Health and Safety at Work Act 1974
2. Corporate Manslaughter
3. Never Safe Enough: The Case of Boxing
D. Pay—​The National Living Wage and Minimum Wages
1. Enforcing the Living and Minimum Wage
2. Employers Evading Paying the National Living Wage: Two Devices, Zero Hours Contracts; Working as an Independent Contractor
3. Employees Colluding with Employers to Evade the National Living Wage
4. Is the National Living Wage High Enough?
3. Using Criminal Law to Enforce Statutory Employment Rights
A. Introduction
B. Charting the Criminal Enforcement of Statutory Employment Rights
C. The Notion of ‘Public Wrongs’ in Criminal Law Theory
1. Introduction
2. Applying the Negative and Positive Versions of Public Wrongs Theory
3. Public Wrongs Theory: Getting from the Positive to the Absolute
D. Conclusion
4. Where Criminal Law Meets Labour Law: The Effectiveness of Criminal Sanctions to Enforce Labour Rights
A. Introduction
B. Defining the Powers
1. Gangmasters and Labour Abuse Authority
2. Employment Agencies and Employment Agency Standards Inspectorate (EASI)
3. National Minimum Wage and the HMRC
4. Working Time and the HSE
5. Office of the Director of Labour Market Enforcement
6. Preliminary Conclusions
C. Complexity of the Legal Response Created
1. Strict Liability Offences
2. Offences Requiring an Element of Mens Rea
3. Breach of Order Offences
4. Civil Liability Offences
5. Summary Conclusions
D. Why Have Some Acts of Non-​compliance Been Criminalized?
1. The Political Choices
E. Effectiveness of Enforcement Action
1. The Nature and Extent of Enforcement
2. The Risk of Enforcement
F. Conclusions
part 2
5. Exploitation at Work: Beyond a ‘Criminalization’ or ‘Regulatory Alternatives’ Dichotomy
A. Introduction
B. The Conceptual Framework
1. Connections between ‘Exploitation’ and ‘Work’
2. What is Exploitation in Work Relations?
3. What are the State’s Obligations?
C. Limitations of Criminal Law Interventions
1. ‘Internal’ Limitations of Criminalization
2. ‘External’ Limitations of Criminalization
D. Criminalization and Regulatory Alternatives: A Key Juncture
E. Multidimensional Regulation: Further Principles and Constraints
F. Conclusion
6. The Duty of Loyalty and the Scope of the Law of Fraud
A. Criminalization as an Additional Deterrent for Civil Wrongs
B. Fraudulent Omissions
1. False or Misleading Representation
2. Fraud by Failure to Disclose Information
3. Fraud by Abuse of Position
C. Relational Contracts
D. Obligations of Disclosure in the Contract of Employment
1. Mutual Trust and Confidence
2. The Duty of Loyalty or Fidelity
E. Duties to Safeguard the Financial Interests of the Other Party
F. Over-​criminalization?
7. Wage Theft as a Legal Concept
A. Introduction
B. What is ‘Wage Theft’?
C. Is ‘Wage Theft’ Theft?
D. Is What the Employee Has Lost ‘Property’ for the Purposes of the Theft Act 1968?
E. Wage Fraud
8. The Criminalization of Workplace Harassment and Abuse:  An Over-​Personalized Wrong?
A. Introduction
B. The Negative Effect of the Criminalization of Workplace Harassment in its Present Form
C. The Over-​personalization of Workplace Harassment as a Criminal Offence
D. Reform Proposals and the Role of the Criminal Law
1. Regulatory Activity at the International Labour Organization
2. Regulatory Activity in the House of Commons Women and Equalities Committee
E. Justifying a Role for Criminalization in Harassment Wrongs: The Role of Human Dignity
1. Employer Duties and the Criminal Law
2. Dignity and the Relationship between Criminal and Civil Law
3. Dignity and the Role of Consent
F. Conclusion: Bargaining in the Shadow of the (Criminal) Law
9. Sex, Work, and Criminalization
A. Introduction
B. Conceptualizing Commercial Sex: From Sex Work to Modern Slavery
1. Commercial Sex
2. Sex Work
3. Commercial Sexual Exploitation
4. Modern Slavery
C. Commercial Sex as Labour: Three Philosophical Issues
1. Commodification of Sex
2. Commercial Sex and Structural Inequalities
3. Commercial Sex and Adaptive Preferences
D. Criminalization and Commercial Sex
1. Criminalization and Commercial Sex: A Normative Account
2. Criminalization and ‘Commercial-​Sex-​as-​Work’
3. Criminalization and ‘Commercial-​Sex-​as-​Exploitation’
4. Criminalization and ‘Commercial-​Sex-​as-​Modern-​Slavery’
5. Criminalization and ‘Commercial-​Sex-​as-​Varied’
E. Conclusion
10. The Work of Sex Work: Prostitution, Unfreedom, and Criminality at Work
A. Introduction
B. The Case for Decriminalization: Safety, Sex, and Work
1. Safety and (De)criminalization
2. Prostitution as Sex, Prostitution as Work
C. Prostitution as Work: Labour and Criminal Law
D. Capitalism, Unfreedom, and Criminal Law
1. Exploitation and Alienation
2. The Continuum of Unfreedom and Class Struggle
3. History, Unfreedom, and Criminality at Work
E. Prostitution as Work and Criminality at Work
F. Conclusion: Prostitution as Work, Work and Freedom
11. Human Rights, Labour Rights, and Criminal Wrongs
A. Introduction
B. Human Rights and Positive Duties
C. Criminalization of Breaches of Labour Rights
1. Severe Labour Exploitation
2. The Modern Slavery Act 2015
3. Collective Labour Rights and Blacklisting
D. Overcriminalization/​Exclusive Focus on Criminalization
E. Conclusion
part 3
12. The Preventive Role of the Criminal Law in Employment Relations
A. Introduction
B. Offences of Omission
C. The Offence of Employing an Illegal Worker
D. The New Preventive Orders
E. Conclusions
13. Licensing of Employing Entities and Criminalization
A. Introduction
B. Licensing
C. Licensing and the Criminal Law
D. Licensing in Employment Settings
1. Gangmasters and Labour Abuse Authority
2. Employment Agencies and Businesses
3. Evaluation
E. Licensing and the Criminal Law in Employment Settings
F. Conclusion
14. Criminalizing Care Workers: A Critique of Prosecution for Ill-​treatment or Wilful Neglect
A. Introduction
B. A Workers’ Crime
C. A Crime against Caring
D. Ontological Confusion
E. Care Worker Prosecution
1. Ill-​treatment and Wilful Neglect in the Courts
2. In the Context of Social Care Work
3. Erasing Employment Relations
F. Problems of Abuse in Care Settings
1. The Organizational Dynamics of Abuse
2. Training and the Regulation of Conduct
3. The Making of Section 20 Criminal Justice and Courts Act 2015
G. Conclusion
15. The Medical Professional as Special before the Criminal Law
A. Introduction
B. The Criminal Law and the Doctor’s Exculpatory and Influential Professional Role
C. The Particular Criminal Liability that the Doctor’s Professional Role Attracts
D. Arguments Supporting the Criminal Law’s Special Treatment
1. Providing Doctors with Special Protection from the Criminal Law is Appropriate Because of the Nature of their Professional Role
2. The Special Attention of the Criminal Law is Required to Protect Patients Because of the Nature of the Doctor–​Patient Relationship
3. The Special Attention of the Criminal Law is Required to Address What are Perceived to be Both Private and Public Wrongs
E. Concluding Thoughts
16. Victim or Perpetrator? The Criminalization of Migration and the Idea of ‘Harm’ in the Labour Market Context
A. Introduction
B. Regulating Immigration—​Migration Control and Migration Status
1. Immigration Law’s Selectivity and Global Inequality of Migration Statuses
2. The Exclusion of Refugees from Regular Migration Opportunities
3. States’ Generation of Illegality
4. Immigration Law Enforcement
C. The Criminalization of Migration
1. Where’s the Harm?
2. Mala Prohibita? Immigration Regulations and Regulatory Crimes
3. Wrongfulness—​Defences for the Blameless—​Refugees and Victims of Trafficking
D. Some Consequences of Criminalization
1. Criminalization Begets Criminalization
2. Criminalization Blocks Regularization
E. Conclusion
17. Doing the Dirty Job: Labour at the Intersections of Criminal Law and Immigration Controls
A. Introduction
B. Criminal Law as an Institution
C. Historicizing Criminalization: Labour Market, Sovereignty, and Criminal Law (Late Eighteenth Century to 1900s)
D. Labour Protection, Citizenship, and Criminalization (early 1900s until 1980s)
E. Neoliberalism, Labour Deregulation, and Criminalization (1980s to Present)
F. Conclusion
18. Modern Slavery, Domestic Work, and the Criminal Law
A. Introduction
B. Background to the Modern Slavery Act 2015
C. Summary of the Legislation
1. The Section 1 Offence
2. Defence for Slavery or Trafficked Victims
3. The Independent Anti-​Slavery Commissioner
4. The Modern Slavery Act: An Assessment
D. Terminology
E. Domestic Labour
1. Feminist Analysis of the Nature of Housework
2. Modern Slavery and Domestic Abuse
3. Slavery, Domestic Abuse, and Patriarchy
F. Conclusion
19. The Persistence of Criminal Law and Police in Collective Labour Relations
A. Introduction
B. The Late Eighteenth and Nineteenth Centuries up to the Conspiracy and Protection of Property Act 1875: The ‘Classical’ Repression Phase
C. From Public Wrongs to Private Rights
D. The Continuing Presence of the Criminal Law
E. From Normal Law to Emergency Regulations
F. From Emergency to Reserve Powers: The Miners’ Strike 1984–​85
G. Trade Unions in the Shadow of ‘Enemy’ Criminal Law
H. Conclusion
part 4
20. Workplace Safety and Criminalization: A Double-​edged Sword
A. Introduction
B. Criminalizing Safety I: Health and Safety Offences
C. Criminalizing Safety II: Corporate Manslaughter
D. The Corporate Manslaughter Offence in Practice
E. Criminalization and Context-​dependency
F. Context-​dependency in Practice
G. Conclusion: Criminalization and the Regulatory Project
21. The Criminalization of Health and Safety at Work
A. Introduction
B. Health and Safety Law and Criminalization
C. The Health and Safety at Work etc Act 1974
D. The New Direction of Recent Policy
E. The Working Time Directive and Regulations
F. Conclusion
22. Accessory Liability for National Minimum Wage Violations in the Fissured Workplace
A. Enforcement in Crisis: The Limits of Primary Liability
B. Enforcement Strategies in the ‘Fissured’ Labour Market
C. The Structure of Criminal Accessory Liability
D. Broadening Liability: Lessons from Australia
E. Conclusion
part 5
23. Class Crimes: Master and Servant Laws and Factories Acts in Industrializing Britain and (Ontario) Canada
A. Introduction
B. The Role and Function of the Criminal Law
C. Master and Servant Law: Disciplining Workers
1. England
2. Canada (Ontario)
D. Factory Acts
1. UK
2. Canada (Ontario)
E. Conclusion
24. Criminalization, Social Exclusion, and Access to Employment
A. Introduction
B. Context, Background, and Themes
C. Use of Criminal Record Information in Employment Decisions
D. Formal Sources of Information on Criminal Backgrounds of Job Applicants
E. Policy and Areas Legislatures Carve Out to Support Certain Criminal Record Checks
1. Protecting Vulnerable Groups
2. Professional Employees
F. Ex-​offenders as Actors in the Labour Market and in Society
G. Employment and Recidivism
H. Employment and Social Inclusion and Exclusion
I. Legal Intervention to Assist the Vulnerable Job Seeker with a Criminal Record
1. Spent Conviction or Clean Slate Legislation
2. Anti-​discrimination Legislation
3. Other Legislative Protections
J. Adequacy of Legislative Intervention
K. Employment, Unfair Dismissal Legislation, and Criminal Behaviour
L. Impact of Informal Sources of Information about a Person’s Conduct on Employment
M. Concluding Comments—​The Need for Continuing Law Reform
25. The Carceral State at Work: Exclusion, Coercion, and Subordinated Inclusion
A. Introduction
B. Subordinated Inclusion: Frameworks and Foils
C. The Carceral State and Labour Market Exclusion
D. Intermediate Steps: Prison Labour, Channelling, and Wage Penalties
E. The Carceral State and Labour Coercion Outside of Prison
F. From Double Binds to Subordinated Inclusion
G. Conclusion
26. Restorative Regulation of Criminality at Work in Canada: Workplace Safety, Penal Law, and Human Capability Enhancement
A. Introduction
B. Penal Law and Worker Protective Regulation: Labour Standards and Occupational Safety
1. Statutory Labour Standards for Worker Protection
2. Earlier Criminal Law Regimes and Workplace Safety
3. The Westray Disaster: A Catalyst for Penal Reform of Workplace Regulation
4. Workplace-​related Reform of the Criminal Code
5. Reform of Specific Occupational Health and Safety Regimes
C. Human Capability Development via Penal Sentencing and Restorative Workplace Approaches
1. Sentencing, the General Part of Criminal Law, and Human Capacity
2. Capabilities Approaches and Labour Market Regulation
3. Sentencing Reform and Capability Enhancement
4. Human Capabilities and Restorative Approaches to Workplace Safety Offences
D. Conclusion
Index

Citation preview

Criminality at Work

Criminality at Work Edited by

ALAN BOGG Professor of Labour Law, University of Bristol

JENNIFER COLLINS Senior Lecturer in Law, University of Bristol

M A R K F R E E D L A N D Q C ( HO N ) ,   F BA Emeritus Research Fellow, St John’s College, Oxford

J O NAT HA N H E R R I N G Professor of Law, University of Oxford

1

3 Great Clarendon Street, Oxford, OX2 6DP, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries © Alan Bogg, Jennifer Collins, Mark Freedland, and Jonathan Herring 2020 The moral rights of the authors have been asserted First Edition published in 2020 Impression: 1 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Crown copyright material is reproduced under Class Licence Number C01P0000148 with the permission of OPSI and the Queen’s Printer for Scotland Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2019951599 ISBN 978–​0–​19–​883699–​5 Printed and bound by CPI Group (UK) Ltd, Croydon, CR0 4YY Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

Foreword It is a great pleasure to introduce this innovative collection which brings labour law and criminal law into productive dialogue in its exploration of how the law interprets and responds to criminality at work. Any collection which brought together such a distinguished group of scholars would in any case be a cause for celebration. In the case of this book, however, there are several more specific reasons to welcome its appearance. First, this book makes a very substantial contribution to the important genre of cross-​ doctrinal scholarship. In stark contrast to much legal practice, and notwithstanding the diversifying impact of the growth of human rights standards stretching across different areas of the common law, the majority of legal scholarship today remains mono-​doctrinal. And in this focus on areas of the common law one at a time, it often achieves depth at the expense of the breadth which is needed to identify issues of significance to law understood not merely as a body of doctrine but as a complex social institution engaged in both regulation and the production of social norms and understandings. Multi-​doctrinal studies such as those attempted in this collection—​studies organized around a social problem or phenomenon rather than formal legal categories—​are in short absolutely central, not only to any genuinely socio-​legal enterprise but to the quality and relevance of legal scholarship more generally. Beyond this advantage, there are further reasons to celebrate a collection which brings together labour law and criminal law in particular. The first of these reasons lies in the central place which master and servant law holds in the historical development of both fields. We need not go as far as Rusche and Kirchheimer’s Marxist view of criminalization and punishment as reflecting underlying structural relations of production to recognize that they function in important ways as labour market institutions. Criminal law has long been used to regulate the powers of both capital and labour; it is time to put this aspect of criminal law at the centre of how we think about its role and functions in the modern administrative state. This in turn implies that the labour law/​criminal law interface is a powerful lens on both a central puzzle and a central conundrum of contemporary criminal law theory, each of which—​like master and servant laws—​holds a central place in Bogg and Freedland’s fine introductory essay. First, the puzzle: The intersection between criminal law and labour law has a central place in any effort to understand the relationship between so-​called ‘mala in se’ and ‘mala prohibita’; ‘real’ or ‘regulatory’ aspects of criminal law. Indeed (as I suspect Kahn-​Freund or Wedderburn would have insisted), master and servant laws and their descendants raise real questions about the integrity of that distinction, given that many offences conventionally understood as ‘mala in se’ have regulatory aspects, and vice versa. Furthermore, the scope and persistence of regulatory criminal laws raise questions about whether the moralized, ‘mala in se’ quality of criminal law so commonly central to contemporary theory—​as powerfully reflected in Duff ’s account on which Bogg and Freedland draw—​is quite satisfactory. Can its impulse to embrace the regulatory offences by reference to a moralized,

vi Foreword ‘mala in se’—​driven paradigm really be justified? Might it be more appropriate to work the other way round, or at least to recognize the equally central place of regulatory offences as aspects of modern criminal law? Second, the conundrum: Criminal law claims to be a form of justice, rather than merely of state power, and hence to have the potential to temper unequal power relations, and to counter forms of oppression and unfairness. This claim, central to the legitimation of modern criminal law, has much plausibility, when we consider that the least powerful are also the most vulnerable to the harms caused by crime. Hence the promise of ‘worker-​ protective’ criminal laws. And the regulatory enforcement potential of criminal law has often made it a tempting means for the pursuit of social justice more generally, in projects of what Aaronson has called ‘progressive criminalisation’. This aspiration to use criminal law as a means of protective regulation for vulnerable labour market actors—​through anti-​ trafficking or ‘modern slavery’ laws, the criminal enforcement of a range of labour rights, or the criminalization of workplace safety breaches or of harassment, to take just a few examples—​is accordingly an important focus of many of the essays in the collection. Criminal law is, however, a double-​edged tool in the pursuit of social justice, and projects of progressive criminalization are accordingly fraught with difficulty, as many of the contributions to this book demonstrate. The most vulnerable are disproportionately liable to not only criminal victimization but also criminalization; and in most systems for which we have enough evidence about enforcement practice to come to a judgment, we can say with confidence that the power-​redistributive efforts of criminal law have, at best, modest success, while, at worst, they can be counterproductive. As in cases such as the attempt to secure justice and a reduction of the harms of human trafficking by criminalizing it, unintended side effects which make life yet harder for victims are rife. (The intersection here with immigration law and policy is of course exemplary of the need for a multi-​doctrinal approach.) Criminal law is, after all, the creature of the same state which has presided over accumulated inequalities of power. In this context, the genuinely progressive aspirations of criminal law reform can, in the process of interpretation and enforcement, turn into something quite different from reformers’ intentions. In the important effort to understand the conditions under which progressive law reform can indeed have a genuine purchase—​not merely towards worker-​protective criminal law but quite generally—​the sort of fine-​grained case studies contained in this collection are indispensable. Their value is enhanced by the way in which this collection is integrated within an overall framework provided by the wide-​ranging, elegant introductory essay, which embraces both normative and aspirational concerns alongside a close attention to the historical development, the socio-​legal context and the practical upshot of the laws in question. I am conscious that this foreword is very much shaped by my perspective as a criminal lawyer, and that a labour lawyer, a public lawyer, or a human rights specialist might well have concentrated on other important aspects of the collection. This simply reflects its richness and breadth. I welcome and celebrate its publication, and trust that it will find the wide and attentive readership that it merits. Nicola Lacey London School of Economics, September 2019

Preface This symposium book is about the Criminal Law in work relations. It had its origins in conversations between us about the possibility of a project of writing which would examine the interface between Labour Law and Criminal Law and, more generally, the whole subject of the role of the Criminal Law in the regulation of work or employment relations. Initial discussions took place in the course of 2015 and 2016, and the fact that this book is due for publication early in 2020 seems to represent quite a rapid period of execution for such a novel and ambitious project. In the course of the period of concentrated activity that this book has required, we have incurred many intellectual, organizational, and personal debts of gratitude which this Preface seeks to identify and to acknowledge. At the outset, our thanks go to the other contributors to this book; and this is not just for producing chapters in a timely way, but for being absolutely integral to the conception of the work as a whole. The formal division between editors and other contributors has been an entirely porous one, masking a reality of continual exchanging of ideas within the group of authors as a whole. For example—​and this is only one of many which could be cited—​Cathryn Costello was quite centrally involved in our initial discussions, which were specially focussed upon the Modern Slavery Act 2015 and the Immigration Act 2016; and the fully general inclusion of all the authors in the design of the book seems to us to have been apparent during and since the workshop which was held in Bristol in June 2018. We have also benefitted hugely from the inputs of other colleagues who have supported us with their inspiration and enthusiasm: first and foremost among them has been Professor Nicola Lacey, whose influence upon our work is evident in many of its chapters, and for whose intellectual generosity, so apparent in the Foreword which she has contributed, we are most especially grateful. Then, we have also other acknowledgements to make which are in part institutional and in part personal. The Universities of Bristol and of Oxford have provided support and facilities for our work: the Bristol Law Faculty generously hosted our workshop, and both universities enabled us to use our research allowances to fund research assistance, most ably and diligently provided by Serena Crawshay-​Williams in Bristol and Cian O’Concubhair in Oxford. We would also like to record our gratitude to the Leverhulme Trust, since Alan Bogg’s Leverhulme Prize money was used to support the Criminality workshop and the provision of editorial assistance in the preparation of the manuscript. Last but by no means least, we are most grateful for the excellent work on the production of our book which has taken place, and is in progress at the time of writing this Preface, at Oxford University Press—​our most sincere thanks go in particular to Jamie Berezin and Catherine Rogers, the commissioning editors at OUP itself; to Arokia Anthuvan Rani, the project manager at Newgen KnowledgeWorks Pvt. Ltd.; and to Libby Holcroft of OUP for the cover design. Alan Bogg Jennifer Collins Mark Freedland Jonathan Herring Bristol and Oxford, January 2020

Table of Contents List of Contributors

xvii

I N T R O D U C T IO N 1. Criminality at Work: A Framework for Discussion Alan Bogg and Mark Freedland

3

PA RT I :   C R I M I NA L I T Y AT WO R K :  M A P P I N G T H E T E R R A I N 2. Workplace Welfare and State Coercion GR Sullivan A. Introduction

35

. Workers with no Right to Work B C. Physical Safety

39 43

D. Pay—​The National Living Wage and Minimum Wages

46

1. Social Democracy 2. Why a Social Democratic Perspective? 3. Social Democratic Criminal Law 1 . The Health and Safety at Work Act 1974 2. Corporate Manslaughter 3. Never Safe Enough: The Case of Boxing

1 . Enforcing the Living and Minimum Wage 2. Employers Evading Paying the National Living Wage: Two Devices, Zero Hours Contracts; Working as an Independent Contractor 3. Employees Colluding with Employers to Evade the National Living Wage 4. Is the National Living Wage High Enough?

3. Using Criminal Law to Enforce Statutory Employment Rights David Cabrelli A. Introduction B. Charting the Criminal Enforcement of Statutory Employment Rights C. The Notion of ‘Public Wrongs’ in Criminal Law Theory 1. Introduction 2. Applying the Negative and Positive Versions of Public Wrongs Theory 3. Public Wrongs Theory: Getting from the Positive to the Absolute

D. Conclusion

4. Where Criminal Law Meets Labour Law: The Effectiveness of Criminal Sanctions to Enforce Labour Rights Catherine Barnard and Sarah Fraser Butlin A. Introduction B. Defining the Powers 1. Gangmasters and Labour Abuse Authority

35 35 37 37 43 43 44 47 48 49 51

53 53 54 57 57 59 62

68

70 70 71 72

x  Table of Contents 2. Employment Agencies and Employment Agency Standards Inspectorate (EASI) 3. National Minimum Wage and the HMRC 4. Working Time and the HSE 5. Office of the Director of Labour Market Enforcement 6. Preliminary Conclusions

74 75 77 78 79

C. Complexity of the Legal Response Created

79

D. Why Have Some Acts of Non-​compliance Been Criminalized?

83

E. Effectiveness of Enforcement Action

88

F. Conclusions

92

1 . Strict Liability Offences 2. Offences Requiring an Element of Mens Rea 3. Breach of Order Offences 4. Civil Liability Offences 5. Summary Conclusions 1. The Political Choices

1 . The Nature and Extent of Enforcement 2. The Risk of Enforcement

79 80 80 81 82 83 88 92

PA RT I I :   L A B O U R W R O N G S A S P U B L IC   W R O N G S 5. Exploitation at Work: Beyond a ‘Criminalization’ or ‘Regulatory Alternatives’ Dichotomy Jennifer Collins A. Introduction B. The Conceptual Framework 1 . Connections between ‘Exploitation’ and ‘Work’ 2. What is Exploitation in Work Relations? 3. What are the State’s Obligations?

97 97 99

99 101 103

C. Limitations of Criminal Law Interventions

103

. Criminalization and Regulatory Alternatives: A Key Juncture D E. Multidimensional Regulation: Further Principles and Constraints F. Conclusion

108 110 115

1 . ‘Internal’ Limitations of Criminalization 2. ‘External’ Limitations of Criminalization

104 106

6. The Duty of Loyalty and the Scope of the Law of Fraud Hugh Collins A. Criminalization as an Additional Deterrent for Civil Wrongs B. Fraudulent Omissions

116

C. Relational Contracts D. Obligations of Disclosure in the Contract of Employment

123 125

. Duties to Safeguard the Financial Interests of the Other Party E F. Over-​criminalization?

129 132

1 . False or Misleading Representation 2. Fraud by Failure to Disclose Information 3. Fraud by Abuse of Position 1 . Mutual Trust and Confidence 2. The Duty of Loyalty or Fidelity

116 119

119 120 121 126 127

Table of Contents  xi

7. Wage Theft as a Legal Concept Sarah Green A. Introduction B. What is ‘Wage Theft’? C. Is ‘Wage Theft’ Theft? D. Is What the Employee Has Lost ‘Property’ for the Purposes of the Theft Act 1968? E. Wage Fraud 8. The Criminalization of Workplace Harassment and Abuse:  An Over-​Personalized Wrong? Alan Bogg and Mark Freedland A. Introduction B. The Negative Effect of the Criminalization of Workplace Harassment in its Present Form C. The Over-​personalization of Workplace Harassment as a Criminal Offence D. Reform Proposals and the Role of the Criminal Law 1 . Regulatory Activity at the International Labour Organization 2. Regulatory Activity in the House of Commons Women and Equalities Committee

134 134 134 135 136 146

151 151 152 156 159 159 161

E. Justifying a Role for Criminalization in Harassment Wrongs: The Role of Human Dignity

163

F. Conclusion: Bargaining in the Shadow of the (Criminal) Law

171

1 . Employer Duties and the Criminal Law 2. Dignity and the Relationship between Criminal and Civil Law 3. Dignity and the Role of Consent

9. Sex, Work, and Criminalization Michelle Madden Dempsey A. Introduction B. Conceptualizing Commercial Sex: From Sex Work to Modern Slavery 1. Commercial  Sex 2. Sex Work 3. Commercial Sexual Exploitation 4. Modern Slavery

164 167 169

173 173 173 174 174 175 176

C. Commercial Sex as Labour: Three Philosophical Issues

177

D. Criminalization and Commercial Sex

181

E. Conclusion

190

1 . Commodification of Sex 2. Commercial Sex and Structural Inequalities 3. Commercial Sex and Adaptive Preferences

1 . Criminalization and Commercial Sex: A Normative Account 2. Criminalization and ‘Commercial-​Sex-​as-​Work’ 3. Criminalization and ‘Commercial-​Sex-​as-​Exploitation’ 4. Criminalization and ‘Commercial-​Sex-​as-​Modern-​Slavery’ 5. Criminalization and ‘Commercial-​Sex-​as-​Varied’

10. The Work of Sex Work: Prostitution, Unfreedom, and Criminality at Work Katie Cruz A. Introduction B. The Case for Decriminalization: Safety, Sex, and Work

177 178 180 182 184 186 187 188

192 192 193

xii  Table of Contents 1 . Safety and (De)criminalization 2. Prostitution as Sex, Prostitution as Work

194 196

. Prostitution as Work: Labour and Criminal Law C D. Capitalism, Unfreedom, and Criminal Law

198 200

. Prostitution as Work and Criminality at Work E F. Conclusion: Prostitution as Work, Work and Freedom

206 209

1 . Exploitation and Alienation 2. The Continuum of Unfreedom and Class Struggle 3. History, Unfreedom, and Criminality at Work

11. Human Rights, Labour Rights, and Criminal Wrongs Virginia Mantouvalou A. Introduction B. Human Rights and Positive Duties C. Criminalization of Breaches of Labour Rights 1 . Severe Labour Exploitation 2. The Modern Slavery Act 2015 3. Collective Labour Rights and Blacklisting

. Overcriminalization/​Exclusive Focus on Criminalization D E. Conclusion

200 203 204

210 210 211 213

214 217 218

224 228

PA RT I I I :   T H E C O N T E M P O R A RY SHA P E O F C R I M I NA L I Z AT IO N P R AC T IC E S :  R I SK , S TAT U S , A N D C HA R AC T E R I N T H E N E O L I B E R A L C R I M I NA L   L AW 12. The Preventive Role of the Criminal Law in Employment Relations Jennifer Collins and Andrew Ashworth A. Introduction B. Offences of Omission C. The Offence of Employing an Illegal Worker D. The New Preventive Orders E. Conclusions

231

13. Licensing of Employing Entities and Criminalization ACL Davies A. Introduction B. Licensing C. Licensing and the Criminal Law D. Licensing in Employment Settings

249

1 . Gangmasters and Labour Abuse Authority 2. Employment Agencies and Businesses 3. Evaluation

. Licensing and the Criminal Law in Employment Settings E F. Conclusion

14. Criminalizing Care Workers: A Critique of Prosecution for Ill-​treatment or Wilful Neglect LJB Hayes A. Introduction B. A Workers’ Crime

231 231 234 240 247

249 250 251 254

255 257 260

263 265

267 267 269

Table of Contents  xiii C. A Crime against Caring D. Ontological Confusion E. Care Worker Prosecution

270 273 274

F. Problems of Abuse in Care Settings

279

G. Conclusion

289

1 . Ill-​treatment and Wilful Neglect in the Courts 2. In the Context of Social Care Work 3. Erasing Employment Relations 1 . The Organizational Dynamics of Abuse 2. Training and the Regulation of Conduct 3. The Making of Section 20 Criminal Justice and Courts Act 2015

15. The Medical Professional as Special before the Criminal Law Suzanne Ost A. Introduction B. The Criminal Law and the Doctor’s Exculpatory and Influential Professional Role C. The Particular Criminal Liability that the Doctor’s Professional Role Attracts D. Arguments Supporting the Criminal Law’s Special Treatment

1. Providing Doctors with Special Protection from the Criminal Law is Appropriate Because of the Nature of their Professional Role 2. The Special Attention of the Criminal Law is Required to Protect Patients Because of the Nature of the Doctor–​Patient Relationship 3. The Special Attention of the Criminal Law is Required to Address What are Perceived to be Both Private and Public Wrongs

E. Concluding Thoughts

16. Victim or Perpetrator? The Criminalization of Migration and the Idea of ‘Harm’ in the Labour Market Context Cathryn Costello A. Introduction B. Regulating Immigration—​Migration Control and Migration Status

1 . Immigration Law’s Selectivity and Global Inequality of Migration Statuses 2. The Exclusion of Refugees from Regular Migration Opportunities 3. States’ Generation of Illegality 4. Immigration Law Enforcement

274 276 277 282 284 287

291 291 292 296 301 301 302 302

304

309 309 311 311 312 314 315

C. The Criminalization of Migration

316

D. Some Consequences of Criminalization

324

E. Conclusion

326

1 . Where’s the Harm? 2. Mala Prohibita? Immigration Regulations and Regulatory Crimes 3. Wrongfulness—​Defences for the Blameless—​Refugees and Victims of Trafficking 1 . Criminalization Begets Criminalization 2. Criminalization Blocks Regularization

17. Doing the Dirty Job: Labour at the Intersections of Criminal Law and Immigration Controls Ana Aliverti A. Introduction

316 319 322 324 325

327 327

xiv  Table of Contents . Criminal Law as an Institution B C. Historicizing Criminalization: Labour Market, Sovereignty, and Criminal Law (Late Eighteenth Century to 1900s) D. Labour Protection, Citizenship, and Criminalization (early 1900s until 1980s) E. Neoliberalism, Labour Deregulation, and Criminalization (1980s to Present) F. Conclusion

18. Modern Slavery, Domestic Work, and the Criminal Law Jonathan Herring A. Introduction B. Background to the Modern Slavery Act 2015 C. Summary of the Legislation 1 . The Section 1 Offence 2. Defence for Slavery or Trafficked Victims 3. The Independent Anti-​Slavery Commissioner 4. The Modern Slavery Act: An Assessment

329 331 334 337 341

343 343 343 344

345 347 348 349

D. Terminology E. Domestic Labour

350 354

F. Conclusion

361

1 . Feminist Analysis of the Nature of Housework 2. Modern Slavery and Domestic Abuse 3. Slavery, Domestic Abuse, and Patriarchy

19. The Persistence of Criminal Law and Police in Collective Labour Relations Alan Bogg, KD Ewing, and Andrew Moretta A. Introduction B. The Late Eighteenth and Nineteenth Centuries up to the Conspiracy and Protection of Property Act 1875: The ‘Classical’ Repression Phase C. From Public Wrongs to Private Rights D. The Continuing Presence of the Criminal Law E. From Normal Law to Emergency Regulations F. From Emergency to Reserve Powers: The Miners’ Strike 1984–​85 G. Trade Unions in the Shadow of ‘Enemy’ Criminal Law H. Conclusion

355 358 360

362 362 364 368 371 375 379 383 386

PA RT I V:   C R I M I NA L I Z AT IO N A N D E N F O R C E M E N T 20. Workplace Safety and Criminalization: A Double-​edged Sword Paul Almond A. Introduction B. Criminalizing Safety I: Health and Safety Offences C. Criminalizing Safety II: Corporate Manslaughter D. The Corporate Manslaughter Offence in Practice E. Criminalization and Context-​dependency F. Context-​dependency in Practice G. Conclusion: Criminalization and the Regulatory Project

391 391 392 395 399 401 403 406

Table of Contents  xv

21. The Criminalization of Health and Safety at Work Michael Ford A. Introduction B. Health and Safety Law and Criminalization C. The Health and Safety at Work etc Act 1974 D. The New Direction of Recent Policy E. The Working Time Directive and Regulations F. Conclusion 22. Accessory Liability for National Minimum Wage Violations in the Fissured Workplace Alan Bogg and Paul S Davies A. Enforcement in Crisis: The Limits of Primary Liability B. Enforcement Strategies in the ‘Fissured’ Labour Market C. The Structure of Criminal Accessory Liability D. Broadening Liability: Lessons from Australia E. Conclusion

409 409 412 417 421 425 429

431 431 433 440 445 449

PA RT V:   C OM PA R AT I V E P E R SP E C T I V E S O N C R I M I NA L I Z AT IO N 23. Class Crimes: Master and Servant Laws and Factories Acts in Industrializing Britain and (Ontario) Canada Eric Tucker and Judy Fudge A. Introduction B. The Role and Function of the Criminal Law C. Master and Servant Law: Disciplining Workers 1. England 2. Canada (Ontario)

455 455 456 458 458 463

D. Factory Acts

464

E. Conclusion

472

1. UK 2. Canada (Ontario)

464 469

24. Criminalization, Social Exclusion, and Access to Employment Marilyn J Pittard A. Introduction B. Context, Background, and Themes C. Use of Criminal Record Information in Employment Decisions D. Formal Sources of Information on Criminal Backgrounds of Job Applicants E. Policy and Areas Legislatures Carve Out to Support Certain Criminal Record Checks

474

. Ex-​offenders as Actors in the Labour Market and in Society F G. Employment and Recidivism

481 484

1 . Protecting Vulnerable Groups 2. Professional Employees

474 474 477 479 479 480 481

xvi  Table of Contents . Employment and Social Inclusion and Exclusion H I. Legal Intervention to Assist the Vulnerable Job Seeker with a Criminal Record

486

J. Adequacy of Legislative Intervention K. Employment, Unfair Dismissal Legislation, and Criminal Behaviour L. Impact of Informal Sources of Information about a Person’s Conduct on Employment M. Concluding Comments—​The Need for Continuing Law Reform

491 491

1 . Spent Conviction or Clean Slate Legislation 2. Anti-​discrimination Legislation 3. Other Legislative Protections

25. The Carceral State at Work: Exclusion, Coercion, and Subordinated Inclusion Noah D Zatz A. Introduction B. Subordinated Inclusion: Frameworks and Foils C. The Carceral State and Labour Market Exclusion D. Intermediate Steps: Prison Labour, Channelling, and Wage Penalties E. The Carceral State and Labour Coercion Outside of Prison F. From Double Binds to Subordinated Inclusion G. Conclusion 26. Restorative Regulation of Criminality at Work in Canada: Workplace Safety, Penal Law, and Human Capability Enhancement Bruce P Archibald QC A. Introduction B. Penal Law and Worker Protective Regulation: Labour Standards and Occupational Safety

1 . Statutory Labour Standards for Worker Protection 2. Earlier Criminal Law Regimes and Workplace Safety 3. The Westray Disaster: A Catalyst for Penal Reform of Workplace Regulation 4. Workplace-​related Reform of the Criminal Code 5. Reform of Specific Occupational Health and Safety Regimes

487

488 488 489

493 495

496 496 497 500 503 505 508 513

516 516 517

517 518 520 521 523

C. Human Capability Development via Penal Sentencing and Restorative Workplace Approaches

524

D. Conclusion

533

1 . Sentencing, the General Part of Criminal Law, and Human Capacity 2. Capabilities Approaches and Labour Market Regulation 3. Sentencing Reform and Capability Enhancement 4. Human Capabilities and Restorative Approaches to Workplace Safety Offences

Index

524 526 528 530

535

List of Contributors Ana Aliverti is Reader in Law at the University of Warwick. Paul Almond is Professor of Law at the University of Reading. Bruce P Archibald, QC, is a Professor Emeritus at the Dalhousie University Schulich School of Law, Halifax, Canada. Andrew Ashworth, CBE, QC (Hon), FBA, is Emeritus Vinerian Professor of English Law in the University of Oxford, an Emeritus Fellow of All Souls College Oxford, and an Adjunct Professor at the University of Tasmania. Catherine Barnard is Professor of EU law and Employment law at the University of Cambridge and Senior Tutor of Trinity College, Cambridge. Alan Bogg is Professor of Labour Law at the University of Bristol, and an Emeritus Fellow of Hertford College, Oxford. David Cabrelli is Professor of Labour Law at the University of Edinburgh. Hugh Collins, FBA, is Cassel Professor of Commercial Law at the London School of Economics and was formerly Vinerian Professor of English Law at All Souls College, University of Oxford. Jennifer Collins is Senior Lecturer in Law at the University of Bristol. Cathryn Costello is Andrew W Mellon Professor of Refugee and Migration Law at the Refugee Studies Centre, University of Oxford, and the Professor II at the Norwegian Centre for Human Rights, Faculty of Law, University of Oslo. Katie Cruz is Senior Lecturer in Law at the University of Bristol. ACL Davies is Professor of Law and Public Policy at the University of Oxford, and Dean of the Oxford Law Faculty. Paul S Davies is Professor of Commercial Law at University College London, and an Associate Member of Maitland Chambers. KD Ewing is Professor of Public Law at King’s College London. Michael Ford, QC, is a Professor of Law at the University of Bristol and a barrister at Old Square Chambers. Sarah Fraser Butlin is a Fellow of Selwyn College, Cambridge, an Affiliated Lecturer in Labour law at the University of Cambridge, and is a barrister at Cloisters Chambers.

xviii  List of Contributors Mark Freedland, QC (Hon), FBA, is Emeritus Professor of Employment Law in the University of Oxford, an Emeritus Research Fellow of St John’s College Oxford, and an Honorary Professor in the Faculty of Laws of University College London. Judy Fudge, FRSC, is LIUNA Enrico Henry Mancinelli Professor of Global Labour Issues at the School of Labour Studies, McMaster University, Canada. Sarah Green is Professor of Private Law at the University of Bristol and Law Commissioner for Commercial and Common Law. LJB Hayes is Professor of Law at Kent Law School. Jonathan Herring is Professor of Law at the University of Oxford, and DW Wolfe-​Clarendon Fellow in Law at Exeter College, Oxford. Michelle Madden Dempsey is the Harold Reuschlein Scholar Chair and Professor of Law at Villanova University, USA. Virginia Mantouvalou is Professor of Human Rights and Labour Law at University College London, Faculty of Laws. Andrew Moretta graduated from King’s College London with an LLM in Labour Law in 2012 and completed his PhD at Liverpool University in 2019. Suzanne Ost is Professor of Law at Lancaster University. Marilyn J Pittard is Professor of Law at Monash University. GR Sullivan is Emeritus Professor of Law at University College London. Eric Tucker is a Professor at Osgoode Hall Law School, York University, Toronto, and a Distinguished Scholar in Residence at the Cleveland-​Marshall College of Law, Cleveland State University. Noah D Zatz is Professor of Law at the University of California, Los Angeles.

 

IN T RODU CT ION

1

Criminality at Work A Framework for Discussion Alan Bogg and Mark Freedland

The purpose of this introductory chapter is to develop a framework for understanding the lessons of a project of over three years’ duration, and to distil what has been learned in the course of the project. The project was undertaken in the conviction that the area of intersection between criminal law and labour law would be an interesting one to investigate, both for labour lawyers and for criminal lawyers. This had been prompted by two important recent Acts of the UK Parliament, the Modern Slavery Act 2015 (MSA) and the Immigration Act 2016 (IA), the former of which introduced a new offence criminalizing modern slavery practices and the latter of which made it a criminal offence to employ an irregular migrant or to work as an irregular migrant. We had a general but imprecise idea of what the outcomes of that investigation might be. The intuition that this was an area ripe for investigation has been very fully vindicated. We emerge from the inquiry with some conclusions about the significance of ‘criminality at work’ which are more exact than, and sometimes different from, our initial hypotheses. This introductory chapter describes a voyage of discovery about the functions of criminal law in the sphere of labour or employment relations: of how these functions have shifted and been reconfigured during different historical periods; and of how normative theories of the criminal law often diverge from the political uses to which the criminal law has been put by elected governments from time to time. From a historical perspective, the criminal law was often used as a regulatory technique for regulating labour markets. It had been used in repressive ways, especially in the nineteenth-​century Master and Servant Acts and Combination Acts, and by the use of public order offences to restrain trade union activity and industrial action. Criminal law in its historically repressive guise has generated a rich and sophisticated literature, especially at the intersection of legal history and labour history.1 However, at the interface between criminal law and labour law, contemporary scholarship reveals a most surprising gap. For example, in a recent work on the ‘autonomy of labour law’, the general interaction between labour law and criminal law is left untouched in a volume that exhaustively explores labour law’s encounter with private law, contract law, fiduciary law, EU law, international law, discrimination law, corporate and company law, public law, migration law, and human rights law.2 1 See, eg, Douglas Hay and Paul Craven (eds), Masters, Servants, and Magistrates in Britain and the Empire, 1562–​1955 (Chapel Hill 2004); Robert J Steinfeld, Coercion, Contract, and Free Labor in the Nineteenth Century (Cambridge UP 2001). 2 Alan Bogg, Cathryn Costello, ACL Davies, and Jeremias Prassl (eds), The Autonomy of Labour Law (Hart 2015). Alan Bogg and Mark Freedland, Criminality at Work In: Criminality at Work. Edited by Alan Bogg, Jennifer Collins, Mark Freedland QC, and Jonathan Herring, Oxford University Press (2020). © Alan Bogg, Jennifer Collins, Mark Freedland, and Jonathan Herring. DOI: 10.1093/​oso/​9780198836995.003.0001

4  Criminality at Work This gap is all the more remarkable in view of the intensive engagement between these two bodies of law at the current time.3 It is worth reflecting briefly upon why there might be such an oversight in the existing literature, given the otherwise prodigious scholarly activity in the fields of labour law and criminal law and theory. Starting with labour law, it is appropriate to begin with two ‘classics’ of modern labour law scholarship in the post-​war period. Here we see that the labour lawyer’s scepticism about the criminal law is deeply embedded in the canon. We focus on the treatment of the criminal law as a technique of labour market regulation in Otto Kahn-​Freund’s Hamlyn Lectures, Labour and the Law,4 and Lord Wedderburn’s The Worker and the Law.5 While both scholars are understandably critical of the repressive function of the criminal law in enforcing labour discipline and suppressing trade union activity in the nineteenth century, they are also very sceptical of any positive role for the criminal law in supporting worker-​ protective norms in the modern era. In Labour and the Law, Kahn-​Freund’s discussion of the criminal law is located principally within c­ hapter 2 on ‘Sources of Regulation’. Quite naturally, for a labour law treatise, the entrée for Kahn-​Freund’s analysis is the penal Master and Servant legislation. Kahn-​ Freund characterized this legislation as providing auxiliary sanctions for enforcing contractual promises in the employment contract: workers were rarely defendants in civil claims for breach of contract brought by employers for ‘they are not worth powder and shot’.6 In regulatory terms, civil enforcement was not cost-​effective for employers. The criminal law functioned as a substitute enforcement mechanism to maintain labour discipline. The criminal law’s repressive function is understood from an instrumental enforcement perspective, with criminal penalties identified as an alternative way of securing compliance with contractual obligations. As worker-​protective norms were introduced through legislation, for example in relation to factory safety, the criminal law was also used as the principal enforcement mechanism here. Despite the worker-​protective provenance of the criminal law, Kahn-​Freund was very sceptical of the criminal law’s value as an enforcement tool. For example, the use of criminal penalties to enforce the employer’s duty to provide employees with written particulars under the Contracts of Employment Act 1963 attracted the withering assessment that it was a ‘clumsy penal sanction’.7 Furthermore, Kahn-​Freund expressed the hope that individual civil enforcement in the tribunals ‘will increasingly replace the circuitous and ineffective enforcement of social standards through criminal proceedings’.8 Effective and speedy enforcement needed to be channelled through individual civil proceedings through the tribunal system, superseding the criminal law as a ‘clumsy’, ‘circuitous’, and ‘ineffective’ regulatory system.

3 It is important to acknowledge the exceptional scholarship on the criminalization of migration statuses and modern slavery discourse, which is of obvious importance and relevance to the wider theoretical concerns of this volume. See, eg, Ana Aliverti, ‘The Wrongs of Unlawful Immigration’ (2017) 11 Criminal Law and Philosophy 375; Judy Fudge, ‘Illegal Working, Migrants and Labour Exploitation in the UK’ (2018) 28 Oxford Journal of Legal Studies 557. 4 Paul L Davies and Mark R Freedland (eds), Kahn-​Freund’s Labour and the Law (3rd edn, Stevens 1983) 5 Lord Wedderburn, The Worker and the Law (3rd edn, Penguin 1986). 6 Davies and Freedland (n 4) 35. 7 ibid 50. 8 ibid 51.

A Framework for Discussion  5 There is a similar tone of engagement in Lord Wedderburn’s The Worker and the Law. Thus, Wedderburn acknowledges the oppressive role of the criminal law under the Master and Servant legislation, and the use of this legislation to penalize trade unionists for leaving or neglecting their work.9 The third edition was published in 1986, and the rediscovery of the criminal law in policing the Miners’ Strike was also noted by Wedderburn.10 In this way, the criminal law’s repressive qualities were no longer a mere historical footnote. Like Kahn-​Freund, Wedderburn was also very sceptical about the worker-​protective role of the criminal law. For example, the use of penal sanctions to enforce the written statement in the Contracts of Employment Act 1963 was criticized in the following terms: ‘But who was going to prosecute? Nothing illustrated better the absence in Britain of any general “labour inspectorate” of the French kind. After two years this sanction was displaced, with an embarrassed shuffle, by a civil remedy.’11 There is also a nuanced discussion of the role of criminal prosecutions in the enforcement of Health and Safety laws.12 Wedderburn identifies the importance of public resourcing for enforcement agencies, and the role of discretion in pursuing prosecutions. The Inspectorate is described as adopting ‘an unduly gentle prosecution policy’.13 This was reflected in the rarity of prosecutions on indictment, and the low level of financial penalties secured from offenders in the magistrates’ courts.14 While some of this was attributable to the restriction of public funding for enforcement, Wedderburn offered an extremely sophisticated diagnosis of its deeper roots. In part, it reflected ‘the “conventionalization” of factory crimes, removing the social stigma’ of this kind of offending.15 We think this ‘conventionalization’ thesis identifies an issue of central importance at the borderland of criminal law and labour law, and it is the role of crimes mala prohibita in specifying and enforcing labour standards.16 We will return to the notion of crimes mala prohibita shortly, and to the debate about whether these are ‘real’ crimes. We should also observe that Wedderburn is sceptical about the potential of corporate criminal liability to secure safe and healthy working practices, advocating instead in favour of prosecution ‘against the human beings responsible’.17 Both Kahn-​Freund and Wedderburn adopt a ‘regulatory’ perspective on criminalization: the overriding focus is on outcomes and whether criminalization promotes compliance with labour standards. For example, Kahn-​Freund once observed that . . . most legislation operates not by the lesson it teaches or the sermon it preaches but by the promise of rewards or the threat of deprivations attached to its observance or breach, that is, by the expectation of its enforcement. Legal norms have their social effect through legal sanctions.18



9

Wedderburn (n 5) 141. ibid  91–​93. 11 ibid 139. 12 ibid 420–​21. 13 ibid 421. 14 ibid 420. 15 ibid. 16 For discussion, see Antony Duff, The Realm of the Criminal Law (OUP 2018) 20–​22. 17 Wedderburn (n 5) 421. 18 Otto Kahn-​Freund, Selected Writings (Stevens 1978) 128. 10

6  Criminality at Work This perspective tends towards a reductive and flattened account of the differences between legal norms, calibrated as a function of severity of sanction and likelihood of enforcement. Wedderburn’s approach is more nuanced. Thus, he is careful to distinguish different types of legal wrongdoing, noting that in criminal wrongs ‘the acts done are regarded as wrongs against society generally, and the sanction is punishment, usually a fine or imprisonment’.19 This is an important recognition that there is a moral dimension to the designation of certain wrongs as criminal, that they have a quality as public wrongs,20 and that the choice between different legal responses to public wrongs (eg crime, tort, administrative regulation) should not simply be based upon an assessment of whether enforcement outcomes will be improved.21 Of course, Wedderburn’s work is not specifically addressed to criminalization, nor the distinctive modalities of criminal law and the criminal justice process in responding to public wrongs. Yet his work is at least useful in identifying this as a potential area where the labour law perspective might be enriched through a dialogue with criminal theory. Furthermore, Wedderburn’s negative assessment of worker-​protective criminalization is far from categorical. For example, the evaluation of criminal law in The Worker and the Law is based upon a critical socio-​legal examination of the existing state of public enforcement. Yet if there was a properly funded labour inspectorate that pursued a vigorous enforcement strategy, and if punishment was sufficiently severe and targeted at human agents so as to prompt organizational changes, then a more positive evaluation of the criminal law might be expected to follow. By contrast, Kahn-​Freund’s reflections on the role of the criminal law seem more pessimistic. What is missing from both accounts is an exploration of the full range of complex regulatory interactions between criminal law and other forms of legal regulation. Criminal law might provide a substitute for civil enforcement. Or it might operate as an auxiliary mode of enforcement of labour standards, in parallel with individual rights of enforcement. Or the criminal law might be reserved for certain persistent or egregious breaches of labour standards. The current and diverse landscape of ‘criminality at work’ highlights how the traditional labour law canon was ill-​equipped to map these complex regulatory typologies. This lacuna in labour law scholarship is matched by a lacuna in the developing scholarship in criminal law and theory. In part, at least, we think that this is attributable to a prevailing idea of criminal regulation of labour dominated by the looming historical presence of penal Master and Servant legislation. According to Nicola Lacey, this legislation was ‘an early example of a hybrid civil/​administrative/​criminal genre’.22 Despite the importance of these statutes in enforcing labour discipline, sometimes through harsh penalties such as whipping and imprisonment with hard labour, there was perhaps a tendency to regard these offences as distinct from ‘real’ crimes at the core of the criminal law.23 The one-​sided imposition of penal sanctions on workmen but not their masters for breaches of the employment contract is such a flagrant departure from basic legal equality that criminal lawyers today are perhaps inclined to treat it as an embarrassing episode in the early history of industrial capitalism. Given its morally aberrant quality, it may be disregarded as an area of enquiry that has little relevance to the questions animating contemporary criminal law scholarship.

19

Wedderburn (n 5) 3. Duff (n 16) ch 7. 21 This is one of the central preoccupations of Duff ’s work, ibid. 22 Nicola Lacey, In Search of Criminal Responsibility: Ideas, Interests, and Institutions (OUP 2016) 43. 23 ibid 44. 20

A Framework for Discussion  7 This organizing idea of ‘real’ criminal law is also relevant to the expansion of ‘regulatory’ crimes concerned with factory conditions and industrial health and safety. As Jeremy Horder has argued, over the course of the nineteenth century there was a significant expansion of the administrative state in different spheres of social and economic life, and this was often reflected in criminalization practices conducted by public inspectorates.24 These ‘regulatory’ crimes were often marginalized from the canon of ‘real’ criminal law, which was treated as constituted by wrongs mala in se.25 According to Horder, the theoretical position of regulatory crime, which may involve ‘strict’ liability imposed on corporate agents, continues to be normatively contested. This reflects wider anxieties about ‘over-​criminalization’ and conflicts between ‘regulatory’ crime and liberal values of autonomy.26 Since many modern labour market offences occupy this contested terrain, viewed perhaps as regulatory infractions that do not really ‘belong’ in an appropriately demarcated criminal law, they have often been marginalized in the leading treatises on general criminal law. Our volume therefore represents an attempt to bring these two disciplines into dialogue with each other, at a time when that dialogue is urgently needed. It is useful to remind ourselves what our initial hypotheses were. They represented what most labour lawyers might have been likely to say if they had chosen to reflect on the criminal law as a form of labour market regulation. In the interests of full disclosure, we confess that these hypotheses were also shaped by a rather negative view of the criminal law’s likely impact on labour market outcomes, especially for the most precarious workers. The following propositions represented our main starting hypotheses for the project: (i) whereas criminal law had over a long historical period been in retreat from the sphere of employment relations, it was now resurging into labour market regulation, especially in recent statutory interventions such as the MSA and IA; (ii) this resurgence was part of a wider phenomenon of ‘over-​criminalization’ driven by the politicization of criminal justice practices;27 (iii) that much of this ‘over-​criminalization’ was of dubious legitimacy either in targeting non-​wrongful statuses (such as migration statuses) or in using preventive measures to criminalize conduct on the basis of risk of harm (eg licensing regimes for labour market intermediaries or preventive ‘labour market enforcement orders’ under the IA); (iv) that criminalization was increasingly driven by ‘thick’ moral judgements about vice and bad character of defendants, especially in the context of modern slavery, which has tended to obscure wider appreciation of the structural determinants of precariousness and vulnerability to abuse; (v) ‘worker protective’ criminalization, attaching criminal penalties to certain labour law violations, appeared to be a new and paradoxical feature of the neoliberal governance of deregulated labour markets, and this paradox required further investigation and explanation; (vi) certain types of labour wrong, such as blacklisting of trade unionists or the coercive trafficking, appear to constitute serious public wrongs warranting justified criminalization, so how should we identify the category of ‘labour wrongs as public wrongs’ as a basis for reasonable legislative choices to criminalize labour wrongs? And on what basis

24 Jeremy Horder, ‘Bureaucratic “Criminal” Law: Too Much of a Bad Thing?’ in RA Duff, Lindsay Farmer, SE Marshall, Massimo Renzo, and Victor Tadros (eds), Criminalization: The Political Morality of the Criminal Law (OUP 2014) 101. 25 ibid 103. 26 ibid 130–​31. 27 The literature on ‘overcriminalization’ in general criminal law is very significant. See, eg, Douglas Husak, Overcriminalization: The Limits of the Criminal Law (OUP 2008).

8  Criminality at Work should we opt for criminal law rather than private law, statutory rights enforced in the civil courts and tribunals, or professional regulation? Our initial task in designing our project was therefore to survey the whole field of legal governance of labour and employment relations through criminal law intervention. We were deliberately inclusive in this search, extending it beyond the well-​recognized territories of regulation of industrial action and enforcement of health and safety norms, and even beyond the new realms of modern slavery and illegal immigration. We have also included in our study the use of criminal law to enforce the labour standards embodied in the National Minimum Wage legislation and the Working Time Regulations, the use of the criminal law to regulate specific occupations and forms of work, and to support the licensing regimes applying to gangmasters and to employment agencies, and the regulatory activity of the Director of Labour Market Enforcement under Part I of the IA. We also sought contributions on the criminalization of harassment and bullying in the workplace, ‘wage theft’, and human rights violations. On the face of it, this deliberately inclusive approach to the scope of criminality at work might have seemed to produce an unsatisfactory outcome. It could be regarded as a miscellany of disparate legal curiosities which was not going to give rise to an important and interesting study of ‘criminality at work’. In this regard, is this not like a study of ‘criminality on the railways’ or ‘criminality as it relates to swimming pools’?28 The results of the study support a more positive assessment of the intellectual potential of ‘criminality at work’ as an interesting and important intersection. The catalogue we have assembled is sufficiently large and important as to suggest that the impulse towards criminalization in the legal governance of work relations has become strong and pervasive. In the course of this study, we have identified certain thematic continuities across these diverse contexts of work criminalization. Furthermore, it is a form of governance that is not restricted to the UK, as our comparative chapters indicate. ‘Criminality at work’ has become essential to an understanding of the general trajectory and dynamics of labour law and the governance of work. In turn, the sphere of work relations has become one of the key locations for understanding the modern development of criminal law. The first step in establishing a dialogue between labour law and criminal law is to interrogate the distinction between crimes mala in se and crimes mala prohibita. The idea of mala in se relates to crimes that are regarded as tracking pre-​legal wrongs such as murder, rape, and interpersonal violence. The contrasting notion of mala prohibita relates to crimes where the relevant conduct is not perceived as wrongful independently of its legal regulation.29 Examples might include criminal liability for undertaking certain employing activities without having obtained a relevant licence, as under the Gangmasters and Labour Abuse Authority regulatory framework. We have already suggested that this division of offences may have contributed to the obscuring of ‘criminality at work’ as a site of regulatory activity. This is because the crimes in ‘criminality at work’ have tended to be seen as mala prohibita offences. Often, neither the 28 The reference to ‘swimming pool law’ is Brian Langille’s: see Brian Langille, ‘Labour Law’s Back Pages’ in Guy Davidov and Brian Langille (eds), Boundaries and Frontiers of Labour Law (Hart Publishing 2006) 15–​16. The example is used to explore why there might be better reason to assemble legal norms in certain contexts (medical law, labour law) than others (swimming pool law, hairdressing law). ‘Criminality at work’ raises the problem of disciplinary demarcation afresh. 29 Duff (n 16) 20–​21.

A Framework for Discussion  9 criminal lawyers nor the labour lawyers regarded these as ‘real’ crimes involving stigma, public censure, and criminal punishment. The designation was fraught with the implication that crimes mala in se were real crimes which generate authentic and enduring law, whereas mala prohibita were wrongs more ephemerally criminalized by that part of the legal system which governs by fiat or decree. It will come as no surprise to learn that in the Anglo-​American legal tradition, mala in se are often regarded as originating in the timeless common law, while mala prohibita are treated as the product of primary or secondary legislation enacted in response to shifting political priorities. In the historical development of the criminal law, significant consequences were also thought to attach to the differentiation between mala in se and mala prohibita. For crimes mala in se, it was essential to recognize and give effect to the requirement of mens rea, a guilty state of mind, as a pre-​condition of valid conviction for the offence in question. On the other hand, there was a greater readiness to view mala prohibita offences as the subject of strict liability. Furthermore, mala in se criminalization is a fundamentally personalized one. The wrongs which it identifies and acts upon are essentially personal and inter-​personal: they are the wrongful conduct of human beings. Mala prohibita criminalization is by contrast less personalized. Offences may be articulated which are more institutional and structural in their nature. Such offences are more likely to be based on ‘corporate’ liability than on establishing the criminal responsibility of a culpable human being. Nevertheless, we think that this distinction has analytical and normative value, provided that certain important points are emphasized. First, not all ‘regulatory’ crime in the labour field is properly described as mala prohibita. For example, the offence in the Health and Safety at Work Act 1974, whereby it is an offence to breach the duty ‘to ensure, so far as is reasonably practicable, the health, safety and welfare at work of all his employees’, is a crime malum in se.30 That is because the statute is simply giving specificity to a duty that the employer would otherwise have apart from the legislative determination, to take reasonable care of the health and welfare of its employees. Second, many criminal wrongs are strongly mala in se, such as the offence of requiring ‘another person to perform forced or compulsory labour and the circumstances are such that the person knows or ought to know that the other person is being required to perform forced or compulsory labour’.31 The offences that are now described as modern slavery crimes partake very strongly of this mala in se character, and they represent clear instances of very serious public wrongs against the civic life of the polity. Third, the distinction between mala in se and mala prohibita is not conceptually sharp, and it is better treated as a matter of degree.32 For example, take the offence in section 1(1)(b) of the MSA. What constitutes ‘forced or compulsory labour’ will sometimes be a matter of reasonable political disagreement, particularly at its edges. Does this extend to situations where an employer has withheld wages from workers, like a form of modern-​day peonage? Or the imposition of lengthy notice periods in conjunction with wide non-​compete clauses in the contract? The boundaries of ‘free’ labour will require further legal determination, usually through statute, even for a crime such as this that is strongly malum in se. 30 Health and Safety at Work Act 1974, ss 1(1) and 33. Duff gives this as an example as a wrong malum in se, because ‘that wrong is not a creation of the law, in the way that mala prohibita are, although it is formally constructed by the law’. Duff (n 16) 314. 31 Modern Slavery Act 2015, s 1(1)(b). 32 Duff (n 16) 321.

10  Criminality at Work Finally, even those crimes that are situated at the mala prohibita end of the spectrum, as is the case with many of those in the domain of work relations, may involve the commission of serious wrongs that warrant criminalization. Anthony Duff has argued that mala prohibita criminalization is legitimate where the system of regulation is justified by reference to the polity’s common good (eg minimize harms that are caused by otherwise valuable activities), and it would constitute a public wrong to fail to comply with the requirements specified in the regulatory scheme.33 Many of the criminal offences that support the ‘floor of rights’ in employment law could be justified in such terms. For example, the right to be paid the National Minimum Wage is implemented through a complex statutory regime of primary legislation and secondary instruments. The payment of a living wage may be understood as contributing to the common good of a labour market providing valuable opportunities for decent work. Section 31(2) of the National Minimum Wage Act 1998 makes it an offence where a person fails to ‘keep or preserve any record in accordance with regulations under section 9’. Duff ’s work on mala prohibita provides an account of how such an offence might be justified. Where such offences are ‘narrowly tailored towards a significant aspect of the common good, and impose only reasonable burdens on those whose conduct they constrain’,34 such crimes may be justified as appropriate legal responses to public wrongs. Given the contribution of decent wages to securing the common good of the polity, and the critical role of legislation in implementing the protective scheme, the offences in section 31 could be justified as important and legitimate instances of mala prohibita criminalization. This discussion highlights the scope for a constructive dialogue between the disciplines of labour law and criminal law and theory. Broadly speaking, we envisage this dialogue operating across three dimensions. The first dimension relates to the identification of public wrongs that are suitable candidates for a criminal law response. Recent work in criminal law theory has identified the argumentative parameters for the deliberative identification of public wrongs in the democratic process. For example, Duff has argued that wrongs warranting criminalization must engage . . . some aspect of the polity’s civil order, in which the polity therefore has in principle the standing to intervene. We must argue that such conduct is, when seen in the context of the polity’s civil order, wrongful, which is also to argue that we have reason to call those who engage in it to formal, public, censorial (and potentially punitive) account.35

In the recent UNISON judgment on tribunal fees, Lord Reed observed: Relationships between employers and employees are generally characterised by an imbalance of economic power. Recognising the vulnerability of employees to exploitation, discrimination, and other undesirable practices, and the social problems which can result, Parliament has long intervened in those relationships so as to confer statutory rights on employees, rather than leaving their rights to be determined by freedom of contract.36



33

ibid 313–​22. ibid 320. 35 ibid 333. 36 R (UNISON) v Lord Chancellor [2017] UKSC 51, [2017] 3 WLR 409 [6]‌. 34

A Framework for Discussion  11 In this way, labour law is composed of a structure of statutory rights identified by the polity as elements of a common good of decent work for all, and which might constitute a category of criminalizable public wrongs. The second dimension concerns the regulatory differences between criminal law and labour law in addressing public wrongs. The modalities of the criminal justice process are directed at the determination of innocence or guilt, the attribution of blame and censure, and the imposition of punishment. In its focus on the culpability of individual agents, this might obscure the cultural, economic, and regulatory vectors that create the conditions for vulnerability to abusive treatment, exploitation, and so forth. By contrast, labour law has been more attuned to these structural determinants, particularly in the discourse around ‘precarious work’ and new forms of modern slavery. For example, labour lawyers have been devoting increasing care and sophistication to conceptualize the notion of the employer in an institutional and structural way. This could provide a helpful perspective on the construction of criminal liability for mala prohibita offences, resisting the countervailing tendencies of a criminal law perspective in personalizing its norms by focusing on the blameworthiness of individual actors. The third dimension is the normative focus in labour law on the regulatory problems of inequality, vulnerability, and precarity in personal work relations. Most obviously, this provides a critical perspective on the criminalization of labour discipline, historically through Master and Servant legislation, but manifesting itself today in disciplinary criminal intervention in the contexts of care work and medical professionals. It also supports a critical perspective on the ways in which the criminal law both creates a new ‘precariat’ and is used to police that ‘precariat’ through its criminal justice institutions.37 For example, the criminalizing of irregular migrant workers and their employers is widely perceived to be creating a swelling tide of precarity, insecurity, and relative deprivation. It creates a criminalized status, exposing workers in situations of extreme vulnerability to penal measures. It also creates a deregulatory cascade by exposing the fundamental rights claims of these workers to the illegality doctrine.38 Where their civil claims are barred for illegality due to criminal liability, their vulnerability is compounded. The precarity-​generating effects of the criminal law are often highly racialized, and there is a powerful body of scholarship on carceral labour market governance in the US context.39 In talking of this dialogue between labour law and criminal law, we are very mindful of the fact that there is not a single genre of criminal law scholarship (nor, it must be emphasized, is there a single genre of labour law scholarship). As is evident from this introductory chapter, we have been influenced greatly by recent work on criminalization and the normative dimensions of the criminalization enquiry, particularly the work of Antony Duff on public wrongs and the ‘realm of the criminal law’. We also acknowledge the influence of recent work on criminalization practices from a more historical and sociologically informed perspective. In particular, we are indebted to the work of Markus Dubber,40 Lindsay Farmer,41 and Nicola Lacey. For example, in Nicola Lacey’s article of 2009 on ‘Historicising 37 Loic Wacquant, Punishing the Poor: The Neoliberal Government of Social Insecurity (Duke UP 2009). 38 Hounga v Allen [2014] UKSC 47, [2014] 1 WLR 2889. 39 See Noah Zatz, c­ hapter 25 of this volume. 40 Markus Dirk Dubber, The Police Power: Patriarchy and the Foundations of American Government (Columbia UP 2005). 41 Lindsay Farmer, Making the Modern Criminal Law: Criminalization and Civil Order (OUP 2016).

12  Criminality at Work Criminalisation’ she emphasizes the historical and political contingency of criminalization practices.42 These practices have evolved and changed over time, in response to shifting configurations of institutions, interests, and intellectual traditions.43 Lindsay Farmer has provided an interpretive account of changing criminalization practices using the idea of ‘civil order’, the contours of which have changed over time. And Markus Dubber has drawn our attention to the idea of ‘police’ as a distinctive form of governance practice, based upon discretionary state intervention, which has been particularly prominent as a form of labour market intervention. These scholarly interventions have been very helpful to us in identifying underlying patterns in contemporary criminalization practices, establishing links between seemingly unconnected phenomena in criminal justice practices. Moreover, we regard these normative and sociological perspectives as complementary. For example, the populist turn in contemporary politics has led populist governments to single out individuals or collectivities as alien to a unitary moral ethos and therefore amenable to criminalization.44 These tendencies have been especially pronounced in relation to certain statuses (eg migration status), and the discretionary power of criminal justice agencies has borne down very heavily on particular racial groups. The sociological perspective allows us to understand the ideational and material factors that are shaping the contemporary politics of criminalization. The normative perspective provides a critical yardstick to identify the ways in which public deliberation has become corrupted and prone to error in its identification of public wrongs amenable to criminalization. Maintaining this methodological plurality also enables us to understand some puzzling features of contemporary discourse around criminalization. For example, in the area of modern slavery, the debates have often been highly polarized, and this polarization has mapped onto camps who are ‘for’ or ‘against’ criminalization. The ‘for’ camp would emphasize that forms of modern slavery constitute serious public wrongs that attack the basic civic standing of its victims in a liberal polity. The ‘against’ camp object to the negative consequences of criminalization for the victims of modern slavery, who are often subjected to coercive forms of immigration control. They also point to the symbolic ramifications of criminalizing the most serious forms of exploitation, which tends to confer a pale legitimacy on other forms of less serious interpersonal exploitation in personal work relations. It would be better, we suggest, to understand this debate as to a certain extent being conducted at cross-​purposes. From a normative perspective, modern slavery is indisputably a very serious form of public wrong. From a sociological perspective, criminalization practices have indisputably had negative consequences in exacerbating rather than ameliorating precarity for the most vulnerable. Furthermore, the criminalization of modern slavery reflects more general political tendencies to penalize vice and bad character (of ‘modern slavers’) and to ignore the role of politics and regulation in creating the vulnerable statuses of its victims in the first place.45 Methodological pluralism enables us to keep both perspectives in view, and

42 Nicola Lacey, ‘Historicising Criminalisation:  Conceptual and Empirical Issues’ (2009) 72 Modern Law Review 936. 43 Lacey, In Search of Criminal Responsibility (n 22). 44 Alan Bogg and Mark Freedland, ‘Labour Law in the Age of Populism:  Towards Sustainable Democratic Engagement’ in Julia Lopez Lopez (ed), Collective Bargaining and Collective Action—​ Labour Agency and Governance in the 21st Century? (Hart Publishing 2019) 30–​32. 45 For sophisticated discussion, see Julia O’ Connell Davidson, ‘ “Things” Are Not What They Seem: On Persons, Things, Slaves, and the New Abolitionist Movement’ (2016) 69 Current Legal Problems 227.

A Framework for Discussion  13 both perspectives have much to offer to public deliberation about the role of criminalization in addressing modern slavery wrongs. The chapters are divided into four thematic parts and a fifth comparative one. The first part, ‘Mapping the Terrain’, provides an overview of the possible structure of a social democratic theory of criminalization, and it maps the current pattern of criminalization practices in UK labour law. The second part, ‘Labour Wrongs as Public Wrongs’, explores the possibilities and limits of public wrongs in the domain of labour relations. It starts from the position that many violations of fundamental labour rights constitute serious public wrongs. As such, these wrongs may warrant a criminal law response. The chapters explore a range of potential public wrongs in labour law, many of which are central to current debates about criminalization in labour relations: wage theft, disclosure obligations, exploitation, workplace harassment, human rights violations, and the public regulation of sex work. In so doing, the chapters provide some determinate shape and content to public wrongs in the general literature on criminalization. Many of the chapters also explore the possibilities for other kinds of public response to these wrongs, such as through private law, human rights law, or professional regulation. The third part, ‘The Contemporary Shape of Criminalization Practices:  Risk, Status, and Character in the Neoliberal Criminal Law’, uses Lacey’s work on modern criminalization practices to understand the contemporary politics of criminalization. The power of Lacey’s framework is that it identifies important thematic continuities that underlie seemingly disparate phenomena: licensing strategies to manage the risks of labour abuse, hybrid civil-​criminal preventive orders, the regulation of medical professionals and care workers, the rise of character and vice as a basis for culpability in modern slavery, the criminalization of migrant workers, and the repressive use of the criminal law to restrict and pacify radical labour activism. The fourth part, ‘Criminalization and Enforcement’, returns to the traditional focus of labour law which is upon whether criminal sanctions contribute to effective enforcement. The chapters explore workplace health and safety and the enforcement of minimum wage entitlements in the ‘fissured workplace’.46 The final part, ‘Comparative Perspectives on Criminalization’, examines the criminalization debates from USA, Canadian, and Australian perspectives. This provides a baseline for assessing whether there are common patterns to criminalization practices. And to the extent that there are distinctive trajectories, to understand the dynamics of divergence. In ­chapter 2, ‘Workplace Welfare and State Coercion’, Bob Sullivan explores the possibilities and the limits of a social democratic criminal law. As a leading criminal law theorist, developing an account of criminal law from a social democratic perspective, Sullivan in this chapter provides an important corrective to the default position in social democratic labour law, which is to regard the criminal law with scepticism. He sets out a range of possibilities for the criminal law to contribute to securing social democratic goals, ensuring a sufficient degree of social and economic welfare for all citizens in the political community. This social democratic perspective also reveals some important differences from the dominant liberal paradigm in Anglo-​American criminal theory. Sullivan’s chapter elucidates an important role for the criminal law in the social democratic state. The first is the positive and important role of regulatory criminalization, based



46

David Weil, The Fissured Workplace (Harvard UP 2014).

14  Criminality at Work upon strict liability, particularly in the spheres of minimum wage enforcement and workplace health and safety. Health and safety might also be protected through a corporate manslaughter offence, although Sullivan notes the practical problems with the current offence in English criminal law. The second is the criminalization of modern slavery offences, although Sullivan regrets the current linkages with immigration enforcement and the wider context of state-​sanctioned precarity for irregular migrant workers. The third is the potential role for criminal law to prohibit certain forms of occupation that create unacceptable risks to the health and safety of the participants, such as boxing or sex work. The social democratic state must tread a careful line here, given the importance of individual freedom in liberal forms of social democratic theory. Overall, this chapter provides an important counterpoint to social democratic labour law, and it provides a compelling vision of what a legitimate social democratic criminal law might look like. In ­chapter  3, ‘Using Criminal Law to Enforce Statutory Employment Rights’, David Cabrelli examines the current terrain of criminal law as a technique of labour market regulation. His chapter identifies a range of possible interactions between the criminal law and civil law in the legal enforcement of labour standards. Sometimes fundamental labour rights, such as the right not to be unfairly dismissed or the right not to be discriminated against, are protected exclusively through a ‘private’ enforcement model at the initiative of the individual right-​holder. Sometimes there will be exclusive enforcement through the criminal law with no private right of civil action, as under the Health and Safety at Work Act 1974. Finally, there may be mixed enforcement regimes where there is a combination of criminal and civil measures linked to specific statutory rights, as with the enforcement of the National Minimum Wage Act 1998. It is possible, of course, that this landscape is simply a chaotic result of a chaotic politics, where legislators respond to short-​term political pressures from time to time without reference to a framework of principle.47 However, Cabrelli’s chapter explores whether this pattern might be intelligible in the light of the notion of public wrongs, a normative theory of criminalization that has been developed most notably by Antony Duff and SE Marshall. This idea of public wrongs has an obvious resonance in labour law, given that the fundamental rights protected in labour law reflect a view of work relations as a matter of shared public concern. While Cabrelli recognizes the potentially capacious category of public wrongs, his detailed exploration of different statutory employment rights provides some important texture to the enquiry. As Duff himself acknowledges, there are many ways in which legal systems can respond in a public way to public wrongs–​–​the criminal law is one response, but there are others such as the law of torts.48 The law of torts is public too. Cabrelli’s methodology involves a particularized analysis of whether certain public wrongs encompassed by statutory employment rights should be criminalized, all things considered. Sometimes there will be value in protecting the autonomy of the right-​holder to retain control over the pursuit of her wrong in a public court, using private law remedies. Where criminal law is utilized, it would be best to do so in an additive rather than a substitutive fashion. In the end, this will depend upon a contextual assessment on a right-​by-​right basis. Cabrelli’s chapter exemplifies how a conversation between labour law and criminal theory can be mutually beneficial to both disciplines.

47 48

Andrew Ashworth, ‘Is the Criminal law a Lost Cause?’ (2000) 116 Law Quarterly Review 225. Duff (n 16) 286–​88.

A Framework for Discussion  15 In ­chapter 4, ‘Where Criminal Law Meets Labour Law: The Effectiveness of Criminal Sanctions to Enforce Labour Rights’, Catherine Barnard and Sarah Fraser Butlin provide a detailed examination of the politics of criminalization in four key areas: the enforcement of working time rights by the Health and Safety Executive; the enforcement of National Minimum Wage entitlements by HMRC; the enforcement role of the Employment Agencies Standards Inspectorate; and the licensing regime administered by the Gangmasters and Labour Abuse Authority. As the authors note, the politics of criminalization is being shaped by public anxieties around the under-​or non-​enforcement of labour standards, and so the authors examine these diverse regimes through an enforcement lens. The Gangmasters and Labour Abuse Authority regime, based in the Gangmasters (Licensing) Act 2004, provides for a system of licensing for ‘gangmasters’ in specific sectors of economic activity. It also provides for a range of serious criminal offences for non-​ compliance with the licensing regime. It is supported by further provisions empowering enforcement action, with obstruction of an enforcement officer constituting a criminal ­offence under the legislation. The licensing regime is also supported by hybrid civil-​criminal ‘Labour Market Enforcement Undertakings’ and ‘Labour Market Enforcement Orders’, breach of which is punishable as a criminal offence. By contrast, the Employment Agency Standards Inspectorate regime is no longer based on a licensing model. Instead, the statutory framework specifies that a failure of employment agencies to comply with the certain specified standards (eg withholding wages from workers) itself constitutes a criminal o ­ ffence. There are also accompanying criminal offences dealing with, for example, the creation of false records. The National Minimum Wage framework provides for a composite mechanism of civil and criminal enforcement. The criminal offences extend beyond the refusal or wilful neglect to pay the minimum wage, to include false record keeping and obstruction of enforcement officers. Finally, the various working time limits in the Working Time Regulations are enforced through criminal offences, in contrast with the dual enforcement regime in the National Minimum Wage framework. These different models display a rich variety, sometimes requiring mens rea and sometimes based on strict liability. They also range from situations where the criminal law is additional to existing civil rights of enforcement (as under the National Minimum Wage framework) to situations where the criminal law is envisaged as an exclusive enforcement mechanism (as with the Working Time Regulations). Barnard and Fraser Butlin examine the political choices that underlie these different legislative models. Sometimes, as with trade union blacklisting, the government eschews a criminalization approach despite the identification of very serious public wrongs perpetrated against identifiable victims. The authors suggest that criminalization of labour breaches has often occurred within the context of a wider deregulatory agenda. Although this is puzzling–​–​for a government that way inclined, would it not be preferable simply to deregulate rather than to go through the charade of criminalization?–​–​the criminalization of breaches is a way of signalling that there is governmental concern with labour abuses. Cynically, this could provide something of an alibi to governments who can maintain the appearance of worker protection without the corresponding substance. In truth, the public discourse around criminalization has been dominated by enforcement, and the need to promote compliance with labour standards. In an era of austerity and constrained public funding for agencies charged with compliance responsibilities, Barnard and Fraser Butlin note that public agencies have not tended to pursue prosecutions except

16  Criminality at Work in the most serious of cases. In these situations, the loss of an individual normative power to seek a private remedy in the ordinary courts is particularly disempowering. This provides an important argument in favour of retaining private rights to seek remedies for rights violations, so that criminal law operates as an auxiliary prop to remedies pursued in the civil legal process. The authors also note that the Director of Labour Market Enforcement is developing a sophisticated methodology on enforcement, identifying an appropriate role for the criminal law within that wider enforcement strategy. We think that this chapter provides some modest support for a subsidiary role for the criminal law in supporting labour rights, provided that the main enforcement occurs through civil reparative mechanisms and criminal enforcement agencies are given appropriate resources. In the second part of the book, ‘Labour Wrongs as Public Wrongs’, the chapters explore the kinds of wrongs that might be appropriate candidates for criminalization. In this respect, we have found the concept of public wrongs very helpful in framing the current debates on criminalization in the labour field. In his recent book, The Realm of the Criminal Law, Duff has defended the following position: To justify a claim that we should criminalize a type of conduct, or that it is legitimately and appropriately criminalized, we must argue that it is a public rather than a private matter: that it is a matter that implicates some aspect of the polity’s civil order, in which the polity therefore has in principle the standing to intervene.49

Even then, having identified that the wrong is public in this sense of engaging the polity’s civil order, there are other ways in which the polity might respond: do nothing; restorative justice; non-​governmental codes; non-​criminal regulation; private law; preventive measures.50 Given those possible alternatives, the legislator must have good reason to respond through the modalities of the criminal law, which involves a specific kind of public response: one that involves an authoritative determination of guilt, the public calling of an offender to account, public censuring of the wrongdoing, and the imposition of punishment.51 This does not provide a scientific formula through which decisions on criminalization can be cranked out mechanistically. Its application requires judgement and political deliberation. These deliberations ‘will inevitably be messier’ than under a simple master principle . . . since we will need to appeal to a range of different values in identifying a range of public wrongs—​values unified only by the fact that they all play a part in defining the polity’s civil order; and we will . . . need to appeal to a range of considerations in deciding whether we have reason to criminalize a type of public wrong rather than responding in some other way (or providing no formal response at all).52

49 ibid 333. 50 ibid 280–​92. 51 ibid 50. 52 ibid 298. We suggest that ‘rather than’ may not be a helpful way of posing the question, since often criminalization will be a public legal response alongside rather than instead of others (such as private law or statutory employment rights).

A Framework for Discussion  17 The chapters in the second part explore some of these messy first-​order issues in specific contexts of criminalization. In ­ chapter  5, ‘Exploitation at Work:  Beyond a “Criminalization” or “Regulatory Alternatives” Dichotomy’, Jennifer Collins reflects upon the treatment of the wrong of exploitation in criminal law and criminal theory. In many ways, exploitation is the core wrong of personal work relations.53 It is also a wrong that defies reduction into a simple definition, and it manifests across a spectrum from the very egregious to the relatively innocuous. At one extreme, it is meaningful to describe the trafficked migrant engaged in forced labour as ‘exploited’. It may also be true of the well-​remunerated law fellow whose good will is used by the unscrupulous Head of House to induce unpaid supererogation in College activities. At its core, according to Collins, it involves interpersonal predation by a stronger party on a weaker party, the abuse of the weaker party’s vulnerability, in order to make a gain at the weaker party’s expense. The offences in the MSA may be understood as targeting the most serious forms of exploitative wrongdoing through criminalization. Collins argues that the debates around criminalization have sometimes been hobbled by a tendency for protagonists to line up ‘for’ or ‘against’ criminalization. She acknowledges the political validity of many of the criminal-​law-​sceptics’ concerns. Specifically, the isolation of egregious interpersonal exploitation may have the effect of legitimizing less extreme forms of exploitation. Moreover, the focus on interpersonal exploitation obscures the structural determinants of vulnerability such as migration laws, visa regimes, labour standards, and trade union density. There has also been a close coupling of modern slavery discourse and immigration control—​an undeniable fact of contemporary political rhetoric and legislative structures, even if that linkage is only a contingent one in normative terms. She suggests that this binary framing of the criminalization enquiry is too limiting. Instead, the role of the criminal law must be examined from a wider regulatory perspective, assessing its real-​world interaction with other forms of non-​criminal regulation and its impact on the most vulnerable workers. She proposes a theoretical structure within which that multidimensional enquiry can be undertaken. According to Collins, regulatory criminalization will have an important role in supporting the effective enforcement of labour standards, alongside the criminalization of the most extreme forms of exploitative abuse. Ultimately, of course, the most effective regulatory mechanisms for eliminating exploitation will involve the eradication of precarity and insecure work. In this respect, we would expect the standard goals and regulatory techniques of labour law to take the lead role, with the criminal law functioning as a vital but subsidiary intervention. In the next two chapters, the authors explore the potential role for the criminal law in enforcing certain contractual obligations in the employment contract. In c­ hapter  6, ‘The Duty of Loyalty and the Scope of the Law of Fraud’, Hugh Collins explores the role of criminal law as an additional sanction against the breach of existing civil obligations in the contract. As he notes, the Master and Servant legislation was justified on this basis. Masters often complained that the civil remedy of compensatory damages was unlikely to deter insubordination and the wilful neglect of work, so that the attachment of penal sanctions to contractual breaches was necessary to secure contractual bargains. In the modern context, Collins examines the ways in which property offences in criminal law might be used to deter contractual 53 Virginia Mantouvalou, ‘Legal Construction of Structures of Exploitation’ in Hugh Collins, Gillian Lester, and Virginia Mantouvalou (eds), Philosophical Foundations of Labour Law (OUP 2018) 188.

18  Criminality at Work breaches. He focuses on the offence of fraud in the Fraud Act 2006, and in particular the role of criminalization where there has been a failure to disclose information. To this end, he examines the current scope of disclosure obligations under the employment contract, focusing on the implied duties of trust and confidence and of loyalty. There is considerable uncertainty about the scope and content of disclosure obligations, which raises serious rule of law concerns where breaches of these contractual obligations are, in effect, criminalized as instances of fraud. There is also a more general reluctance in English criminal law to criminalize omissions rather than positive acts, and a failure to disclose is an omission. In the end, Collins’ careful examination of the fraud offence in employment contract situations counsels caution. There is a real risk of over-​extension of criminal sanctions in enforcing ‘relational’ contractual obligations in this area. Criminalization is likely to be a disproportionate response, especially where disclosure obligations can be supported through equitable remedies such as an obligation to account for secret profits as in Attorney General v Blake.54 As Collins is astute to recognize, the heart of the matter lies in the echoes of Master and Servant law in the contemporary landscape of fraud as a mechanism for enforcing contractual duties of fidelity and loyalty. This evokes an era where employment obligations were treated as quasi-​familial incidents of status. In this respect, we might say that the wrong of non-​disclosure in a relational contract ought not to be treated as a public wrong but as a private matter between the parties, especially where the existing remedies in private law are sufficient to secure corrective justice between the parties. There are extra reasons to be cautious about criminalization where the (public) criminal law is bolstering the (private) disciplinary powers of employing entities. In ­chapter 7, ‘Wage Theft as a Legal Concept’, Sarah Green analyses the fraud offence from the different vantage point of ‘wage theft’. The social concept of a ‘wage theft’ has significant traction in public discourse at the current time. It encompasses a wide range of dishonest or ‘sharp’ practices: false labelling of individuals as ‘self-​employed’ and hence outside the scope of the National Minimum Wage framework, failure to pay holiday pay, unlawful deductions, and an absence of transparency in relation to wage entitlements. It is linked to wider public concerns about the effective enforcement of the statutory minimum wage regime. Green’s chapter then examines whether the social concept of ‘wage theft’ maps onto the legal definition of ‘theft’ in section 1 of the Theft Act 1968. She argues that the legal label of theft is ill-​suited to the constellation of practices associated with the social label of ‘wage theft’. This is because of the disjunction between the proprietary status of ‘wages’ and the elements of the offence of theft in English law. In short, unpaid wages will often not count as ‘property belonging to another’ at the time of the dishonest appropriation by the employer, and hence there is a difficulty with identifying a complete and coincident mens rea and actus reus. That is because unpaid wages will ordinarily give rise to an action for damages in contract, rather than giving rise to a debt claim (which is proprietary). Nor is this a merely technical difficulty. Green argues that the offence of theft is also morally unsuited to tackling wage theft, because it does not label the wrong of ‘wage theft’ accurately–​–​at its core, theft is concerned with the dishonest assumption of another’s property rights. Green further argues that it would be more appropriate to tackle ‘wage theft’ through the offence of fraud under the Fraud Act 2006, specifically ‘fraud by abuse of position’.55 This

54 55

[2008] EWCA Crim 2304, [2008] 10 WLUK 181. Fraud Act 2006, s 4.

A Framework for Discussion  19 would be a more appropriate label for the wrong. It would be better, perhaps, in the interests of ‘moral clarity’ through ‘fair labelling’, for criminalization to be reframed around the rallying call of ‘wage fraud’ rather than ‘wage theft’.56 There are interesting parallels and contrasts with Hugh Collins’ treatment of fraud as a basis for augmenting the employee’s duties under the employment contract. The obligations related to wages are of a more obviously public nature, reflected in the elaborate protective statutory regime that has developed around the wage, its level, and its payment. This chapter also identifies important overlaps between regulatory criminalization (eg using the criminal law to support the enforcement of statutory norms) and crimes mala in se (tackling under-​payment or non-​payment through fraud). It is moreover a testament to the importance of general principles of criminal law, such as the principle of fair labelling, to more particular enquiries into criminalization in the labour field. In c­hapter  8, ‘The Criminalization of Workplace Harassment and Abuse:  An Over-​ personalized Wrong?’, Alan Bogg and Mark Freedland consider the legal treatment of workplace harassment and abuse. In the wake of social and political activism focused on sexual harassment, the legal regulation of harassment has been under intensive scrutiny. In the English context, harassment is already regulated by an extensive body of legal norms. The Protection from Harassment Act 1997 is an interesting example where legislation provides for parallel tortious and criminal liability for the same wrong. The authors develop the idea that criminal liability has had a ‘dragging’ effect on the civil liability regime, which has undermined the effectiveness of the legal response. This is because the paradigm of criminal liability has treated harassment as a personalized wrong, reflecting the inherent propensity of the criminal process to allocate censure and blame to culpable agents. The consequence of this has been to obscure the structural and organizational context within which harassment and abuse takes place: it propagates in circumstances of insecurity, precarity, and non-​decent work. Should criminalization have any role to play in addressing workplace harassment and abuse? Examining regulatory activity at the national and international levels, they argue that the value of ‘dignity’ could be used to formulate a principled criminalization response. This leads to three proposals: (i) that criminalization focus on the organizational responsibilities of employers, based on a corporate liability model, along the lines of a ‘health and safety’ model; (ii) that the standing of individual workers to enforce their own rights and seek civil remedies against their harassers is constitutive of respect for their human dignity; and (iii) that dignity points away from the victim’s ‘consent’ being part of the liability enquiry: consent should not be relevant to criminal or civil liability for harassment and abuse. As such, criminal law has an appropriate role alongside other types of legal intervention. The authors end with a proposal for ‘collective bargaining in the shadow of the criminal law’, whereby trade unions act as ‘private prosecutors’ in appropriate cases. This could be expanded to other areas of criminalization, such as offences under the National Minimum Wage framework. The next two chapters offer contrasting perspectives on the role of criminalization in regulating sex work. Sex work is an important and difficult test-​case for ‘criminality at work’, and it has already been touched upon by Sullivan as a possible candidate for criminal restriction in the social democratic state. Is sex work so inherently exploitative and dangerous 56 This lovely phrase is John Gardner’s ‘Rationality and the Rule of Law in Offences against the Person’ (1994) 53 Cambridge Law Journal 502, 515.

20  Criminality at Work under current structural conditions that it should not be permitted at all as a form of ‘work’? If so, should we criminalize the buying of sex, the selling of sex, or target the role of intermediaries such as traffickers? Or does criminal prohibition dehumanize the participants in sex work by not crediting their agency with any public salience? In ­chapter 9, ‘Sex, Work, and Criminalization’, Michelle Madden Dempsey provides a philosophical examination of the various dimensions of commercial sex as a form of work. She then offers perspectives on the legitimate role of the criminal law in regulating commercial sex, based upon three limiting principles: (i) a principle of ‘minimalism’ such that the criminal law should only be used as a last resort; (ii) a principle of ‘modest legal moralism’ such that the criminal law should be reserved as a legal response to public wrongs; and (iii) a ‘presumption of non-​ interference’ based in the liberal harm principle, so that state coercion is limited to situations of direct or indirect harm. Given these three limiting principles, does the criminal law have a legitimate role in regulating, restricting, or prohibiting sex work? Dempsey provides a nuanced taxonomy of commercial sex relations, and she argues that the application of the limiting principles varies depending upon where commercial sex is situated in the taxonomy. Where commercial sex is treated as a form of work, the criminal law would protect sex workers from rape, abuse, and violence through its core mala in se crimes. There might also be scope for regulatory criminalization, supporting the enforcement of Health and Safety laws (although Dempsey notes some practical difficulties in this regard). Where commercial sex is viewed as irredeemably exploitative in virtue of structural ‘patriarchy-​precarity’ constraints, Dempsey suggests that criminalization would legitimately target the purchasers, pimps, and other intermediaries who profit from the exploitation of sex workers. Where commercial sex is viewed as akin to modern slavery, so that the agency of participants is utterly negated, the criminal law response will focus on the organized criminal gangs that orchestrate such practices. The difficult case is where commercial sex is ‘varied’, consisting of a variety of personal work relations that range from the fully voluntary through to the most extreme forms of modern slavery organized by criminal gangs. In this varied context, Dempsey argues that criminalization of buyers of sex may be warranted where not criminalizing would increase market demand and this increase in demand leads to increasing levels of exploitation and modern slavery. This is a form of ‘indirect harm’ argument that rebuts the ‘presumption of non-​interference’, and it would appear to be consistent with criminalization-​limiting principles of a liberal criminal theory. In ­chapter  10, ‘The Work of Sex Work:  Prostitution, Unfreedom, and Criminality at Work’, Katie Cruz analyses the legal treatment of sex work from the perspective of Marxist feminism. For Cruz, the work of sex work must be understood in its wider structural context of exploitative capitalist relations. She argues that sex work should be understood as work. Furthermore, the features of exploitation and unfreedom associated with sex work do not vitiate its identity as a form of work, and therefore as an activity that warrants the application of protective norms of labour law. This appears to mark an important difference with Dempsey’s taxonomy of commercial sex work. For Cruz, in fact, all work under capitalism is structurally coupled with exploitation and unfreedom. Given this structural coupling, it is problematic to use the exploitation in sex work as a basis for excluding it from the domain of personal work relations. Otherwise, personal work relations would be an empty category in capitalist arrangements.

A Framework for Discussion  21 Cruz’s chapter outlines the evidence of the harms of criminalization practices on sex workers themselves, harms which are highly gendered, racialized, and reinforce the degrading stigma that attaches to particular groups who undertake sex work. The starting point for a humane legal intervention–​–​that is to say, a legal intervention that eschews dehumanization of any of the law’s subjects–​–​is decent work for all who are engaged in paid labour. There is accordingly no basis for excluding any worker from this protective floor of labour standards, and it would appear perverse to do so on the basis that the work is seriously exploitative. The withdrawal of the law’s protection from those who are in such a position would appear to compound their abuse, not ameliorate it. In terms of the role of criminalization in supporting worker-​protective goals, Cruz argues that there are no simple answers to this question. Furthermore, we speculate that she would be sceptical that a normative theory divorced from an appreciation of the material conditions of work can provide answers to that enquiry. What matters is the lived experience of sex workers in their encounters with different types of law, their capacities to resist injustice and unfreedom, the activities of criminal justice officials, and the impact of the criminal law on freedom in their working lives. The answers to these questions should be shaped by the political agency of the workers themselves, empowered to give voice to their own experiences of the criminal law and its impact in their lives. Theory cannot be a substitute for agency. And to be useful, theory cannot be abstracted from the structures of power and subjugation that envelop work relations under capitalism. In the final chapter of the second part, ‘Human Rights, Labour Rights, and Criminal Wrongs’, Virginia Mantouvalou assesses the contribution of the criminal law in protecting the fundamental human rights of workers. Fundamental human rights identify the basic moral entitlements of workers. These entitlements have a special normative significance in the polity. Accordingly, serious violations of those entitlements by powerful private actors would seem to count as public wrongs against the polity’s ‘civil order’, as constituted by its most fundamental rights. Mantouvalou discusses the position of criminal law under the European Convention of Human Rights (ECHR), and the interpretation of the state’s positive obligation to protect as encompassing a positive obligation to criminalize certain public wrongs against fundamental entitlements. This has been most marked in relation to the wrongs under Article 4 ECHR, but it is by no means limited to Article 4. She also emphasizes how the European Court of Human Rights has developed an approach to criminalization that is sensitive to its effects in the real world. It is not sufficient to have criminal laws on the statute books. Those criminal laws must be investigated and enforced effectively, with adequate support and protection for victims, and the state also has a duty to dismantle the structural determinants of vulnerability such as those constituted by its own visa regimes. Mantouvalou identifies the blacklisting of trade unionists as a particularly serious human rights violation warranting a criminal law response. This is necessary to mark and communicate the serious nature of the public wrong involved in blacklisting, and the catastrophic harms wrought by this secretive and insidious practice on the lives of trade unionists. In defending criminalization as an appropriate response, she suggests that this should be treated as an addition to civil modes of redress for victims of blacklisting. Criminalization should not be a substitute for private law-​type norms focused on redress and compensation, but should form part of a broad and multifaceted legal response. This echoes the message from many of the chapters in the first and second parts of the book.

22  Criminality at Work In the third part of the book, ‘The Contemporary Shape of Criminalization Practices: Risk, Status, and Character in the Neoliberal Criminal Law’, we use Nicola Lacey’s work on modern criminalization practices to understand the contemporary terrain of criminalization in labour law. Our starting point for this project was the sense that the criminal law was re-​emerging as a mechanism of regulation and control in work relations. This re-​emergence could be detected across a range of different work contexts, such as: the regulation of specific occupations such as care work and medical professions; modern slavery; immigration; and worker-​protective criminalization through preventive orders and licensing regimes. Were we to regard these developments as coincidences that simply converged on employment relations by happenstance? Or is there more to be said in revealing deeper continuities across seemingly independent regulatory practices? According to Lacey, contemporary criminalization practices have often been directed at a ‘hybrid character/​risk model’, where particular statuses (eg migration statuses) have a particular salience in the attribution of criminal responsibility.57 These criminalized statuses are politically configured as presenting dangers to the moral community of ‘virtuous’ or ‘normal’ citizens. The politicization of criminalization has also been aligned with the use of preventive measures to abate anticipated risks of harm, based around a political concern to assure citizens of their security.58 This has been reflected in the expanding use of pre-​emptive interventions, such as hybrid civil-​criminal orders and licensing regimes.59 In broad terms, we suggest that the chapters in the third part provide support for Lacey’s account of the development of a modern notion of criminal responsibility. This is an area in which there is scope for fruitful dialogue between labour lawyers and sociologically oriented criminal theorists. In ­chapter  12, ‘The Preventive Role of the Criminal Law in Employment Relations’, Jennifer Collins and Andrew Ashworth assess the range of preventive interventions in labour market regulation. These include offences imposing omissions liability on employers, the effect of which is to impose safeguarding duties on private actors as agents responsible for avoiding or mitigating harms; the offence of employing an illegal worker under the IA which has co-​opted private employers into the public responsibility of immigration enforcement; and preventive orders in relation to trafficking and labour market enforcement. In relation to preventive orders in labour market enforcement, the authors query whether there are good reasons to resort to criminal law in situations where civil enforcement mechanisms might be improved. This also raises uncomfortable questions for governments that have presided over onerous tribunal fees impeding access to justice, and the pursuit of deregulation of labour standards and restriction of trade unions. Is the enforcement crisis being produced by ‘bad’ employers or the political choices of our elected representatives? The most effective form of prevention would be worker-​protective labour law, supported by access to justice and strong trade unions. Preventive criminal law is a response to an avoidable crisis that has been brought about by specific political choices. Nevertheless, the penetration of these preventive orders into labour law fits with the more general expansion of preventive criminal measures in different spheres of social life.



57 Lacey, In Search of Criminal Responsibility (n 22) 157–​61. 58

Peter Ramsay, The Insecurity State (OUP 2012).

59 Lacey, In Search of Criminal Responsibility (n 22) 46–​48.

A Framework for Discussion  23 The next chapter by ACL Davies, ‘Licensing of Employing Entities and Criminalization’, continues in the vein of preventive criminal law, examining the role of licensing regimes in the labour field. She focuses in particular on the ‘Gangmasters’ licensing regime, which is operative in the agricultural and food processing industries. This licensing regime is supported by a structure of criminal offences, some of which are very serious and provide for high maximum penalties of ten years’ imprisonment. Davies notes the difficulties in providing a principled justification for criminalization of licensing requirements, given that non-​compliance is only indirectly linked to the commission of serious wrongs against vulnerable workers. However, she considers that this form of regulatory criminalization can be justified in principle. Furthermore, from a regulatory perspective, she identifies some advantages of a licensing regime in promoting compliance with basic labour standards. Of course, mechanisms of public enforcement are always vulnerable to political decisions about the allocation of resources. Nevertheless, the licensing of labour market intermediaries might usefully be extended to other sectors where there are specific problems with precarity and abuse. In our view, this chapter provides a strong example of why some mala prohibita crimes should be viewed as ‘real’ crimes, making a vital positive contribution to the wellbeing of workers. In ­chapters 14 and 15, the role of the criminal law in regulating specific professions is examined. In ­chapter 14, ‘Criminalizing Care Workers A Critique of Prosecution for Ill-​ treatment or Wilful Neglect’, LJB Hayes offers a critique of the ‘care worker offence’ in section 20 of the Criminal Justice and Courts Act 2015, which criminalizes ill-​treatment or wilful neglect in circumstances of ‘paid work’ by care workers. According to Hayes, this new offence has obscured the political and managerial choices that render care work badly paid and precarious. Judicial narratives are all too often framed around thick moralized judgements of vice and bad character, marginalizing alternative narratives that identify the terrible pressures that precarious workers experience in their working lives. These pressures are being generated by poor management, underinvestment in training, high staff turnover and high rates of sickness absence, and chronic starvation of public resources for social care. In this way, Hayes’ arguments suggest that we might even regard the section 20 offence as a contemporary variant of nineteenth-​century Master and Servant laws, using criminal law to enforce labour discipline and to police the precarious. In ­chapter 15, ‘The Medical Professional as Special Before the Criminal Law’, Suzanne Ost explores the position of medical professionals such as doctors in criminal law. In certain respects, doctors appear to have an unusually privileged position under English criminal law. For example, there are a range of special and general defences available to doctors who are discharging their medical responsibilities. Furthermore, judges are often receptive to the professional judgements of doctors, reflected in expert medical opinions, in fleshing out the content of abstract norms in the criminal law. It might even reflect comity between the professions of law and medicine. Certainly, it provides a stark contrast with the social, economic, and cultural chasm between judges and care workers, and their different experiences of the criminal law under the Criminal Justice and Courts Act 2015. On the other hand, doctors are also particularly exposed to the risk of very serious criminal liability, particularly in respect of gross negligence manslaughter. While we would be unlikely to identify doctors with the precariat, as we might be inclined to do with care workers, doctors are also working under intense pressures in circumstances of austerity. In this way, the personalizing dynamic of criminal liability deflects attention towards individualized medical

24  Criminality at Work mistakes and errors of judgement and away from the economic context. In these situations, Ost suggests that the structural dimension might be better captured through forms of organizational liability for employers. This reflects the proposal by Bogg and Freedland on criminal liability for workplace harassment. This organizational model of criminal liability might also provide a better approach for dealing with problems of abuse in the care sector. The following three chapters each address the complex dynamics of criminalization in the spheres of migration and modern slavery. In c­ hapter  16, Cathryn Costello provides powerful arguments against criminalization practices in the migration sphere. She does so by testing the extensive criminalization of irregular migrants against standard liberal principles of criminalization. In her view, it is very difficult to identify any direct wrongs or harms to others that arise in virtue of ‘irregular’ migration. She is also doubtful that a malum prohibitum offence can be justified. Against these weak arguments in favour of criminalization, she identifies compelling reasons against criminalization. Criminalization leads to a deregulatory cascade through the operation of the illegality doctrine, which undermines the legal protection of migrants’ fundamental rights. It also reinforces an image of migrants as dangerous threats to the civil order, which leads to hostility, discrimination, and stigmatization of migrants. In truth, the criminalization of migrants represents a context where there has been a decisive rupture with liberal principles of criminalization. It leads us into the dark political territory of migrants as warranting criminalization simply in virtue of their dangerous status, which presents undefined threats to the moral and social order. In this respect, it is highly significant that the IA has now criminalized ‘illegal working’ as such. Under previous regimes, this conduct would have been criminal under general immigration offences, but this targeted status-​based labelling of the crime is a new development that is fraught with political significance. In ­chapter 17, ‘Doing the Dirty Job: Labour at the Intersections of Criminal Law and Immigration Controls’, Ana Aliverti offers a more historical and sociological perspective on the inclusionary and exclusionary effects of criminal law, determining who is inside and who is outside the polity’s ‘civil order’. In so doing, she draws upon Lindsay Farmer’s work which tracks shifts in the structure and content of criminal law as being reflective of shifting notions of ‘civil order’. The resulting analysis illuminates the shifting social functions of criminalization during different historical phases, especially as it relates to migration control and labour market governance. Legal norms have both reflected and constructed political and social boundaries, often based around racialized notions of ‘good’ and ‘bad’ migrants. The most recent phase of criminal law intervention conjoins highly deregulated labour markets and highly juridified migration regimes. This has been linked to the rise of populism and more authoritarian and nationalistic forms of statecraft. Nevertheless, as Aliverti reminds us, we should avoid determinism in reading these historical shifts. ‘Civil order’ is always a contested idea, and like Cruz she emphasizes the critical role of political mobilization and agency in resisting the highly exclusionary and disciplinary configuration of criminal law. In this respect, recent changes in the illegality doctrine may reflect the use of legal mobilization to reassert the fundamental rights of irregular migrant workers. In this way, litigation can become an act of political resistance to an unjust ‘civil order’, and the courts may provide a forum where more emancipated visions of ‘civil order’ can be asserted by the oppressed. In ­chapter 18, ‘Modern Slavery, Domestic Work and the Criminal Law’, Jonathan Herring provides a critical account of the political and regulatory consequences of modern slavery

A Framework for Discussion  25 discourse. In this respect, his work echoes and amplifies the powerful critical work of Judy Fudge and Kendra Strauss.60 The criminalization of ‘modern slavery’ frames the regulatory problem around evil wrongdoers, and represents enslaving and trafficking as a manifestation of a (usually foreign) vice perpetrated by organized criminal gangs. This marginalizes the responsibility of states in maintaining labour law and migration regimes that create the structural conditions for exploitation and abuse. Herring then traces the continuities with domestic abuse and the forms of abuse that are criminalized in the MSA. He argues persuasively that modern slavery practices cannot be severed from a wider appreciation of patriarchal structures in society. This opens up an interesting agenda for future research, which is to consider the overlaps between family law, labour law, and criminal law, in theorizing and critiquing modern slavery. In ­chapter 19, ‘The Persistence of Criminal Law and Police in Collective Labour Relations’, Alan Bogg, KD Ewing, and Andrew Moretta examine the role of criminal law and police power in regulating and restricting collective labour activities. In so doing, they challenge the dominant view that the historical development of collective labour law involved the progressive withdrawal of criminal law from trade union affairs. Like Aliverti, the authors draw upon sociologically and historically based accounts of criminal law to examine the development of collective labour relations across different historical periods. They argue that a wider framing of criminal law interventions, to include discretionary ‘police’ measures, reveals a much more persistent presence for criminal law as a mode of repression and control. This has been most pronounced during periods of ‘emergency’, such as the General Strike, wartime, and the Miners’ Strike during the mid-​1980s. More recently, the authors suggest that there has been a ‘normalization’ of emergency, with the radical trade unionist emerging as a new dangerous status that is the object of covert police surveillance and disciplinary control. In the fourth part of the book, we return to the theme that animated Kahn-​Freund and Wedderburn, and continues to shape contemporary debate around criminalization, which is its deterrent value in promoting effective enforcement of labour s­ tandards. In ­chapter 20, ‘Workplace Safety and Criminalization: A Double-​Edged Sword’, Paul Almond argues that the contribution of criminalization to better health and safety in workplaces has been limited by certain contextual features of this regulatory method. His chapter focuses on the role of criminal law in the Health and Safety legislation and the corporate manslaughter offence. In particular, he argues that criminal law interventions are ­gravitationally oriented towards individualized notions of fault, capacity, choice, and responsibility. Once the ­liability enquiry is structured in this highly personalized way, the regulatory ­capacities of the criminal law to secure effective and enduring structural change are limited. This echoes the concerns of Alan Bogg and Mark Freedland with the overly personalized framing of the harassment wrong in criminal law. It remains an open question whether the criminal law can accommodate approaches to responsibility that are more attuned to structures, cultures, and organizational norms. In ­chapter 21, ‘The Criminalization of Health and Safety at Work’, Michael Ford adopts a more historical perspective on the development of criminalization practices in Health and Safety law. In so doing, he emphasizes the divergence in perspective as between criminal 60 Judy Fudge and Kendra Strauss, Migrant Domestic Workers in the U.K. (Foundation for Law, Society and Justice 2016).

26  Criminality at Work lawyers and labour lawyers on fundamental matters of value. Criminal lawyers have tended to focus on the development of criminalization-​limiting principles as an exercise in normative theory, whereas labour lawyers have tended to focus on instrumental outcomes in terms of whether health and safety outcomes are improved. If criminal law works in that instrumental sense, then so much the better, and that supersedes niceties about the justifiability of criminalization. Interestingly, Ford identifies the central importance of criminalization as a tool of deregulation in the modern era, following the removal of a civil right to seek compensation for breach of statutory duties under the Health and Safety legislation. By the channelling of enforcement exclusively through the criminal law, the individual worker is thereby disempowered in her standing to control the legal process and its outcomes. In this way, the criminal law can sometimes be a tool for weakening enforcement, not strengthening it. We must be cautious, therefore, in deploying the terminology of ‘worker protective criminalization’. In the final chapter of the fourth part, ‘Accessory Liability for National Minimum Wage Violations in the Fissured Workplace’, Alan Bogg and Paul S Davies examine whether ‘­accessory liability’ could provide a way of attributing criminal liability to lead companies in supply chains where those lead companies are not functioning as ‘employers’. For e­ xample, company X subcontracts a particular economic activity to company Y, and Y then employs workers to fulfil the requirements of its commercial contract with X. Accessory ­liability criminalizes those who intentionally assist or encourage the commission of principal ­offences, thereby extending the web of criminal liability beyond principal parties. This could provide a principled way of responding to enforcement problems in the ‘fissured workplace’. This would be consistent with the requirements of fair labelling and culpability in general criminal law, by avoiding the need for fictional attribution of ‘employer’ status to entities that are not employing. In the fissured workplace, the responsibility for employer compliance with labour standards appears to rest increasingly with powerful economic entities that condition the affordability of employment rights through their commercial bargaining with these smaller firms that act as the direct employers. Yet these lead companies are not acting as employers or, in criminal law terminology, ‘principals’. Drawing upon the Australian experience with accessory liability, Bogg and Davies argue that accessory liability could provide a principled tool to improve enforcement outcomes in relation to wage theft. This is because criminal liability is being targeted at large companies occupying strategically important positions in particular economic sectors. This could prompt widespread changes in employer compliance by modifying the economic incentives created by the commercial behaviour of lead firms. They propose a statutory definition of accessory liability that could be used as a basis for legislative reform. The four sets of chapters described in the foregoing paragraphs have, we feel, provided a rich and multifaceted discourse about the role of criminal law in the regulation of work relations. We have been specially fortunate in having been able to create a fifth part of this book which consists of four chapters of comparative reflection, one from the USA (Noah Zatz’s ­chapter 25), one from Canada (Bruce Archibald’s ­chapter 26), and two which are comparative within themselves, one as between Australia, the UK, and the USA (Marilyn Pittard’s ­chapter 24), and the other as between Canada and the UK (Judy Fudge and Eric Tucker’s ­chapter 23).

A Framework for Discussion  27 This variety of comparisons itself begins to suggest the sense in which this comparative part of the work makes a particularly important contribution to the book as a whole, and, we believe, constitutes a very fitting way to conclude it. That is because these chapters completely avoid the rather formulaic approach to comparative law in which each chapter narratively presents the state of the law in one or more jurisdictions. Instead, each of these four chapters represents in itself a study of one or more aspects of ‘criminality at work’, choosing to do so in relation to one or more jurisdictions other than the UK which is the primary location of this work as a whole. In combination with each other, the four chapters provide a microcosm of the book as a whole. Indeed, they serve to extend the scope and ambition of the work, and they very fully bear their share of the task of accomplishment of the project as a whole. Jointly and separately, these chapters express and reflect the continuity, the internal interrelatedness, and the experimental character, of the book as a whole. In so doing, they move effortlessly and convincingly between the particular and the general, between past and present, and above all between normative critique and the expression of positive aspirations for what the criminal law might achieve in the regulation of work relations. The first chapter of the fifth part of the book is a historical and comparative one; Judy Fudge and Eric Tucker are interested in the historical asymmetry which they discern in the evolution of labour law, both in the UK and in Canada, between the use of the criminal law as on the one hand an instrument of the discipline and repression of the workforce and on the other hand as a means of the protection of workers. The worker-​ disciplinary or worker-​repressive tendency is represented in the history of both labour law systems by the Master and Servant legislation which originated in the UK but was replicated in Canada. They regard this species of criminalization as having had ‘shallower roots’ in Canada than in the UK and as having faded out sooner in Canada, though not before ‘employers were actively developing other labour control strategies less directly dependent on law and the state’. However, for these authors there is nevertheless a similar historical asymmetry in both jurisdictions between worker-​repressive and worker-​protective criminalization; for the relative mildness of worker-​protective criminal law, discernible in both jurisdictions, is in their view somewhat more marked in Canada than in the UK. They make this analysis of the historical trajectory of the Factories Acts and other workplace Health and Safety legislation in both legal systems. Both systems display the tendency to regard the offences created by such legislation as merely regulatory ones rather than as true crimes, and to ‘conventionalize’ them as such: but in Canada this treatment of them as mala prohibita rather than as mala in se actually has the constitutional effect of relegating them to the sphere of provincial police regulation rather than that of federal criminal law enforcement. The resulting asymmetry seems to them to represent an undue deference to freedom of contract and to the power and prerogative of employers, from which, they conclude, ‘weakly enforced regulatory law was born’. The next two chapters of this part provide between them a comparative study of an important aspect of ‘criminality at work’ not covered elsewhere in the book. This is because they both consider the ways in which and the extent to which conviction and punishment of workers of and for criminal offences, whether work-​related or not, may bring about critical constraints upon their access to employment, or at least to decent employment. In her

28  Criminality at Work chapter on ‘Criminalization, Social Exclusion, and Access to Employment’, Marilyn Pittard provides a critical analysis of the impact of criminalization practices on access to employment in Australia, the USA, and the UK. As increasing numbers of citizens are subject to criminal conviction, often the very ones who have already experienced social and economic disadvantage, this can create a further barrier to labour market participation and to full integration into society. Thus the criminal law can be instrumental in creating a stratified precariat based upon ‘discreditable’ statuses, such as those of irregular migrant or ‘ex-​convict’, creating structural traps in forms of employment with poor conditions and prospects. The chapter explores the ways in which labour laws may on the one hand aggravate this form of social exclusion—​for example by intensifying the various requirements for criminal record checks and ‘right to work’ checks upon those seeking employment, or may on the other hand relieve such social exclusion by allowing past criminal convictions eventually to be concealed and forgotten. The author’s normative concern is to find the right balance in this respect between, on the one hand, the necessary safeguarding of the employing enterprise, its workers, and the recipients of its services, and, on the other hand, the maximizing of the possibility of rehabilitation into society of those with criminal convictions, especially those who have served terms of imprisonment—​reintegration into employment being one of the primary means and conditions for such rehabilitation. It is notable how, in Marilyn Pittard’s chapter, the USA often figures as the jurisdiction in which practices of restriction of access to employment or good-​quality employment on the basis of previous criminality are the most severe. In Noah Zatz’s extremely innovative chapter, it emerges that these manifestations are merely the tip of a menacing iceberg. He presents a very striking configuration of the relationship between the criminal law and the world of work: in the USA, custodial punishment for criminal offences, bearing especially heavily upon the population of colour, is so extensive as to identify that country as a ‘carceral state’; and it is one in which this enormous resort to imprisonment has a dually repressive effect upon the workforce. On the one hand, it excludes many people from work while they are in prison and after they have emerged: but on the other hand, and almost paradoxically, it coerces many people into work, and moreover into poor quality work, by requiring them to accept such work either as prison labour or as an alternative to imprisonment. The significant insight of this chapter is that this dual phenomenon, taken to the extreme as it has been in the USA, actually transforms the nature of the labour market, and that we have to include a consciousness of this transformative effect of criminalization in our basic understanding of labour law’s framing of the regulation of personal work relations. It is encouraging to be able to conclude this account of the comparative part of our book on a slightly less sombre note. In an equally innovative but very different mode, Bruce Archibald’s chapter suggests a way of enriching the normative theorization of the interface between labour law and criminal law in Canada. He homes in on the role of the criminal law in enforcing worker-​protective labour standards, in particular with regard to workplace health and safety. Focusing specially on penal policy in respect of violations of health and safety standards by employing enterprises and by individual members of the staff of those enterprises, he contends that there is real scope for bringing to bear the principles and tenets of restorative justice upon the practice of applying criminal or quasi-​criminal sanctions in this regulatory domain.

A Framework for Discussion  29 This might in his view generate some more nuanced and creative regulatory approaches than those which are sometimes manifested in high-​profile corporate criminal prosecutions and by the imposition of blockbusting fines upon such corporations. Interestingly, he argues that certain of the currently much-​discussed human capabilities approaches to legal regulation might be deployed to develop and flesh out a methodology of restorative justice in this particular context, thus contributing an important element to the evolution of an overall notion of ‘responsive workplace law’. He indicates, very cogently in our view, that progress in these directions is more likely to take place in unionized environments where workers’ collective voice is vigorous and likely to be heard. All this serves, in our suggestion, to show how the fifth and final part forms a cumulative high point of the work as a whole, and an appropriate conclusion to it. We now return briefly to our starting hypotheses and reflect upon these in light of the findings in the book: (i) Whereas criminal law had over a long historical period been in retreat from the sphere of employment relations, it was now resurging into labour market regulation, especially in recent statutory interventions such as the MSA and IA.   We think that this proposition has been borne out. Furthermore, the resurgence of criminal law extends beyond these rather dramatic statutory episodes, which could all too easily be dismissed as high-​profile aberrations. We also see this resurgence in the context of health and safety enforcement, political discourse around minimum wage underpayment and non-​payment, the possible expansion of licensing regimes for labour market intermediaries, the criminal regulation of care workers and doctors, and even the regulation of trade union internal affairs and ‘leverage’ protest activities. The chapters in the first part of the book provide a compelling overview of the full extent of this terrain. (ii) This resurgence was part of a wider phenomenon of ‘over-​criminalization’ driven by the politicization of criminal justice practices.   We are somewhat cautious about the ‘over-​criminalization’ presentation. Others have voiced concerns with this discourse:  how much criminal law is too much?61 There are also formidable difficulties with achieving a precise quantification of ­criminal law growth.62 These are exacerbated when we are trying to isolate ‘labour law’ offences, without an agreed definition of what this boundary might encompass. It would be better, perhaps, to assess whether particular instances of criminalization are legitimate and effective in securing welfarist goals in the labour market. Sullivan’s chapter provides a very powerful case for a social democratic criminal law. Whatever misgivings have been expressed in this volume about ‘regulatory’ criminalization, this has usually reflected empirical criticisms about enforcement outcomes rather than ‘in principle’ objections to regulatory criminalization. Where regulatory criminalization is supported through adequate public resourcing for enforcement ­agencies, and where this leads to higher levels of employer compliance, we would regard it as a welcome addition to the field of labour regulation. Indeed, further expansion of the

61 Horder (n 24). 62 James Chalmers and Fiona Leverick, ‘Quantifying Criminalization’ in RA Duff, Lindsay Farmer, SE Marshall, Massimo Renzo, and Victor Tadros (eds), Criminalization:  The Political Morality of the Criminal Law (OUP 2014) ch 2.

30  Criminality at Work criminal law might be valuable in certain contexts, as with accessory liability for lead companies in supply chains, a new offence of wage fraud, criminal liability for blacklisting and other human rights abuses, or a new harassment crime focused on corporate wrongdoing. On other hand, some expansion has been very detrimental for the most vulnerable workers, for example in relation to the new care worker offences. And some proposals to extend the criminal law would appear to have dubious value, as in using the criminal law to bolster the employee’s contractual duty of loyalty or to suppress legitimate industrial protest.   We think that the traditional formulation of worker-​protective labour law provides a very useful rule of thumb for assessing new instances of proposed criminalization. Briefly stated, this identifies the function of labour law as being to regulate and equalize the inequality of power between an employer and an individual worker. Where criminalization is seeking to further this goal, by assisting the weaker party, the regulation is likely to have a presumptive legitimacy from a labour law perspective. By contrast, where criminalization impedes this goal, for example where the criminal law has the purpose or effect of intensifying the subordination of the weaker party, we would regard such legislation as presumptively illegitimate. This helps to account for the deep suspicion and distaste for anything that smacks of Master and Servant law or the suppression of worker combinations. Contemporary manifestations of this include the examples of detrimental criminalization set out in the preceding paragraph. (iii) That much of this ‘over-​criminalization’ was of dubious legitimacy either in targeting non-​wrongful statuses (such as migration statuses) or using preventive measures to criminalize conduct on the basis of risk of harm (eg licensing regimes for labour market intermediaries or preventive ‘labour market enforcement orders’ under the IA).   We agree that some of this criminalization has been of very dubious legitimacy, and it does not surprise us that we have identified the migration field as providing the most egregious examples of this. However, the work on licensing and other preventive measures in labour market governance suggests that a well-​designed and properly resourced ‘preventive’ criminal law could be a valuable addition to worker-​protective labour law.   We also think that the labour law field provides interesting new insights into the mala prohibita/​mala in se distinction as it is used in criminal law theory. The different permutations of labour offences surveyed in this work support Duff ’s suggestion that the difference at stake is one of degree rather than kind. Furthermore, the field of regulatory criminalization provides some powerful examples of legitimate mala prohibita offences that have made an important and positive difference to the lives of the most vulnerable workers. We think that the interesting debates around modern slavery criminalization demonstrate this. We would do far better to invest political will into the effective enforcement of basic labour standards for all workers, perhaps through mala prohibita crimes, than through the selective criminalization of the most egregious forms of labour exploitation through mala in se crimes. At the very least, we hope that this volume assists in the rehabilitation of mala prohibita crimes as worthy objects of study in general criminal law scholarship; and as being crimes that have an important welfare-​enhancing role to play in a social democratic criminal law.

A Framework for Discussion  31 (iv) That criminalization was increasingly driven by ‘thick’ moral judgements about vice and bad character of defendants, especially in the context of modern slavery, which has tended to obscure wider appreciation of the structural determinants of precariousness and vulnerability to abuse.   We think that this phenomenon is an important and troubling feature of the new profile of criminalization in labour law. It is a phenomenon that is very powerfully analysed by Nicola Lacey in her work on changing patterns of criminal responsibility. This volume has revealed that the criminal law’s role in obscuring regulatory structures and organizational practices is more widespread than we had anticipated. It extends beyond modern slavery to include doctors’ liability for gross negligence manslaughter, criminal liability for care workers, harassment, Health and Safety laws, and corporate manslaughter. In this way, the labour law perspective is very valuable in keeping these wider structures in full view when framing ‘criminality at work’. This counteracts the new criminal law moralism centred on notions of vice and bad character. It would tend to support greater experimentation with approaches to criminal responsibility based upon corporate or organizational attribution. (v) ‘Worker protective’ criminalization, attaching criminal penalties to certain labour law violations, appeared to be a new and paradoxical feature of the neoliberal governance of deregulated labour markets, and this paradox required further investigation and explanation.   We continue to think that this feature is of the utmost importance, though we now see more clearly that the conjunction is less paradoxical than it at first appeared to be. Aliverti’s chapter on ‘civil order’ and migration provides some important insights into this phenomenon. So does Noah Zatz’s chapter on the carceral state. This context of migration criminalization seems to represent a particularly toxic form of disciplinary control. On the one hand, a deregulated labour market entrenches the private disciplinary power of employers. On the other hand, a criminalized regime of migration control entrenches the disciplinary control of the state in policing the boundaries between ‘good’ citizens and ‘bad’ non-​citizens. This conjunction is particularly troubling where private employers are co-​opted into public enforcement of border controls. Criminal law as labour market governance becomes a hyper-​charged form of public-​ private disciplinary control to keep the precariat in their place.   We should also note the complex interactions between criminalization and deregulation. Ford’s chapter demonstrates how criminalization can itself be a deregulatory technique, where the standing of the individual worker to seek compensation in the civil courts is removed. Where enforcement becomes the ‘property’ of the state, it is then for the state to decide whether to investigate and prosecute criminal breaches. Enforcement also becomes very vulnerable to political choices about the allocation of resources to public agencies. It is important, therefore, to remember that context is very important in distinguishing ‘deregulatory criminalization’ from ‘regulatory criminalization’. This will often require us to look beyond what is in the statute books to what happens ‘on the ground’ in real workplaces. (vi) Certain types of labour wrong, such as blacklisting of trade unionists or the coercive trafficking, appear to constitute serious public wrongs warranting justified criminalization, so how should we identify the category of ‘labour wrongs as public wrongs’ as a basis for reasonable legislative choices to criminalize labour wrongs? And on what

32  Criminality at Work basis should we opt for criminal law rather than private law, statutory rights enforced in the civil courts and tribunals, or professional regulation?   The chapters in the second part of the book provide some very useful engagements with the question of public wrongs and how to identify them. As is to be expected in a normative enquiry of this kind, and in a messy (and interesting) moral universe of pluralistic and incommensurable values, there is continuing disagreement about what counts as a public wrong. The two contrasting chapters on sex work exemplify the richness of those debates. Here we think labour lawyers have much to offer to criminal lawyers, in providing some context and content to the abstract idea of public wrong. Labour lawyers can also offer insights into how different types of regulation operate in workplaces, reflecting the strong socio-​legal basis of much labour law scholarship. On the question of criminalization in comparison with other regulatory modes, most of the chapters emphasize a ‘criminal law plus’ model. That is to say, there are often serious problems where criminal law operates as a substitute for other forms of legal protection. The criminal law usually works best where it is one element in a wider regulatory structure that also provides for individual rights of enforcement, collective bargaining, statutory employment rights, licensing, and other forms of preventive regulation. There is a moral dimension to this determination: it is not simply an exercise in measuring the comparative efficiency of different regulatory types. In the harassment chapter, for example, dignity is identified as a foundational constitutional value that is honoured where the individual is given standing to enforce her own rights in the legal process. In this way, private enforcement has its own moral value. In turn, we might reimagine the criminal process so that opportunities for voice for workers and their trade unions are incorporated into the criminal justice system. The chapters also generally highlight the importance of interrogating democratic law-​ making practices in criminalization decisions. How are criminalization decisions made and implemented in the legislative process? Which Select Committees have oversight over these decisions, and how effective is that oversight? In what ways could the constitutional context to law-​making be reshaped so that this process is more sensitive to the normative principles that ought to limit the reach of criminalization? A new conversation with constitutional lawyers, labour lawyers, and criminal lawyers would, we think, be very welcome and productive. We recognize that all of this generates more questions than it does answers. We regard that as an intellectual vindication of the study of ‘Criminality at Work’, and we commend this volume to our readers as the beginning of a new and exciting frontier of disciplinary engagement. The need for reasoned and evidence-​based reflection on the criminal law as a regulatory technique in work relations could not be more acute. It is perhaps too much to hope that this might provide a basis for a different kind of politics of criminalization, given the troubling overlaps between criminalization and populism at the current time. Still, the intellectual richness and humanity on display in this volume makes us think that it is worth the struggle.

PART I

CR IMINA L IT Y AT WOR K : MA PPING T HE T E R R A I N

2

Workplace Welfare and State Coercion GR Sullivan

A.  Introduction Imagine the UK has a social democratic government that aspires to provide adequate welfare for all. For our purposes welfare is conceived materially, with good physical and mental health at the core. Of course, good mental and physical health requires more than food, clothing, housing, education, and free or affordable healthcare. There would be fair and respectful treatment of persons by other persons and organizations given forms of empowerment over the welfare of others.

1. Social Democracy A material account of welfare follows from the nature of social democratic governance supposed here. There would be no advocacy of any version of the good life. The key task would be ensuring sufficient resources for sustaining a healthy and aspirational life for all. Laws requiring fairness and respect in the treatment of persons would be enforced effectively. What is required to secure such a state of affairs is not described here. Instead the existing legal protection of some crucial welfare interests of workers will be scrutinized to see if they are protected in England and Wales1 to a level that would satisfy a social democratic government of the kind supposed. ‘Social Democracy’ is a suitable term for all democracies which enact equality of civil status for all adult citizens and make significant governmental interventions in the economy if necessary, to protect the welfare of workers (and their dependents) generally and/​or for a particular group of vulnerable workers. A  neoliberal government favouring non-​ interference in markets, low taxes, and frugal welfare payments would not be a social democracy in these pages. A government elected to bring about a significant socialization of the economy would be in our social democratic club if it respected core democratic rights and freedoms.2 It might prefer to describe itself as a socialist democracy, which would be fine. The kind of social democratic governance assumed here is broadly UK governance from 1951 to 1979. Nostalgia must be avoided. There have been social gains since 1979, particularly rights and protections related to gender, ethnicity, and sexuality. Yet, until the 1 England and Wales comprise a unified criminal law jurisdiction and, despite Welsh devolution, broadly the same distributional funding arrangements for central and local government, which is not the case for Scotland and Northern Ireland. When dealing with aspects of immigration law, the rules apply to all areas of the UK. 2 For a thoughtful account of the strategies and commitments required for a successful alliance between socialism and democracy see Keith Ewing, ‘Democratic Socialism and Labour Law’ (1995) 24(2) Industrial Law Journal 103. GR Sullivan, Workplace Welfare and State Coercion In: Criminality at Work. Edited by Alan Bogg, Jennifer Collins, Mark Freedland QC, and Jonathan Herring, Oxford University Press (2020). © Alan Bogg, Jennifer Collins, Mark Freedland, and Jonathan Herring. DOI: 10.1093/oso/9780198836995.003.002

36  Criminality at Work Thatcher government, administrations of different stripes and tones of voice were in practical terms more engaged in delivering welfare for all sectors of society than any post-​1979 administration, albeit in some cases because of prudential considerations rather than beneficence. By and large political parties argued their respective cases using social and economic data derived from official and academic sources. From their differing perspectives, competing parties advocated realistic objectives rather than wishful thinking fuelled by nativist delusions.3 The Atlee government of 1945–​1951 was a democratic socialist and epoch-​changing government that left its mark for many decades.4 Although its industrial/​economic policies were not carried forward, key social provision–​–t​he National Health Service; free milk, orange juice, and meals for school children; the large scale building of social housing; controlled rents and security of tenure for private tenants; financial support for the unemployed; old age and death related pensions; steep rates of taxation on high earnings and investment income–​–​were not swept aside. Until Thatcher, much of this welfarism survived. Also sustained till then was a degree of restraint in executive and bankers’ pay, far from the stratospheric rates of today. From Thatcher onwards, there have been significant reductions in welfare provision, a process checked somewhat by the Labour governments of 1997–​20105 but then a process re-​engaged and intensified by the coalition government of 2010–​2015 and the current Conservative government. There have been cuts to and the freezing of many benefits (including disability benefits), the reduction of tax credits, and major cuts in support services. The underfunding and chaotic delivery of Universal Credit have compounded these hardships, sending many households into sudden crisis. A dearth of social housing has left millions paying high rents and enduring the short-​term leases and poor accommodation typical of many private rentals.6 Changes in the nature of employment have not helped. Over time, but particularly from 1979 onwards, many reasonably paid jobs in industry, mining, utilities, and transport have gone.7 Services constitute 80 per cent of the economy. Many service jobs are low skilled and low paid. Zero hours contracts entail fluctuating incomes. Minimum wage rates are evaded by employers who turn their workers into independent contractors.8 The circumstances of

3 Contrast the tone and content of the debate between the competing sides in the referendum on EU membership in 1975 and level of debate during the referendum on the same question held in 2016. For an account of how, what has been called, ‘Democracy of Feeling’ has become ever more salient see William Davies, Nervous States: How Feelings Took Over the World (Jonathan Cape 2018). 4 AJP Taylor wrote of the election of this government, ‘[i]‌mperial greatness was on the way out; the welfare state was on the way in. The British Empire declined; the condition of the people improved.’ AJP Taylor, English History 1914–​1945 (OUP 1965) 600. 5 The Labour governments of 1997–​2010 promised more generous welfare, a promise to some extent delivered particularly for children in poverty. But these governments brought no changes of consequence to trade union legislation, or to the building of social housing and retained the rates of personal and corporate tax they inherited. 6 For a full account of these matters together with proposals for improvements see Prosperity and Justice—​A Plan for the New Economy: Final Report of the IPPR Commission on Economic Justice (5 September 2018)  accessed 31 July 2019. For a scathing critique of economic justice in the UK see Philip Alston, ‘Statement on Visit to UK’ (16 November 2018) accessed 31 July 2019. 7 Michael Kitson and Jonathan Michie, ‘The Industrial Revolution: The Rise and Fall of UK Manufacturing, 1870–​2010’ (June 2014) Centre for Business Research, University of Cambridge WP No 459  accessed 31 July 2019. 8 IPPR Commission Report (n 6).

Workplace Welfare and State Coercion  37 persons working in the UK without a right to work are frequently dire and not in accord with domestic and international human rights standards.

2.  Why a Social Democratic Perspective? Social democracy of the kind envisaged here may no longer be politically feasible in the UK yet standards of welfare achieved from 1951 to 1979 can provoke critical reflection on our present situation. Two of the three welfare interests which will be discussed, namely rates of pay and health and safety, allow direct comparisons between then and now. Wages should be set at a level allowing a full-​time worker to meet for herself and her dependents, the costs of accommodation and household expenses, the funding of a healthy diet, and suitable clothing. Work hours should not be excessive. All reasonable steps should be taken to safeguard the physical and mental health of workers. Employers should be compliant with health and safety standards. The third area of discussion is the legal regime relating to workers without a right to work. Here no direct comparisons are possible: such workers and their conditions were not a salient issue in the period 1951–​1979. One would expect a social democratic government to comply with domestic and international human rights standards, in particular to respect asylum entitlements and be receptive to well-​founded claims for exceptional leave to enter or remain. It is against these expectations that these three areas of workplace law and practice will be critically assessed. A particular focus will be on what the criminal law can contribute to the meeting of these expectations. But first we must ask to what extent can social democratic values influence the content and form of the criminal law?

3.  Social Democratic Criminal Law The academic study of Anglophone criminal law has been enriched by a group of scholars often dubbed criminal law theorists. Broadly, they take the core of criminal law scholarship to be justifying retributivist responses to particular tokens of wrongdoing.9 Retributivist theorizing is most at ease when dealing with completed wrongful harms: the liberalism of these theorists10 makes awkward acceptance of those offences whose rationale is preventative rather than retributivist and which may criminalize conduct quite remote from the realization of any harm to a legally protected interest.11 Trials for crimes such as murder and rape vividly exemplify the need to respond both morally and effectively to serious wrongdoing. Yet the modern criminal law is also a major 9 Prominent criminal law theorists of this stripe are Antony Duff, John Gardner, Douglas Husak, Michael Moore, and Andrew Simester. There are theorists of a different stripe such as Nicola Lacey and Alan Norrie who insist that conceptual analysis and moral philosophy cannot tell the whole story of the criminal law, a story that must include consideration of the social and political circumstances prevailing at the time of any legislative or doctrinal development to be studied. 10 For discussion of the linkage between liberalism and retributivist theorizing see Paul Roberts, ‘Criminal Law Theory and the Limits of Liberalism’ in AP Simester, Antje Du Bois-​Pedain, and Ulfrid Neumann (eds), Liberal Criminal Law Theory: Essays for Andreas von Hirsch (Hart 2014). 11 For a robust defence of preventative crimes, see Jeremy Horder, ‘Harmless Wrongdoing and the Anticipatory Perspective on Criminalization’ in GR Sullivan and Ian Dennis (eds), Seeking Security: Pre-​empting the Commission of Criminal Harms (Hart 2012).

38  Criminality at Work resource in the regulation of activities that are lawful and useful, in many cases indispensable, yet which in the public interest require regulation. Much of this regulation takes the form of offences of strict liability. The justification of strict liability requires a move from deontology to consequentialism. The move will be resisted by retributivists. It may even be wished away as in the case of Michael Moore: ‘offences of strict liability are not real crimes’.12 But they most certainly are; many strict liability offences can be punished by imprisonment. By contrast, Barbara Wotton, a consequentialist all the way down, dismissed retribution. For her, a move to strict liability across the board was a move to a long overdue modernity.13 This was famously disputed by HLA Hart. He, like Wotton, was a Benthamite when it came to punishment. The practice was justified if conducive to general welfare. But achieving welfare had to be tempered by fairness and justice. Punishment should be restricted to those who broke the law despite having, ‘a fair opportunity and the capacity’ to comply with the law.14 For Hart strict liability was problematic. He accepted that punishment for negligence15 satisfied the fair opportunity and capacity conditions. But contra Hart, there is a role for strict liability in a social democratic criminal law, given its effectiveness in securing compliance with regulatory criminal law and the welfare gains that that entails. Many defendants will be corporations or other organizations subjected to sanctions limited to fines or remedial measures. No individual should be imprisoned for an offence of strict liability unless at a formal, post-​conviction hearing it is proved that the defendant was culpable to a degree warranting a custodial sentence.16 Subject to that, strict liability is justifiable as a regulatory goal, bearing in mind the power, influence, and contumacy of powerful actors resistant to regulation.17 Strict liability has a place in two of the areas of regulation to be discussed, namely health and safety and pay. That is not the case for all aspects of the legal regime relating to the protection of workers with no right to work. The Modern Slavery Act 2015 (MSA) enacts offences of slavery/​servitude and trafficking carrying maximum penalties of life imprisonment.18 Rightly, these offences are not strict: they require proof of culpability. Questionably, ordinary rather than gross negligence19 will suffice as culpability. As sentencing ‘steers by the maximum’ any conviction for these offences will entail lengthy custodial sentences.20 As with strict liability offences no individual found guilty of these offences on the basis of

12 Michael Moore, ‘The Specialness of the General Part of the Criminal Law’ in Dennis J Baker and Jeremy Horder (eds), The Sanctity of Life and the Criminal Law (CUP 2013) 69, 89. For critique see GR Sullivan, ‘Professing Criminal Law’ (2014) 12 Ohio State Journal of Criminal Law 267, 271–​74. 13 Barbara Wootton, Crime and the Criminal Law: Reflections of a Magistrate and Social Scientist (2nd edn, Sweet and Maxwell 1981) ch 3. 14 HLA Hart, ‘Punishment and the Elimination of Responsibility’ in HLA Hart (ed), Punishment and Responsibility (OUP 1968) 158, 181. 15 HLA Hart, ‘Negligence, Mens Rea and Criminal Responsibility’ in ibid 136. 16 Following the decision of the Court of Appeal in R v Newton (1982) 77 Cr App R 13 there must be a post-​trial hearing attended by counsel to resolve any factual matter germane to sentence not agreed by the parties, proved at trial, or implicit in the guilty verdict. 17 AP Simester, ‘Is Strict Liability Always Wrong?’ in AP Simester (ed), Appraising Strict Liability (OUP 2005) 21. 18 Modern Slavery Act 2015, ss 1 and 2. 19 D will be liable if he ‘knows or ought to know’ that the circumstances of V amount to slavery/​servitude or trafficking. 20 Particularly in the case of the trafficking offence it is easy to imagine circumstances where D arranges or facilitates the transportation of V but is no more than obtuse in failing to be aware that V is a victim of trafficking.

Workplace Welfare and State Coercion  39 negligence should be imprisoned unless at a formal, post-​conviction hearing,21 D is proved to be culpable to a degree of warranting imprisonment.22

B.  Workers with no Right to Work Workers lacking a legal right to work in England and Wales will predominantly be from non-​EU countries. Their conditions of work may be so extreme as to amount to ‘slavery, servitude or forced or compulsory labour’ in contravention of Article 4 of the European Convention on Human Rights (ECHR). This provision requires contracting states to take proactive steps to protect workers from these conditions,23 steps which should include criminal sanctions.24 The MSA, section 1 makes it an offence for D to subject V to such conditions.25 The Act also criminalizes trafficking people (transporting from one country to another, or from one part of a country to another part of the country) for the purposes of exploitation,26 an offence in line with the commitments the UK undertook on becoming a party to the Council of Europe Convention on Action against Trafficking in Human Beings 2005 (hereafter ‘Trafficking Convention’)27 and when opting-​in to the EU Directive on Trafficking in Human Beings and Protecting its Victims.28 The slavery/​servitude offence and the trafficking offence carry a maximum penalty of life imprisonment. The offences cover not only situations where D knows he is responsible for V being held in slavery/​servitude or is aware that he is responsible for trafficking V to be exploited but questionably29 also incriminate him for the slavery/​servitude or trafficking ­offences if he ought to have known the circumstances of V made for servitude or to being trafficked for exploitation. It is immaterial if V should consent to servitude or being trafficked for exploitation. For the slavery/​servitude offence it does not matter where in the world the offence is committed. The same applies to the trafficking offence provided that where the exploitation is sexual or relates to the removal of organs, the conduct would amount to an offence if it were carried out in England or Wales. The MSA reaches beyond criminalization and ostensibly reflects a strong institutional commitment to protection and assistance of victims. There is an Anti-​Slavery 21 Newton (n 16). 22 For many years there was consensus among academic criminal lawyers that negligence was an unsuitable culpability standard for serious offence. Hart broke with that consensus, but with the caveat that D must have the capacity to attain reasonable standards of care. A narrower view is taken by AP Simester who would disallow convictions for serious offences based on negligence unless an inference of bad character can be drawn from the token of negligence at issue in the case: ‘A Disintegrated Theory of Culpability’ in Baker and Horder (n 12) 45. That view is taken here. 23 In K and another, R (on the application of) v Secretary of State for the Home Department [2018] EWHC 2951 (Admin), [2019] 4 WLR 92 [37], Mr Justice Mostyn confirmed that Art 4 imposed a duty on government to take active steps to protect and support persons subjected to slavery and servitude. 24 Siliadin v France (2005) App no 73316/​01. 25 MSA 2015, s 1. The focus here is on what the Act can do to protect persons from non-​EU countries who lack entitlement to work. The Act also provides important protection for British and EU citizens subjected to servitude or trafficked for exploitation. 26 MSA 2015, s 2.  Exploitation includes slavery and forced/​compulsory labour and also includes indecent photographs of children, offences contrary to Part 1 of the Sexual Offences Act 2003, unlawful removal of organs, securing services by force, threat, or deception, securing services from children or vulnerable persons. 27 CETS No 197. The Convention was ratified by the UK on 17 December 2008 and came into force on 1 April 2009. 28 2011/​36  EU. 29 See n 20.

40  Criminality at Work Commissioner and independent child trafficking advocates.30 The Secretary of State must issue guidance relating to the identification, assistance, and support of persons found in slavery/​servitude or trafficked for exploitation31 and may make regulations concerning guidance and support.32 Victims may be granted civil legal aid to support claims for leave to enter or to remain or to pursue employment law claims.33 Provision is made for reparation orders to be paid to victims.34 To a limited extent, defences are made available to persons found to be slaves or trafficked for exploitation if they have committed criminal offences arising out of their dire circumstances.35 A recent report by the Gangmasters and Labour Abuse Authority (GLAA)36 indicates that labour exploitation is widespread. As from May 201737 the GLAA has identified 1335 forced or exploited workers, organized fuller and back-​dated payments for individual workers, and made 107 arrests. Unfortunately, this good and useful activity must be contrasted with failures to assist and support. In K v Secretary of State for the Home Department38 Mostyn J inquired of counsel for the Home Office why three years from the passing of the MSA there was no guidance relating to the identification and support of victims, despite the duty the Act imposed for the provision of guidance. Counsel could not assist the court as the Home Office declined to explain the omission.39 The case concerned trafficking victims seeking asylum whose weekly subsistence allowance of £65 had been cut by the Home Office to £37.50, the same rate paid to asylum seekers who were not trafficking victims. Those trafficking victims not seeking asylum continued to receive a £65 subsistence allowance. Mostyn J found that the terms of the contract between the Home Office and the Salvation Army which regulated the delivery of support to trafficking victims by the Salvation Army was squarely at odds with the reduction in the subsistence allowance paid to trafficking victims seeking asylum. The contention of the Home Office that the contract did permit these differing levels of support was ‘irrational and perverse’.40 That sufficed to resolve the case by payment of reparations. Had 30 MSA 2015, ss 40 and 48. 31 MSA 2015, s 49. 32 MSA 2015, s 50. 33 MSA 2015, s 47. 34 MSA 2015, s 8. Reparation is only payable to V if D has been convicted and a confiscation order made against him. In Taiwo v Olaigbe [2016] UKSC 31, [2016] 1 WLR 2653 [34] Lady Hale advised reconsideration of this condition. 35 MSA 2015, s 45. Protection is given to persons subjected to slavery or trafficked for exploitation: s 45(5). S 56 (1), (2) make clear that D is a victim of slavery if she is a victim of any conduct which constitutes an offence under s 1 (the slavery offence) and a victim of human trafficking for exploitation if the victim of any conduct which constitutes an offence under s 2 (the trafficking offence). These provisions extend the protection of s 45 to circumstances occurring before the the MSA came into force. To secure a defence, adult offenders must face compulsion to commit the offence charged, children will be excused if it was reasonable to commit the offence. Many serious offences are excluded from the defence. In R v O [2008] EWCA 2835, [2008] 9 WLUK 11, the Court of Appeal held that the trial of V for offences relating to documents was in breach of her right to a fair trial under Art 6 of the ECHR because she was a person entitled to protection under the Trafficking Convention. In R v L and others [2013] EWCA Crim 991, [2014] 1 All ER 113, the trial and conviction of V for offences relating to cannabis production was an abuse of process because his trafficked circumstances ‘extinguished’ his culpability. These authorities may offer more protection for trafficked persons than the substantive defence provided for in s 45. 36 GLAA, ‘The Nature and Scale of Labour Exploitation across All Sectors’ (May 2018)  accessed 8 May 2018. 37 The GLAA was created by the Immigration Act 2016 and was granted police powers in May 2017. 38 K (n 23). 39 ibid  [6]‌. 40 ibid [23]. The Home Office interpretation of the relevant documentation was utterly implausible. It is difficult to believe that it was made in good faith.

Workplace Welfare and State Coercion  41 it been necessary, Mostyn J would have further ruled that the Home Office had breached Article 14 of the ECHR and section 149(1) of the Equality Act 2010. The major protective gap left unremedied by the MSA is that being found in slavery/​servitude or trafficked for exploitation does not enhance the victim’s UK immigration status. Victimhood of this kind gives no right to enter41 or remain in the UK if V is 18. To remain, V must establish a right to asylum on long-​settled principles or be granted discretionary, exceptional leave to remain for a given period. Victims under the age of 18 are permitted to remain in the UK, but on reaching the age of seventeen and a half must make arrangements to return to the country of origin unless granted exceptional leave to remain.42 This state of affairs is quite deliberate. When the UK ratified the Trafficking Convention and opted in to the EU directive, the protection these measures require was not incorporated into UK law. Instead, there was an executive policy statement by the Secretary of State for the Home Department known as the National Referral Mechanism (NRM). Discretionary leave to enter or remain for persons trafficked for exploitation was only to be granted for ‘compelling’ cases, such as, using an example in the policy guidance, a terminal illness for which there is no treatment in the country of origin. A decision against a seventeen-​year-​old boy known as Stephen discloses the operational reality of the NRM. The Home Office accepted that Stephen when ten was homeless in Hanoi and trafficked to England for forced labour in a cannabis factory. He was convicted for cannabis-​related offences at sixteen and afterwards placed in foster care where he flourished, making very credible applications to universities. The Home Office refused exceptional leave to remain: S had shown: [C]‌onsiderable personal fortitude in relocating (sic) to the UK to establish a life here. There was no reason why he could not demonstrate the same resolve to re-​establish his life in Vietnam.43

The NRM came under scrutiny in R (PK Ghana) v Secretary of State for the Home Department.44 V was trafficked from Ghana, put into forced labour, given little to eat, and suffered mental and physical abuse. The experience left its mark: epilepsy, post-​traumatic stress disorder, depression, and psychosis. V was receiving treatment for all these conditions. Leave to remain for a time sufficient to complete treatment was refused: treatment was available in Ghana, if not of the same standard in England.45 Although V was a slave within section 1 of the MSA, there was nothing in the Act that helped him. Help came from Article 14(1)(a) of the Trafficking Convention. The Secretary of State, after initial resistance, conceded that the UK was bound by the terms of the Convention. Article 14(1)(a) requires the identification of victims of trafficking and an assessment of their personal circumstances so to provide protection and assistance. The Secretary of State’s guidance in the NRM omitted any reference to these obligations and was 41 Leave to enter may be sought by persons present in the UK, who lacked any permission to enter. 42 MSA 2015, s 53 provides for a limited exception for domestic workers who are able to find domestic work with a non-​exploitative employer. 43 Excerpt from Home Office letter, as reported by Amelia Gentleman, ‘Home Office Accused of Cruelty for Ordering Cannabis Slave Back to Vietnam’ The Guardian (2 February 2018) accessed 31 July 2019. 44 [2018] EWCA Civ 98, [2018] 1 WLR 3955. 45 ibid [31].

42  Criminality at Work therefore unlawful. The correct test was not whether V’s circumstances were compelling but whether his circumstances made it necessary to grant leave to remain in order to provide assistance and protection. On the face, PK (Ghana) is helpful for victims of trafficking. Yet the fate of trafficked persons will largely depend on what period of leave to remain an immigration enforcement officer deems necessary for assistance and protection. Mrs May’s ‘hostile environment’ for immigrants runs deep at the Home Office.46 On a particular day, in a particular case, an immigration enforcement officer, conscious of the Home Office deportation target,47 might decide that no period of leave is necessary, whatever the merits of the applicant’s claim. Even applicants seeking to establish rule-​based entitlements for indefinite leave to remain, such as asylum, meet all manner of difficulties, not least unreliable decision making by ground-​ level officials.48 The Immigration Act 2016, building on the Immigration Act 2014, reinforces the hostile environment for immigrants.49 It covers all those workers who are not in servitude or trafficked, who lack the right to work in the UK. For these workers, the very act of working is a crime. Sanctions relating to work done by persons lacking the right to work are not confined to employers. The worker too can be imprisoned or fined. The earnings of the worker can be confiscated as the proceeds of crime. Whether V, who has been brought to the UK, is simply a person who has evaded immigration controls or has breached terms of leave or is, by contrast, a person subjected to slavery/​servitude, or trafficked for exploitation, will frequently be a close-​run thing decided initially by a labour abuse prevention officer, an immigration officer or a police officer. Measures such as the MSA and the ratification of the Trafficking Convention, can be used to claim that the UK is actively engaged in combatting the worst forms of abuse connected with the movement of people. That claim can be challenged by the reality of an oppressive system of immigration control, leading to many breaches of legal/​human rights. This is not a state of affairs that a social democratic government, mindful of its domestic human rights law and international legal obligations, should allow.

46 Exemplified by the Windrush scandal. Persons of Afro-​Caribbean heritage and persons from Nigeria and Ghana, legally present in the UK, found themselves wrongfully detained and in many cases deported because the Home Office considered that they had provided insufficient evidence of their legal status. Additionally persons deemed to lack adequate documentation lost jobs, homes, and access to medical treatment under the terms of the Immigration Acts 2014 and 2016. See Comptroller and Auditor General, ‘Handling of the Windrush Situation’ (5 December 2018) accessed 31 July 2019. 47 For 2017–​2018 some 12,800 enforced returns against a backdrop of limiting net migration to the UK to less than 100,000 persons. 48 A recent Law Society report on the law and practice of immigration reveals that at least 50 per cent of the appeals against initial refusals of leave to enter or remain are found to be flawed by the lower or upper tier immigration tribunals or the Court of Appeal. An application for indefinite leave to enter or remain as a refugee costs £2,927 per person. The unit cost to the Home Office is £252. The appellate process will be costly and the result of a successful appeal is the quashing of the refusal of leave. And so, the process may begin again for those with sufficient resources. The Legal Aid, Sentencing and Punishment Act 2012 in substance removed legal aid for immigration cases. In 2012–​13, legal aid was granted for 22,496 immigration cases, in 2016–​17 there were three legally aided immigration cases. ‘Serious Flaws in UK Immigration System, Law Society Warns’ BBC News (12 April 2018) accessed 31 July 2019. 49 The Acts criminalize all manner of things necessary for normal life such as working, accessing free healthcare, renting property, obtaining a driving licence for immigrants lacking leave to enter or remain. No financial provision is made, however desperate the circumstances of an individual may be, unless there are dependent children.

Workplace Welfare and State Coercion  43

C. Physical Safety 1.  The Health and Safety at Work Act 1974 One might think that there would be general agreement that the workplace should be made as safe as is reasonably practicable. Assessments concerning how to make a workplace safe would, one might think, be a matter of expertise, uninfluenced by political leanings. Yet health and safety is a site of political conflict. Remarkably, individuals who raised safety concerns relating to construction projects or encouraged unionization within the construction industry became persons of interest to Special Branch. Officers passed details of these individuals to private security firms, who in turn passed on their details to employers. This led to blacklisting and unemployment for a significant number of construction workers.50 Strikes have been called by rail unions because of plans to make train drivers, deprived of a guard, solely responsible for passenger safety, regardless of train length.51 Funding for the Health and Safety Executive to oversee compliance with the Health and Safety at Work Act 1974 has been cut. The ensuing reduction of visits to construction sites is a concern for the GLAA.52 Particular health and safety interventions have been imaginatively reported and derided in the press. The criminal courts when interpreting the Health and Safety at Work Act have supported safety standards. In R v Chargot Ltd,53 V was killed in the course of work for D Ltd on a building site. His dumper truck suddenly overturned without any apparent explanation. The House of Lords ruled that the fatality of itself proved a failure to provide for V’s safety unless D Ltd could prove that it was not reasonably practicable to prevent the accident, something that could not be done in the absence of an explanation for it. In terms of optimizing safety, the decision is welcome. Nonetheless, the case is a striking example of absolute liability, the most stringent form of strict liability. The prosecution could not pinpoint anything omitted which would have prevented the accident. Such severity should be confined to fiscal sanctions against companies. At the time of Chargot, criminal liability under the Act also established a breach of statutory duty on which a civil claim could be based without the need for proof of negligence. The right to sue for breaches of the Act has been removed.54 This is unfortunate. There is a lot to be said for the analysis in Chargot as the basis of a civil law claim.

2. Corporate Manslaughter The Health and Safety at Work Act is regulatory criminal law so expansive interpretation is to some extent acceptable. Some large-​scale disasters gave rise to a strong sentiment that a 50 Metropolitan Police role in blacklisting construction workers; Dominic Casciani, ‘Metropolitan Police Admits Role in Blacklisting Construction Workers’ BBC News (23 March 2018) accessed 31 July 2019. 51 Strike action has been taken by staff in City Thameslink, Southern Trains, South Western Railways, and Northern Trains. 52 In a recent report the GLAA estimated that 50 per cent of construction workers were migrant workers who were likely to be allotted the most dangerous tasks: GLAA, ‘Labour Exploitation’ (n 36). 53 [2008] UKHL 73, [2009] 1 WLR 1. 54 Enterprise and Regulatory Reform Act 2013, s 47.

44  Criminality at Work regulatory response was inadequate to the occasion. The Zeebrugge tragedy in particular, focused attention to corporate cultures (‘P&O was infected with the disease of sloppiness from top to bottom’).55 The idea grew that organizations in their character as organizations could be culpable and could be blamed and punished as culpable agents. That idea led to the offence of corporate manslaughter. Corporate manslaughter reaches beyond workers but includes workers. If a worker is killed in the course of her work, if her employer is a company or organization, it may be guilty of manslaughter. On the face of it, this offence should have salience. Fines are unlimited: the reputational damage arising from a conviction for manslaughter is something to avoid. To date, however, the impact of the offence has been underwhelming.56 No company of any size and complexity has been convicted. What needs to be proved is unduly complex. Establishing a collective gross breach of a duty of care by a large organization’s senior management may involve scrutiny of the conduct of many individuals over an open-​ended period of time, commencing well before the proximate causes of the death(s) in question. Arguably, the concept behind the offence—​corporate agency—​is ontologically and normatively misconceived.57 Be that as it may, a simpler offence might be more effective. Such an offence might require proof that at least one person in the company, whatever their status, had the culpability for gross negligence manslaughter when becoming at least partly responsible for causing a fatality. Organizational liability for manslaughter would then follow, unless the organization could prove that it had done everything that was reasonably practicable to avoid deaths of the kind that had occurred.

3.  Never Safe Enough: The Case of Boxing A social democratic government might ask itself whether certain forms of work are so inherently unsafe, that they should not be forms of lawful employment. Some jobs can never be safe: the military, police, and emergency services are obvious examples. A degree of exposure to danger is a given. The social need for such jobs is plain. But there are other dangerous choices lacking any obvious public benefit. For instance, V may choose to be a sex worker and sometimes that choice will be an unconstrained choice. Arguably, the choice to do sex work should be disallowed,58 not least because of the physical dangers, and the prevalence of addictions and mental stress among sex-​workers.59 Here, we will consider the choice to be a professional boxer, fighting in tournaments supervised by the British Boxing Board of Control.60 Such a career choice is a dangerous choice. Typically, a commitment to 55 J Sheen, ‘MV Herald of Free Enterprise:  Report of Court No 8074 Formal Investigation’ (September 1987)   accessed 31 July 2019. 56 Celia Wells, ‘Corporate Criminal Liability—​A Ten-​Year Review’ [2014] Criminal Law Review 849. 57 GR Sullivan, ‘Expressing Corporate Guilt’ (1995) 15 Oxford Journal of Legal Studies 281; ‘The Attribution of Culpability to Limited Companies’ (1996) 55 Cambridge Law Journal 515. 58 The form of disallowance should involve criminal liability for organizers and customers but not for the sex worker. 59 There are also important arguments arising from the fact that the great majority of sex workers are women, a fact with implications of that for gender equality. 60 Founded in 1929 and the governing and rule setting body for the sport in the UK. The existence of such an organization with associated disciplinary procedures is important in the judicial acceptance that an injury is a sporting injury from which no criminal liability ensues: R v Barnes [2005] 1 WLR 910.

Workplace Welfare and State Coercion  45 such a career will involve fighting from the late teens/​early twenties to the mid-​thirties and even beyond. There will be some risk of premature death, a considerable risk of incurring serious physical injuries, and a very considerable risk of cognitive impairment the more fights a boxer has. The choice to box professionally is seen as a route to self-​fulfilment and prosperity by talented, and not so talented, amateur boxers. Big fights have world-​wide television audiences, raise large profits, and create a range of ancillary jobs for non-​boxers. Why should this form of commerce be closed off? For the forms of liberalism that dominate Anglophone criminal law theory, there should be no question of closing it off. Mill’s enormously influential harm principle,61 lays particular stress on the sanctity of autonomous choices which impact on the agent, and the agent alone. Whereas premature death, serious injury, and depersonalizing cognitive impairment are considerable setbacks to interests, on Mill’s liberal conception of state authority, they are not harms of any concern to the criminal law. If a pair of boxers, D and V, voluntarily agree to fight at a tournament, they commit no crime even in the case where D kills V by a powerful and carefully aimed blow, something endorsed though not explained by adherence to Mill’s harm principle.62 The harm principle offers a thin, denatured account of the effects of human interactions. Take V whose boxing career has ended. He has early onset dementia because he has taken many blows to his head. It is a progressive condition, which will lead to irascibility and unpredictability, destructive of close human relationships, and costly in terms of medical, familial, and social care. Sometimes, the brain damage incurred in a brutal fight will entail complete paralysis and a state of total dependence. It would be perfectly appropriate for a social democratic government to commission an inquiry into the harms and benefits of boxing and seek electoral support for its abolition if it concluded that overall, material welfare was diminished rather than enhanced by allowing professional boxing.63 This is most unlikely to happen in England and Wales in any foreseeable future. Not so long ago, any fatality or a profound injury occurring in a British boxing ring would give rise to public debate about the banning of the sport. Fatalities and profound injuries still occur but the debate is largely absent. The medical consensus that once favoured the abolition of the sport64 is no longer so settled.65 There is now a general reluctance to regulate the lifestyle choices of others. There is so much to say in favour of this development, that one must row upstream if advocating any curtailment of choices.66 It is accepted that boxers choose to be boxers. One should also accept that the choice is a lawful choice to engage with a lawful activity.

61 John Stuart Mill, On Liberty (JW Parker & Son 1859). 62 The long-​standing legality of boxing is not institutionally based on adoption of Mill’s harm principle but on a conception of boxing as an acceptable, even honourable activity—​‘the noble art of self-​defence’—​contrasting say, with organized, painful yet safe sado-​masochism. See R v Brown [1993] UKHL 19. 63 The argument for abolition is put here in terms of the welfare of boxers, yet undeniably a campaign for abolition will attract the support of persons who morally disapprove of boxing as an uncivilized practice. If boxing were to be abolished, sanctions should be visited on organizers and profit takers rather than on boxers. 64 British Medical Association, ‘The Boxing Debate’ BMA ( 1993) accessed 31 July 2019. 65 Raanan Gillon, ‘Doctors Should Not Try to Ban Boxing—​But Boxing’s Own Ethics Suggests Reform’ (1998) 24 Journal of Medical Ethics 3. 66 Which is why so much of the literature advocating the abolition of the sex trade seeks to establish that very few women working in the sex trade have chosen such work.

46  Criminality at Work This acceptance of the legality of boxing cannot be explained by referencing clear and settled law. There is no legislation or leading case establishing the legality of boxing. The leading case on its precursor–​–​bare knuckle prize fighting–​–u ​ nequivocally rules that this form of fighting is unlawful.67 A distinction is drawn between prize fighting and boxing but in very guarded terms: ‘boxing with gloves [does not make for an assault] if done in the ordinary way, and not with ferocity and severe punishment’.68 This circumspection conforms to the general principle of English law that consent may only be given for harm that is less than actual bodily harm unless a special exception applies.69 Organized sporting activity is such an exception, but the case-​law70 consistently draws a line between injury incidentally arising during the course of play and deliberately inflicted injury, the latter being outside the exemption. Indeed, deliberately inflicted injury is taken to be unlawful even if within the rules of the game.71 Yet, in boxing, directly knocking out or otherwise incapacitating the opponent is the surest way to win the fight. In the leading case of R v Brown, Lord Mustill confessed himself bemused by the acceptance that boxing is lawful.72 But it surely is because it is organized, publicized, and practised so openly, without any challenge by law enforcement authority.73 These social facts have obtained for well over a century. They seem more embedded than they have ever been.74

D.  Pay—​The National Living Wage and Minimum Wages The National Minimum Wage Act 1998 (NMWA) made mandatory a minimum rate of pay. An employer in breach of this obligation will not only incur civil fiscal sanctions,75 but will commit a criminal offence on refusal or wilful neglect to pay the required rate or comply with record-​keeping obligations.76 The rate for workers of twenty-​five and over has been rebadged as the National Living Wage, with lesser rates for younger workers and apprentices. The current rates, together with upgrades scheduled for April 2019 are as follows: NLW, £7.83 (£8.21); workers aged 21–​24, £7.38 (£7.70); workers aged 18–​20, £5.90 (£6.15); workers between 16 and 17, £4.20 (£4.35); and apprentices £3.70 (£3.90). It has 67 R v Coney (1882) 8 QBD 534. 68 ibid (Cave J). 69 Brown (n 62). 70 The leading case is Barnes (n 60). 71 R v Bradshaw (1878) Cox CC 83. 72 In Brown (n 62) the infliction of pain for sexual gratification was found to be unlawful as some of the injuries inflicted amounted to actual bodily harm. None of the injuries required medical attention. Yet it does seem to be lawful, even to cause somebody’s death, when fighting for money at an organized boxing event. 73 This assumption that social facts can generate legal norms is hard to square with the view that any norm to be a valid member of a legal system must be traceable back to a rule of recognition or grundnorm. See Brian Simpson, ‘The Common Law and Legal Theory’ in William Twining (ed), Legal Theory and Common Law (Basil Blackwell 1986) 8, for the argument that not all legal norms can be captured by the enclosed, seamless criteria for legal validity as proposed by Hart and Kelsen. 74 Indeed, forms of fighting more violent than boxing such as cage fighting are openly organized and televised without any legal intervention. One assumes there are some limits. Irving Kristol famously challenged libertarians what they would make of commercially organized gladiatorial combats between competitors agreeing to fight until the death of themselves or the opponent. Consistent followers of Mill’s harm principle should have no qualms about such contests provided the competitors and the audience are volunteers. However, Joel Feinberg, the leading modern expositor of the harm principle, conceded the force of Kristol’s example and had no objection to the proscription of such contests as ‘free-​floating evils’: Joel Feinberg, Harmless Wrongdoing (OUP 1998) 321–​23. 75 NMWA 1998, s 21. 76 NMWA 1998, s 31.

Workplace Welfare and State Coercion  47 been estimated that 5 million workers and apprentices are receiving their respective minimum rates of pay.77 The contribution that a minimum wage makes to general welfare is disputed by economists. For neo-​classical economists, a minimum wage will cut hours for workers in post, and supress recruitment until market equilibrium is restored.78 That view is now challenged by institutional economists more interested in economic practice than theory.79 In the case of the of the UK economy, the introduction of minimum wage rates has coincided with historically high levels of participation in the labour market. Yet the Living Wage and the lesser rates are low wages. The lot of many workers is monotonous drudgery for scant reward: ‘[h]‌ardship has forced people into worse jobs, demanding fewer skills and lower capital investment, so that Britain’s productivity has stalled to a degree not seen since the Industrial Revolution.’80 One obvious response is a much higher minimum wage. Before considering that, we must first examine the extent of compliance by employers with the legal duty to pay the Living and Minimum wage.

1.  Enforcing the Living and Minimum Wage The possibility of coercive reparations and sanctions including criminal sanctions is essential to achieve general compliance and to respond effectively to cases of non-​compliance. An estimate of non-​compliance made in September 2017 for workers over twenty-​five reported figures of 305,600 workers receiving less than the National Living Wage, or alternatively 278,000 such workers, according to which data base was used for the estimate.81 These are ball-​park figures and do not include the informal economy. The usual suspects–​–​retail, hospitality, cleaning, security, transport, deliveries, and child care–​–​feature as non-​compliant black spots but, surprisingly perhaps, the largest group numerically were salaried workers (paid by the month) but with no specified hourly rate.82

77 Low Pay Commission, National Minimum Wage: Low Pay Commission Report 2018 (CM9717, Crown 27 November 2018) [6.33] accessed 31 July 2019. 78 George J Stigler, ‘The Economics of Minimum Wage Legislation’ (1946) 36(3) American Economic Review 358. This remains a major point of reference in the minimum wage debate. 79 David Card and Alan Kreuger, Myth and Measurement: The New Economics of the Minimum Wage (Princeton University Press 2015). The argument continues. The World Bank has circulated a draft report (2018) advocating the abolition of minimum wages to better empower human beings to meet the challenges of machine learning and robotics. 80 William Davis, ‘What Are They After?’ (2018) 40 London Review of Books 4. Not only is the pay low but working conditions can be abysmal. James Bloodworth in his Hired: Six Months Undercover in Low Wage Britain (Atlantic 2018) convincingly demonstrated among other things what a harsh employer Amazon is. Workers wear a wristband that pinpoints the exact location of a warehouse worker and tracks their hand movements. A company, Humanyse, whose business is ‘people analytics’ manufactures an identity badge which contains a microphone to monitor conversations between employees, an infra-​red sensor to pinpoint where individual employees are, and an accelerometer to track when they move. The product is widely sold, but the company refuses to reveal which organizations they supply. Such intense workplace surveillance should engage for European workers, Art 8 of the ECHR. 81 Low Pay Commission, ‘Non-​Compliance and Enforcement of National Minimum Wage’ (September 2017) [1.13]–​[1.20] https://​assets.publishing.service.gov.uk/​government/​uploads/​system/​uploads/​attachment_​data/​ file/​645462/​Non-​compliance_​and_​enforcement_​with_​the_​National_​Minimum_​Wage.pdf accessed 31 July 2019. 82 ibid [2.8].

48  Criminality at Work Ensuring the payment of living and minimum wages is the ultimate responsibility of the Department of Business, Energy and Industrial Strategy while the enforcement process, including the payment of reparations, is entrusted to HMRC. The HMRC allows ‘self-​ correction’. If a firm or individual makes an accurate disclosure of payment arrears and pays out those arrears, the file is closed if there is an assurance of future compliance. Where that process fails the revenue may impose a civil penalty resulting in payment of double the arrears, up to a limit of £20,000 for each worker, payment to each worker of arrears, and naming and shaming if the arrears for individual workers exceed £100. For the worst cases, criminal proceedings (which will be conducted by the Crown Prosecution Service (CPS)) may be brought. This is rare. Since 1998 to September 2017, only thirteen criminal prosecutions have been brought. Such proceedings are expensive and may be something of a damp squib as the penalty is capped at a fine, not exceeding level 5. Arguably, civil penalties are just a form of business overhead and insufficient as a deterrent. The criminal armoury has been recently strengthened with the introduction of a Labour Market Enforcement Undertaking (LMEU), which requires firms or individuals to take practical steps to correct non-​compliant behaviour including failure to pay the ­minimum wage.83 Where steps are not taken or an undertaking order is refused, the business can be prosecuted for breach of an LME order (LMEO). The maximum penalty is a two-​year custodial sentence and/​or an unlimited fine. Moreover, a court convicting for the ­offence of underpayment can attach an LMEO. It remains to be seen how much the threat or the use of imprisonment will be used for cases of failure to pay the minimum wage.

2.  Employers Evading Paying the National Living Wage: Two Devices, Zero Hours Contracts; Working as an Independent Contractor A zero hours contract (where V must make herself available to work for D Ltd but is not guaranteed any given amount of work) is a device that can be used to decrease pay and related benefits. Zero hours contracts came under review in the Taylor report.84 If V cannot say at any point in time what her annual pay might be, this will make calculation of any tax credits and benefits that may be due very difficult. She may well be paid the minimum wage for each hour worked but still be paid far less annually compared to the days when she was guaranteed forty hours plus overtime rates for additional hours worked. Unfortunately, there is nothing in the minimum wage legislation to help her. It enforces an hourly rate. It does not enforce a minimum income for full-​time workers. Failing legislation which will formulate a sharp distinction between full-​time and part-​time work and which will stipulate a minimum annual income for full-​time work, zero hours contracts can be used to evade in substance if not in form, payment of the minimum wage. Another way that V may be deprived in substance of an effective minimum wage is to be refused employment as such, but offered instead payment for working as an independent contractor. For example, the delivery firm Hermes has contracts with some 15,000 ‘life-​style 83 Immigration Act 2016 (as amended), s 14(4)(b). 84 Matthew Taylor and others, ‘Good Work: The Taylor Review of Modern Working Practices’ (Department for Business, Energy & Industrial Strategy, July 2017) accessed 31 July 2019.

Workplace Welfare and State Coercion  49 couriers’ who are paid per delivery, earning less than the minimum wage. HMRC is investigating the legality of these contracts.85 If V is looking for work as, say, a taxi drive or a courier, working as a service provider rather than an employee, apart from receiving less than the Living Wage, there will be no National Insurance payments and no holiday and sickness pay. Payment for services is not covered by minimum wage legislation and can be anything agreed between D Ltd and V. Some successes have been achieved before tribunals by V proving that she was an employee rather than a service provider.86 In Pimlico Plumbers Ltd v Smith87 the claimant, a plumber by trade, was found to be a worker for the company within the meaning of section 230(3) of the Employment Rights Act 1996 rather than an independent trader providing services. The Supreme Court made this finding following a careful scrutiny of a voluminous contract that sought to create an independent trader status for the claimant while at the same time subjecting him to a strict working regime to be carried out according to the letter. Several features of Pimlico Plumber’s practices swung the case the claimant’s way:  a branded uniform; a branded van with a tracker; an identity card; detailed instructions from the control room.88 Very likely there will now be attempts by companies and their advisors to organize themselves in a way that avoids any direct replication of the features that did down Pimlico yet retain a workforce denied the status and protections enjoyed by workers/​ employees.89 Far better would be legislation that encapsulates the clear distinction between persons who provide, as required, specialist advice and ad hoc services to organizations in general and are not tied to one organization, and, by contrast, persons who on a regular basis make the products or provide the services that comprise the core business of the company and its associated companies. The latter are workers/​employees in substance, a fact that should be recognized in law.90 The National Centre for Social Research estimates that there are 2.8 million workers in what is called the gig economy, many on zero hours contracts, or made to work as self-​ employed service providers. It seems that circa 700,000 of the persons in these circumstances earn less than the minimum wage.

3.  Employees Colluding with Employers to Evade the National Living Wage Collusion may occur in many different circumstances: V1, on reaching the age of twenty-​five, is to be paid the National Living Wage. The increased pay will lead to a loss of credits and benefits, leaving her worse off. She agrees 85 Robert Booth, ‘Hermes “Trained Managers to Mislead Tax Inspectors” ’ The Guardian (4 May 2018) 12 accessed 31 July 2019. 86 Tribunal verdicts to that effect have been achieved against Addison Lee, City Sprint, e Courier, Excel, and Uber. 87 [2018] UKSC 29, [2018] 4 All ER 641. 88 ibid [48]. 89 For instance, where workers can be assigned to different locations, each location could be designated as a franchise area and the workers would be described as franchisees. One might also anticipate greater use of a contractual term allowing the ‘service provider’ to appoint a substitute to perform the services should need arise, a term which may not reflect any intention of using substitutes. 90 In the ‘Good Work Plan’ (Department for Business, Energy & Industrial Strategy 17 December 2018)  accessed 31 July 2019, the government has undertaken to examine the issue of workers versus service providers with a view to remedial legislation if deemed necessary.

50  Criminality at Work with her employer to reduce her hours as reported but in reality, to work the same hours. She works at the National Living Wage rate for the work hours recorded, but ‘off the books’ she receives less than that rate for the hours worked. She is content with the arrangement. V2 is an ambitious young chef. He is anxious to do a stint at a renowned London restaurant with a view to upping his skills, enhancing his profile, with the ultimate objective of attracting financial backing to open his own restaurant. He agrees to accept £5 an hour in cash. The restaurant keeps no record of his employment. V3 has a strong desire to work in fashion. She sees the following notice from French Connection ‘PR and Marketing Placement. Successful applicant will for 12 months provide trend reports, manage media data base, monitor press reports, and file press clippings. Zone 1–​6 travel card and lunch money’. She is delighted to be chosen for the position. In the cases of V1 and V2 we can be sure that the employer is in breach of the NMWA. The same assurance cannot be offered for the employer of V3 and discussion of the position of V3 and her employer is deferred for now. Are V1 and V2 criminally liable along with their employers? In principle and in terms of social policy, should they be found criminally liable? There would seem to be no obvious grounds for distinguishing between these two cases. However, it seems that V1 will breach the Act as a principal offender whereas V2 may not commit any offence. In V1’s case she, ‘allowed to be made in a record [of the amount of remuneration paid] an entry which she knew to be false in a material particular’ thereby infringing the NMWA.91 By accepting cash payments less than the living wage, V2 does not commit any offence under the Act, but he has agreed with his employer to accept lesser payments, thereby assisting the employer’s ­offence of underpayment.92 Arguably then, he could be charged with conspiracy to breach the Act and liable for the offence of underpayment as an accomplice. In the very unlikely event of the CPS agreeing with HMRC to prosecute V2, he could defend himself by reference to the case of R v Tyrell93 which ruled that if an offence is created for a vulnerable class of persons (in the case itself girls under the age of sexual consent) it would undermine the rationale of the offence to convict members of the protected class as accomplices, irrespective how willing their collusion with the principal offender. If that lets V2 off the hook, the same should apply for V1. There is much to be said for extending Tyrell to principal offenders where the defendant commits the offence in circumstances where essentially, in terms of the policy of the offence, she is the victim of the offence rather than the perpetrator. But what of V3 and her employer? There are clearly entirely legitimate offers of work experience as an intern, such as the opportunity offered by the Guardian newspaper to aspiring journalists to spend a fortnight observing and to some extent participating in the various stages of producing a daily newspaper and its online version. Other newspapers offer longer stays, up to three months in some cases. Some magazines offer engagements of up to twelve months and the same applies to the fashion industry, book publishing, positions with theatre and dance companies, and assistance to MPs. A recent study reveals that 48 per

91 NMWA 1998, s 31(3). Some of the offences created by this section are confined to the employer but the false record offence can be committed by ‘any person’. 92 NMWA 1998, s 31(1). 93 [1894] 1QB 710.

Workplace Welfare and State Coercion  51 cent of internships are unpaid, 27 per cent offer expenses, and 12 per cent, no pay, no expenses.94 Generally, the longer a person stays, the more useful she will become and find herself working rather than watching others work. A person such as V3 can be both privileged and exploited. Many young persons wish to work in fashion, but not that many could afford a point of entry into that industry by working for nothing for twelve months, particularly in London.95 Insisting that work should be paid for at some state enforced minimum rate militates against the exploitation and widens access for persons from all socio-​economic backgrounds. A private members bill96 proposes to limit unpaid work experience to four weeks with the applicable minimum wage rate to be paid once that period has elapsed. Whatever the fate of that bill, the HMRC could at this point in time usefully explore and police the line between work experience and work. It could presume that any internship/​work placement position becomes work per se after four weeks has elapsed, work that must be paid for at the applicable minimum rate.

4.  Is the National Living Wage High Enough? The Living Wage Foundation recommends that employers in London should pay employees at least £10.20 and elsewhere in the UK the hourly rate should not be less than £8.75. The Foundation estimates that because of wage stagnation since 2007, a person on the minimum wage in 2017 (£7.50 then) would have lost £1,809 in annual purchasing power to meet the costs of housing, childcare, transport, and food. The current increase in the NLW is more than offset by the freeze on household benefits entailing average, annual losses of £200 for minimum wage households. By 2020 it is estimated that freezes and cuts on benefits will entail annual losses of £1,000 for these households. Housing costs are a major cause of impoverishment. Home ownership is out of reach for millions with an average house price outside London of £232,554 and an average London price of £482,241.97 The average salary of a first-​time house buyer is £53,000, more in the bigger cities, and £84,000 in London.98 In 1980, when the ‘right to buy’ council homes was introduced, there were 6.5  million council houses, which housed approximately 40 per cent of the population.99 Now there are 2 million council and financially supported social houses and flats. Over 40 per cent of ex-​council houses are in the hands of private landlords. Outside London, the average council/​social housing rent in 2017 was a weekly £102; for privately owned accommodation it was £192. For London, the corresponding figures are 94 See further, Carl Cullinane and Rebecca Montacute, ‘Pay as You Go’ (Sutton Trust 23 November 2018) accessed 31 July 2019. 95 The Sutton Trust estimates that a single person needs at least a monthly net sum of £1,019 to live in London. 96 Sponsored by Lord Christopher Holmes in the House of Lords and Alec Shelbroke MP in the House of Commons. 97 ‘National Statistics UK House Price Index Summary: September 2018’ (HM Land Registry 14 November 2018)   accessed 31 July 2019. 98 Richard Partington, ‘First-​time Buyers: Average Salary Requirement Rises 18% in UK Cities’ The Guardian (27 September 2018) accessed 31 July 2019. 99 The large-​scale house building programme initiated by the Atlee government, together with that government’s creation of the NHS, established the British state as the major provider of shelter and security for a large part of the general population. Conservative governments until Thatcher also built many council houses though often to a lower specification compared to the houses built by Labour.

52  Criminality at Work £132 and £309.100 For persons between the ages of 25–​34, 27 per cent own their own homes, while 46 per cent rent from private landlords. It has been estimated that 33 per cent of those currently paying rent to private landlords will be doing so at the point of retirement and beyond.101 The accommodation provided is often far from ideal. Tenancies are typically for six months only.102 A core aspiration of our imaginary social democratic government was that an individual adult wage earner should be able to feed, clothe, and shelter his or her dependents. Such a modest aspiration was delivered by successive UK governments over many post-​war decades. We are a long way from that now.103 The prevalence of foodbanks and the rise in homelessness starkly evidence the inadequate provision of state-​funded welfare.

100 The rent payments as published by Countrywide Estate Agents, ‘Monthly Lettings Index—​January 2018’ (12 February 2018) accessed 31 July 2019. 101 Assuming merely incremental increases in state and private pensions, very large amounts of housing benefit will have to be paid to these retirees to keep a roof over their heads. 102 For an instructive account of the economic and social consequences of the demise of council housing, see John Broughton, Municipal Dreams: The Rise and Fall of Council Housing (Verso 2018). 103 Two recent films, I Daniel Blake (2016) and Sorry We Missed You (2019) both, directed by Ken Loach, vividly expose the truth about the condition of the working poor in twenty-​first century Britain.

3

Using Criminal Law to Enforce Statutory Employment Rights David Cabrelli*

A.  Introduction It is a trite statement that individual and collective employment rights without enforcement are paper tigers. To the labour lawyer whose work has predominantly focused on the amalgam of the private law and statutory regulation of the contract of employment, a seductive, albeit more difficult, claim is one positing that such enforcement should fall within the exclusive province of the civil law. It is ‘seductive’, to the extent that the intuition of such a specialist would tend to suggest that–​–​being rooted in the contract of employment1—​the employee–​employer relationship is located within the private sphere. This gives rise to the proposition that civil, rather than criminal, proceedings ought to guarantee the vindication of employment rights, that is, that traditional means of enforcing private law, namely civil enforcement, should take the lead. However, it is at once ‘difficult’ because it clearly dismisses a role for the criminal law without explanation or demur. Hence, despite its simplicity, the claim that compliance with statutory employment rights is a matter for the civil law should be suspended until it has been more intensely scrutinized. This approach is particularly important in light of Lord Reed’s reminder of the public character2 of statutory employment rights in R (on the application of UNISON) v Lord Chancellor.3 Having a public nature, something more than simply civil enforcement might be demanded. The following section of this chapter (section B) is of an expository character, being devoted to a mapping out of the key employment rights with a statutory source where an exclusive or shared role for the criminal law is encountered. Afforded equal status in this process is an account of those statutory employment rights where the criminal law is far ­removed from debates concerning enforcement. This descriptive exercise will naturally ­entail a review of the principal labour law statutes.4 Having produced such an account, * Professor of Labour Law, University of Edinburgh. I  am indebted to James Edwards, Andrew Cornford, Anthony Duff, and the editors for discussions regarding the content and writing of this chapter. 1 See Mark Freedland, The Personal Employment Contract (OUP 2003); Mark Freedland and others (eds), The Contract of Employment (OUP 2016). 2 For discussion, see Michael Ford, ‘Employment Tribunal Fees and the Rule of Law:  R (Unison) v Lord Chancellor in the Supreme Court’ (2018) 47 Industrial Law Journal 1, 9–​15, 44–​45; Alan Bogg, ‘The Common Law Constitution at Work: R (on the application of Unison) v Lord Chancellor’ (2018) 81 Modern Law Review 509, 515–​16. 3 [2017] UKSC 51, [2017] 3 WLR 409, 433C–​E (Lord Reed). 4 The Employment Rights Act 1996 (ERA), the Trade Union and Labour Relations (Consolidation) Act 1992 (TULRCA), the Equality Act 2010 (EqA), the National Minimum Wage Act 1998 (NMWA), the Health and Safety at Work, etc Act 1974 (HSWA), and the Working Time Regulations 1998, SI 1998/​1833 (WTR). David Cabrelli, Using Criminal Law to Enforce Statutory Employment Rights In: Criminality at Work. Edited by Alan Bogg, Jennifer Collins, Mark Freedland QC, and Jonathan Herring, Oxford University Press (2020). © Alan Bogg, Jennifer Collins, Mark Freedland, and Jonathan Herring. DOI: 10.1093/oso/9780198836995.003.003

54  Criminality at Work section C takes a theoretical and analytical turn. It assesses whether contemporary theories of criminal law have the capacity to illuminate the regulatory environment. As part of that process, one of the most compelling criminal law theories is probed within the rubric of the descriptive narrative provided in section B. This exercise is intended to determine whether the legislative position in the UK can be justified in terms of this normative theory. The larger enterprise of which this third section forms part will be conducted by pursuing an evaluative framework throughout, taking labour laws with their foundation in legislation–​–​ as opposed to the common law–​–​as the primary focus.

B.  Charting the Criminal Enforcement of Statutory Employment Rights A cursory glance at the content of the ERA, TULRCA, EqA, NMWA, etc would suggest that Parliament has a tendency to prioritize the civil enforcement5 of contraventions of labour laws, with a casuistic, individual litigation-​driven model generally laid down. Under this framework, compliance and enforcement is channelled principally through the employment tribunal system,6 although the courts operate as a subsidiary or supplementary option: ‘[i]‌n general, statutory employment rights in Britain are enforced by individuals taking cases to the employment tribunals’.7 Three simple illustrations suffice. Section 111 of the ERA prescribes that the right not to be unfairly dismissed at work under section 94 of the ERA is enforceable by an employee presenting a complaint to an employment tribunal. Likewise, claims brought in respect of discrimination law in the workplace are to be presented as a complaint to an employment tribunal under section 120 of the EqA. Meanwhile, the enforcement of unlawful deductions from wages claims under Part II of the ERA8 and equal pay claims under Chapter 3 of Part 5 of the EqA9 is shared between the employment tribunal and the courts. The ubiquity of civil remedies in labour law statutes might suggest a presumption against criminalization in principle, but the most that can be claimed is that the prevailing legislative pattern provides evidence of an underlying political preference in favour of civil enforcement. As noted at the outset, the fact that in practice we encounter such a preference for civil law remedies may also give rise to the normative argument that we ought to eschew a role for criminalization in the field. However, that would be wide of the mark, since it entails the derivation of an ‘ought’ from an ‘is’, that is, the argument that we should reject the criminalization of statutory employment rights because in practice, their enforcement is predominantly confined to the civil law—​is palpably false. Instead, 5 See generally, Gillian S Morris, ‘The Development of Statutory Employment Rights in Britain and Enforcement Mechanisms’ in Linda Dickens (ed), Making Employment Rights Effective (Hart Publishing 2012) 10–​14; for the negative aspects of this model, see Lizzie Barmes, Bullying and Behavioural Conflict at Work (OUP 2016) 38–​47; Dave Renton, Struck Out (Pluto Press 2012) 2–​21. 6 See Linda Dickens, ‘Employment Tribunals and Alternative Dispute Resolution’ in Linda Dickens (ed), Making Employment Rights Effective (Hart Publishing 2012) 29. 7 Morris (n 5) 10. 8 See ERA 1996, ss 13, 23, and 24. An unlawful deduction also constitutes a breach of contract, thus enabling an employee to raise a court action seeking damages. 9 The EqA 2010, s 129 enables a protected individual to present a complaint to an employment tribunal. A common law court action for damages is prescribed by the EqA 2010, s 128: see Abdulla v Birmingham City Council [2012] UKSC 47, [2012] ICR 1419.

Using Criminal Law to Enforce Employment Rights  55 a role for the criminal enforcement of statutory employment rights is undoubtedly legitimate. A separate point is that civil and criminal enforcement in this context should not be viewed as mutually exclusive. As we will see, shared enforcement is not that unusual in the case of employment rights. One of the criticisms of an insistence on a civil litigation framework of enforcement of labour laws centres on its inability to secure more powerful or redistributive outcomes. For example, where certain rights are fundamental in nature or represent a reflection of the collective goals or shared values of a community, there is an argument that more symbolic10 forms of enforcement and remedies are needed. In short, here, civil enforcement is not enough. From the perspective of securing compliance, the absence of any involvement for criminal adjudication will be of concern in many fields of the law. But even more so in the case of labour law, which can only be properly understood in light of its objectives and social/​vocational mission,11 namely to push back against the inequality of bargaining power intrinsic to the employment relationship.12 Of course, the classic counterpoint to the general proposition that labour law is predominantly enforced civilly13 finds its expression in section 33 of the Health and Safety at Work Act 1974 (HSWA). This statutory provision makes it an offence for an employer to fail to discharge various duties or to contravene any of the terms of the multitude of Health and Safety Regulations.14 The only adjunct to criminal proceedings is public agency enforcement by the Health and Safety Executive (HSE).15 As such, orthodoxy is turned on its head. This demands an explanation. Or at the very least an attempt at such a justification, which is a task to which section C of this chapter is dedicated. It may be useful for analytical purposes to evoke a tripartite taxonomy of statutes conferring employment rights, reflective of the varying dimensions of complexity involved in the enforcement of labour laws. The first category is embodied by those statutes that confer employment rights where enforcement is exclusively private or carried out or assisted by a public agency, for example the Equality and Human Rights Commission. At the polar opposite is the third category, where we find exclusively criminal enforcement undertaken by the police and Crown Prosecution Service (CPS) or investigation delegated to specialist agencies such as the Gangmasters and Labour Abuse Authority with prosecutions subsequently conducted by the CPS. The criminal forms of intervention can range from the prescription of more traditional forms of offences carrying the sanctions of imprisonment or fines, to quasi-​criminal public agency enforcement (eg the Certification Officer’s powers in TULRCA, the powers of the HSE under the HSWA, or those of the HMRC in terms of the NMWA), licensing requirements,16 breach of which triggers criminal sanctions, and preventive measures and orders to control managerial 10 However, criminal enforcement may not necessarily be stronger or more powerful than civil litigation, and may, in fact, be weaker, eg owing to the higher standard of proof in the case of criminal trials. 11 See Hugh Collins, ‘Labour Law as a Vocation’ (1989) 105 Law Quarterly Review 468. 12 Paul Davies and Mark Freedland (eds), Kahn-​Freund’s Labour and the Law (3rd edn, Stevens 1983) 18. 13 For other exceptions to the privatized model of justice, see Ford (n 2) 5–​8. 14 The various penalties are set out in the HSWA 1974, sch 3A. HSWA 1974, s 47 expressly disavows civil liability for breach of statutory duty. 15 See Steve Tombs and David Whyte, ‘Reshaping Health and Safety Enforcement: Institutionalising Impunity’ in Linda Dickens (ed), Making Employment Rights Effective (Hart Publishing 2012) 67. 16 See ACL Davies’ c­ hapter 13 in this volume.

56  Criminality at Work behaviour.17 Finally, the second category lying somewhere in between the first and third is the most complex, where the relevant statute lays down both civil and criminal liability, that is, there are overlaps. This second category demonstrates how it is misconceived to present the enforcement of employment rights as necessarily entailing a binary choice between the criminal and civil. We can pick out a limited sample of the most central employment rights conferred by statute18 and whether they correspond to categories 1, 2, or 3. For example, statutory rights such as the right not to suffer direct discrimination or indirect discrimination,19 the right to be supplied with a written statement of the main particulars of employment,20 the right not to be unfairly dismissed,21 and the failure to inform and consult on collective redundancies22 are each enforced civilly through private means or with the assistance of an administrative agency. But there are many other examples. Meanwhile, criminal proceedings are the only way of enforcing rights such as the right of employees to health and safety at work,23 the right of young workers not to be compelled to work in excess of forty hours in a working week,24 and the right to be provided with adequate rest breaks owing to the pattern of an employee’s work.25 Finally, we encounter the statutory entitlement where enforcement is shared between the civil and criminal law, for example in the case of the right to be paid the minimum wage or National Living Wage.26 Two observations can be derived from the preceding exposition of rights falling within the tripartite categories. First, the criminal enforcement of statutory labour laws is scant in comparison with the instances of civil proceedings. This may not be surprising inasmuch as it conforms to the general pattern discussed above. But we need a more reflective and better understanding of the possible reasons for this finding. Here, a pack of explanations jostle for our attention. For instance, an explanation may lie in a lack of specialism on the part of the police and CPS. Another may alight on the absence of proper funding and resourcing in the case of the specialist public agencies tasked with enforcement. Alternatively, we may resort to normative theories of criminalization for assistance, which is a task we turn to in section C below. A second remark is that the aforementioned illustration where enforcement is divided between civil and criminal law is particularly intriguing. What the existence of shared enforcement in the case of the minimum wage suggests is that the basic distinction between civil and criminal liability may be somewhat anachronistic and bear a greater amount of weight than it is currently equipped to carry. In the next section, we delve deeper into these issues, and in much broader terms, we enquire whether the prevailing configuration of enforcement we find in UK labour law statutes is defensible in terms of some overarching normative theory of criminalization.

17 A similar issue arises where there is a crossover between criminal and civil enforcement, eg where a civil breach is clothed with criminal liability in certain circumstances. 18 This analysis is not intended to be exhaustive of each and every statutory employment right in UK law. 19 EqA 2010, ss 13, 19, and 120. 20 ERA 1996, ss 1, 2, 3, and 11. 21 ERA 1996, ss 95 and 111. 22 TULRCA 1992, ss 188, 188A, and 189. 23 HSWA 1974, ss 2–​7 and 33(1)(a). 24 WTR regs 1998, 5A(1), and 29(1). 25 WTR regs 1998, 8, and 29(1). 26 NMWA 1998, s 31(1) and ERA 1996, s 23.

Using Criminal Law to Enforce Employment Rights  57

C.  The Notion of ‘Public Wrongs’ in Criminal Law Theory 1.  Introduction Having provided a sketch of various statutory employment rights that are exclusively enforced via the criminal law or civil law, and the circumstances where we encounter a complex mixture of both civil and criminal compliance measures and techniques, a natural successive point is to assess whether some existing criminal law theory can make sense of, and/​or provide a normative account of, and justification for the forms of enforcement prescribed in the case of a limited range of the statutory employment rights discussed in section B. It should be stressed that this search for a theory is motivated by the desire to explore whether criminalization would be justified in normative terms in relation to the selected range of employment rights. It is an exercise that is analytically distinct from enquiring whether a theory of criminalization can supply an exploratory account for the political choices made by law-​makers in Parliament about criminalization from time to time. This is a distinction of material importance, since there is every possibility that there might be a misalignment between what a normative theory would suggest ought or ought not to be criminalized and what we actually find has been criminalized in practice. If we embark on our enquiry by evoking normative principles and concepts that can be derived from the academic literature that identify when it might be legitimate to criminalize a practice, it is evident that a number of options present themselves as potential candidates. However, for reasons of economy of space, we will restrict the discussion to one: the ‘public wrongs’ theory of criminalization will be taken as the focus of our enquiry. Historically, the public wrongs framework claimed that criminal laws ought to regulate a form of human behaviour where it constituted a wrong done to the public. This does not seem entirely persuasive as a univocal theory of criminal law for two self-​explanatory reasons. Even a mere cursory glance at the law reveals that certain wrongs are controlled by the criminal law even though they (i) cause no direct harm to the public or (ii) have no inherent properties which are public in character.27 In recent years, a more refined incarnation of the ‘public wrongs’ theory has been put forward by Duff and Marshall.28 It is this form of the ‘public wrongs’ theory that can be treated as a conceptually more formidable account for the criminalization of behaviour. It is taken as an evaluative tool to test whether it might have the capacity in normative terms to provide support to those areas of labour law (i) that are the subject of civil enforcement, (ii) that have been earmarked exclusively for criminal enforcement, and (iii) where a breach of employment rights entails both civil and criminal enforcement on a shared basis. The notion of ‘public wrongs’ advanced by Duff and Marshall has undergone a series of evolutions since it was first versed in 1998.29 The most recent articulation focuses on the 27 Although this may lead us to conclude that illustrations of criminalization where there is (a) no harm to the public or (b) nothing public about their nature, are simply illegitimate under the public wrongs theory, rather than sufficient to question its legitimacy. 28 Antony Duff and Sandra E Marshall, ‘Crimes, Public Wrongs, and Civil Order’ (2019) 13(1) Criminal Law and Philosophy 27. 29 Sandra E Marshall and Antony Duff, ‘Criminalization and Sharing Wrongs’ (1998) 11 Canadian Journal of Law and Jurisprudence 7; Antony Duff, ‘Towards a Theory of Criminal Law’ (2010) 84 Proceedings of the Aristotelian Society, Supplementary Volume 1; Antony Duff and Sandra E Marshall, ‘Public and Private Wrongs’ in James Chalmers, Fiona Leverick, and Lindsay Farmer (eds), Essays in Criminal Law in Honour of Sir Gerald Gordon (Edinburgh University Press 2010) 70; Antony Duff, ‘Responsibility, Citizenship, and Criminal Law’ in Antony

58  Criminality at Work positive and negative embodiments of that theory, as well as their respective implications.30 The positive form of the public wrongs theory is intended to help us understand what we have a reason to criminalize, rather than what we ought to criminalize: a legitimate justification is simply a necessary, albeit not a sufficient, condition for the criminalization of conduct. The fact that conduct X concerns us as living together as members and citizens of a political community—​referred to as the ‘civil order’ in shorthand—​and X is a wrongful kind of conduct within that civil order, means that we have sufficient justification to criminalize it. Turning to the negative version, this is deployed by Duff and Marshall in an additive and more profound way. It is used to cast light on what we have no reason to criminalize, but also what ought not to be criminalized by the law. It claims that we have no legitimate justification to criminalize a conduct A, and we ought not to criminalize it, if it is not (i) a public matter in the sense that it does not concern the ‘civil order’ or (ii) a wrongful kind of conduct within that civil order: in this way, for an activity not to be criminalized, it may be (a) public, but not wrongful, or (b) wrongful, but not public. As such, the value of the negative version of the public wrongs theory to labour law lies in its negative normative edge, that is, how it purports to convey what workplace practices or exercises of the managerial prerogative we ought not to criminalize and helps us understand better those areas of labour law where a conscious decision has been made to eschew the operation of the criminal law. First, a few clarificatory remarks to make about ‘public wrongs’ theory. The notion of a public wrong does not tell us anything about the effect of the relevant conduct on the public, in the sense of whether the public are somehow wronged. It also fails to clarify whether the wrong is in some sense ‘collective’ in nature.31 Likewise, it says nothing about ‘the intrinsic character [of that conduct] as a wrong’, that is, whether it is public or private.32 For example, an inherently private act may constitute conduct that is (i) a ‘public matter’ which is (ii) wrongful. Instead, we are concerned with ‘public wrongdoing’ not ‘moral wrongdoing’33 in terms of the specific framework adumbrated by Duff and Marshall. What this means is that ‘public wrongs’ theory does not correspond to the public/​private dichotomy in traditional legal thought,34 or the debate conducted by labour lawyers as to whether we can distinguish public and private activities according to a spatial conceptualization.35 In this way, the fact that certain key workplace behaviours regulated by statutory labour laws such as dismissal, suspension, etc might take place exclusively in the private sphere between two parties to a horizontal employment relationship does not automatically translate into any claim about the proper or improper criminalization of such activities.36 Instead, the key issue here is Duff and Stuart Green (eds), Philosophical Foundations of Criminal Law (OUP 2011) 125; Antony Duff, ‘Legal Moralism and Public Wrongs’ in Kimberly K Ferzan and Stephen J Morse (eds), Legal, Moral, and Metaphysical Truths: The Philosophy of Michael S Moore (OUP 2016) 95; Antony Duff, The Realm of Criminal Law (OUP 2018) 146–​84; Ambrose Y K Lee, ‘Public Wrongs and the Criminal Law’ (2015) 9 Criminal Law and Philosophy 155, 158. 30 Duff and Marshall, ‘Crimes, Public Wrongs, and Civil Order’ (n 28) 18; Duff, The Realm of Criminal Law (n 29) 235–​37. 31 Marshall and Duff, ‘Criminalization and Sharing Wrongs’ (n 29). 32 ibid 2; Duff, The Realm of Criminal Law (n 29) 146. 33 James Edwards and Andrew Simester, ‘What’s Public About Crime?’ (2017) 37 Oxford Journal of Legal Studies 105, 107. 34 See R v Disciplinary Committee of the Jockey Club (ex parte Aga Khan) [1993] 1 WLR 909 (CA). 35 For example, see X v Y [2004] EWCA Civ 662, [2004] ICR 1634; Pay v Lancashire Probation Service [2003] EAT/​1224/​02/​LA, [2004] ICR 187; Pay v UK [2009] IRLR 139 (ECtHR). 36 Otherwise domestic violence would also slip out of the net. See also the discussion in Edwards and Simester (n 33) 108–​10.

Using Criminal Law to Enforce Employment Rights  59 whether the conduct is (i) a public matter, or (ii) that within the context of that civil order, the conduct is a ‘wrong’.

2.  Applying the Negative and Positive Versions of Public Wrongs Theory Duff tells us two things about the (i)  ‘public’ as opposed to the (ii) ‘wrongfulness’ variable: first, that for a practice to be a ‘public matter’, it must be something that directly bears on the distinctive goals and values of the civil order concerning us as living together as members and citizens of a political community.37 Secondly, workplace conduct will not be a public matter if it is not the ‘business of members of the community [simply] in virtue of their membership [and our mutual citizenship]’.38 As for the wrongfulness criterion (ii), we are informed that this will be met if the conduct either violates or is inconsistent with the shared values of the polity in that civil order.39 This translates into a requirement for a political community with a ‘sufficiently rich shared understanding between at least most of its members about the goals and values that define that community and its civil order’.40 But how do we ascertain whether this negative formulation of Duff and Marshall’s public wrongs theory harbours the capacity to provide a normative justification (i) running counter to criminalization in the case of statutory labour law wrongs that are exclusively enforced via the civil law, (ii) for those workplace practices or omissions that have been left largely to the criminal law for enforcement, and (iii) for areas of shared enforcement? This exercise will involve ascertaining (a) which labour rights do or do not qualify as public wrongs and/​or (b) whether existing legislative patterns of criminalization track those normative distinctions.41 In analysing the salience of the public wrongs theory, we adopt dismissal, discriminatory conduct, and the right to a written statement as three illustrative samples of statutory employment rights that attract civil enforcement, whereas health and safety is taken as the classic embodiment of statutory labour laws that have been earmarked for criminal prosecution. We also invoke an employer’s infringement of the right to receive the National Minimum Wage as an illustration of shared civil and criminal enforcement.42 If we apply the (i) ‘publicness’ and (ii) ‘wrongfulness’ criteria to the selected five statutory rights in reverse order, turning first to the wrongfulness criterion (ii), the question is whether each of these five illustrations do not represent an affront to, or rupture from, such shared values of the community. In the case of discriminatory treatment, the answer is ‘no’, as it clearly does. This is undoubtedly the appropriate response if we call to mind the various powerful social justifications for this statutory body of labour law, for example respect

37 Duff and Marshall, ‘Crimes, Public Wrongs, and Civil Order’ (n 28) 18. 38 Duff, ‘Towards a Theory of Criminal Law?’ (n 29) 5. 39 Marshall and Duff, ‘Criminalization and Sharing Wrongs’ (n 29)  18–​22; Duff and Marshall, ‘Public and Private Wrongs’ (n 29) 70–​72; Duff, ‘Responsibility, Citizenship, and Criminal Law’ (n 29) 125, 139–​40; Antony Duff, Answering for Crime (Hart Publishing 2007) 141–​43; Lee (n 29) 158. 40 Duff and Marshall, ‘Crimes, Public Wrongs, and Civil Order’ (n 28) 18. 41 Of course, there may be a mismatch between (a) and (b), but this does not discredit public wrongs as a normative theory. Instead, it simply means that what has and has not been criminalized by Parliament falls short of the standards set by this theory. 42 We have limited the enquiry in these terms for reasons of economy of space.

60  Criminality at Work for individual dignity43 or fundamental choices;44 the prevention of serious, persistent, and ubiquitous forms of relative disadvantage;45 the recognition of a person’s immutable characteristics;46 or historic stigmatization of certain practices or characteristics.47 As for health and safety infringements giving rise to the harm and injury of workers, we must ask whether they do not represent a breach of the common values underpinning a social and political community. Like discriminatory conduct, the answer again must be negative, bearing in mind that the objective of health and safety laws is the protection of the physical and psychological integrity of workers, which is clearly a fundamental issue.48 In the case of the failure of an employer to issue a statutory written statement, this can be viewed as generating a rift with the community’s common values insofar as an employee has a basic entitlement to understand the basis of his/​her relationship with the hirer of his/​her labour, that is, exactly where he/​she stands. As for the right to be paid the minimum wage, if a society places weight on the dignity of workers and the capacity of wage controls to generate redistributive results,49 there is a strong case to be made for the proposition that a polity’s values will be breached where the earning capacity of some of the workforce is squeezed by the employer so that they struggle to meet their basic needs. In such a context, the deprivation of a wage floor would strike at the society’s shared values and be wrongful. Finally, turning to unreasonable no-​cause dismissals, the position is less clear-​cut, as all depends on one’s personal opinions and moral convictions on the relative strength of countervailing social, economic, and political arguments. Here, we can evoke competing claims concerning the beneficial value of laws restricting dismissals insofar as they limit the scope for failures in the operation of the labour market, as against the contention that strong controls over dismissals entail higher unemployment, lower labour force participation,50 and less efficient, dynamic, and flexible labour markets. As such, the assessment of whether dismissals fall down at the wrongfulness criterion (ii) is more finely balanced, but if one subscribes to, and assumes the existence of a social community wedded to a ‘job security’51 or ‘job property’52 43 Benjamin Eidelson, ‘Treating People as Individuals’ in Deborah Hellmann and Sophia Moreau (eds), Philosophical Foundations of Discrimination Law (OUP 2013) 203; Rory O’Connell, ‘The Role of Dignity in Equality Law: Lessons from Canada and South Africa’ (2008) 6(2) International Journal of Constitutional Law 267. 44 Robert Wintemute, Sexual Orientation and Human Rights Framework (Clarendon Press 1997); John Gardner, ‘On the Ground of Her Sex(uality)’ (1998) 18 Oxford Journal of Legal Studies 167, 169–​74. 45 Tarunabh Khaitan, A Theory of Discrimination Law (OUP 2015) 137–​39. 46 Catherine Dupre, The Age of Dignity:  Human Rights and Constitutionalism in Europe (Hart Publishing 2015); Samuel A Marcosson, ‘Constructive Immutability’ (2001) 3(2) University of Pennsylvania Journal of Constitutional Law 673; Jessica A Clarke, ‘Against Immutability’ (2015) 125 Yale Law Journal 1; Sandra Fredman, Discrimination Law (2nd edn, OUP 2011) 130–​34. 47 Iyiola Solanke, Discrimination as Stigma: A Theory of Anti-​Discrimination Law (Hart 2017) ch 4; Robin A Lenhardt, ‘Understanding the Mark: Race, Stigma, and Equality in Context’ (2004) 79 NYU Law Review 803; Iyiola Solanke, ‘Infusing the Silos in the Equality Act 2010 with Synergy’ (2011) 40 Industrial Law Journal 336, 350–​55. 48 Thus, the claim that they are merely mala prohibita or regulatory criminal law is rejected: see Duff, The Realm of Criminal Law (n 29) 245–​47. 49 Guy Davidov, ‘A Purposive Interpretation of the National Minimum Wage Act’ (2009) 72 Modern Law Review 581, 586–​94; Brisham Rogers, ‘Justice at Work: Minimum Wage Laws and Social Equality’ (2015) 92 Texas Law Review 1543. 50 Juan C Botero and others, ‘The Regulation of Labor’ (2004) 119 The Quarterly Journal of Economics 1339, 1379. 51 Keith D Ewing, ‘Job Security and the Contract of Employment’ (1989) 18 Industrial Law Journal 217; Hugh Collins, ‘Regulating the Employment Relation for Competitiveness’ (2001) 30 Industrial Law Journal 17, 29. 52 Wanjiru Njoya, Property in Work: The Employment Relationship in the Anglo-​American Firm (Ashgate 2007) 6–​7; Claire Mummé, ‘Property in Labour and the Limits of Contracting’ in Ugo Mattei and John D Haskell (eds), Research Handbook on Political Economy and Law (Edward Elgar 2015) 413–​15.

Using Criminal Law to Enforce Employment Rights  61 outlook, the argument can be made that such dismissals would jar quite severely against the common values of that society. In the final analysis, this would mean that the wrongfulness criterion (ii) would be satisfied. Of course, the prevalence of competing contestations here might suggest that public wrongs theory’s implicit faith in being able to identify a civil order with uniform, shared, or common values is entirely misplaced in the industrial relations context, which is a site of sharp political conflict characterized by pluralism and disagreement. In this way, the immersion of Duff ’s theory within the crucible of labour law may give rise to serious doubts about the plausibility of concepts such as the ‘civil order’ and ‘common values’, as well as the operational capacity of ‘public wrongs’ theory more generally. However, Duff would counter this by invoking the pluralist and liberal nature of his conception of the ‘civil order’. Such a formulation recognizes that citizens may reasonably disagree about what is ‘good’, but nonetheless reach consensus on certain basic, minimum values of a substantive and procedural hue relating to the operation and composition of their political community.53 It follows from the preceding analysis that each of the selected five examples of managerial behaviour or failures are indeed ‘wrongful’. We now turn to the requirement that such workplace practices must not be a ‘public matter’ (i). It is not entirely persuasive to argue that no-​cause dismissals, discriminatory workplace behaviour, the failure to adhere to a wage floor, and health and safety norms in employment have no bearing on the distinctive goals and values of the civil order. On the contrary, each of them are surely sites of workplace activity qualifying as the business of members of that social community. In particular, in the case of discriminatory practices, it can be claimed that by setting out one’s stall, an employer has voluntarily transposed its field of activities into the public sphere and as such, the values of the political community in which its employees reside are engaged.54 Seen from this perspective, the case that such examples of workplace conduct ought not to be criminalized also founders on the rock of the ‘publicness’ criterion (i). These rights can be contrasted with the statutory right to receive a written statement, which would appear to have a less obvious public character. Here, the relevance of an employer’s non-​compliance with the shared understandings of the community and defining that polity is not as clear-​cut. However, on balance, there is a strong case for the proposition that the role of the statutory right to a statement in enabling employees to appreciate where they stand vis-​à-​vis their employer incorporates a material and vital core of publicness. It is surely a matter of shared public concern that societal groups who may be subject to the rational or irrational exercise of coercive social power–​–​such as employees by their employers–​–​have full transparency and disclosure of fundamental contractual terms. And if such an analysis is correct, we cannot claim that there is a normative obligation not to criminalize such failures of disclosure. Having established that the negative strand of public wrongs theory provides no support for the proposition that each of our five statutory employment rights should not be criminalized, its positive incarnation–​–​whereby we have a reason to criminalize them—​is also by definition satisfied. In terms of the positive strand, the (i) ‘publicness’ and (ii) ‘wrongfulness’ tests must be met, and the preceding analysis has shown how they indeed are in the case of the five selected examples. However, it is worth recalling that although we have

53 Duff, The Realm of Criminal Law (n 29) 167–​82. 54

Khaitan (n 45) 201–​09.

62  Criminality at Work positive grounds to criminalize each of these five practices, this does not lead automatically to the proposition that we ought to criminalize them. Instead, that would be a false move. The crucial point here is that the positive incarnation of Duff and Marshall’s public wrongs theory stops short of exacting a full ‘normative bite’. How one transitions from (a) having a positive legitimate basis for the criminalization of these forms of workplace conduct to (b) being under an all-​things-​considered absolute obligation to do so (hereinafter referred to as the ‘positive to the absolute’) is an issue to which we now turn. The various countervailing considerations to criminalization must first be assessed, including any possible regulatory alternatives.

3.  Public Wrongs Theory: Getting from the Positive to the Absolute A pressing issue for resolution is whether we are persuaded to take the relevant leap from the (a) positive to the (b) absolute case for criminalization. If not, then the current legal position described in section B above that dismissals, the right to receive a written statement, and discriminatory practices are not matters for the criminal law would be theoretically justified in terms of public wrongs theory, that is, the policy decision to leave these matters to individuals to litigate in the courts and tribunals. But, if we elicit the opposite response, then the conclusion can be drawn that Parliament’s decision not to marshal the criminal law in these three instances is philosophically misconceived in terms of the public wrongs theory.55 Meanwhile, in the case of health and safety infractions, our analysis would be turned on its head, for example if we pose the question whether we ought to make the transition from the positive to the absolute, and our answer is negative, then the subjection of health and safety breaches to the criminal law in the current legal regime can be cast as wrong-​headed as a matter of public wrongs theory. On the subject of such a transition, Duff says the following: first, we do not need to take that step if we have ‘reasons, principled or pragmatic . . . against doing so . . . [or if we] conclude that on balance we have better reason not to criminalize . . .’56 Secondly, we may decide criminalization is only one among other possible ways of responding to public wrongs . . . [eg we may provide] no formal response to [them], and instead leav[e]‌any response to be an informal social matter, or rather [make] it a matter of tort law rather than of criminal law.57

In light of this, do the variety of factors and justifications cited by Duff truly assist, and/​or persuade us to take the relevant jump in the case of the five illustrative employment rights? The response calls into question, in general, the nature of the relevant and appropriate factors and reasons that must be taken into account in evaluating whether to take that step. The difficulty is that the guidance from Duff regarding the relevant issues to factor in at 55 An alternative conclusion is that public wrongs theory is intrinsically defective. But this is a suspect move, since it is not intellectually persuasive to extrapolate from what we see (or do not see) ‘on the ground’ in the case of (the political choices that have or have not been made regarding) criminalization to broader propositions about the lack of value or flaws in public wrongs theory or normative theorizing about criminal law more generally. 56 Duff and Marshall, ‘Crimes, Public Wrongs, and Civil Order’ (n 28) 4. 57 ibid.

Using Criminal Law to Enforce Employment Rights  63 the normative leap stage is so general as to be rather scant; for example Duff ’s most recent work identifies only six factors58 that we should consider. Beyond that, we are left guessing, showing how the positive version of the public wrongs theory is somewhat impoverished as soon as it reaches the second stage of questioning whether there is an absolute obligation to criminalize: it is here that the public wrongs theory seems to break down.59 The fundamental point here is that, in comparison with this second stage concerning whether to take the step from the positive to the absolute case for criminalization, the first stage gateway in the public wrongs theory is extraordinarily wide—​it is difficult to see how any statutory labour right would fail to meet the (i) ‘public’ and (ii) ‘wrongful’ criteria.60 In effect, we must resort to our own capacities for lateral reasoning and call to mind a number of factors that we anticipate would assume a degree of relevance. They can be each grouped into justifications in favour of adopting the relevant leap from the positive to the absolute case for criminalization and secondly, those against doing so. Two sets of distinctions should be made at this juncture. First, there are general factors/​reasons and secondly, rights-​particular factors/​variables. The latter are by definition context-​dependent on the particular right. As for the general variables, these are of relevance and application to each of the five chosen examples. We begin with a consideration of what might constitute the abstract reasons for criminalization. On the positive side, first, it may be that there is something so egregious about infringements of the five statutory employment rights that we ought to adopt punitive61 rather than corrective measures. In that respect, we may conjecture that this is where the harm principle makes its presence felt within the context of the public wrongs theory:62 the more severe the harm done to the employee by these practices or omissions, the more convincing the claim that we ought to take the relevant leap. Secondly, we may wish to subrogate the state to enforcement to ensure consistency in policy and to depart from the individual litigation-​ driven model for the enforcement of these five statutory employment laws. Thirdly, and a point closely related to the second, is that we may believe that state involvement via the criminal law has a degree of merit in propagating a greater level of deterrence of dismissals, etc. The symbolic communicative role of the criminal law in signposting the wrongfulness of a particular activity is important,63 particularly in light of its supplementary deterrent function.64 As is the relevance of the criminal law in securing a (a) formal and authoritative judicial declaration of the culpability of the employer,65 as well as in a public and collective (b) calling to account of the employer.66 Of course, if we do reach the relevant threshold that we demand the criminalization of such five managerial practices, we may also decide that there are powerful reasons to elicit stronger forms of criminal enforcement in the case of deliberate or intentional practices on the part of employers, such as direct discrimination and harassment (as opposed to indirect discrimination). In contrast, given the nature of the 58 Duff, The Realm of Criminal Law (n 29) 297. Each of the six issues are addressed in the discussion in the remainder of this section. 59 Duff decides not to provide a detailed account of how we ought to make this transition: ibid 298. 60 All of the normative work is being done at this second stage. 61 Duff, The Realm of Criminal Law (n 29) 297. 62 Duff himself recognizes this at various points: ibid 296, 298, and 308. 63 ibid 293. 64 ibid 297. 65 ibid. 66 ibid.

64  Criminality at Work harm, in the case of managerial behaviour resulting in physical injury to employees, we may form the opinion that it is irrelevant whether the conduct was unintentional and that criminalization is the only appropriate response. Alternatively, we may decide on a regime where enforcement is shared between the criminal and civil law. Such a shared regulatory response may be appropriate in the case of objectionable forms of behaviour where we feel that we ought to reserve a role for individuals themselves to be involved in enforcement, (eg to leave scope for individual negotiation and settlement) or that the civil law can operate as an auxiliary aid to criminal enforcement, for example as recognized in the case of the minimum wage in the current legal regime. On the negative side of the equation, we can evoke certain factors or values (integral to the civil order of the polity) which suggest that we ought to resist any recourse to the criminal law. For example, the costs of criminal enforcement may be too high and criminalization may be disproportionate in light of the magnitude and extent of the wrong. It may also be impractical to criminalize. A further consideration is that an anxiety may take root that the value of liberty would be undermined if we criminalize,67 that is, that we are illegitimately depriving individuals of their freedom to take remedial action by compelling them to relinquish their private control of enforcement and hand it over to the state. Such an expropriation would be tantamount to the denial of the hugely significant and symbolic sovereignty/​autonomy and prerogative of the individual to retain the decision whether to vindicate his/​her rights where he/​she has been subjected to a wrong, that is, whether it is the individual or the state who should ‘own’ any legal action and ultimately exercise control over the instituting of legal proceedings.68 The devolution of control over enforcement to the state also carries the risk that the relevant prosecution officials decide not to take action for one reason or another. But even where they do act, there is also the danger that they engage in plea bargaining, which would be possible in the shadow of the criminal law. In essence, the state prosecution authorities would have the capacity to do deals. This may engender a negative perception towards any role for state enforcement. Another concern we may harbour is that the powers of the state are so far-​reaching that it would be heavy-​ handed to involve it in the enforcement of these five statutory employment rights, that is, that it is disproportionate to bring the criminal law into play in light of the harm done by the conduct which they regulate. Of course, there is also the higher standard of proof associated with the criminal law, which may lead us down the path of securing fewer positive outcomes. One other possible issue to take into contention is that there is an insufficient basis to criminalize dismissal, discrimination, managerial omissions to issue a written statutory statement or pay the minimum wage, or health and safety breaches because a less restrictive alternative is available in the guise of mala prohibita offences or even civil penalties,69 that is, regulatory law/​administrative enforcement mechanisms as a less compelling form of punishment. Finally, we also have to factor in the reality that the criminalization of each of these workplace practices or failures will have a chilling effect on managerial decision-​making more generally.

67 ibid 297. 68 See Arthur Ripstein, Private Wrongs (Harvard University Press 2016) 269–​72; Alan Bogg, ‘Labour, Love and Futility: Philosophical Perspectives on Labour Law’ (2017) 33(1) International Journal of Comparative Labour Law and Industrial Relations 7, 21. 69 Duff lists and critiques each of these less drastic alternatives in The Realm of Criminal Law (n 29) 280–​92.

Using Criminal Law to Enforce Employment Rights  65 If we place each of these positive and negative factors into the balance, we can form a provisional holistic and general view as to whether the scales balance out in favour of, or against, a role for criminal enforcement. However, this would only be a preliminary conclusion, as we must also identify a swathe of context-​particular issues that we consider significant in the sense described above. If we begin with dismissals, our response to the question whether the unfair dismissal right should be enforced by the criminal law would be largely dependent on our level of acceptance of dismissals in society and our attitude to a dismissal-​free workplace. If the attitude to such exercises of the managerial prerogative is one of intolerance, this would support the establishment of a role for the criminal law. We70 may also harbour a desire to criminalize dismissals to avoid the opportunities for employers to engage in efficient repudiatory breaches of the contract of employment, that is, to ‘profit from their own wrong’. Moving to the negative aspects, a ban on dismissals supported by criminal sanctions would represent a startling and radical intrusion into the managerial prerogative and the employer’s liberty and property rights.71 Another closely related factor to take into account from a managerial property rights angle is whether it is relevant that the employer retains the capacity to substitute labour for capital. The point being made here is whether some margin should be afforded to the employer as a badge of recognition that the employer has given the employee a job where it could otherwise have deployed capital instead. There is also the consideration that the employer needs to stay competitive to ensure employment security. Another salient question is whether it is in any way relevant that the employer’s power of dismissal can be viewed as reciprocal to the employee’s power of resignation, in the sense that it is correlative to the employee’s fundamental freedom to quit? Might this reciprocity temper any claims that we have an absolute duty to criminalize a no-​cause and unreasonable dismissal regime? Turning to discriminatory conduct, there is a potent argument that the intentional forms of ‘prohibited conduct’ in the EqA should be supported by punitive measures, as these forms of behaviour involve more direct and piercing forms of harm, bordering on corrosive malice. We may also criminalize them as an unequivocal manifestation of public solidarity with those possessing the legally protected characteristics adumbrated in section 4 of the EqA. However, it is not entirely clear whether the state has a legitimate entitlement to be involved in the policing of discrimination. On the one hand, we can appreciate how claimants might wish to retain control over legal proceedings instead of relying on the state (through its officials) to take the decision whether to prosecute. Since such practices represent a personal attack on a claimant’s individual dignity and identity, it may be that dispossessing them of the freedom to initiate and steer the litigation process is too much for us to contemplate. On the other hand, we may be reluctant to expect claimants to shoulder the costs of civil enforcement. There is also the public benefit associated with the deterrence of discrimination that can be secured through the intervention of the criminal law, as well as the authoritative judgment that the employer is ‘guilty’ of discrimination and ought to be ‘punished’.

70 Of course, the ‘we’ here depends on the polity’s civil order and the outcome of deliberations and debates between those citizens who have engaged in ‘reasonable disagreement’: ibid 167–​82. 71 See the discussion in Mummé (n 52) 400.

66  Criminality at Work In the case of the rights to be paid the minimum wage and to receive a written statutory statement, it is an obvious point that the employer’s breach causes no physical harm or threat to the employees concerned. However, of course, their financial position is adversely affected in the case of a minimum wage breach. But whether this is sufficient to claim with any authority that such a failure should be criminalized is debatable. It may be that there is a stronger case to be made for criminalization where the employer’s failure to meet the wage floor is a conscious decision, but again, it is open to discussion whether it is warranted to (i) devolve decisions concerning enforcement to the state or (ii) enjoin prosecution authorities to make fine judgement calls about deliberate and inadvertent omissions. In fact, in this context and in the case of managerial failures to issue a written statement, state involvement does seem somewhat over-​blown and disproportionate. It is also problematic that there are clearly other less restrictive alternatives to criminal enforcement which would achieve the requisite objectives, for example civil enforcement, civil penalties, administrative agency, or regulatory enforcement. Finally, we come to health and safety breaches and whether the criminal law should play a part in enforcement. We may argue that the extent of the physical harm done to employees is such that resorting to financial recompense through civil remedies would be bankrupt as a regulatory response.72 However, herein lies a puzzle. If that is true in the case of contraventions of health and safety regulation, then why is it untrue in the case of tort law which also regulates physical injury, where we do find that enforcement is left to the civil law? From another angle, if we decide that it is appropriate to eschew state enforcement, that may be misconceived where the actions of the employer that have led to the physical injury of its employees are reckless or deliberate. There are a variety of possible conceptual frameworks for understanding the outcomes when we place the general and rights-​specific factors above in the balance within the context of the five examples of statutory employment rights. The model adopted is to evoke three particular states that we can plot across a spectrum from a more heightened, to less intensive, set of justifications for us to invoke the criminal law: we can dub these the ‘thick’, the ‘medium’, and the ‘thin’ forms. If we pose each of these general and rights-​ specific questions, and reach the conclusion that the thick form is applicable in the case of each of the five selected employment rights, this will indicate that we have a strong case to criminalize these practices or omissions. At the opposite end of the pole, if the product of this exercise is the thin form, then we can say that we most certainly do not have any legitimate grounds to claim that we ought to criminalize them. And if our deliberations result in the medium form, a relevant series of justifications can be made for both criminal prosecution and civil proceedings, that is, shared enforcement.73 The choice between these three conceptual models is significant since it influences the way we understand the enforcement of these employment rights and the extent to which we can agree that the

72 However, the criminal enforcement of health and safety laws does not adhere to the standard pattern insofar as the police and DPP have no involvement in the process. Instead, the HSE can conduct investigations and enforce violations through enforcement notices (see the HSWA 1974, ss 22 and 23 and HSE v Chevron North Sea Ltd [2018] UKSC 7, [2018] 1 WLR 964). 73 The fact that public wrongs theory does not provide a comprehensive account of when it is appropriate to switch from having a reasonable ground, to having an absolute duty, to criminalize a practice, renders it of limited assistance in identifying the various employment rights where shared enforcement is legitimate.

Using Criminal Law to Enforce Employment Rights  67 current legal position (as regards such enforcement) is aligned with, or falls below the standards set by public wrongs theory. It would also hint at the law being ripe for reform or further development.74 We can benchmark the five statutory employment rights against each of the aforementioned rights-​specific and universal factors, reasons, and variables. The first conclusion that we can take from this exercise is that there are weighty general and context-​dependent reasons for us not to criminalize employers who engage in decisions to dismiss for no cause or unreasonably. For example, it is stretching credulity to suggest that an employer should be punished in such a context, and it is suggested that the employee’s liberty to engage in litigation should not be removed in favour of the state having exclusive control over criminal proceedings, since the latter seems disproportionate to the harm caused. As does any engagement of the higher criminal standard of proof. Likewise, there is a strong likelihood that the deterrence of no-​cause or unreasonable dismissals can be secured through less restrictive means, such as civil litigation. As such, in the context of dismissal, on the application of the public wrongs theory, the overall balance of factors, reasons, and variables accords with the ‘thin’ state. This conforms to the contemporary legal position, which underscores how the deliberations of law-​makers in Parliament meet the expectations set by Duff ’s normative theoretical framework. To that extent, we can claim that Duff and Marshall’s approach to the theorization of what properly belongs to the criminal law possesses a degree of intellectual ballast in the case of no-​cause or unreasonable dismissals. Moving on to health and safety breaches, when we measure them against the yardsticks cited above, we find that they accord with the ‘thick’ state. This means that the current legal position and public wrongs theory conform with one another, that is, predominantly criminal enforcement. Turning to the legal position prioritizing the civil enforcement of managerial omissions to issue written statutory statements of terms and conditions, we also find that it is consistent with the application of public wrongs theory, that is, there is little to suggest such failure ought to be criminalized. As for our findings in the case of rights affording employees protection from discriminatory treatment, the positive version of public wrongs theory suggests shared enforcement between the civil and criminal law when assessed against the yardsticks of deterrence, punishment, preservation of the ‘victim’s’ individual liberty, the desirability of subrogating the state to enforcement, the aptness of the application of the criminal standard of proof, and whether criminalization will have a chilling effect on workplace practices. However, this fails to adhere to the current legal reality, which demonstrates that it falls short of the benchmark set by the public wrongs theory. Likewise, in the case of failures to pay the National Minimum Wage, where the application of the aforementioned variables, factors, etc point towards civil enforcement in contrast to the shared enforcement enshrined in the existing law. The conclusion can be drawn that—​by a slim majority of three to two—​the relationship between the legal position with regard to enforcement and the predictions of the public wrongs theory is, in general, a positive one. As such, the end result of this benchmarking exercise is to provide moderate support for the proposition that the positive incarnation of the public wrongs theory possesses a degree of predictive power in the case of statutory labour laws. 74 However, where there is a misalignment, this should not call into question the normative credentials of public wrongs theory: see nn 41, 55 and 79.

68  Criminality at Work

D.  Conclusion The objective of this chapter was twofold. First to chart some of the areas of statutorily imposed labour law where we encounter criminal enforcement, including where this is an absence of such enforcement, and also to probe where we find overlaps between them. This was primarily a descriptive exercise. Secondly, the account of ‘public wrongs’ in Duff ’s work was harnessed as an evaluative instrument to determine whether criminalization is normatively justified in relation to a limited range of five statutory employment rights. This enabled us to gauge the validity of the actual political choices made by law-​makers concerning the areas earmarked for criminal enforcement in labour law (as well as those areas not chosen for criminal enforcement, or identified for shared enforcement) and to test whether they could be supported by normative theories of criminalization, such as Duff ’s ‘public wrongs’ theory. Chiming with the views of thinkers such as Christie,75 Hulsman,76 and Bianchi,77 like Michael Ford,78 the writer’s intuition was one of scepticism. It was thought that the search for an all-​encompassing normative theory of criminalization—​such as Duff ’s ‘public wrongs’ model—​that would have relevance in the context of statutory labour laws was an exercise born of futility.79 However, to the surprise of the author, the antipodean framing of ‘public versus private’ in the theorization of what properly belongs to the criminal law is indeed of some utility in predicting the prevailing legislative position concerning statutory employment rights. Whilst the results stemming from the adoption of the negative incarnation of public wrongs theory were rather banal, in the case of the positive version, we were taken much further down the track in the search for an account of what should be criminalized in the case of statutory labour laws. Nevertheless, in the final analysis, several notes of caution should be struck so that the workplace relevance of public wrongs theory is not overstated:  first, that the Duff and Marshall account of public wrongs suffers from a marked lack of detail at the critical point at which it is essential to turn from a positive to absolute case for criminalization. The identity of the relevant factors/​variables/​justifications are impressionistic and the outcome of their application to the five employment rights is inherently subjective. This necessitates an element of guesswork as to the filling in of the relevant normative blanks,80 meaning that it is perhaps overly sanguine to make an unreserved positive case for its predictive capacity. This speaks to the reality that the findings drawn from this inherently indeterminate exercise are open to debate, as we may not all agree on how the relevant judgements have been made or the various processes have been applied. Secondly, having surveyed how five prominent statutory labour rights are enforced, and applied the public wrongs construct to them as a sorting principle, what broader claims can be extrapolated from the findings of the exercise adopted is unclear. Hence, further comprehensive research ought to be carried out to test this chapter’s conclusion that this theory possesses a moderate degree of plausibility 75 Nils Christie, ‘Conflicts as Property’ (1977) 17 British Journal of Criminology 1. 76 Louk H Hulsman, ‘Critical Criminology and the Concept of Crime’ (1986) 10 Contemporary Crises 63. 77 Herman Bianchi, Justice as Sanctuary: Toward a New System of Crime Control (Indiana University Press 1994). 78 See Michael Ford’s c­ hapter 21 in this volume. 79 However, an explanatory account of what is found to be criminalized in various areas of law (eg such as labour law) is not what criminal law theorists are looking to achieve, as their objectives are primarily normative. 80 This exercise demonstrates how normative criminal theories and other legal disciplines and modes of enquiry must engage in a dialogue to complete the picture.

Using Criminal Law to Enforce Employment Rights  69 as an account of when criminalization is appropriate in the case of statutory labour rights. Finally, and more profoundly, there is the anxiety that by overly stressing the public nature of some statutory employment rights, the symbolism associated with the adoption of a lexicon of ‘public wrongs’ might underplay the collective and solidaristic origins and nature of labour laws. Here, the concern is with over-​juridification and the general movement away from the social and economic power associated with the collective, which has traditionally underpinned labour law. It would appear to overlook how labour law was forged in the historical crucible of class and social struggle and political discourse. The public wrongs theory is also vulnerable to the charge of being overly focused on the individual private employment relationship and whether the relevant regulated activity is public and wrongful, thus harbouring the potential to undercut the collective objectives that labour law is ultimately seeking to achieve.

4

Where Criminal Law Meets Labour Law The Effectiveness of Criminal Sanctions to Enforce Labour Rights Catherine Barnard and Sarah Fraser Butlin

A.  Introduction Over the last decade, there has been an increased focus on the criminalization of labour law breaches. In particular, we have seen a rise in the use of the coercive arm of the state and a growing number of civil preventive measures in the labour market, backed up by criminal sanction upon breach in regulatory offences, as well as specific offences for certain labour law breaches.1 A key question for labour lawyers is whether criminalization actually works to enforce labour rights.2 In order to answer this question, we consider first, why certain labour law breaches have been criminalized and, in light of the nature and extent of enforcement action that has been taken, does criminalization work? In respect of the first question—​why certain labour law breaches have been criminalized (and why others have not)—​we seek to explore this issue by considering the nature of the offences that have been created and the political framework from which criminalization arose (section C). We will examine four regimes: the working time provisions, partly regulated by the Health and Safety Executive (HSE); National Minimum Wage protection, with some regulation by the HMRC; the work of the Employment Agencies and the Employment Agencies Standards Inspectorate (EASI); and the regulation of gangmasters through the Gangmasters and Labour Abuse Authority (GLAA). We have selected these four areas as the lens for our analysis because they operate across a spectrum of how rights are enforced, ranging from requiring a licence for gangmasters to operate, the breach of which is a criminal offence, through to the mixed criminal and civil enforcement mechanisms of the Working Time Regulations. These diverse regimes help us to examine the location of the boundary between civil and criminal law enforcement because there is a panoply of approaches, each varying across and within the regimes. We contrast these regimes with the legislative scheme dealing with blacklisting where there were considerable public calls for its criminalization and yet the civil route was chosen. This gives us an opportunity to address the question as to why policy-​makers may have opted for particular legal regimes.

1 On which see further c­ hapter 1 of this volume. 2 We leave the questions as to the theoretical justifications for the application of criminal law to the criminal lawyers; compares in particular c­ hapters 5 and 12 of this volume. Catherine Barnard and Sarah Fraser Butlin, Where Criminal Law Meets Labour Law In: Criminality at Work. Edited by Alan Bogg, Jennifer Collins, Mark Freedland QC, and Jonathan Herring, Oxford University Press (2020). © Alan Bogg, Jennifer Collins, Mark Freedland, and Jonathan Herring. DOI: 10.1093/oso/9780198836995.003.004

Where Criminal Law Meets Labour Law  71 As the political rhetoric shows, one fundamental feature of criminalization is that it sends a signal to employers requiring them to take the labour law breach particularly seriously. But to what extent is this really achieved? This brings us to the second part of the chapter where we consider the question: does criminalization work (section D)? We explore the nature and extent of enforcement action that has been taken by the various enforcement agencies and the level of awareness of enforcement that employers and employees have. We argue that while in general criminalization might send out a particular message that the conduct is not acceptable, the lack of effective enforcement—​and the absence of a civil alternative—​means that, in reality, the criminalization of certain labour law breaches may in fact be political posturing. At best, it acts as a distraction from the more fundamental question of how to achieve compliance. From this point of view, the recent Strategy document produced by the new Director of the Labour Market Enforcement Unit setting out a clear vision for both the basis of criminal enforcement and a strategy might offer a more robust way of using the criminal powers. At worst, where the criminal law operates as a substitute for a private civil right against the employer, the criminalization of a particular offence denies the individual worker the opportunity to bring a civil action and thus leaves them without a remedy at all. However, before we address the two questions around which this chapter is structured, we begin by mapping the powers in our selected areas.

B.  Defining the Powers This section describes the complex mix of regulatory offences and civil liabilities that apply within our four chosen areas. It also details the powers of enforcement across those regimes. We consider this exercise to be necessary: there is a lack of familiarity with the different regimes and a detailed analysis of them enables us to draw some conclusions in relation to what the structure of the legislation can tell us about where the boundary is drawn between criminal and civil liability. Only after a careful description of powers is it possible to conduct some analysis first in relation to the complexity of the legal response and second on the political choices as to which enforcement mechanism to choose. Third, we assess the effectiveness of the chosen mechanism. We are not addressing here the regime within the Modern Slavery Act 2015, as this is dealt with in depth in c­ hapter 18. Moreover, we wish to focus our discussion on labour law breaches where criminalization is more contentious. Nevertheless, the two core offences within the Modern Slavery Act 2015 should be noted for comparative purposes in respect of the regimes that we do consider in this chapter. Section 1 makes it an offence to hold someone in slavery or servitude, or to require someone to perform forced and compulsory labour. Regard must be had to all the circumstances including the personal circumstances that make someone vulnerable and the work required of that person. Section 2 creates a human trafficking offence whereby an offence is committed if a person arranges or facilitates the travel of another person with a view to the victim being exploited. There must be intent to exploit. We turn now to examine our four chosen regimes and regulatory bodies.

72  Criminality at Work

1.  Gangmasters and Labour Abuse Authority The GLAA started life as the Gangmasters Licensing Authority. It was given the task of licensing gangmasters and enforcing the requirements of the Gangmasters (Licensing) Act 2004 (GLA 2004). The provisions of GLA 2004 continue to apply in relation to licensing gangmasters, which is undertaken by the GLAA. In relation to licensing, the GLA 2004 applies to agricultural work, gathering shellfish, and processing or packaging agricultural produce, shellfish, or fish (section 3(1)). Section 12 of the GLA 2004 provides that it is an offence to act as a gangmaster without a licence and, in England and Wales, on indictment the maximum sentence is of ten years’ imprisonment. Under section 13 of the GLA 2004 it is an offence to use an unlicensed gangmaster, carrying with it a maximum sentence, in England and Wales, of fifty-​one weeks’ imprisonment or a fine or both. In addition, where there is a failure to comply with the Gangmasters (Licensing Conditions) Rules 2009, such a failure is actionable by the individual, insofar as it causes damage (section 7). The GLA 2004 gives enforcement officers the power to arrest (section 14); to require the production of records, explanation of records, and further information (section 16(1)(a)–​(c)); and to enter premises (section 16(1)(d)). They may also obtain a warrant to enter premises and remove records and documents (section 17). The obstruction of an enforcement officer acting in the exercise of his functions constitutes a criminal offence with a maximum sentence of fifty-​one weeks’ imprisonment or a fine or both (section 18). The grant of a licence to act as a gangmaster is discretionary (section 7(1)) and the Licensing Standards are applied. There are eight Licensing Standards which cover a range of issues, including the requirement that a gangmaster is a fit and proper person, dealing with pay and tax, preventing forced labour and the mistreatment of workers, accommodation, working conditions, health and safety, recruiting workers, and contractual arrangements and sub-​contracting. For our purposes, the following ‘critical’ standards are of most interest: 2.2: to pay at least the National Minimum Wage or Agricultural Minimum Wage; 3.1: not to subject workers to physical or mental mistreatment or to make threats to workers or others; 3.2: not to restrict a worker’s movement, retain identity papers, not be required to pay more than any sum borrowed and not force or coerce a worker to work against their will; 3.3: not to withhold wages; 4.1: where accommodation is provided, to ensure that it is safe; 7.1: not to charge a fee for work finding services or making them conditional on the use of other fee generating services. The requirement to give a worker the rest periods, breaks, and annual leave to which they are entitled, and the requirement that a worker does not work more than a forty-​eight-​ hour week without express agreement are ‘non-​critical’ standards (Standards 5.1 and 5.2). The right to belong to a trade union and the requirement not to discriminate are also non-​ critical standards (Standards 5.3 and 5.7).

Where Criminal Law Meets Labour Law  73 The GLAA undertakes inspections of all new business, as part of random checks and following a risk assessment.3 When a gangmaster is inspected then the GLAA applies a scoring system. A failure to meet a non-​critical standard results in eight points and a failure of a critical standard is given thirty points. A licence will usually be revoked where the overall mark is over thirty points. Appeals against the refusal or revocation of a licence, the attachment of a condition, or the refusal to transfer a licence are heard by the Gangmasters Licensing Authority Appeal Tribunal.4 In addition to its licensing function, the GLAA was given new powers through the Immigration Act 2016 which, among other things, created Labour Abuse Prevention Officers (LAPOs). LAPOs, who work within the GLAA, have been given additional powers under the Police and Criminal Evidence Act 1984 (PACE) to arrest and search in relation to ‘labour market offences’ (as defined by section 3 of the Immigration Act 2016) across the entire labour market, not just in the specific GLA 2004 sectors. Labour market offences (‘trigger offences’) include offences under the Employment Agencies Act 1973, the National Minimum Wage Act 1998, and the GLA 2004. Specific offences under the Modern Slavery Act 2015 are also included. The consultation and subsequent response suggests that these new powers were being given to seek to tackle ‘high harm, low volume cases of exploitation falling short of those tackled by the National Crime Agency’, an issue to which we return later in the chapter.5 LAPOs only came into operation at the end of April 2017. So far, there have been no prosecutions under LAPO powers, although it appears that there has been an increase in operations by the GLAA by virtue of these powers and six Labour Market Enforcement Undertakings (LMEU) have been issued.6 Finally, under section 14 of the Immigration Act 2016, the GLAA along with the HMRC and the EASI, has the power to request a LMEU. Before doing so, they must believe that a person has committed or is committing a trigger offence. They must then give notice to the relevant person identifying the trigger offence, the reasons for believing that the offence has been or is being committed, and inviting the person to give a relevant undertaking to comply with any prohibitions, restrictions, or requirements. Such prohibitions etc must be just and reasonable (section 15(1)(b)). In addition, the authorities may apply to a court, the magistrates’ court in the England and Wales, for a Labour Market Enforcement Order (LMEO) (section 18). The circumstances in which such an application may be made are broadly similar to when the authority may invite the person to give an LMEU; the two are not mutually exclusive. An LMEO may be sought where it is shown, on the balance of probabilities, that the person has committed or is committing a trigger offence and the court considers it just and reasonable to make the order. This is an order which prohibits or restricts a person from doing anything set out in the order or requires them to do something.

3 See GLAA, ‘How we Inspect and Prosecute’ accessed 2 August 2019. 4 The Gangmasters (Appeals) Regulations 2006. 5 Department of Business Innovation and Skills, ‘Tackling Exploitation in the Labour Market: Government Response’ (January 2016) [16] accessed 2 August 2019. 6 In 2017–​18, the GLAA conducted over 100 operations, with more than 80 of those across sectors outside of the traditional agricultural sector. As a result, the GLAA arrested over 100 people for suspected labour market offences: See 2008 UK Annual Report on Modern Slavery (October 2018)  accessed 2 August 2019.

74  Criminality at Work Provisions are made in relation to variation and discharge of such undertakings and orders. Breaches are punishable by a maximum sentence of two years imprisonment. The Code of Practice (CoP) indicates that LMEUs may be used in parallel or as an alternative to existing powers.7 The appropriate approach is to be determined according to their assessment of the best means of preventing or reducing the risk of further labour market offences being committed by the business in question. The CoP goes on to say that LME undertakings and orders are an additional tool in more serious and persistent cases: namely in cases where existing civil sanctions will not prevent or stop the non-​compliance but prosecution is not yet proportionate. Therefore LME undertakings and orders constitute a mid-​ way sanction and sit at the boundary of civil liability and criminal sanctions.

2.  Employment Agencies and Employment Agency Standards Inspectorate (EASI) The mission of the EASI is described as ‘to work with agencies, employers and workers to ensure compliance with employment rights, particularly for vulnerable agency workers’. They are part of the Department for Business, Energy and Industrial Strategy (BEIS) and are the enforcement body ensuring compliance with the Employment Agencies Act 1973 (the 1973 Act) and the Conduct of Employment Agencies and Employment Businesses Regulations 2003 (the 2003 Regulations). The EASI was established after the licensing regime for employment agencies was abolished in 1994.8 The 2003 Regulations, made pursuant to the powers in section 5 of the 1973 Act, include the following key provisions: • to prevent an agency from making the provision of work to a work seeker being conditional on using other chargeable services or hiring or purchasing goods (Regulation 5); • to provide protection from detriment to workers who terminate a contract with the agency or employment business (Regulation 6); • to restrict the use of transfer fees (Regulation 10); • to restrict the withholding of wages from workers (Regulation 12); • to require the agency or employment business to provide, in writing, specific information to hirers and work seekers and obtain agreement to terms and conditions with work seekers (Regulations 13–​19, 21). All breaches of the Regulations are actionable by individuals where damage has been caused (Regulation 30). In addition, by section 5(2) of the 1973 Act, failure to comply with the Regulations constitutes a criminal offence. In addition, section 6 of the 1973 Act makes it an offence for an employment agency or business to demand or receive a fee from any person for finding or seeking to find them employment.

7 Home Office, ‘Code of Practice on Labour Market Enforcement Undertakings and Orders’ (Crown November 2016) accessed 2 August 2019. 8 Deregulation and Contracting Out Act 1994.

Where Criminal Law Meets Labour Law  75 Furthermore, in addition to the criminal offences, section 3A provides that the Secretary of State may apply to an employment tribunal for a prohibition order where a person ‘on account of his misconduct or for any other sufficient reason’ is unsuitable to run an employment agency. It is a criminal offence to fail to comply with such an order (section 3B). Any appeal against an enforcement order is made to the Employment Appeal Tribunal (EAT).9 Finally, criminal offences were created by section 9(3) and section 10 of the 1973 Act where an employment agency or business obstructs an officer or is involved in creating false records. The EASI may now also take advantage of, and operate, the LMEU/​LMEO regime. In June 2017, the EASI published the ‘Enforcement Policy Statement’,10 which sets out the stages and principles of enforcement. It emphasizes that where it has found a breach, it will usually seek compliance by issuing a warning letter setting out the requirements for compliance, the time scale in which to complete this, and any supporting evidence that will be required. Prosecution or a prohibition order may follow. It expressly notes that ‘the purpose of prosecution is condemnatory and a deterrent to others’.11

3.  National Minimum Wage and the HMRC So far we have focused on the mainly criminal powers which can be used against those actors (gangmasters, agencies) conducting their business in a way which contravenes the regulations. We turn now to the situation where the act itself is unlawful: non-​payment of the minimum wage in section B.3 and non-​compliance with the rights in the Working Time Regulations in section B.4. The NMW gives workers the right to be paid at least the National Minimum Wage. The national living wage (NLW) must be paid to workers over the age of 25. An individual worker can bring a claim for failure to pay the NMW/​NLW either as a breach of contract claim in the county court or the employment tribunal, or as a claim for deduction from wages under section 23(1)(a) of the Employment Rights Act 1996 in the tribunal. In addition, the HMRC may also enforce non-​payment of the NMW/​NLW. The HMRC enforcement regime starts with an investigation and may result in a Notice of Underpayment being issued. This notice sets out the arrears of NMW to be repaid and the penalty for non-​ compliance. Where the employer complies with the Notice then enforcement action comes to an end, otherwise payment may be pursued via the civil courts under section 19D(1)(c) of the 1998 Act or in the employment tribunal under section 19D(1)(a) of the 1998 Act.12 The BEIS Policy on HMRC enforcement of the NMW specifically states that ‘where a compliance officer discovers that the NMW has not been paid to a worker or group of 9 Of which there appears to have only been one such appeal: Webster and Burman v DTI EAT/​539/​99. 10 Department for Business, Energy and Industrial Strategy, ‘Employment Agency Standards Inspectorate (EAS):  Enforcement Policy Statement’ (June 2017)  accessed 2 August 2019 (hereafter EAS Enforcement Policy). 11 ibid [10]. 12 Department for Business, Energy and Industrial Strategy, ‘National Minimum Wage Law: Enforcement: Policy on HM Revenue & Customs enforcement, prosecutions and naming employers who break National Minimum Wage law’ (November 2017) accessed 2 August 2019 (hereafter NMW Law Enforcement).

76  Criminality at Work workers, his aim is to ensure that workers receive what they are entitled to as soon as practicable’.13 While this reflects corrective justice concern that is usually understood in civil liability terms, the punitive elements of the enforcement regime have grown in importance.14 Notably the Employment Act 2008 amended the NMW Act 1998 so that notices of underpayment could be issued and penalties were required to be paid (section 19A(1) of the NMW Act1998). The penalty was increased via the National Minimum Wage (Variation of Financial Penalty) Regulations 2014 such that the percentage penalty is now 100 per cent of the total underpayment for all workers, with the maximum penalty being £20,000. Where the underpayments are of more than £20,000 then they will also receive one notice of underpayment per worker or group of workers who have been underpaid by £20,000 or more. Section 152 of the Small Business, Enterprise and Employment Act 2015 (SBEEA) will, when it comes into effect, mean that the maximum financial penalties for NMW underpayment are calculated on a per worker basis such that the more workers an employer underpays, the greater the maximum penalty.15 In addition to these civil liability provisions, under section 31 of the 1998 Act, criminal offences have also been created:

(a) refusing or willfully neglecting to pay the NMW (section 31(1)); (b) failing to keep or preserve relevant records (section 31(2)); (c) offences in relation to the making of false records (sections 31(3) and (4)); (d) intentionally delaying or obstructing an officer or refusing or neglecting to answer questions, provide information, or documents (section 31(5)).

These are summary offences with a maximum sentence of a level 5 fine. The HMRC now also has the LMEU and LMEO regime as a further means of enforcement, which is discussed at B.5, but only in relation to the criminal offences not the matters of civil liability. Finally, the introduction of a ‘naming and shaming’ scheme, which came into effect on 1 January 2011, represented a further mechanism for enforcement. Initially employers were only named where they owed workers at least £2000 and the average per worker was at least £500. This resulted in only one referral. However the scheme was revised on 1 October 2013 with these restrictions removed and between 8 June 2014 and 24 February 2015, 157 employers were named.16 Since then the numbers of employers who have been named and 13 ibid [2.2.3]. 14 On the shifts in enforcement policy generally, Department for Business Innovation and Skills, ‘National Minimum Wage Compliance Strategy’ (March 2010)  [4.8] accessed 2 August 2019. 15 SBEEA, s 152  accessed 2 August 2019. Note the proposals beforehand: David Cameron, ‘Free Movement within Europe Needs to Be Less Free’ Financial Times (26 November 2013) www.ft.com/​intl/​cms/​s/​0/​add36222-​56be-​11e3-​ab12-​00144feabdc0.html#axzz2lpEpnxHv accessed 2 August 2019, and Department for Business, Innovation and Skills (BIS) press release, ‘National Minimum Wage Penalties Increased on Rogue Employers’ (15 January 2014) < www.gov.uk/​government/​news/​ national-​minimum-​wage-​penalties-​increased-​on-​rogue-​employers> accessed 2 August 2019. 16 BIS press release:  ‘National Minimum Wage Rogues to be Publicly Named and Shamed under New Plans’ (23 August 2013)  accessed 2 August 2019; ‘Government Names Employers who Fail to Pay Minimum Wage’ (8 June 2014)  < www.gov.uk/​government/​news/​government-​names-​employers-​ who-​fail-​to-​pay-​minimum-​wage> accessed 2 August 2019; ‘Government Names Employers who Fail to Pay the National Minimum Wage’ (27 November 2014)  < www.gov.uk/​government/​news/​government-​names-​ employers-​who-​fail-​to-​pay-​the-​national-​minimum-​wage> accessed 2 August 2019; ‘Government Names and Shames 37 National Minimum Wage Offenders’ (15 January 2015)  < www.gov.uk/​government/​news/​

Where Criminal Law Meets Labour Law  77 shamed have continued to grow with 260 named on 8 December 201717 and 179 named on 9 March 2018.18 In his May 2018 strategy report, the Director of Labour Market Enforcement and Exploitation expressed concern about what evidence was available on the impact of naming and shaming. His preliminary view was that while there should not be any distinction in relation to financial penalties according to the size of breach, the reputational penalties ‘could be tailored to greater effect, increasing both the deterrence and compliance effect’.19

4.  Working Time and the HSE The Working Time Regulations 1998 (WTR) provide workers with the right to daily rest (Regulations 10(1) and (2)), weekly rest (Regulations 11(1)–​(3)), rest breaks (Regulations 12(1) and (4)), paid annual leave (Regulations 13, 14, and 16) along with specific provision in relation to the armed forces (Regulation 25(3)) and young workers (Regulation 27(2)). These entitlements to rest are enforced in the employment tribunal by an individual worker. An individual worker also has the right not to suffer a detriment or to be unfairly dismissed for objecting to an unlawful requirement or for complaining about a breach of the Regulations (Regulations 31 and 32). In addition, rights are provided to workers in relation to the maximum weekly work (Regulation 4(2)), night work (Regulations 6 and 7), and work patterns that put a person’s health and safety at risk (Regulation 8). There is a requirement for employers to keep relevant records (Regulation 9) and compensatory rest (Regulation 24). These are the ‘relevant requirements’ that the HSE is under a duty to ‘make adequate arrangements for the enforcement of ’ except where workers are employed in premises for which a local authority is responsible. Importantly the HSE does not enforce time off, rest break entitlements, or paid annual leave entitlements. Crucially, the WTR themselves do not provide for remedies for individual workers in relation to the rights that the HSE enforces (although there is some evidence of confusion in respect of this matter before tribunals).20 This stands in stark contrast to the situation under the NMW.

government-​names-​and-​shames-​37-​national-​minimum-​wage-​offenders> accessed 2 August 2019; ‘Government Names and Shames Largest Ever Number of National Minimum Wage Offenders’ (24 February 2015) < www. gov.uk/​government/​news/​government-​names-​and-​shames-​largest-​ever-​number-​of-​national-​minimum-​wage-​ offenders> accessed 2 August 2019. 17 BIS press release, ‘£1.7m Back Pay Identified for a Record 16,000 Workers as 260 Employers are Named and Shamed for Underpaying Minimum Wage Rates’ (8 December 2017)  < www.gov.uk/​government/​news/​17m-​ back-​pay-​identified-​for-​a-​record-​16000-​workers-​as-​260-​employers-​are-​named-​and-​shamed-​for-​underpaying-​ minimum-​wage-​rates> accessed 2 August 2019. 18 BIS press release, ‘Nearly 180 Employers Named and Shamed for Underpaying Thousands of Minimum Wage Workers’ (9 March 2018) < www.gov.uk/​government/​news/​nearly-​200-​employers-​named-​and-​shamed-​for-​ underpaying-​thousands-​of-​minimum-​wage-​workers> accessed 2 August 2019. 19 Director of Labour Market Enforcement David Metcalf, United Kingdom Labour Market Enforcement Strategy 2018/​19 (Crown May 2018) 28 accessed 2 August 2019. 20 See also Barber and others v RJB Mining (UK) Ltd [1999] ICR 679 in which the High Court granted declaratory relief for breaches of reg 4(1), maximum working week.

78  Criminality at Work By Regulation 29(1) employers are guilty of an offence where they fail to comply with a relevant requirement. Such offence may be tried either way (as a summary offence or on an indictment in the Crown Court) and are liable to a fine upon conviction. Employers will also be guilty of an offence under section 33 of the Health and Safety at Work Act 1974 where they, for example, contravene a requirement imposed by an inspector, contravene a requirement or prohibition imposed by an improvement notice or prohibition notice, prevent or obstruct the work of an inspector, or provide false information to them. Broadly an HSE Working Time Officer will investigate complaints and then decide whether to refer the issue to an inspector for more formal enforcement or to provide advice and later follow up.21 The HSE, and the WTR, has not been included in the authorities which may use the new LMEU/​LMEO regime. It is understood that the HSE operates in a similar way to the HMRC in seeking informal compliance first, followed up with prohibition notices and prosecutions being the last resort. However, there is very limited information on enforcement22 with no breakdown of offences provided in the data on prosecutions.23

5.  Office of the Director of Labour Market Enforcement For completeness, it should be noted that the Office of the Director of Labour Market Enforcement (ODLME) was established through the Immigration Act 2016. The ODLME coordinates and brings together HMRC, the EASI, and GLAA. The HSE has not been included but we have been unable to establish any specific justification as to why.24 The aim of this new body, according to the government, is to address illegal immigration and ‘the businesses who exploit cheap labour from overseas’ which damage the labour market and ‘push down wages’ for others.25 The reform was also motivated by concerns about uncoordinated enforcement that can impose excessive burdens.26 A cornerstone in the strategy for ODLME has been the creation of a new post of Director of Labour Market Enforcement and Exploitation. This post has been created to ‘achieve a single set of priorities across enforcement bodies and more flexible allocation of resources’.27 21 Health and Safety Executive, ‘The Working Time Regulations 1998: Guidance on the Legislation Appendix: 6 Flow Diagram’ www.hse.gov.uk/​foi/​internalops/​ocs/​001-​099/​1_​6-​apendices/​app6.pdf accessed 2 August 2019. 22 ibid. 23 Health and Safety Executive, ‘Enforcement Statistics in Great Britain, 2018’ (Crown 31 October 2018) accessed 2 August 2019. 24 See now s 3 of the Immigration Act 2016. 25 The point emerges from David Cameron’s speech when introducing the strategy: ‘PM Speech on Immigration’ (Home Office 21 May 2015) accessed 2 August 2019. The link between the creation of the new agency and illegal immigration was repeated by Theresa May in the House of Commons on 28 May 2015, when she said its aim was ‘to crack down on the exploitation that fuels illegal immigration’. HC Deb 28 May 2015, col 211. 26 See, eg, BIS, ‘Recruitment Sector Legislation:  Consultation on Reforming the Regulatory Framework for Employment Agencies and Employment Businesses’ (January 2013) accessed 2 August 2019 and BIS, ‘Reforming the Regulatory Framework for the Recruitment Sector: Government Response to Consultation’ (July 2013) accessed 2 August 2019. 27 BIS, ‘Tackling Exploitation in the Labour Market: Consultation’ (October 2015) (hereafter BIS, ‘Tackling Exploitation:  Consultation’) [66] https://​assets.publishing.service.gov.uk/​government/​uploads/​system/​uploads/​attachment_​data/​file/​471048/​BIS-​15-​549-​tackling-​exploitation-​in-​the-​labour-​market.pdf accessed 2 August 2019.

Where Criminal Law Meets Labour Law  79 The Director is responsible for setting the enforcement strategy and ensuring delivery of that strategy by the individual enforcement bodies. Individual enforcement agencies remain separate but the Director’s role is to enable better coordination and focus, ‘form[ing] a coherent view of the nature and extent of exploitation and non-​compliance’.28 It is notable that the Director’s aims appear to be different to that of the government: he does not consider his role to be to address illegal immigration, but rather to address illegal working practices.29

6. Preliminary Conclusions When one considers the bodies that have been created, and the powers given to them, across the different regimes, it becomes clear that there are a number of different permutations and models of criminalization that have been established. Intriguingly, the approach to licensing, and the criminalization of breaches of a licence, has shifted from being removed in the employment agencies context in 1994 in search of deregulation, to being introduced in relation to gangmasters in 2004. All the regimes provide for the criminalization of the obstruction of enforcement officers but beyond that the scope, nature, and mechanisms of criminal and civil liability vary considerably. This complexity and variety is further emphasized when the nature of the offences that have been created within each of the regimes is considered: the complexity of the interplay between regulatory offences and civil liability is striking and it is to this that we now turn.

C.  Complexity of the Legal Response Created 1.  Strict Liability Offences The nature of the offences that have been created are varied and broadly operate across a spectrum (see Figure 4.1). At the most stringent level are standalone regulatory offences, including operating as a gangmaster without a licence, failing to comply with the Employment

Gangmasters

Offence to operate without a licence: GLAA enforcement

Employment Agencies Offence of failure to comply with regulations: EASI enforcement

Working Time Offences re max working week, night work, and work patterns: HSE enforcement

NMW

Offence of refusal or wilfully neglecting to pay NMW/NLW: HMRC enforcement

Figure 4.1  Strict liability offences 28 ibid [71]. 29 Director of Labour Market Enforcement David Metcalf, United Kingdom Labour Market Enforcement Strategy—​Introductory Report 1 April 2016–​31 March 2017 (Crown July 2017) 28 accessed 2 August 2019.

80  Criminality at Work Agencies Regulations, or demanding or receiving a fee for finding or seeking employment for someone and failing to comply with a relevant requirement of the WTR. These are strict liability offences: the simple failure to comply, or the act undertaken, is sufficient to prove the offence. The maximum sentences vary considerably however, from ten years’ imprisonment in relation to gangmasters to a simple fine in relation to employment agencies and working time offences.

2.  Offences Requiring an Element of Mens Rea The next ‘level’ of offence is the offence established by the NMW Act 1998. Such an offence is only made out where the person has refused or wilfully neglected to pay the NMW. In other words, there is a specific element of mens rea required. The challenges involved in an offence including such a fault element is highlighted by the discussions which took place regarding whether a new offence of aggravated labour law breach should be created where: [E]‌mployers deliberately, persistently and brazenly commit breaches of labour law and fail to take remedial action. Their pattern of exploitative behaviour neither meets the threshold for Modern Slavery offences . . . nor can it be dealt with satisfactorily through repeated use of existing penalties or offences.30

Despite broad support for this new offence in the consultation responses, following discussions with the Crown Prosecution Service around difficulties in proving the necessary intention, the government decided to move forward with the LMEU/​O rather than creating an offence involving an intention to deprive a worker of their rights.31 Thus it can be seen that the offence created in relation to the NMW is unusual, compared to the other regimes.

3.  Breach of Order Offences The next level down are the ‘breach of order’ offences: in essence, an auxiliary backstop where informal enforcement methods fail to achieve compliance (see Figure 4.2). Here we see the offence of failing to comply with an enforcement order made on application by the EASI or a prohibition notice of the HSE.32 Once again, the gangmasters provisions are stricter in that a licence will (in theory) simply be revoked. The NMW provisions are different to the other regimes: a failure to comply with an NMW enforcement notice results only in civil law liability, not the creation of a regulatory offence. There are similarities in this division when one considers the LMEU/​O provisions (see Figure 4.3):

30 BIS, ‘Tackling Exploitation: Consultation’ (n 27) [90]. 31 In contrast to the offence created by s 2 of the Modern Slavery Act 2015, which does require intent to exploit. 32 Finally, offences are created in relation to obstructing the work of each of the regulatory bodies and the failure to keep proper records, or the creation of fraudulent records. These are of less interest to us for our purposes because they are collateral to the investigative powers of the authorities.

Where Criminal Law Meets Labour Law  81 Gangmasters

Employment Agencies

Working Time

NMW

Offence to operate without a licence: GLAA enforcement

Offence of failure to comply with regulations: EASI enforcement

Offences re max working week, night work, and work patterns: HSE enforcement

Offence of refusal or wilfully neglecting to pay NMW/NLW: HMRC enforcement

No prohibition notice. Only revocation of licence.

Prohibition notice. Breach = criminal offence

Prohibition notice. Breach = criminal offence

Prohibition notice leading to civil liability

Figure 4.2  Breach of order offences

Gangmasters

Employment Agencies

Working Time

NMW

Offence to operate without a licence: GLAA enforcement

Offence of failure to comply with regulations: EASI enforcement

Offences re max working week, night work, and work patterns: HSE enforcement

Offence of refusal or wilfully neglecting to pay NMW/NLW: HMRC enforcement

No prohibition notice. Only revocation of licence.

Prohibition notice. Breach = criminal offence

Prohibition notice. Breach = criminal offence

Prohibition notice leading to civil liability

LMEU/Os

LMEU/Os

No LMEU/Os

LMEU/Os

Figure 4.3  LMEU/​Os

The LMEU/​Os are applicable to gangmasters, employment agencies, and in relation to the NMW. However they are not to be used in relation to working time matters.

4.  Civil Liability Offences The final element to consider in the spectrum is the availability of civil enforcement measures that sit alongside the regulatory offences (see Figure 4.4). The picture in respect of civil liability regimes is much more mixed: with both NMW and EASI, individual workers have civil law statutory remedies that sit in parallel with the criminal provisions. In these contexts, criminal liability is auxiliary to the civil remedies. However, with the WTR an individual worker can seek the enforcement of their right to paid annual leave, rest breaks, daily and weekly rest but there are no free standing statutory rights to enforce the maximum weekly work and night work provisions, other than by way of High

82  Criminality at Work Gangmasters

Employment Agencies

Working Time

NMW

Offence to operate without a licence: GLAA enforcement

Offence of failure to comply with regulations: EASI enforcement

Offences re max working week, night work, and work patterns: HSE enforcement

Offence of refusal or wilfully neglecting to pay NMW/NLW: HMRC enforcement

No prohibition notice. Only revocation of licence.

Prohibition notice. Breach = criminal offence

Prohibition notice. Breach = criminal offence

Prohibition notice leading to civil liability

LMEU/Os

LMEU/Os

No LMEU/Os

LMEU/Os

Civil remedy where non compliance causes damage

Civil remedy where non compliance causes damage

Civil remedies re daily and weekly rest, rest breaks, and annual leave only: individual ET claims

Civil remedy for NMW/NLW

Figure 4.4  Civil enforcement measures

Court declaratory relief for breach of statutory duty. Indeed, the rights that the Regulations provide are such that those rights that an individual can enforce and those enforced by the HSE are mutually exclusive. In those circumstances, criminal liability becomes substitutive rather than auxiliary or additive. In respect of gangmaster matters, individuals may bring a claim for a breach of the Gangmaster Rules where they have suffered damage33 but have no means of enforcing Licensing Standards per se. In those circumstances, all they can do is to complain to the GLAA so that the GLAA can enforce those requirements. The GLAA also has no powers to pay over to the individual, for example, unpaid holiday pay. Consequently the criminal law plays a greater independent role within the scheme of enforcement.

5. Summary Conclusions Taking a step back, it is apparent that the ecosystem of labour market regulation and the interplay between civil and criminal liability is highly complex. At first sight, there appears to be a greater synergy between the GLAA and EASI regimes on the one hand and the WTR and NMW regimes on the other. This could be explained by the fact that the former two regulate the operation of organizations, as opposed to the latter that regulate activities. However, that analysis breaks down when one considers the applicability of LMEU/​ Os, particularly in light of the mixed civil and criminal approach that is taken in relation to LMEU/​Os. It also breaks down when the availability of civil remedies to individuals is



33

Galdikas v Houghton and others [2016] EWHC 1376, [2016] 6 WLUK 226.

Where Criminal Law Meets Labour Law  83 considered. When considering who can enforce the rights and the nature of the regulatory offences themselves, the distinctions across the whole spectrum are muddled and complex. This raises the question as to why the regimes are so different. We do this by considering the political rhetoric surrounding the regimes

D.  Why Have Some Acts of Non-​compliance Been Criminalized? In the previous section we sought to map the rules and means of enforcement across our four chosen areas. We turn now to analyse the structure of the regulatory offences and whether the offences are civil or criminal. We also consider the agencies seeking to enforce the rules. We then turn to examine the political rhetoric surrounding the offences to seek to understand why criminalization has (or not) been selected as an enforcement tool. We shall argue that there is no obvious single explanation as to why the particular smorgasbord of regulatory choices has been made in each of our chosen areas but we suggest the greater use of criminalization is a quick but rather blunt way of demonstrating state disapproval of conduct which might in the past have been stamped on by an effective trade union movement and in certain contexts, by an effective labour inspectorate.

1.  The Political Choices (a) Gangmasters (and blacklisting) The political discourse around the GLAA is emphatically focused on the serious criminal activities that are associated with some gangmasters. Even before the Morecambe Bay cockle picking disaster, debates on the need for the regulation of gangmasters had highlighted the role of professional criminal gangs operating in this area.34 During the second reading of the Bill establishing the licensing regime, Jim Sheridan MP focused on the exploitation that workers suffered and noted that ‘[o]‌ther serious criminal activities are associated with these gangmasters. According to the National Criminal Intelligence Service, the worst gangmasters are known to be involved in both human trafficking and drug smuggling.’35 These sentiments were repeated throughout the debates in both Houses, perhaps best expressed by Roger Williams MP during the third reading: ‘These people form a criminal fraternity, and the more we can do to break its back, the better.’36 The emphasis on criminalization of gangmasters is particularly striking when one compares this to the political discourse dealing with blacklisting. In both contexts, the public pressure to criminalize was considerable and yet different routes of enforcement were taken. The power to make regulations prohibiting blacklists was given to the Secretary of State via section 3 of the Employment Relations Act 1999. However, it was not until after the discovery by the Information Commissioner of the activities of the Consulting Association that the Employment Relations Act 1999 (Blacklisting) Regulations 2010 were brought into force. The Consulting Association held a database of 3213 individuals with information

34

See, eg, Mark Simmonds, HC Deb 10 September 2003, vol 410, cc 331. HC Deb 27 Feb 2004, vol 418, cc 518. 36 HC Deb 21 May 2004, vol 421, cc 1227. 35

84  Criminality at Work about their trade union involvement that was covertly collected and supplied to a group of building companies. In the light of the gangmasters’ regime, it is perhaps surprising that the blacklisting regime is entirely one of civil liability, enforced through the county court or employment tribunal. The rights created are individual rights, with no overarching enforcement body available to enforce rights on behalf of, or in place of, individual workers. There had been numerous calls from a number of sectors, including during Parliamentary debates, for blacklisting to be criminalized, and these calls continue today. Parliamentary debates continue to focus on the lack of proper punishment for breaches, that ‘no one has ever been brought to book properly’.37 As Jack Dromey put it ‘no user company has been punished for blacklisting. No director has ended up in the dock, and that is completely wrong . . .’.38 Initially these calls for criminalization were rejected because the practice was believed to be uncommon, mirroring the rationale for declining to implement any Regulations on blacklisting before 2010. Indeed, this view appears to remain in some parts of the political arena. Vince Cable, responding in an Opposition Day motion debate in 2013, said that if blacklisting practices were continuing then that would need to be investigated and ‘we might well need legislative change’ to strengthen the Regulations. The Explanatory Memorandum to the Regulations said that ‘[b]‌lacklisting is an uncommon practice and few legal actions are therefore likely to arise’.39 At the same time, a further seam of argument was, and is, that financial penalties are adequate and provide a ‘powerful deterrent’ such that there is no need for either criminalization nor a separate public body to enforce rights. This may provide much of the explanation as to why blacklisting has not been criminalized because when the gravest labour market exploitation is considered, there may be very serious criminal activities also in play such as drug trafficking, prostitution, and child labour. By comparison, the gravest blacklisting resulted in significant economic and occupational losses but does not overlap with other matters that are obviously criminal. Nevertheless, this does not tell the whole story. We think that there is a further significant political rationale that explains more clearly why blacklisting has not been criminalized, namely the alignment of blacklisting protection with anti-​discrimination law. The Explanatory Memorandum makes clear the distinction: There are provisions in existing trade union law (principally found in section 137, 138, 146 and 152 of the Trade Union and Labour Relations (Consolidation) Act 1992) which protect individuals against discrimination in employment on grounds of their trade union membership and activities. The blacklisting of trade unionists is a practice which facilitates anti-​union discrimination.

This is repeated in the Final Impact Assessment produced in January 2010: ‘[b]‌lacklisting facilitates and encourages unlawful discrimination against trade unionists and compromises a fair system of employment relations’. Thus by categorizing blacklisting as an adjunct to discrimination, it places it within the framework of statutory torts and marries it up with

37 Chuka Umunna, Labour MP for Streatham, HC Deb 5 September 2017, vol 628, col 68WH. 38 Jack Dromey, Labour MP for Birmingham Erdington, HC Deb 5 September 2017, vol 628, col 77WH. 39 The Employment Relations Act 1999 (Blacklists) Regulations 2010, SI 2010/​493, Explanatory Note [12] accessed 2 August 2019.

Where Criminal Law Meets Labour Law  85 the provisions of the Equality Act 2010. Blacklisting is said to be unacceptable and regulation is required to ‘stamp out this practice’. By aligning blacklisting with discrimination, it locates blacklisting within the civil law domain.

(b) NMW, WTR, and employment agencies In respect of the other three regimes: NMW, WTR, and the regulations relating to employment agencies, the discourse is different again. It sits between the civil and criminal law. The concern is both to ensure minimum standards of protection for vulnerable workers against unscrupulous employers and also to ensure that ‘good employers’ are not undercut by bad ones, who seek to profit from exploiting workers.40 So it is that in these schemes there is a need . . . to balance the potential burdens on business of the structure of the regulation itself against the need to ensure that individuals receive their rights and are adequately protected; and that the economic as well as the social benefits we envisage flowing from a national minimum wage do indeed ensue.41

The discourse is also mixed in relation to the basis for the recent creation of the ODLME. In the Government’s Response to the Consultation prior to the Immigration Act 2016, it was noted that ‘[w]‌here employers are exploiting their workers by non-​compliance with employment law, the Government steps in to enforce legislation in some circumstances, for example where it is believed that there is a higher risk of exploitation or vulnerability’.42 It goes on to note: [F]‌eedback from enforcement agencies suggests that there has been a change in the nature of non-​compliance with labour market regulation over time. This has seen a shift from the more general abuses of employment regulation towards increasing organized criminal activity involving serious and organized crime gangs infiltrating labour supply chains across a number of sectors to exploit workers.43

This is reflected in the Explanatory Memorandum to the Immigration Act 2016 which states: [T]‌he Government believes that labour market exploitation is an increasingly organized criminal activity and that government regulators that enforce workers’ rights need reform and better coordination. The Conservative Party Manifesto also committed to introduce tougher labour market regulation to tackle illegal working and exploitation.44

40 See, eg, HC Deb 27 October 1998, vol 318, cc 213–​42 on WTR and HC Deb 16 December 1997, vol 303 cc 162–​239 on NMW. 41 Secretary of State for Trade and Industry on Second Reading of NMW Bill (See HC Deb (n 40) cc 170). 42 Immigration Act 2016, Commentary on Provisions of the Act, at p 5  accessed 2 August 2019. 43 At [42]. 44 Immigration Act 2016, Commentary on Provisions of the Act accessed 2 August 2019.

86  Criminality at Work In addition, it has been emphasized that the aim of the ODLME is to address illegal immigration and ‘the businesses who exploit cheap labour from overseas’ which damage the labour market and ‘push down wages’ for others.45 It must be recalled that the establishment of the ODLME arose in parallel with a clamp down on illegal migration and the political discourse was primarily concerned with stamping out illegal migration, rather than enhancing labour market protections.46 The political focus on criminality fits within that framework. However, the tone of the recent strategy promulgated by the Director of the ODLME is quite different. While he discusses serious and organized crime, this is always tied together with the work on modern slavery and that of the Independent Anti-​Slavery Commissioner. Consequently, . . . the purpose of the Strategy . . . is straightforward. First, to protect vulnerable workers. Second, to try to ensure that the majority of good compliant firms and other organizations are not undercut, and possibly driven out of business, by unscrupulous non-​compliant competitors.47

The delineation by the Director between serious crime and modern slavery on one hand and the less serious wrongs on the other is hardly surprising and reflects wider political rhetoric.48 However, it appears that the focus may be on the more serious forms of exploitation rather than the high volume, but less serious, breaches and this represents a significant strategic choice by the architects of this criminalization regime. The Director of the ODLME’s report makes explicit that the basis for the strategy is the interaction between the deterrence and compliance theories. He emphasizes in c­ hapter 2 that ‘the intended aim of enforcement tools is to encourage compliance and deter employer non-​compliance . . . the most effective enforcement approach is likely to be some mix of [compliance and deterrence]’.49 He notes that the three enforcement bodies take very different approaches with the EASI taking a primarily compliance (educative) approach, the GLAA taking a mixed approach, and the HMRC treating identified non-​compliance with a deterrence-​based approach.50 His conclusion is that: The compliance approach is premised on the idea that violations of employment regulations are the result of employer ignorance and incompetence. I recognize that a large part of noncompliance can be tackled in this way . . . I have concerns that both the chances of being investigated by enforcement officers and the scale of financial penalties for those found to be non-​compliant are too low. Not only does this provide little incentive for employers to comply, but some employers are actively discounting these costs as part of their wider business model.51 45 ‘PM Speech on Immigration’ (n 25). The link between illegal immigration and the ODLME was repeated by Theresa May in the House of Commons: HC Deb 28 May 2015, col 211. 46 Judy Fudge, ‘Illegal Working, Migrants and Labour Exploitation in the UK’ (2018) 38(3) Oxford Journal of Legal Studies 557. 47 Metcalf, ‘Labour Market Enforcement Strategy 2018/​19’ (n 19) [1.3]. 48 Theresa May PM, ‘Defeating modern slavery: article by Theresa May’ (31 July 2016) accessed 2 August 2019. 49 Metcalf, ‘Labour Market Enforcement Strategy 2018/​19’ (n 19) 32. 50 ibid 33. 51 ibid 33.

Where Criminal Law Meets Labour Law  87 Thus, rather than emphasizing either the punitive nature of criminal sanctions, as we see in much of the political discourse, or the worker protective perspective, the emphasis of the LMEU is on deterrence. In many ways the deterrence approach is better aligned to the work and employment context because, particularly in relation to lower level violations, the economics involved and the relative power between worker and employer play a more significant role. In addition, the Director acknowledged the challenges in relation to holiday pay and encouraged enforcement bodies to pursue these issues where they could.52 He also noted that ‘operational focus through complaints-​led investigations’ should remain in sectors that would not necessarily score highly on their harm metric but where there was, ‘a high volume of complaints about low harm breaches, such as hairdressing for the HMRC NMW/​NLW team’. The strategy document appears to be suggesting focusing civil penalties and ‘educative’ measures on high volume/​low exploitation breaches and reserving the criminal penalties for very serious breaches, in which collaboration with the Anti-​Slavery Commissioner and the police would be encouraged. However, the Director’s analysis regarding the naming and shaming process does not quite fit into this structure: As discussed above, the publicity arising from the naming rounds can often focus on the larger, household-​name employers as BEIS lists employers who have failed to pay minimum wages in descending order of total wage arrears. However, this often masks the relative financial severity of the non-​compliance taking place for individual workers. Taking the December 2017 naming round as an example, the Best Connection Group topped the list and owed £469,000 in unpaid wages, equating to £183 per worker. Across the 260 employers named in this round, a total of £1.764 million was owed to around 16,400 workers, averaging less than £108 each. In fact, almost 11,000 workers were owed less than £50 each. By contrast, five workers had each been underpaid by over £11,000 and a further 15 by over £3,000 (BEIS, 2017e).53

The report goes on to suggest differentiating the public measures taken against companies where the sums per worker are smaller. However, for us, this raises concerns that the focus on more serious exploitation may limit the impetus to tackle high volume, lower exploitation breaches.

(c) Summary conclusion Putting the LMEU work to one side, and to which we will return towards the end of the chapter, the political rhetoric is very mixed and does not provide a particularly clear basis to explain the choices as to whether to go down a criminal or a civil route or a hybrid approach. The one exception to this is in respect of modern slavery where the Modern Slavery Act and the Immigration Act 2016 need to be understood as an integrated legislative approach where criminalization of employer conduct is at the forefront of the regime. Outside modern slavery, we can surmise various explanations for the mixed approach. It may well be that the criminalization of labour law breaches has arisen in the context of increased deregulation. In other words, as red tape has been cut to free businesses to operate

52 53

ibid 106. ibid 58.

88  Criminality at Work as they wish, there has been a need to find new and different types of enforcement. It could also be argued that with the significant reduction in the coverage and density of trade union membership in the labour market other mechanisms have been developed to address the failure of collectivism. But we suspect that there is a third and probably more persuasive justification: effectiveness of the enforcement. Criminalization is portrayed as necessary when the breach is sufficiently serious, both for the individual and for society as a whole. In other words, criminalization of labour breaches sends a signal to employers that certain rules are important and must be complied with. Thus, success, or otherwise, of criminalization is to be measured by whether employers change their behaviour. Behavioural change is very difficult to show. However, employers will not change their behaviour if the risk of enforcement, whether actual or potential, is limited. That leads us to consider the enforcement of rights in the regimes that we are considering. Enforcement is not a direct proxy in evaluating effective deterrence. However, we argue that only if employers are likely to be caught might that prompt a change their behaviour. So in the next section we look at the effectiveness of the various enforcement regimes.

E.  Effectiveness of Enforcement Action So, in this section we consider the nature and extent of enforcement action. We consider this in two senses: the nature and extent of actual enforcement and the perception of the risk of enforcement.

1.  The Nature and Extent of Enforcement The most recent BEIS policy on enforcement of the NMW by the HMRC was published in November 2017.54 It states that the HMRC will instigate criminal investigations ‘when appropriate to bolster our overall enforcement strategy’. They emphasize that in relation to prosecutions, the HMRC will focus on the most serious cases where the NMW offences form part of a pattern of criminality and where prosecution ‘will do most to deter employers from deliberately flouting the law’. Civil enforcement may be taken alongside prosecutions to ensure that workers are repaid.55 The funding for this enforcement work by the HMRC in 2017–​18 was £25.3 million, equating to over 399 full-​time equivalent staff members.56 It is clear that enforcement overall has increased considerably both in terms of criminal prosecutions and civil enforcement. There have been fourteen successful prosecutions by the HMRC since 2007, four of which were in 2016–​17.57 The arrears recovered for workers has also significantly increased. The sums recovered were consistently between £3.2 million

54 BIS, ‘National Minimum Wage Law Enforcement Policy’ (n 12) [4.2]. 55 ibid [4.1]. 56 Metcalf, ‘Labour Market Enforcement Strategy 2018/​19’ (n 19). 57 Alison Thewliss, ‘Minimum Wage: Prosecutions:Written question –​58353’ (asked 20 December 2016, answered by Jane Ellison 9 January 2017) accessed 2 August 2019. See also Metcalf, ‘Labour Market Enforcement Strategy 2018/​19’ (n 19) 23–​24.

Where Criminal Law Meets Labour Law  89 and £4.6 million per year from 2009–​10 to 2015–​16. However, in 2016–​17 this increased to £10.2 million and increased again in 2016–​17 to £10.9 million. There does not appear to be a correlation between the amount of arrears and the number of closed cases with arrears. There has, however, been a significant increase in the number of workers for whom arrears have been recovered, rising from 19,245 in 2009–​10 to 98,150 in 2016–​17.58 In the employment tribunal only 1383 claims were brought between April 2012 and April 2017. There is very limited information available on enforcement by the HSE, or the funding provided to it for enforcement work,59 with no breakdown of offences provided in the data on prosecutions.60 When we considered the Public Register of Notices concerning working time, in the previous five years, it revealed just nine entries of which two were immediate prohibition notices and seven improvement notices. 61 Only three of the notices expressly dealt with working time matters, the others detailed various health and safety matters with working time simply highlighted as a further breach but without any detail. In addition, the historic register revealed one further case. No cases dealing with working time were found on the Prosecutions Register. However one reference in a press report was found to a prosecution in 2003.62 There appears to be very limited enforcement of working time by the HSE with no, or virtually no, hard enforcement applied. This is particularly striking when one considers the numbers of claims that have been brought in the employment tribunals for unpaid holiday pay. The official statistics for the employment tribunals show that from April 2012 to April 2017, there were 247,291 claims that included a claim relating to working time. In Q1 of 2017–​18 there were 1728 claims brought. The EASI’s enforcement action has also been very limited. The government funding for enforcement is strikingly small: just £0.5 million core budget.63 The Annual Reports from 2009 to 2016 show a fluctuating, but seemingly reducing, level of enforcement activity with the majority of enforcement activity being sending warning letters.64 The number of those prohibited or prosecuted is very low indeed. In 2016–​17 there was just one prosecution and one prohibition.65 The current list of prohibited individuals stands at just ten.66 In the political discourse for these three regimes, the emphasis was on a balance between protecting vulnerable workers and limiting the burden of regulation particularly on good employers. It is interesting to note that this appears to be reflected in the enforcement

58 BIS, ‘National Living Wage and National Minimum Wage: Government Evidence to Low Pay Commission on Compliance and Enforcement’ (July 2017) accessed 2 August 2019. 59 HSE, ‘WTR Guidance’ (n 21). 60 HSE, ‘Enforcement Statistics’ (n 23). 61 accessed 2 August 2019. 62 ‘News: Working Time Regulations—​First HSE Prosecution’ Business Advice Directory (29 September 2003) accessed 2 August  2019. 63 BIS, ‘National Minimum Wage Law Enforcement Policy’ (n 12) [4.1]. 64 For further detail, see Table 4.1. See also BIS, Employment Agency Standards Inspectorate: Annual Report 2015–​2016 (July 2016) accessed 2 August  2019. 65 Metcalf, ‘Labour Market Enforcement Strategy 2018/​19’ (n 19) 26 66 BIS, ‘Guidance: People Prohibited from Running an Employment Agency or Business’ (23 February 2018) accessed 2 August 2019.

90  Criminality at Work strategies: there have been very few prosecutions and a strategy of seeking informal compliance first has been utilized. However, it is particularly interesting that where there has been extremely limited enforcement by the HSE, there has been the greatest level of enforcement within the employment tribunals. While the type of claims brought within the tribunal and those by the HSE are different, it is fair to posit that the level of compliance with the right to annual leave is likely to be similar to the other working time rights. With regard to the GLAA, detailed data are provided in Table 4.1 (below, page 91). In 2014 it was said that the emphasis of the GLAA was moving to cases where criminal activity is alleged.67 Initially this seemed to be the case with a reduction in revocations and signs of an increase in prosecutions. However, it has not now been followed through with a consistent increase in prosecutions. In the Strategy report it is noted that in each of 2015–​ 16, 2016–​17, and 2017–​18 the GLAA took five prosecutions but no detail is provided as to which sections the offences were charged under.68 Even this represents a reduction from a high of twenty-​five offences charged in 2012–​13 and nine in 2014–​15. Consequently, the enforcement action of the GLAA appears to be becoming more limited in scope and number of successful actions. As noted above there have been no prosecutions under the new powers of LAPOs.69 The limited number of prosecutions by the GLAA is somewhat surprising, particularly given the strong political rhetoric and the concerns as to serious criminal involvement in gangmasters. However, this is perhaps reflective of two things: firstly the use of modern slavery provisions for the most serious crimes (ie those adverted to in the Parliamentary debates),70 and secondly, the strategy adopted by the ODLME which merges the GLAA concerns with those of the HMRC and HSE seeking to balance both the protection of vulnerable workers and preventing unfair competition by unscrupulous, non-​compliant competitors. However, the government funding provided to the GLAA in 2017–​18 is noteworthy, at £7.1 million equating to 104 FTE staff members.71 Nevertheless, the apparent reduction in enforcement generally by the GLAA is surprising. Drawing these points together it is clear that when the data on prosecutions is considered the degree of actual criminal enforcement is very low indeed. The focus of the authorities is now on the most serious infringements of workers’ rights and with it, an inevitable emphasis on issues of modern slavery. That may, of course, be right in policy terms. However, it is not reflective of the legislative provisions which sanction simple breaches of, for example, the WTR by way of criminal penalty. The legislative choice is one of criminalization; the practical implementation is one of encouraging compliance with the criminal law operating as an auxiliary backstop.

67 Department of Environment and Rural Affairs, Report of the Triennial Review of the Gangmasters Licensing Authority (Crown April 2014) . 68 Metcalf, ‘Labour Market Enforcement Strategy 2018/​19’ (n 19) 25. 69 Further statistics are available in Annual Report on Modern Slavery (n 6). 70 Although it is notable that the number of prosecutions under the Modern Slavery Act have also been limited compared to the number of modern slavery crimes recorded: Virginia Mantouvalou, ‘The UK Modern Slavery Act 2015 Three Years On’ (2018) 81(6) Modern Law Review 1017. 71 Metcalf, ‘Labour Market Enforcement Strategy 2018/​19’ (n 19).

828

714

603

781

No data available

Section 12 Offence

14

8

11

5

9

5

2012–​13

2013–​14

2014–​15

2015–​16

2016–​17

Year

2010–​11

2011–​12

2012–​13

2013–​14

2014–​15

2015–​16

194

23

46

229

407

0

0

1

15

4

3

Section 13 Offence

782

186

320

1479

2146

2065

0

0

0

0

0

Section 18 Offence

GLAA enforcement data1

275

133

179

471

602

917

647

1 GLAA, ‘Convictions Total’ accessed 2 August 2019.

£83,000

4

2

5

1

2

1

0

3

4

7

3

1

1

Prosecutions

0

1

2

2

0

8

2

Prohibitions

9

15

16

18

13

15

30

31

Revocation without immediate effect

No data available

No data available

£169,184

£128,523

£295,010

£204,000

Monies recovered

Revocation with immediate effect

3

643

2011–​12

243

2236

Warning letters sent

2017–​18

958

2010–​11

164

Total infringements found on all cases

3

1714

2009–​10

Targeted inspections in high risk areas /​follow up inspections

EASI enforcement data

2016–​17

Complaints received

Period

Table 4.1  Enforcement statistics

Where Criminal Law Meets Labour Law  91

92  Criminality at Work

2.  The Risk of Enforcement It is clear that very few workers, or indeed employers are aware of the work of the enforcement bodies HMRC, GLA, HSE, and EASI. In a recent study concerning the enforcement of employment rights by EU-​8 migrant workers, the authors individually interviewed twenty EU-​8 migrant workers and held five focus groups.72 Most of the participants were Polish, Lithuanian, or Latvian nationals. They had diverse occupational backgrounds, ranging from agricultural, retail, and administrative work, to dentistry and computer science. Most had been resident in the UK for two years; some had been working in the UK since accession of their home countries to the EU in 2004. None of them knew of the existence of the HMRC, GLA, EASI, or the HSE. In 2011 a Department for Business, Innovation and Skills (BIS) study found little awareness among employers about the potential for inspection and enforcement by the HMRC of the NMW and no awareness of the potential for criminal proceedings.73

F.  Conclusions In light of this analysis, what can be said of the success or otherwise of criminalization? The ideas and rationale for criminalization are muddled. Outside very serious cases of labour exploitation (such as via the Modern Slavery Act 2015), there is relatively little clarity on what criminalization is seeking to achieve in respect of breach of labour standards which are less serious. When the structure of the legislation is considered, there is little correlation between the nature of offences and means of enforcement, still less the relative budgetary provisions for the different enforcement agencies, and the political rhetoric that surrounded the establishment of the different regimes. The position is particularly confused and confusing when individual rights are set against enforcement by the enforcement bodies: there is little logic to when or why individuals can bring a claim and when it must go via an enforcement body. In this sense, the criminal law can become a two-​edged sword for the victims: they lose the right to bring a claim and are disempowered in the process. Of course, the enforcement bodies have an important role to play in protecting the most vulnerable in society and to bring claims and enforce rights that they would not otherwise be able to deal with. However this only holds good if (i) enforcement is in fact happening and (ii) at the very least, the enforcement bodies are known amongst those vulnerable workers as ‘their’ protective body. But we know that this is not the case. As we have shown above, actual enforcement is surprisingly limited. Moreover, the picture is also one of employers having very little awareness of the risk of enforcement action, either through potential criminal processes or from experiencing civil law action. When considering the degree to which civil liability is imposed on organizations that too is limited, particularly when the most vulnerable groups of workers are considered. It is 72 Catherine Barnard and Sarah Fraser Butlin, ‘Beyond Employment Tribunals: Enforcement of Employment Rights by EU-​8 Workers’ (2018) 47(2) Industrial Law Journal 226. 73 BIS, ‘Employment Relations Research Series No 121: Research into Employers’ Attitudes and Behaviours towards Compliance with UK NMW Legislation’ (October 2011)  accessed 2 August 2019.

Where Criminal Law Meets Labour Law  93 now accepted that while tribunal fees were in place, they placed a significant restriction on workers’ ability to pursue a claim in the tribunal.74 However, there are other obstacles that also exist and the level of enforcement in terms of bringing a claim is relatively low.75 When EU-​8 workers are considered, as an example of a vulnerable group, very few bring claims: a study published by one of the authors in 2016 showed that between 2010 and 2012 inclusive, the employment tribunals disposed of at most 1548 claims brought by EU-​8 migrant workers out of an annual case disposal total of over 100,000.76 The reasons are many and various,77 but the important point for our purposes is that civil enforcement of rights is also generally low. So we are left with a picture of considerable and growing criminalization when one considers the ‘law on the books’. But when one considers ‘law in practice’, there is little enforcement taking place. Although deterrence cannot be measured by enforcement statistics, the very limited level of enforcement tends to suggest that in truth the criminalization of labour law breaches is simply ineffective political signalling: politically it can be said that labour law breaches are being taken seriously but little is done in practice. Nevertheless, we detect a source of optimism. On 9 May 2018 the Director for Labour Market Enforcement published the UK Labour Market Enforcement Strategy 2018 to 2019.78 This Strategy document set out the severe limitations in the data available dealing with the nature and scale of non-​compliance with labour rights. However, he also identified modern slavery as the ‘extreme end’ of the enforcement spectrum and sat it primarily within the remit of the police and national crime agency. The Director also set out his framework for effective enforcement focusing on prioritization, deterrence, sustainability, and dealing with system-​wide effects. Importantly he emphasized that the most effective enforcement approach was a mix of compliance, that is supporting compliance by employers and promoting rights and the awareness of rights, and deterrence. The Strategy noted the Director’s concern that the chances of being investigated by enforcement officers and the scale of financial penalties for those who are non-​compliant are too low. This, he noted, gave little incentive to employers to comply and that some employers were actively discounting these costs as part of their whole business model. The Director’s proposals included:

• • • • • •

a significant increase in the size of civil penalties; an increased use of reputational penalties; an increase in the use and publicity of prosecutions; encouraging the use of LMEU/​Os; shifting towards more proactive enforcement; and greater use of joint working.

We would suggest that this Strategy represents a significant and positive change in the direction of travel. Firstly, the Director of the LMEU appears to have a clear focus on what 74 See R (on the application of Unison) v Lord Chancellor [2017] UKSC 51, [2017] 3 WLR 409. 75 On which see further Nicole Busby and Morag McDermont, ‘Access to Justice in Employment Disputes: Private Disputes or Public Concerns’ in Ellie Palmer, Tom Cornford, Yseult Marique, and Audrey Guinchard (eds), Access to Justice: Beyond the Policies and Politics of Austerity (Hart 2016) 175–​96. 76 Catherine Barnard and Amy Ludlow ‘Enforcement of Employment Rights by EU-​8 Workers in Employment Tribunals’ (2016) 45(1) Industrial Law Journal 1. 77 ibid. 78 Metcalf, ‘Labour Market Enforcement Strategy 2018/​19’ (n 19).

94  Criminality at Work criminalization is seeking to achieve. Secondly, he has carefully divided the ‘lower tier’ breaches from the most serious cases of modern slavery. This is critical to their success: the underenforcement of lower tier rights will only serve to undermine the rights themselves. It may be that the LMEU represents a wind of change bringing true enforcement, both actual and potential, and will result in changes to employer behaviour.

PART II

L A B OU R WRONG S AS PU BL IC  W RONG S

5

Exploitation at Work Beyond a ‘Criminalization’ or ‘Regulatory Alternatives’ Dichotomy Jennifer Collins*

A.  Introduction Identifying a legally appropriate response to exploitation in work relations is normatively complex. The coercive arm of the state, propelled by political, social and institutional pressures, is increasingly utilized to criminalize exploitation in its various guises. Although exploitative conduct is already criminalized via other offences, specific exploitation offences have recently been enacted, spurred by demands for ‘fair labelling’.1 The line between criminal law and civil law measures has become blurred, with new civil preventive orders backed up by criminal sanction on breach.2 Licensing regimes have been imposed in expanded areas of the labour market, to be enforced by regulators with increased powers to penalize breaches.3 A criminal and civil law response to a particular problem is not new or remarkable. The existence of civil law measures does not preclude criminalization, and vice versa. Nor is the division between civil and criminal law measures always strict.4 However, the theme of exploitation in work relations has produced an increasingly polarized set of positions ‘for’ or ‘against’ criminalization. Of late, complaints have been made about the deleterious effects of criminalization of exploitation in work relations. Judy Fudge has argued that the legal characterization of ‘unfree labour’ in both the Modern Slavery Act 2015 and the Immigration Act 2016 supports criminalization ‘as the “best” instrument’ for censuring ‘morally culpable individuals’.5 Her argument is that the effects of criminalization, deployed in a political environment that welcomes a ‘law and order’ policy agenda, over-​bear other areas of law (such as labour law) which are worker-​protective in ameliorating labour exploitation.6 A further unwelcome effect of criminalization exists for workers bearing irregular work status who can be seen as a ‘threat to working people’s living standards’.7 In this way criminalization * Senior Lecturer in Law, University of Bristol Law School. I am very grateful to Andrew Ashworth, Alan Bogg, Mark Freedland, and Virginia Mantouvalou, and to participants at Criminality at Work in June 2018 for comments and suggestions. 1 Modern Slavery Act 2015, ss 1–​2; CN v UK (2013) 56 EHRR 24. 2 See Jennifer Collins and Andrew Ashworth, ‘The Preventive Role of the Criminal Law in Employment Relations’ in this volume. 3 Immigration Act 2016, s 11(2). There has also been appeal to regulatory measures without reinforcement from the criminal law, eg Modern Slavery Act 2015, s 54. 4 Richard Macrory, ‘Sanctions and Safeguards: The Brave New World of Regulatory Enforcement’ (2013) 66 Current Legal Problems 233; Richard Macrory, Regulatory Justice: Making Sanctions Effective (TSO 2006). 5 Judy Fudge, ‘Modern Slavery, Unfree Labour and the Labour Market:  The Social Dynamics of Legal Characterization’ (2017) 27 Social and Legal Studies 414, 429. 6 ibid. 7 ibid 430. Jennifer Collins, Exploitation at Work In: Criminality at Work. Edited by Alan Bogg, Jennifer Collins, Mark Freedland QC, and Jonathan Herring, Oxford University Press (2020). © Alan Bogg, Jennifer Collins, Mark Freedland, and Jonathan Herring. DOI: 10.1093/oso/9780198836995.003.005

98  Criminality at Work of exploitation can be linked with an anti-​immigration agenda, further corroding the protective remit of labour law and human rights law for society’s most vulnerable workers. Together these arguments have led Fudge to harden her position ‘against’ criminalization of exploitation in work relations in favour of alternatives.8 This valuable work, which highlights an under-​explored and under-​theorized intersection between the criminal law and other branches of the law, provides impetus for the arguments of this chapter. First, I unpack and critically analyse two sets of criticisms of using the criminal law to regulate exploitation in work relations. The ‘internal’ critique is that criminalization can only penalize the most extreme forms of exploitation and unfree labour in work relations, neglecting long-​standing reasons for worker-​precarity. The second ‘external’ critique is that criminalization of exploitation, deployed in the current political climate of a hostile environment, unduly privileges criminalization over other areas of law and enforcement, which impedes the protection of vulnerable workers. Responding to the internal critique, I argue for detailed analysis of the reinforcement function of the criminal law in regulating exploitation at work through regulatory criminal law measures. There is no reason to think that criminal law is by definition linked to immigration controls: this is a politically contingent issue which is not inherent in the choice of criminal law as a regulatory technique. In fact, the criminal law might operate as an auxiliary means of enforcing basic labour standards, such as breach of standards found in the National Minimum Wage 1998. To date the potential variety of criminal law responses available has been overlooked and criminal law theorists can shed new light on this matter. On the ‘external’ critique, I open up new critical space for deciding how a criminal law response to exploitation in work relations ought to be fashioned. Building on and developing Andrew Simester and Andreas von Hirsch’s work on principled criminalization, I argue that careful thought needs to be given to the ways in which decisions between regulatory channels ought to be made once an in-​principle case for criminalization has been established.9 This is not the only complex question which arises in thinking about the criminalization of exploitation in the context of work relations. But it is a key point at which concerns about ameliorating negative consequences for vulnerable workers can be taken into account. At this point, a rigid position ‘for’ or ‘against’ criminalization ought to be resisted in favour of a ‘multidimensional regulatory model’.10 Specifically, I develop this notion of a richer and more nuanced regulatory mix by elaborating further principles and constraints to criminalizing exploitation in work relations. This must include appraising criminalization in its wider regulatory context, as well as more nuanced arguments about the appropriateness of criminal law interventions which include ‘regulation plus crime’ measures. The issue of enforcement is flagged as pivotal, and herein I identify concerns about current enforcement of criminal law measures relating to exploitation in work relations. This chapter is structured as follows. Section B sets out the conceptual framework, before turning in section C to consider limitations of criminalizing exploitation at work. Section 8 Judy Fudge, ‘The Dangerous Appeal of the Modern Slavery Paradigm’ Open Democracy (25 March 2015)

accessed 3 October 2018. 9 AP Simester and Andreas von Hirsch, Crimes, Harms and Wrongs: On the Principles of Criminalization (Hart Publishing 2011) 193–​98. Both civil law and criminal law channels could be considered defensible. 10 Judy Fudge and Deirdre McCann, Unacceptable Forms of Work:  A Global and Comparative Study (International Labour Office 2015) accessed 3 October 2018.

Exploitation at Work  99 D highlights uncertainty involved in considering regulatory alternatives to criminalization. Section E develops arguments which ought to inform this decision-​making as multidimensional, reflecting on the precise nature of criminal law measures proposed and on broader issues of enforcement.

B.  The Conceptual Framework In this section I build the conceptual framework by addressing three key questions. First, how does exploitation connect to the sphere of ‘work’ (broadly encompassed)? Second, what do we mean by exploitation in work relations? Third, does the state have a mandate to curb strongly exploitative practices and/​or interpersonal exploitation, and must it do so using the criminal law?

1.  Connections between ‘Exploitation’ and ‘Work’ Exploitation thrives in situations where one party is vulnerable, allowing an exploiter the opportunity to exert control over another person.11 Work relations provide a context of vulnerability for workers in at least three ways. First, workers occupy a subordinate position in the labour market—​a position of situational vulnerability.12 This view is supported by Kahn Freund’s foundational argument that the justification for labour law is ‘to be a countervailing force to counteract the inequality of bargaining power which is inherent and must be inherent in the employment relationship’.13 Second, the vulnerability of workers to exploitation is compounded for those who are deemed by law to have irregular, semi-​ precarious or precarious work status, as identified by the Council of Europe Commissioner for Human Rights and the European Court of Human Rights.14 Workers who require permission to continue to reside in a country may have this permission withdrawn at any time, such as by evidence provided by an employer.15 This is vulnerability created by labour market structures which disempowers certain groups of workers from enjoying full rights in employment relations.16 As argued powerfully by Virginia Mantouvalou, the position of vulnerability for workers is created by law—​a form of ‘legislative precariousness’.17 Nicola Countouris identifies several other ‘legal determinants of precariousness’ in personal work relations, including through classifying a worker as informal or self-​employed 11 Allen Wood, ‘Unjust Exploitation’ (2016) 54 The Southern Journal of Philosophy 92, 96. 12 Borrowing Farmer’s language, see Lindsay Farmer, Making the Modern Criminal Law (OUP 2016) 167. 13 Paul Davies and Mark Freedland (eds), Kahn-​Freund’s Labour and the Law (3rd edn, Stevens 1983) 18. 14 Council of Europe Commissioner for Human Rights, ‘The Human Rights of Irregular Migrants in Europe’ (CommDH/​IssuePaper 2007) 1, 7; Siliadin v France App no 73316/​01 (ECtHR, 26 October 2005), Rantsev v Cyprus and Russia App no 25965/​04 (ECtHR, 7 January 2010), CN v the UK (n 1), noted by Siobhán Mullally and Clíodhna Murphy, ‘Migrant Domestic Workers in the UK: Enacting Exclusions, Exemptions, and Rights’ (2014) 36 Human Rights Quarterly 397, 399. 15 Alan Bogg and Tonia Novitz, ‘Links Between Individual Employment Law and Collective Labour Law’ in Cathryn Costello and Mark Freedland (eds), Migrants at Work:  Immigration and Vulnerability in Labour Law (OUP 2016) 362. 16 See Clíodhna Murphy, ‘The Enduring Vulnerability of Migrant Domestic Workers in Europe’ (2013) 62 International & Comparative Law Quarterly 599. 17 Virginia Mantouvalou, ‘The Right to Non-​Exploitative Work’ in Virginia Mantouvalou (ed), The Right to Work: Legal and Philosophical Perspectives (Hart Publishing 2015) 60.

100  Criminality at Work (precariousness through ‘employment status’) or through denying job security (‘temporal precariousness’).18 A third source of vulnerability is that some workers, especially precarious workers such as undocumented migrant workers, have a kind of ‘double vulnerability’ to exploitation. The first layer comes through their subordinate position, outlined above, which may be exacerbated by law. But a second layer arises because they may be less able to protect themselves from exploitation, for example, by reporting what they experience and/​ or by having others to advocate on their behalf.19 Where an immigration agenda is vigorously pursued alongside identification of victims of modern slavery, for example, there is a strong disincentive for victims to report through the UK’s National Referral Mechanism (NRM). Where victims do report through the NRM they may obtain a positive Reasonable Grounds NRM decision. This grants temporary status to remain in the UK but does not regularize work status.20 Two leading anti-​trafficking charities note that ‘all a referral into the NRM has done is disrupted an exploitative situation’, with no guaranteed results. They note that vulnerable workers who ‘are dependent on any wages they do receive, no matter how low’ are likely to ‘decline a referral into the NRM, preferring to avoid contact with the authorities’.21 This lack of accountability analysis extends widely given an increasingly precarious workforce in England and Wales.22 For example, gig economy workers on zero-​hour contracts face significant challenges in establishing employee or worker status, depriving them of basic protections such as National Minimum Wage and Living Wage or working time protections.23 Incursions into statutory rights are just one example of ‘regulatory indeterminacy’ in the labour market.24 FLEX notes that the use of subcontracting and outsourcing in employment relationships ‘obscure[s]‌responsibility for workers’ rights’.25 These factors pull together to create significant barriers to workers of various employment statuses actually reporting exploitation. The argument here is that vulnerabilities of workers (inherent in the work relation, created/​exacerbated by law, compounded by real or perceived restrictions on workers’ ability to challenge or report their treatment) and the attendant control held by employers, provides opportunity for exploitation. This is therefore a far more modest statement than scholarship around exploitation which draws on the work of Karl Marx.26 The Marxist contention is that the relationship between capital (owned by employers) and workers is exploitative, though not necessarily unjust. Control of capital as a means of production makes workers

18 Nicola Countouris, ‘The Legal Determinants of Precariousness in Personal Work Relations:  A European Perspective’ (2012) 34 Comparative Labor Law and Policy Journal 21. 19 Murphy, ‘Enduring Vulnerability’ (n 16) 603. 20 The Anti-​Trafficking Monitoring Group and Human Trafficking Foundation submission of the UN Special Rapporteur on Contemporary Forms of Slavery (March 2017)  accessed 3 October 2018. 21 ibid. 22 See ‘Symposium issue’ (2012) (34)(1) Comparative Labor Law and Policy Journal. 23 CAC, Independent Workers’ Union of Great Britain (IWGB) v RooFoods limited T/​ A Deliveroo Case Number: TUR1/​985 (2016) (14 November 2017). See Employment Appeal Tribunal on Uber, Appeal No UKEAT/​ 0056/​17/​DA. 24 Fudge and McCann (n 10) 10. 25 Focus on Labour Exploitation (FLEX), ‘Tackling Exploitation in the Labour Market Consultation’, 17  accessed 3 October 2018. 26 See Alan Wertheimer, Exploitation (Princeton University Press 1999).

Exploitation at Work  101 vulnerable. On this view vulnerability necessarily results in exploitation because ‘in order to work, and therefore in order to live, the vulnerable people must sell their labour-​power to capitalists on terms advantageous to capitalists’.27

2.  What is Exploitation in Work Relations? Exploitation is a normatively complex idea, reflecting its complex reality. My own view is that we should not aspire to reduce the phenomenon down to a single neat definition or normative account. We can identify a cluster of serious (and less serious) wrongs of exploitation which manifest in work relations to various degrees at different times.28 Slavery, servitude, and forced labour might be identified as the core of serious exploitation in work relations. But the reality may be that even apparently consensual relationships between employers and workers represented contractually may be tainted by exploitative wrongdoing. Exploitation in work relations can change over time, prompting Skrivankova’s point that exploitative work relations should be viewed across a continuum ‘ranging from the positive extremity (desirable situation) of decent work to the negative extremity of forced labour’.29 It follows that some forms of exploitation at work are not very serious at all. Likewise, certain forms of very serious wrongs, such as rape, perpetrated against a trafficked worker—​ who we can clearly identify as exploited—​usefully stand alone from exploitation. This analysis carries implications for criminal lawyers who require clear definitions of serious wrongful conduct for a proposed offence. For the purposes of this chapter I propose a focus on certain core, serious, and clearly defined examples of exploitation in work relations. Exploitative conduct at the penumbra of a core of exploitation may fail to ­satisfy principles for criminalization or constraints to criminalization. That the criminal law does not penalize the full range of serious exploitative conduct has been a critique of its r­ esponse to exploitation in work relations, and I appraise this argument in section C below. It may also be the case that criminal lawyers are inclined to focus on i­ nterpersonal ­exploitation as a variety of exploitation, perhaps spurred by the presentation of ‘core’ exploitation ­offences found in sections 1–​4 of the Modern Slavery Act 2015. This may include extension of which agents count as personal. Criminal liability may include corporate employers and agencies who may be in the chain of exploitation.30 For example, the 2015 Act also introduced transparency obligations on companies in relation to exploitation in supply chains, albeit enforced by civil law measures. Is this an implicit normative view of the type of exploitation (ie interpersonal) which could properly count as a criminal ­offence? Labour lawyers will be hesitant to confine a notion of exploitation to the interpersonal variety, reflected in a concern not to limit a conception of the employer to a human ‘master’ or ‘master and servant law’.

27 Wood (n 11). 28 For example, see FLEX, ‘Shaky Foundations:  Labour Exploitation in London’s Construction Sector’ (2018) accessed 3 October 2018. 29 Klara Skrivankova, Between Decent Work and Forced Labour:  Examining the Continuum of Exploitation (Joseph Rowntree Foundation 2010) 18. 30 See JHA Anupam, ‘The Law on Trafficking in Persons: The Quest for an Effective Model’ (2018) 8 Asian Journal of International Law 225.

102  Criminality at Work The question remains: what is serious exploitation as it manifests in work relations? The equivalent of ‘exploit’ is to ‘use’. The challenge is to explain when exploitation amounts to unfair or wrongful use of another person.31 In the context of the employment relations, this value-​ judgment in turn expresses a political view on the boundary between legitimate and illegitimate work.32 The following three elements are key building blocks of a working definition of serious exploitation in work relations. First, exploitation typically requires a worker’s vulnerability vis-​à-​vis an employer, employment agency, corporation, or state agency. This vulnerability may be created by law or otherwise.33 In the work context, this position of vulnerability provides the means for an employer, employment agency, corporation, or state agency to take advantage. The gist of exploitative wrongdoing is tied to the fact that, by targeting an individual who is in a position of vulnerability, an exploiter hopes to gain easy advantage over them. A secondary reason for doing so is that a worker in a position of vulnerability may be less likely to report their ill-​treatment and/​or it may be difficult to build a case based upon their evidence. I argued in section B(1) above that workers are necessarily in a position of vulnerability in relation to employers and state agencies, and that this vulnerability is heightened for precarious workers, such as workers who lack legal work status. Second, serious exploitation involves abuse of that vulnerability through predatory conduct or predatory working practices. The concept of ‘predation’ provides us with grounds to say that an employer, employment agency, or corporation has used a vulnerable worker. Predatory conduct may include recruitment of workers who are deceived about the nature of the work on offer and/​or the standard of working conditions, or through charging fees for arranging work which may place a worker in a position of debt-​bondage.34 Predatory working practices might include the insertion of ‘sham’ substitution clauses in a worker’s terms of employment; poor working conditions which breach health and safety regulations; a refusal to pay holiday pay, overtime, or to give adequate annual leave; ‘cash-​in-​hand’ payments which are not accompanied by a pay slip, or non-​payment of National Minimum Wage.35 It may appear that a worker is ‘on board’ with these arrangements. Third, exploitation in work relations involves predatory abuse of a vulnerable person which leads to gain for an exploiter, agency, or corporation.36 Intent to gain is likely to be a reason—​often the main reason—​why a person, agency, or corporation engages in exploitative conduct, demonstrating an exploiter’s benefit from taking unfair advantage.37 But it is relevant in a strong account of exploitative wrongdoing that an exploiter makes a gain in practice as an output of exploitation. The first point to note is that an exploiter may not be the only party who gains. There may be some advantage to those who are exploited through having a job rather than being unemployed, for example. Alan Wertheimer characterizes

31 Jennifer Collins, ‘Exploitation of Persons and the Limits of the Criminal Law’ [2017] Crim LR 167. 32 Mimi Zou, ‘The Legal Construction of Hyper-​Dependence and Hyper-​Precarity in Migrant Work Relations’ (2015) 31 International Journal of Comparative Labour Law and Industrial Relations 141, 159–​60, citing Robert Steinfield, ‘Coercion/​Consent in Labour’ (2010) COMPAS Working Paper No 66, 13. 33 cf Mantouvalou (n 17) 49. Gangmaster and Labour Abuse Authority (GLAA), The Nature and Scale of Labour Exploitation Across All Sectors Within the United Kingdom (May 2018) 9 accessed 3 October 2018. 34 Examples given in this paragraph are highlighted in GLAA (n 33). 35 Alan Bogg, ‘Sham Self-​Employment in the Supreme Court’ (2012) 41(3) Industrial Law Journal 328. 36 Wood (n 11). 37 cf Stephen Wilkinson, Bodies for Sale: Ethics and Exploitation in the Human Body Trade (Routledge 2003) 20.

Exploitation at Work  103 this as ‘mutually advantageous’ exploitation.38 Second, there may be no actual gain to an exploiter in practice—​exploitation will have various effects in the real world. This is less likely in the work context, but possible. It is important to focus on the position of the parties, how a vulnerable person has been used through predatory conduct, as well as whether the transaction leads to gain for one party that is greater than the other.39

3.  What are the State’s Obligations? What are the state’s obligations in relation to criminalizing exploitation at work? Only a brief summary can be provided here.40 A large number of international obligations set out the right of individuals to be protected from various forms of exploitation. Most notably there is protection against the most serious forms of labour exploitation (slavery, servitude, and forced labour)41 and against sexual exploitation.42 A bold development in the European Court of Human Rights’ jurisprudence has been to read into Article 4 of the European Convention on Human Rights (ECHR) a right not to be subjected to human trafficking.43 There is additional and specific protection for children from exploitation.44 Criminal law measures may need to be implemented in order to secure these rights where existing measures are held to be insufficient.45 The Strasbourg court in Siliadin imposed positive duties upon states ‘to adopt criminal law provisions which penalize the practices referred to in Article 4 [ECHR] and to apply them in practice’ in order to keep pace with international developments.46 Alongside duties on the state to enact criminal law measures, international human rights obligations and EU legislation may require effective investigatory machinery and procedures in relation to exploitation.47 Questions about criminalizing exploitation must be grounded in a normative basis for state action, which addresses the proper limits of state power. I have argued elsewhere that republican political theory most clearly points in the direction of a foundational duty upon the state to address non-​domination.48

C.  Limitations of Criminal Law Interventions There are rules and principles specific to individual legal domains which guide the establishment of an in principle case for legal intervention. In ‘neo-​classical’ criminal law we begin from the position of ensuring respect for an individual’s personal autonomy to make 38 Wertheimer (n 26). 39 cf Wood (n 11). 40 For more fulsome treatment, see Collins (n 31) 170–​73. 41 Universal Declaration of Human Rights, art 4; ECHR, art 4; ILO Declaration of Fundamental Principles and Rights at Work. 42 The European Parliament approved the Committee on Women’s Rights and Gender Equality’s Report on Sexual Exploitation and Prostitution and its Impact on Gender Equality (2013/​2130(INI)) in March 2014. 43 Rantsev (n 14). 44 See UN Convention of the Rights of the Child, arts 32, 34, 35; Council of Europe, Convention on the Protection of Children Against Sexual Exploitation and Sexual Abuse, art 1. 45 In CN v UK (n 1). 46 Siliadin (n 14). 47 Palermo Protocol, arts 4, 5; Protocol of 2014 to the Forced Labour Convention, art 1; Council of Europe, Convention on Action Against Trafficking in Human Beings CETS 197 2005, art 18. 48 Collins (n 31).

104  Criminality at Work free and unconstrained choices, with state intervention only in principle justifiable where there has been serious and public wrongdoing, harm or the risk of harm, and a fault requirement. If criminal law measures can be defended beyond satisfying these threshold requirements, they will have both a censuring and preventive rationale.49 Inherent in a system of criminal law is a regulatory function. But the criminal law in England and Wales also has a direct regulatory dimension through the means of ‘regulation plus crime’ offences. A large body of regulatory criminal law exists, though these are not openly categorized as such. Here, the criminal law’s role is one of reinforcement, ensuring that breaches of non-​criminal regulatory measures are censured.50 The argument explored in this section is that there is tension in using the criminal law’s censuring and preventive functions in relation to exploitation in work relations because of its potential to produce counter-​productive effects for vulnerable workers. What is the precise nature of these criticisms and are they convincing? My focus here is on general effects of criminalization rather than the specific effects of a particular criminal law measure.

1.  ‘Internal’ Limitations of Criminalization Two arguments highlight ‘internal’ limitations of criminalization. The first is that criminalization of exploitation in work relations penalizes only the most extreme forms of exploitation in employment relations. This leaves other less serious manifestations of exploitation in the work sphere, which are nevertheless characterized by ‘hyper-​dependency’ or ‘hyper-​ precarity’, either to non-​criminal measures or unregulated.51 The second is that criminalization neglects longstanding reasons for worker-​precarity, and for some workers, their exclusion from full labour law protections.52 For example, migrant workers working in certain sectors occupy a position of precariousness which criminalization does not address. The point underlying the first criticism is that it is extremely important how exploitation is legally defined for the purposes of criminal law offences.53 If an offence definition defines exploitative conduct too narrowly/​at the most extreme end of a spectrum of exploitation, criminal law censure will only address a narrow range of employers, employment agencies, corporations, or state agencies involved in exploitation at work. This is not a radical suggestion to many criminal lawyers.54 The criminal law must select the most serious cases of public wrongdoing which causes harm or the risk of harm, alongside other thresholds to criminalization. The most robust case for criminalization must be put forward, and this will involve an argument about serious wrongdoing, and so on. Even then, countervailing considerations to criminalization must be considered. For example, criminalization must not come at the expense of protecting rule of law ideals and an individual’s right to personal autonomy.55 The costs of criminalization must not be disproportionate to the wrongful 49 Andrew Ashworth and Lucia Zedner, ‘Just Prevention: Preventive Rationales and the Limits of the Criminal Law’ in RA Duff and Stuart Green (eds), Philosophical Foundations of Criminal Law (OUP 2011). 50 ibid 283. 51 Zou (n 32) 160. 52 Judy Fudge, ‘Modern Slavery and Migrant Domestic Workers: The Politics of Legal Characterization’ (The Foundation for Law, Justice and Society 2016). 53 Julia O’Connell Davidson, ‘Editorial: The Presence of the Past: Lessons of History for Anti-​Trafficking Work’ (2017) 9 Anti-​Trafficking Review 1, 7. 54 cf Farmer (n 12). 55 John Stanton-​Ife, ‘What is the Harm Principle For?’ [2014] 10(2) Criminal Law and Philosophy 329.

Exploitation at Work  105 harm (or harm prevention) targeted by the criminal law.56 There will also be constraints like privacy to consider.57 Another set of considerations is the possibility of appropriate regulatory alternatives. It must be demonstrated ‘that the criminal law offers the most appropriate method of regulation, being preferable to alternative methods of legal control that are available to the state’.58 Given the real-​life complexity of exploitation in the sphere of work (of various types and degrees, changing over time, see section B above), it is not surprising that some exploitative wrongdoing will fall through the cracks when constructing the strongest normative account of exploitative wrongdoing for the purposes of criminalization arguments. This critique ought to remind us that the legal and political characterizations of a problem are linked.59 Criminalization can too easily be used as a political ‘trump card’ in relation to exploitation, signalling that exploitation has been adequately dealt with.60 While the criminal law can be used as ‘a kind of politics of recognition, used as a way of demonstrating a commitment to the vulnerable or insecure’, as Lindsay Farmer argues, there are difficulties in using extreme examples of exploitation in work relations as the legal representation of this class of wrongdoing.61 Julia O’Connell Davidson powerfully notes in relation to the human trafficking discourse that an [E]‌mphasis on raw physical suffering found in ‘modern slavery’ discourse helps to rein the term ‘trafficking’ back onto territory where politicians who have no interest in defending or extending the rights of either migrants or workers are more comfortable.62

Political discourse can inhibit an appropriate legal response by assuming that by punishing some exploitative conduct all serious manifestations of exploitation are addressed. The second criticism is that the criminal law is ill-​equipped to deal with sources of vulnerability to exploitation. But this prompts the need for further reflection on the reinforcement function of the criminal law through regulatory criminal law measures. If the criminal law should not be confined to censuring serious exploitative wrongdoing, might it make a greater contribution to addressing the vulnerability of highly precarious workers? There are ways in which the criminal law can deal with systemic failures, too, such as through secondary offences. One would expect a well-​ordered state to ameliorate the legal determinants of vulnerability alongside criminalization measures—​pursuing both as part of a coherent and joined up public strategy. I argue that there is a need for a multifaceted legal response to serious exploitation in work relations, which may include, but must extend beyond, criminalization.

56 Ashworth and Zedner, ‘Just Prevention’ (n 49) 244. 57 Simester and von Hirsch (n 9). 58 ibid 189. 59 See Fudge, ‘The Dangerous Appeal of the Modern Slavery Paradigm’ (n 8) 5; Bridget Anderson, ‘Migrant Domestic Workers: Good Workers, Poor Slaves, New Connections’ (2015) 22 Social Politics 636, 645. 60 Andrew Ashworth, ‘Is the Criminal law a Lost Cause?’ (2000) 116 Law Quarterly Review 225. 61 Farmer (n 12) 106. 62 O’Connell Davidson (n 53) 7.

106  Criminality at Work

2.  ‘External’ Limitations of Criminalization One important and under-​theorized insight to come from recent scholarship is a strong critique of the ways in which criminalization of exploitation interacts with other legal disciplines.63 Around the topic of exploitation in work relations, labour law, criminal law, immigration law, and human rights law overlap. These branches operate ‘simultaneously along and across different scales and institutions and they have varying degrees of influence on one another’.64 The issue as to how these legal interventions ‘fit’ together is key to establishing effects on precarious workers.65 Fudge has argued that criminalizing forms of exploitation in work relations has been inextricably linked with criminalizing breaches of immigration rules.66 This has been the case rather than offering protection to seriously exploited workers regardless of legal status.67 In Fudge’s words, the fatal flaw is that ‘[t]‌he human rights of exploited workers are brought under the gravitational sway of an agenda that strengthens the government’s powers to control and punish at the same time as it closes borders’.68 The Modern Slavery Act 2015 and the Immigration Act 2016 are used as illustrative examples. As is well known, both Acts were justified by the government in part or in whole as measures to address labour exploitation. The Modern Slavery Act 2015 targets exploitation through the enactment of two key offences penalizing modern slavery and human trafficking, both attracting maximum sentences of life imprisonment.69 Fudge argues that the 2015 Act is driven by a concern with ‘cross-​border slavery and trafficking and not with forced labour or trafficking for the purpose of labour exploitation that occurs within the United Kingdom’.70 The precise basis for this claim is not altogether clear. In part, it appears linked to the 2015 Act’s failure to define slavery and in its failure to offer labour law protections.71 That a criminalization approach is inextricably linked with sanctioning breaches of immigration rules is better supported by the Immigration Act 2016, which created an offence of illegal working.72 It is an offence for a person ‘subject to immigration control’ to work when disqualified from doing so because of their immigration status, if they have reasonable cause to believe that this is the case. The offence attracts a maximum sentence of six months’ imprisonment and/​or a fine.73 An amended offence of employing an illegal worker extends to those who are disqualified from employment by reason of the employee’s immigration status if an employer has reasonable 63 Fudge, ‘Modern Slavery, Unfree Labour and the Labour Market’ (n 5) 13–​14. 64 Judy Fudge, ‘Migrant Domestic Workers in British Columbia, Canada: Unfreedom, Trafficking and Domestic Servitude’ in Joanna Howe and Rosemary Owens (eds), Temporary Labour Migration in the Global Era:  The Regulatory Challenges (Bloomsbury 2016) 157. 65 Fudge, ‘Modern Slavery, Unfree Labour and the Labour Market’ (n 5) 12. 66 ibid 17; Ana Aliverti, ‘Making People Criminal: The Role of the Criminal Law in Immigration Enforcement’ (2012) 16(4) Theoretical Criminology 417. 67 Candice Morgan, ‘Modern Slave or Illegal Worker? The Haze Around Modern Slavery and its Implications’ British Politics and Policy Blog (17 Nov 2017) accessed 3 October 2018, argues that law enforcement officers ‘are being presented with two separate agendas: on the one hand, “hostile environments” and the removal and deportation of the undocumented person; on the other, the identification and rescue of the “modern slave”.’ 68 Fudge, ‘The Dangerous Appeal of the Modern Slavery Paradigm’ (n 8). 69 Modern Slavery Act 2014, ss 1–​4. 70 Fudge, ‘Modern Slavery, Unfree Labour and the Labour Market’ (n 5) 14. 71 ibid 14. 72 Immigration Act 2016, s 34, inserted into Immigration Act 1971, s 24B. 73 ibid, s 34(3)(4). To be amended to fifty-​one weeks’ imprisonment when s 281(5) of the Criminal Justice Act 2003 comes into force. See Immigration Act 2016, s 34(5) on confiscation orders upon conviction.

Exploitation at Work  107 cause to believe that this is the case.74 The offence carries a maximum sentence of five years’ imprisonment, increased from two.75 An offence of increased severity sits alongside possible civil law penalties of either a £15,000 or £20,000 fine.76 Fudge’s argument is that in these offences punishing and preventing ‘illegal work by migrants is entwined with the goal of preventing labour exploitation’.77 The driver, Fudge argues, is political: [G]‌overnments that embrace a ‘law and order’ agenda and that tend to demonize migrant workers as a threat to their own citizens imbue the criminal law jurisdiction with a great deal more force than that of labour law when it comes to addressing the exploitation of migrant domestic workers.78

The effect of the political and social environment is to repel worker-​protective measures found in labour law and human rights law.79 This is exacerbated by the illegality doctrine, where a crime committed by a worker bearing illegal work status sometimes bars enforcement of labour standards in the civil court.80 Should criminalization of exploitation in work relations be sidelined on this basis? Two points bear further emphasis. First, criminalization deployed in a political and legal context of stringent immigration control has powerful effects upon precarious workers. It creates further barriers to vulnerable workers reporting their exploitation, particularly where workers lack regular migration and/​or work status. Mantouvalou notes that undocumented migrant workers will be concerned not to alert the authorities to their presence in the UK, given the threat of deportation.81 Vulnerability is further increased where the same body or bodies are charged with both the protection of vulnerable workers from exploitation and the enforcement of immigration rules, or where there is information-​sharing between bodies responsible for enforcement.82 At a minimum robust rights in the criminal law must be given to those who are identified as victims of exploitation and/​or trafficking.83 This has not been the case in the Immigration Act 2016.84 Second, some evidence suggests that the enforcement of immigration rules using the criminal law can take priority over the protection of vulnerable workers who hold irregular work status via so-​called ‘exploitation’ offences.85 A recent EU Agency for Fundamental Rights report highlighted that in the EU, ‘Police generally tend to treat severely exploited third-​country nationals in an irregular situation as “illegal” workers first, rather than as

74 Immigration, Asylum and Nationality Act 2006, ss 21, s 35(3). 75 ibid s 35(4). 76 ibid s 15. 77 Fudge, ‘Modern Slavery, Unfree Labour and the Labour Market’ (n 5) 14. 78 ibid 4. 79 Zou (n 32)  160; Judy Fudge and Kendra Strauss, ‘Migrants, Unfree Labour, and the Legal Construction of Domestic Servitude: Migrant Domestic Workers in the UK’ in Cathryn Costello and Mark Freedland (eds), Migrants at Work (OUP 2014) 161–​62. 80 cf Hounga v Allen [2014] UKSC 47, [2014] 1 WLR 2889 [34], [52]. 81 Mantouvalou (n 17) 51. 82 Anderson (n 59) 644. 83 See Diane Taylor, ‘Modern Slavery Ruling May Lead to More Victims Getting Leave to Remain’ The Guardian (14 February 2018) accessed 3 October 2018. 84 Anne Davies, ‘The Immigration Act 2016’ (2016) 45 Industrial Law Journal 431. 85 Mullally and Murphy (n 14) 424.

108  Criminality at Work victims of crime’.86 There is a wider question about whether it is normatively defensible to prioritize immigration crimes over exploitation crimes. Irregular workers may be ‘immediately verifiable’ whereas exploitation is often hidden making identification difficult, coupled with strong pressure upon law enforcement agencies to reduce immigration.87 But there may be a stronger argument from principle that exploitation offences are more serious than immigration crimes, and that their enforcement should therefore be prioritized. Candace Morgan’s point is important: ‘There seems to be a complete disconnect in relation to the prospect that those exploiting the “modern slave” are able to do so because of the hostility towards the rhetorical “migrant”.’88 Criminalization of exploitative practices at work does not operate in a legal vacuum. The interaction of criminal law measures with other legal interventions has important effects for precarious workers. Nevertheless, I argue that the criminal law is not by definition linked to immigration control: this is a politically contingent issue which is not inherent in the choice of criminal law as a regulatory technique. Punishing traffickers via the Modern Slavery Act is significantly different from criminalizing victims as in the Immigration Act 2016, even if systemic difficulties which perpetuate exploitation remain unaddressed by the 2015 Act. There is a wider range of criminal law measures available than mala in se offences, and these measures may be used to ameliorate the determinants of vulnerability to exploitation. I argue that there is critical space within criminal law theory to consider the broader legal picture when interrogating a case for criminalization of exploitation in work relations.

D.  Criminalization and Regulatory Alternatives: A Key Juncture How should a criminal law response to exploitation in work relations be fashioned? Is it possible to consider how branches of the law can be used productively, so as not to undermine protections offered to vulnerable workers from exploitation? This is an emerging research agenda. First, there is a choice as to how to characterize an issue legally. Consideration must be given to ‘the different aims and functions of . . . areas of law (distributive or corrective justice, regulation, and so on)’.89 While there may be justification for civil law and/​or regulatory measures and criminal law measures it is of critical importance that criminalization should not be seen as inevitable. The politicization of the issue of legal characterization of exploitation must be resisted. Second, progress can be made in considering limits to criminal law interventions by (artificially) disaggregating a principled criminalization process into three stages. First, an in principle case for criminalizing a form of exploitation in work relations must be established. If we put to one side ‘co-​ordination offences’, serious public wrongdoing is a necessary pre-​ requisite to criminalization. This threshold requirement is tied up with the criminal law’s strong communicative function. Anthony Duff and Stuart Green note that criminalization communicates to an offender, any victims and ‘the wider community that a public wrong 86 EU Agency for Fundamental Rights, ‘Severe Labour Exploitation: Workers Moving Within or Into the EU’ (Publications Office EU 2015) 90. 87 ibid 90. 88 ibid 90. 89 Farmer (n 12) 121.

Exploitation at Work  109 was committed’.90 It would be manifestly unfair, even ‘morally defamatory’, to ‘publicly convey that [a defendant] acted reprehensibly having established nothing of the kind’.91 Arguably, while the community is not necessarily the victim of wrongdoing, it holds responsibility for censuring as ‘the appropriate body to bring proceedings and impose punishment’.92 This is certainly a plausible argument, but it requires explanation of the reasons why the community is better placed than an individual to hold responsibility for administering punishment. This may be because the state can punish more effectively than ‘lay’ punishment administered by citizens. It may also reflect a concern to ensure that fair processes are upheld and a defendant’s rights protected, since the community would normally hold responsibility for this. A fault requirement ought also to accompany a requirement for serious public wrongdoing. Criminal censure should be reserved for those who act culpably, so excluding situations where harm is caused accidentally.93 Added to these requirements is the need for harm.94 A harm threshold is crossed if the criminalization of conduct will proportionately prevent harm or the risk of harm to others. A harm principle can be used to justify almost any criminal sanction on the basis that it will prevent the risk of harm to others.95 A second stage is required: further constraints or countervailing reasons against an in principle case for criminalization must be considered. ‘Negative constraints’ include ensuring that criminalization respects the need for legal certainty, proportionality, and a person’s right to autonomy, among others.96 We noted in section C above that regulatory alternatives to criminalization must be considered, with the criminal law identified as the ‘most appropriate method of regulation, being preferable to alternative methods of legal control that are available to the state’.97 There are not firm criteria for making this decision. Third, the issue of how criminal law measures are interpreted and enforced in practice is part of a principled system of criminalization.98 A criminalization process involves not only choosing the appropriate regulatory channel, but includes evaluating how that channel will be interpreted in practice and its effects in the real world. I want to highlight the critical importance of these second and third stages in appraising the criminalization of exploitation in work relations—​consideration of good reasons not to criminalize, and the issue of enforcement. The question to be asked when considering alternatives is this: what other regulatory alternatives exist and why should criminal law measures be regarded as more appropriate than these measures (effectiveness being side-​lined as it can too easily be assumed that effectiveness must be the touchstone, without reflecting on whether it is appropriate to use the criminal law in this way)?99 These alternatives may 90 RA Duff and Stuart Green, ‘Introduction’ in RA Duff and Stuart Green (eds), Defining Crimes: Essays on the Special Part of the Criminal Law (OUP 2005) 9. 91 James Edwards, ‘Publication Review: Crimes, Harms, and Wrongs by AP Simester and Andreas von Hirsch’ [2012] Criminal Law Review 75. 92 Grant Lamond, ‘What is a Crime?’ (2007) 27 Oxford Journal of Legal Studies 621. 93 Andrew Ashworth, ‘Conceptions of Overcriminalization’ (2008) 9 Ohio State Journal of Criminal Law 507, 411. 94 JS Mill, On Liberty (first published 1859, Harmondsworth 1979). On the meaning of ‘harm’, see Andreas von Hirsch and Nils Jareborg, ‘Gauging Criminal Harm: A Living-​Standard Analysis’ (1991) 11 Oxford Journal of Legal Studies 1. 95 Andrew Ashworth and Lucia Zedner, Preventive Justice (OUP 2014) 104. 96 Simester and von Hirsch (n 9) 189, 211: ‘The criminal law is a powerful, expensive, and invasive tool. It should not be deployed lightly.’ 97 ibid 189. 98 ibid 189. 99 Andrew Ashworth, ‘Positive Duties, Regulation and the Criminal Sanction’ [2017] Law Quarterly Review 606.

110  Criminality at Work be regulatory (including civil penalties), licensing, or labour law measures, among others. It is an issue of decision-​making between regulatory channels. The least intrusive measures ought to be used, ensuring citizens’ personal autonomy and other core rights. We must consider the different modes of vindicating these rights as well as procedural protections in play when considering the strength of alternatives to criminalization. This requires appraisal of the content, type and scope of a proposed criminal law measure against a proposed alternative whose content, type and scope must also be appraised. There is scope for a great deal of ‘untidiness’ here. It is an area of considerable conceptual ambiguity providing opportunity for arbitrary decision-​making, driven by political interests. I argue that this is a key juncture in the characterization of legal interventions regarding serious exploitation. Moving beyond the tendency for positions to harden ‘for’ or ‘against’ criminalization when considering exploitation in work relations, I want to argue that it is at this site that criminal lawyers can emphasize that criminalization and regulatory alternatives do not have to be thought of as alternatives. Fudge and McCann have argued that it is important to establish a multidimensional regulatory model for unacceptable forms of work, and that idea bears development in this context. This involves designing and implementing ‘regulatory interventions that are integrated and dynamic, and that will have the broadest and most sustainable effect’.100 The best mode of regulatory design and implementation must be scrutinized in tackling unacceptable forms of work.101 Developed in the exploitation at work context, this will require clear focus on what forms of exploitation are considered unacceptable, as well as consideration of enforcement techniques and the amount of discretion given to state bodies/​enforcement agencies. It might be that the best regulatory mix involves criminal law and other forms of law operating in tandem with each other. Choices between different areas of the common law in curating a defensible legal response to exploitation need to be considered, and reasons provided for taking one path or the other, or both. For example, would criminalization, or not criminalizing, leave the vulnerable exposed to serious exploitation in work relations? Given what is at stake, criminal lawyers need to sharpen their analysis of regulatory alternatives and enforcement questions around the theme of exploitation at work.

E.  Multidimensional Regulation: Further Principles and Constraints I have argued that principled decision-​making between regulatory channels is a key and under-​theorized juncture in appraising criminalization arguments about exploitation in work relations. I have also argued that the best regulatory mix may involve the criminal law and other forms of law operating in tandem. The aim of this section is to advance this analysis by elaborating further principles. First it is important to be precise about what we are comparing. What is currently missing from critiques of the role of criminalization in regulating exploitation at work is a more

100 Fudge and McCann (n 10) 53. See Deirdre McCann and Jill Murray, ‘Prompting Formalization Through Labour Market Regulation: A “Framed Flexibility” Model for Domestic Work’ (2014) 43(3) Industrial Law Journal 319, 330. 101 Fudge and McCann (n 10).

Exploitation at Work  111 nuanced view of the range of criminal law measures available.102 The choice is not simply between criminal law measures and various forms of civil law interventions. It is conceivable that the criminal law may have a role to play in censuring serious exploitation in work relations and a regulatory role to play. As noted above, the criminal law might be used as an auxiliary means of backing up the enforcement of basic labour standards. Moreover, a different set of considerations will be relevant when comparing whether a criminal law regulatory response is more appropriate than civil law measures, than in comparing the strength of a punitive response against civil law measures. I return to this below. Turning to the distinction drawn between ‘mala in se’ offences and regulatory offences, can further principle be supplied when considering the strength of regulatory alternatives and criminalization? Beginning with ‘mala in se’ type proposals in relation to exploitation, the case for criminal sanction is strongest where an in principle case is supported by serious wrongfulness, harm or the risk of harm, and a high level of culpability. For the most serious forms of exploitation in work relations, it is hard to think that there is a case for civil and/​or regulatory measures being used without supporting criminal law measures. This must be the case in relation to the criminalization of modern slavery via the Modern Slavery Act 2015.103 We require what is distinctive from the criminal law—​a strong condemnatory state response which penalizes serious wrongdoing. Criminalization is not necessarily a last resort, only justifiable when other measures are ineffective.104 Simester and von Hirsch remind us that this is because criminalization communicates the ‘normative status’ of a serious offence. Furthermore, state action may be required, despite the presence of effective alternatives, in order to bear the burden of instigation, prosecution, and punishment on behalf of private individuals. This contributes to the rule of law, which requires consistent enforcement of legal rights in similar cases.105 Of course relevant objections must be addressed. These may include the high costs of criminalization; criminalization’s effectiveness; and ensuring legal certainty (which could be undermined if a wrong of exploitation cannot be readily defined at law or if judges were unable to adjudicate on the true nature of the exploitative wrong because the offence was drafted too broadly.)106 Where there is strength to criminal law measures, four further issues ought to be taken into account. First, of concern is the scope for easy slide between regulatory channels. The offence of employing an illegal worker is a case in point.107 The move from (administrative) regulatory measures to a serious offence with a maximum of five years’ imprisonment for an employer’s failure to comply with checking immigration status is unjustifiable.108 There must be a guarantee of justifying the specific measure that is being proposed. Second, the issue of enforcement is key. Matters of procedure and prosecutorial policy—​ ‘the everyday practices of criminalization’—​shape the substantive criminal law and keenly determine the effects of criminalization on vulnerable workers.109 One of the advantages 102 Fudge, ‘Modern Slavery, Unfree Labour and the Labour Market’ (n 5); Morgan (n 67). 103 On the duty to enact criminal offences, see Siliadin (n 14). 104 Simester and von Hirsch (n 9) 197–​98. 105 ibid 198. 106 James Edwards, ‘Justice Denied: The Criminal Law and the Ouster of the Courts’ (2010) 30 Oxford Journal of Legal Studies 725. 107 Immigration Act 2016, s 35. 108 Ashworth, ‘Positive Duties, Regulation and the Criminal Sanction’ (n 99). 109 ibid 421.

112  Criminality at Work of a coercive state response is that it involves a state-​led criminal process and higher procedural protections.110 The onus of bringing proceedings is borne by the state rather than placed on precarious workers who may struggle to access legal services. Where a confiscation order and a conviction have been secured under sections 1, 2 or 4 of the Modern Slavery Act 2015, section 8 requires a court to consider whether a reparation order can be given in order to provide victims of slavery and trafficking with compensation. No slavery and reparation orders had been made as of 23 November 2017.111 As part of a wider legal commitment to precarious workers who have been exploited in serious ways, more ought to be done to ensure that victims are compensated. This requires reconsideration of the scope of section 8.112 It may also support an argument for civil law measures, which stand alongside or independently from criminalization. This is because higher procedural protections in the criminal law may mean that enforcement of criminal law measures is more difficult.113 A third principle is to detach criminalization of exploitation from immigration enforcement. It is disappointing that this has not been the case in the Immigration Act 2016. Section 13 amends the Gangmasters (Licensing) Act 2004 to permit the Gangmasters and Labour Abuse Authority (GLAA) to request assistance from the chief police officer, the Director General of the National Crime Agency, or an immigration official if this would facilitate the GLAA’s fulfilment of its functions. The recipient of a request ‘must respond to it in writing within a reasonable period’.114 The GLAA’s remit has expanded considerably. It has powers of enforcement in relation to the Employment Agencies Act 1973 and the National Minimum Wage Act 1998. But its powers also now extend to enforcing the serious ‘core’ exploitation offences of the Modern Slavery Act 2015 where there is a connection to the labour market.115 Anne Davies notes that in linking the GLAA with immigration enforcement, a worker-​protective function may be undermined, ‘since workers may be afraid to seek help from the GLAA if they fear this might ultimately lead to their deportation’.116 Ana Aliverti’s research has found that, in practice, the enforcement of immigration offences ‘reveals that criminal law is often used against foreigners in breach of immigration rules when immigration measures . . . are ineffective or impracticable’.117 Moreover, whether or not immigration offences have a principled basis in criminal law must be carefully established rather than assumed. Aliverti has elegantly argued that a large number of immigration crimes in fact amount to a misuse of the criminal law because they criminalize conduct that is not seriously wrongful. Aliverti highlights that coercive immigration controls have been used both as a threat and in order to circumvent difficulties in deporting ‘irremovable’ migrant workers or in preventing their arrival into the UK.118 But this is to let the aim of selective and politically motivated state enforcement overpower the need for principled development of the modern criminal law. The criminal law requires a high level of legal 110 Note that the standard of proof in criminal law can vary in serious cases, see Salabiaku v France (1988) 13 EHRR 379. 111 HL Deb 23 November 2017  accessed 3 October 2018. 112 Argued by Baroness Hale, Taiwo v Olaigbe; Onu v Akwiwu [2016] UKSC 31, [2016] 1 WLR 2653 [34]. 113 See Danilenkov v Russian Federation App no 67336/​01 (30 July 2009). 114 Immigration Act 2016, s 22A(6). 115 Modern Slavery Act 2015, ss 1, 2. 116 Davies (n 84) 434. 117 ibid 425. 118 Aliverti (n 66) 426.

Exploitation at Work  113 certainty and commitment to rule of law values, for example. Effective protections against immigration offences for those who have been victims of a substantive exploitation offence ought also to be considered. Fourth, there may be political will for criminalization of the most serious forms of exploitation in work relations via ‘mala in se’ offences, but less political appetite for labour law protections which are adequately enforced. Criminal lawyers must play their part in highlighting the importance of a multidimensional approach to exploitation in work relations. A ‘multidimensional regulatory model’ is required for the most egregious forms of unacceptable work.119 This involves protecting and raising labour standards as a priority. The regulatory capacities of the criminal law in addressing exploitation in work relations must also be scrutinized in relation to regulatory alternatives. In this part of the criminal law a deterrent function is at the fore, with criminalization offering a ‘decisive coordinating and signalling device’.120 It is of importance that regulatory measures with criminal sanctions attached are not justified on the basis of effectiveness and efficiency alone. The appropriateness of a regulation plus crime measure as opposed to a regulatory alternative must be defended. In determining the criminal law’s appropriateness, we need to isolate ‘what is distinctive about the criminal law’s regulatory aspect’ as compared with regulatory alternatives backed up by, for example, administrative sanctions.121 We turn first to regulatory measures backed up by criminal sanction. The criminal law is used as a flexible tool to support licensing regimes which are regulated by various licensing authorities, for example.122 Use of regulatory criminal law provides scope to regulate poor employment practices which can be linked to exploitative relationships at work. The International Labour Organization (ILO) has identified a strong link between forced labour as a core example of exploitation and ‘multiple simultaneous violations’ of labour law standards.123 Multiple forms of less serious exploitation (eg wage theft, low-​level harassment) provide structural underpinning to more serious forms of exploitation such as forced labour. The current political focus on forced labour and human trafficking obscures other types of serious exploitation in work relations. There is a wide spectrum of exploitative conduct and regulatory offences may be better suited to responding to this complexity.124 A key principle is that regulatory offences ought to be reserved for serious breaches of regulatory standards. The Macrory Sanctions Review argued that, although the criminal law is a core part of any regulatory regime, its use must be constrained.125 It may be a reason in favour of a regulatory offence that an individual or company blatantly disregards regulatory standards, gaining in the process, for example.126 Deliberate and ongoing breaches 119 Fudge and McCann (n 10). 120 Nicola Lacey, ‘Criminalization as Regulation: The Role of the Criminal Law’ in Christine Parker, Colin Scott, Nicola Lacey and John Braithwaite (eds), Regulating Law (OUP Oxford) 144, 154. 121 ibid 145. Administrative sanctions may involve variable or fixed monetary administrative penalties or statutory notices, see Richard Macrory, ‘Sanctions and Safeguards: The Brave New World of Regulatory Enforcement’ (2013) 66 Current Legal Problems 233, 248. 122 Simester and von Hirsch (n 9). On flexibility, see Ian Ayres and John Braithwaite, Responsive Regulation (OUP 1992) 36. 123 Fudge and McCann (n 10) 38: Labour Inspection Convention, 1947 (No 81), and the Labour Inspection (Agriculture) Convention, 1969 (No 129); the Employment Policy Convention, 1964 (No 122); the Private Employment Agencies Convention, 1997 (No 181); and the Employment Relationship Recommendation, 2006 (No 198); (2013) Strengthening Action to End Forced Labour  9–​10. 124 See section B above for discussion; EU, ‘Severe Labour Exploitation’ (n 86) 34–​36. 125 Macrory, Regulatory Justice (n 4). 126 Macrory ‘Sanctions as Safeguards’ (n 4) 240.

114  Criminality at Work of employment standards, noted above as linked to forced labour as a form of exploitation, might also provide a reason in favour of criminal law measures. So could deliberate ongoing failure to have the correct licence in place, with associated financial gains, in a regulated industry. The Sanctions Review was clear that an ‘inadvertently secured financial gain through non-​compliance with a regulatory scheme’ was not a clear candidate for criminal law measures.127 The concern here is that criminalization could seem ‘heavy-​handed’, jeopardizing the criminal law’s communicative function in censuring and preventing wrongdoing.128 Macrory notes that ‘because under-​taking a criminal prosecution is a serious and resource-​heavy activity, there might be instances of under-​enforcement’.129 It should be noted that enforcement is once again key. Arguably there is an access to justice point in favour of criminal law sanctions for victims of regulatory breaches, as the state will bring action and must follow due process.130 However, this prompts an important question about who is going to undertake the issue of enforcement. Are we expecting the police and the Crown Prosecution Service to do so, in the present climate where they are overstretched? Some specialist enforcement agencies, such as the GLAA and HMRC (NMW/​NLW), have received significant increases of funding to assist in the broad task of labour market enforcement in recent years.131 But there are concerns that these bodies are too closely tied to immigration enforcement and that this will undermine their role in penalizing serious exploitation in work relations. There is growing evidence that regulatory measures detached from criminal sanctions may better enable the strategic targeting of unacceptable forms of work.132 One notable example is the idea of ‘regularisation’.133 Clíodhna Murphy has highlighted the importance of the Irish government’s implementation of an ‘Atypical Scheme for non-​EEA crew in the Fishing Fleet’ as a model of good practice in drafting legal measures which are responsive in tackling serious exploitation in personal work relations.134 This involves an administrative measure to regularize a worker’s immigration and employment status for a set period of time. The fishing industry has been identified as rife with labour exploitation given precarious employment contracts, seasonal work, and unregulated working conditions and holidays.135 Murphy notes that the scheme is ‘an attempt to address the vulnerability to labour exploitation of a group of migrant workers primarily through the medium of ordinary employment and immigration regulation’ rather than through criminalization, such as the measures adopted in the Modern Slavery Act 2015.136 Murphy argues that ‘a focus on protective labour law regulation and enforcement’ is likely to be more responsive to the realities of exploitation which is on a continuum, whilst recognizing that these measures will not 127 ibid 242. 128 ibid 242. 129 ibid 242. 130 On uncertainty regarding the future of licensing and the GLAA, see Davies (n 84) 434–​45. 131 With recommendations from the Director of Labour Market Enforcement for increased enforcement resources for the Employment Agency Standards (EAS), see David Metcalf, ‘United Kingdom Labour Market Enforcement Strategy 2018/​19’ (HM Government: May 2018) 54. 132 Fudge and McCann (n 10) 61 ff. 133 EU, ‘Severe Labour Exploitation’ (n 86) 44. 134 Clíodhna Murphy, ‘Tackling Vulnerability to Labour Exploitation Through Regulation: The Case of Migrant Fishermen in Ireland’ (2017) 46(3) Industrial Law Journal 417. 135 ILO, Caught at Sea: Forced Labour and Trafficking in Fisheries (ILO 2013); Kitty Holland, ‘Ghanian Men in Possible Trafficking Case are in State Care’ The Irish Times (16 April 2018) accessed 3 October 2018. 136 Murphy, ‘Tackling Vulnerability to Labour Exploitation Through Regulation’ (n 134).

Exploitation at Work  115 address the underlying issue of power imbalance in work relations as a source of vulnerability for migrant workers. Much will turn on the exercise of the powers of regulators and on how the regulators themselves will be held to account. The Macrory report notes that: ‘A system that allows greater use of civil penalties gives considerable power to regulators and deliberately by-​passes the automatic involvement of independent courts or tribunals in each and every case.’137 Arguably they are better placed than the police and CPS to enforce labour standards while being cognizant of the wider regulatory environment.138

F.  Conclusion I have examined arguments about limitations of criminalization in responding to exploitation in work relations. For example, the prioritization of enforcement of immigration crimes over exploitation offences, perhaps because irregular workers are more ‘immediately verifiable’, is objectionable. But recognizing limitations prompts the need for reflection on how the criminal law’s response to exploitation in work relations ought to be shaped moving forward. A rigid position ‘for’ or ‘against’ criminalization ought to be rejected in favour of fleshing out how future decisions to utilize the criminal law to respond to exploitation in work relations ought to be fashioned. The criminal law is a powerful coercive tool, but it also has a regulatory dimension which can complement some of the limitations of administrative measures which stand alone. The enforcement of criminal law measures (both ‘mala in se’ and regulatory criminal law) is an important aspect of a principled criminalization enquiry. The criminal law can play a part in enhancing labour standards, too. As I have shown, there is critical space within criminal law theory to interrogate this potential.

137 Macrory, ‘Sanctions as Safeguards’ (n 4) 257. 138 See Fudge, ‘The Dangerous Appeal of the Modern Slavery Paradigm’ (n 8) 6ff, for an argument in favour of a central labour inspectorate, and for the GLAA to be detached from enforcing immigration controls.

6

The Duty of Loyalty and the Scope of  the Law of Fraud Hugh Collins

A.  Criminalization as an Additional Deterrent for Civil Wrongs In the UK, criminal law has always provided one of the major tools for the regulation of the employment relation and labour markets. The criminal justice system serves at least three distinct functions in connection with employment. First, criminal law may provide the sanction to back up a government policy aimed at regulating the size and operation of the labour market. In the past, the criminal justice system controlled internal migration of workers within the national borders,1 whereas today it creates draconian measures to buttress prohibitions on cross-​border economic migration.2 As well as fines, imprisonment, and deportation, under laws to control immigration, all the earnings from work obtained by the irregular migrants are liable to confiscation under the Proceeds of Crime Act 2002, subject only to a loose test of the proportionality of the punishment of deprivation of property.3 Secondly, in recent times probably the most frequent use of criminal sanctions in connection with employment has been as an ingredient of a particular kind of regulatory strategy. From time to time the legislature has created an enforcement agency to secure compliance with a regulatory standard, such as a minimum wage, maximum working hours, or health and safety standards in the workplace. In order to be effective, this enforcement agency needs certain powers, such as entry of premises and inspection of documents, and it needs to be able to order employers to comply precisely with regulatory standards. These orders must be backed up by the threat of a sanction. At the apex of an ‘enforcement pyramid’4 of sanctions will be the threat of a criminal sanction, in order to ensure that ultimately employers have no choice but to comply with the regulatory regime. Thirdly, the criminal law has been used as an additional deterrent against breach of civil obligations by either the employer or the employee. Why are criminal sanctions thought to be sometimes necessary in addition to claims for breach of contract or the commission of a tort? The general answer seems to be that, in line with the justification for many criminal offences, in the law-​maker’s view, the seriousness of the harm caused by the breach of the 1 Statute of Labourers 1351 (Edward III); Statute of Artificers 1563 (5 Eliz 1 c 4). 2 Immigration Act 2016, s 34 creates a new offence in Immigration Act 1971, s 24B of illegal working, punishable by up to fifty-​one weeks’ imprisonment. Immigration, Asylum and Nationality Act 2006, s 21 (as amended by Immigration Act 2016, s 35), sets out the offence of employing an illegal worker. 3 Proceeds of Crime Act 2002; R v Carter [2006] EWCA Crim 416, [2006] 2 WLUK 133; Paulet v UK App no 6219/​08 (13 May 2014); R v Waya [2012] UKSC 51, [2013] 1 AC 294. 4 Ian Ayres and John Braithwaite, Responsive Regulation: Transcending the Deregulation Debate (OUP 1992). Hugh Collins, The Duty of Loyalty and the Scope of the Law of Fraud In: Criminality at Work. Edited by Alan Bogg, Jennifer Collins, Mark Freedland QC, and Jonathan Herring, Oxford University Press (2020). © Alan Bogg, Jennifer Collins, Mark Freedland, and Jonathan Herring. DOI: 10.1093/oso/9780198836995.003.006

The Duty of Loyalty and the Law of Fraud  117 civil obligation merits an approach that maximizes the deterrence provided by the law.5 That seems to have been a reason for enacting the Master and Servant Act6 that imposed criminal sanctions for workers leaving their jobs without permission. Employers argued that the serious harm to their businesses caused by industrial action or employees quitting for better wages offered by other employers had to be met by threats of criminal sanctions, because the civil sanctions of claims of damages for breach of contract against impecunious industrial workers were insufficient to deter this behaviour. In modern times, criminal sanctions have been introduced selectively to provide additional protections for workers. The Modern Slavery Act 2015 provides an example of the introduction of criminal penalties as a deterrent sanction for what are almost certainly torts like false imprisonment and breaches of contract. In this third function of criminal law in connection with employment, the addition of a bespoke criminal offence tries to augment the deterrent effect of the civil law (and any other existing criminal offences). Though rarely mentioned in books on labour law, the law of fraud and related offences has been applied regularly to employment relations. It has performed this third function of criminalization. It supplies a deterrent in addition to possible remedies in civil law such as damages for conversion, deceit, and breach of contract. Today, many instances of frauds committed by employees can also be charged as theft, because, if the deceit was successful, the employer’s property would had been appropriated by the employee. Under the old Larceny Acts, however, the need for employees actually to take the employer’s property away physically led to the development of new offences such as embezzlement and false accounting that could encompass the typical financial frauds conducted by managers, stewards, and other white-​collar workers.7 When the law of theft was reformed in the Theft Act 1968, it replaced the law of embezzlement with an offence of obtaining property by deception,8 and revised other offences such as false accounting.9 That reform proved inadequate. For instance, it did not include obtaining services by deception, so that although an employee who lied to obtain a job might be guilty of an offence of fraud (though only if wages counted as property),10 an employer who persuaded an employee to work for him with no intention to pay for the work would not be guilty of a criminal offence at all. Although that ‘loophole’ (a term that suggests perhaps inaccurately that an accidental mistake was made initially in 1968) was closed by a fresh statute in 1978,11 the handling of fraud and related problems in the new legislation was marked by the adoption of rules that simultaneously and quite remarkably created offences that were at the same time both vague and highly fact specific, as Parliament tried to fill apparent gaps as they popped up in the criminal courts. For instance, there was the crime of ‘making off without payment’ to cover the instance of someone leaving a restaurant without paying the bill,12 and the offence of ‘obtaining a

5 Andrew Ashworth, Principles of Criminal Law (Clarendon 1991) ch 2. 6 Master and Servant Act 1823 (4 Geo IV c 34), mostly repealed by Employers and Workmen Act 1875 (38 & 39 Vict c 90). 7 Larceny Act 1861, ss 82 and 83; Larceny Act 1916, ss 18 and 19. 8 Theft Act 1968, s 15. 9 Theft Act 1968, s 17. 10 If not, there was a specific offence in Theft Act 1968, s 16 of using deception to obtain the opportunity to earn money in employment. 11 Theft Act 1978, s 1. If the employer avoids making a false or misleading statement, this conduct of cheating on wages may be covered by Fraud Act 2016, s 11 (‘obtaining services by deception’). 12 Theft Act 1978, s 3.

118  Criminality at Work pecuniary advantage’ to cover cases where an entry in a bank account had been fraudulently adjusted.13 Now, following detailed work by the Law Commission,14 the whole law of fraud has been reformed by the Fraud Act 2006. The statute is intended as a codification of the field. Broadly in line with the recommendations of the Law Commission, section 1 creates three ways in which the offence of fraud may be committed. These three versions are set out in detail in sections 2–​4. My initial purpose is to map and understand where the Act criminalizes the conduct of employees and employers in order to provide an additional deterrent against a breach of contract. What emerges in the discussion is that many of these criminal offences under the Fraud Act 2006 are parasitic on the existence of civil wrongs in the form of breach of contract or tort. In other words, the actus reus of the offence is often comprised in part by a breach of a civil obligation in tort or contract. Unfortunately, in this area of the law of the contract of employment, it is often contested whether or not a civil obligation exists in the first place and what kind of obligation it imposes. As a consequence, in tension with the values of the rule of law and Article 7 of the European Convention on Human Rights, the effect of the Fraud Act 2006 is to criminalize conduct of employers and employees where there is a great deal of uncertainty about what conduct is prohibited. Indeed, there is perhaps no greater area of uncertainty and confusion in the common law of the contract of employment than questions about the existence, scope, application, and the correct label for what is variously described as a duty of loyalty, a duty of fidelity, a duty of confidence, a duty of mutual trust and confidence, or a duty of good faith owed by employees to employers.15 In order to clarify the scope of the criminal law, the chapter therefore devotes considerable space to an investigation of the civil law regarding the duty of loyalty. In particular, it compares the different obligations of disclosure and duties to protect the financial interests of the other party that apply to employers and employees. In pursuit of that endeavour, the discussion draws on the idea that contracts of employment are properly classified as ‘relational contracts’. Within relational contracts there are typically implied obligations of loyalty and good faith in performance that do not arise in ordinary commercial and consumer transactions. These implied obligations are necessary to achieve an efficient functioning of this kind of contract. The question is when these implied obligations provide the necessary civil obligation, breach of which may satisfy an ingredient of the criminal offence of fraud. Having attempted to clarify the scope of the criminalization of breach of civil wrongs brought about by the Fraud Act 2006 in connection with employment, the final task to be considered is whether or not this criminalization appears to be a necessary and proportionate deterrent to reinforce the existing sanctions for civil wrongs. Ashworth notes that many property offences involve trivial harms, so that it is arguable that the principle that criminal law should only be used for significant challenges to social order often seems to be ignored in this context, especially given that there are usually civil remedies available to a victim.16 In some instances of fraud, it is arguable that there has been no harm caused at all. For example, if a worker obtains a job by lying about his or her experience, although that



13

Theft Act 1978, s 16. Law Commission, Fraud (Law Com No 276 2002). 15 Marathon Asset Management v Seddon [2017] EWHC 300 (Comm), [2017] ICR 791 [111]. 16 Ashworth, Principles (n 5) 325. 14

The Duty of Loyalty and the Law of Fraud  119 action is probably a criminal fraud, if the worker in fact performs a good job in return for the wages, it is unclear what harm to the employer may have been caused. Obtaining a job by a lie is and should be a civil wrong; the issue is rather whether it should also, in addition, be a criminal offence. It will be argued, in conclusion, that the criminalization of breaches of the duty of loyalty by employees represents an unconscious appeal to the moral values of a bygone era when unfaithful servants were believed to deserve severe punishment.

B. Fraudulent Omissions The three ways in which the offence of fraud may be committed under the Fraud Act 2006 extend beyond the common usage of the word fraud in several respects. Here we will focus in particular on the extension to omissions in the form of a failure to inform the other party of facts, thereby intending to cause the other party a loss or a risk of loss. This emphasis is partly dictated by the consideration that omissions constitute the area of greatest uncertainty about the scope of civil obligations, so that for these criminal offences that are parasitic on a civil wrong, this will also be the greatest area of uncertainty about the scope of the criminal law. The emphasis on disclosure of information is also influenced by the wider concern of this chapter to assess whether the criminalization of civil wrongs is proportionate. Since omissions are only exceptionally regarded as deserving of punishment in criminal law,17 the criminalization of omissions to disclose information is likely to be the area that proves most questionable when we come to assess it according to a test of proportionality.

1.  False or Misleading Representation Under the Fraud Act 2006, section 2 it is an offence dishonestly to make a false representation, whether express or implied, in order to make a gain or to cause loss or a risk of loss to another. A representation is false if the person making it knows that it is untrue or misleading. It is worth analysing a few elements of this offence that are shared by the other two versions of the offence of fraud committed by omissions. Like all property offences in the criminal law, fraud requires proof of dishonesty. The Supreme Court recently reformulated the guidance to be given to a jury on the question of the meaning of dishonesty. Speaking for the court, Lord Hughes said: When once his actual state of mind as to knowledge or belief as to facts is established, the question whether his conduct was honest or dishonest is to be determined by the fact-​ finder by applying the (objective) standards of ordinary decent people. There is no requirement that the defendant must appreciate that what he has done is, by those standards, dishonest.18

17 Andrew Ashworth, ‘Criminalising Omissions’ in Andrew Ashworth, Positive Obligations in Criminal Law (Hart Publishing 2013) ch 2; Andrew Ashworth, ‘Positive Duties, Regulation and the Criminal Sanction’ (2017) 133 Law Quarterly Review 606. 18 Ivey v Genting Casinos (UK) Ltd t/​a Crockfords [2017] UKSC 67, [2018] AC 391 [74].

120  Criminality at Work Thus, a finding of dishonesty depends on the objective standards of ordinary decent people, as applied to the facts as they were known to and understood by the defendant. Like the two other versions of the offence, section 2 creates an inchoate offence in the sense that the false statement has to be made with a view to achieving a gain or causing a loss, but it is not necessary for any gain or loss to be proven to have occurred. Nor need the loss be a certain foreseeable consequence of the fraud, if successful, because merely intending to create a risk of loss is sufficient. Thus, in a case where an employer is uncertain that it will be able to pay the wages at the end of the week, but nevertheless asserts expressly or impliedly that it will be able to pay wages, knowing that statement to be untrue or misleading, the employer has intentionally caused a risk of loss and so (subject to the question of dishonesty) the crime is complete, even though, at the end of the week, fortuitously the employer has sufficient funds to cover the wage bill. Similarly, if an employee lies about his or her experience in order to get a job, the crime has been committed by the hope of making a gain in the form of wages, even if the employer would have hired the worker for the job without the lie. In other words, the actus reus of the offence that is described in the statute is in fact an attempt rather than a successful and completed obtaining of another’s property or services. According to Ashworth, oddly it seems possible that there could be a valid charge of an attempt to commit this offence of (an attempted) fraud.19 It should be noted further that, unlike the civil law of deceit or misrepresentation, the criminal offence of fraud does not require the statement to be demonstrably false in fact. It is sufficient if the representation is misleading and the defendant is aware that it is likely to mislead someone else. In extending the criminal law to misleading statements, this provision replicates a similar extension in regulatory offences, such as those protecting consumers against misleading advertising and mis-​selling of products and services.20 But fraud offences differ from those consumer offences owing to the requirement of dishonesty and the potential severity of the sanction. Conviction of the three versions of fraud results, on summary conviction, to imprisonment for a term not exceeding twelve months or a fine, and on conviction on indictment, to imprisonment for a term not exceeding ten years, or to a fine, or to both.21

2.  Fraud by Failure to Disclose Information The Fraud Act 2006, section 3 provides that: A person is in breach of this section if he—​ (a) dishonestly fails to disclose to another person information which he is under a legal duty to disclose, and (b) intends, by failing to disclose the information—​ (i) to make a gain for himself or another, or (ii) to cause loss to another or to expose another to a risk of loss.



19

Andrew Ashworth, ‘A New Generation of Omission Offences?’ [2018] Criminal Law Review 354, 355–​56. The Consumer Protection from Unfair Trading Regulations 2008, SI 2008/​1277. 21 Fraud Act 2006, s 1. 20

The Duty of Loyalty and the Law of Fraud  121 Some of the elements of this version of fraud, such as the dishonesty and its inchoate character, are the same as the core offence in section 2. But instead of an action such as telling a lie, this version of fraud consists of conduct in the form of an omission: a failure to disclose to another person information which the defendant is under a legal duty to disclose. In general, therefore, the scope of this offence will depend on two crucial issues: first, was there a legal duty to disclose information, and second, was the defendant dishonest by reference to the standards of ordinary decent people. The statute provides no further elaboration on the issue of when there will be a legal duty to disclose information. In its preparatory work, the Law Commission clearly assumed that the incidence of a legal duty to disclose information would be dependent on civil law. The Commission report stated: Such a duty may derive from statute (such as the provisions governing company prospectuses), from the fact that the transaction in question is one of the utmost good faith (such as a contract of insurance), from the express or implied terms of a contract, from the custom of a particular trade or market, or from the existence of a fiduciary relationship between the parties (such as that of agent and principal).22

The possible application of section 3 to employment arises primarily because the express or implied terms of the contract of employment may impose duties of disclosure of information on either party. If there is a breach of such a term by reason of a failure to disclose information, in order to constitute a criminal offence, the remaining issues would be whether that party intended to make a gain or expose the other to a risk of loss, and, crucially, whether there was dishonesty.

3.  Fraud by Abuse of Position Section 4 of the Fraud Act 2006 states: (1) A person is in breach of this section if he—​ (a) occupies a position in which he is expected to safeguard, or not to act against, the financial interests of another person, (b) dishonestly abuses that position, and (c) intends, by means of the abuse of that position—​ (i) to make a gain for himself or another, or (ii) to cause loss to another or to expose another to a risk of loss. (2) A person may be regarded as having abused his position even though his conduct consisted of an omission rather than an act.

This third way that fraud can be committed shares the features of the other offences in that it criminalizes the inchoate form of the actus reus and the fault element is a requirement of dishonesty. The actus reus is the abuse of a position in which a person is expected to



22

Law Commission, Fraud (n 14) [7.28].

122  Criminality at Work safeguard, or not to act against, the financial interests of another. Like the previous one, this offence can be committed by an omission, so that a failure to safeguard the financial interests of another could also amount to the offence. That feature of the offence raises the question of what kinds of position will incur this positive obligation. It is clear from the report of the Law Commission that an obligation to safeguard the financial interests of another is likely to arise in positions that would be described in civil law as fiduciary positions. Nevertheless, the Commission insists that a fiduciary relationship should not be necessary.23 The Commission states explicitly that the necessary relationship can be present between employer and employee.24 The report gives the example of an ‘employee who, without the knowledge of his employer, misuses his or her position to make a personal profit at the employer’s expense’.25 The Commission describes the kinds of relationship that trigger section 4 as ones where ‘the victim voluntarily put the defendant in a privileged position, by virtue of which the defendant is expected to safeguard the victim’s financial interests or given the power to damage those interests’.26 Since every contractual relationship tends to create the possibility of one party damaging the financial interests of the other by, for instance, not paying for the goods or services ordered, the scope of this offence is potentially extremely broad unless sufficient weight is attached to the notion of what the Law Commission called a ‘privileged position’, or what the statute refers to a ‘position’ that can be abused. In agreement with that view, in R v Valujevs27 the Court of Appeal interpreted section 4 to require a position ‘akin to a fiduciary relation’. In their view, that requirement would not be satisfied in ordinary commercial and employment relations. In that particular case, however, it was said that there was evidence of the necessary privileged position. The case concerned an unlicensed gangmaster, who received payments from customers that should have been passed on to the workers as wages, but who withheld substantial amounts on unwarranted grounds such as payment of rent and other charges. In law, the position of the gangmaster was arguably akin to that of agent for the workers, so that the duty of an agent to protect the interests of a principal was present to some extent. The relevant legislation on gangmasters provided support for the view that in this relationship gangmasters were legally charged with responsibilities that arose from their position of control over the workers.28 Although a position akin to a fiduciary should suffice for section 4(1)(a), it is not a necessary condition, as Jennifer Collins points out.29 According to the text of the statute, it should also apply to positions where one person has the power to harm the economic interests of

23 Nor, incidentally, is it sufficient for the offence to be committed that there should have been a breach of fiduciary duty. It should be appreciated that the requirement of dishonesty removes the strict liability characteristic of many fiduciary duties in civil law. Not all breaches of fiduciary duty will therefore be regarded as fraud under this section, because only some of them are likely to be regarded as dishonest by ‘ordinary decent people’. For instance, a fiduciary who believes that the beneficiary had agreed or would agree to a particular use of his or her position is unlikely to be found to be dishonest, even though the misuse of the fiduciary position would result in a civil remedy being available to a beneficiary. 24 Law Commission, Fraud (n 14) [7.38]. 25 ibid [7.36]. 26 ibid [7.37]. 27 R v Valujevs [2014] EWCA Crim 2888, [2015] QB 745. 28 Gangmasters (Licensing) Act 2004; Gangmasters Licensing Authority Licensing Standards 2012. 29 Jennifer Collins, ‘Fraud by Abuse of Position and Unlicensed Gangmasters’ (2016) 79 Modern Law Review 354, 359.

The Duty of Loyalty and the Law of Fraud  123 another and there is an expectation that the power should not be exercised. Employers and employees will normally find themselves in a position where they have the practical ability to harm the interests of the other. An employer can abuse its position in various ways such as demanding extra unpaid work or imposing unilateral wage cuts, and employees can shirk on the job or damage their employer’s property. The question is whether the contract of employment creates the kind of privileged position that is associated with expectations that both parties will and should refrain from acting in a self-​interested way that may involve harming the other. Whilst the contract of employment does not create a fiduciary relationship, as will be discussed below, it is best described as a type of relational contract, in which legal duties of good faith signal the existence of such an expectation.

C. Relational Contracts This brief review of the provisions of the Fraud Act 2006, especially as they relate to liability for omissions in the context of employment relations, has revealed that the scope and application of the variations on the criminal offence of fraud in sections 3 and 4 of the Fraud Act 2006 depend to a considerable extent on the existence of obligations in civil law.30 One important question is when, if at all, does the civil law impose an obligation on the parties to a contract of employment to disclose information? A second question is when, if at all, does the civil law impose an obligation on the parties to a contract of employment to safeguard the financial interests of the other party or at least to warn them against potential harm to their material interests? Before considering those questions in detail, it is useful to explore the characteristics of the contract of employment as a relational contract, for that will cast considerable light on the proper scope of liability for omissions in that context. The contract of employment is to some extent governed by the general law of contract. Notoriously, the general law of contract in the common law rarely imposes implied duties to disclose information.31 Nor does the law of contract require the parties to look after the interests of the other party. Instead the general law of contract permits each party to give priority to its own interests. Yet there are some kinds of contract where that general approach does not hold. There are special kinds of contracts, such as partnerships and agencies, where the parties owe some fiduciary obligations that are likely to require duties of disclosure and protecting the interests of the other. In addition, there are relational contracts where the right exclusively to look after one’s own interests is modified by an implied obligation to perform in good faith.32 It is often asserted that contracts of employment are relational contracts.33 Sometimes that label means little more than the point that these contracts involve personal relationships 30 Though often criminal courts are reluctant to investigate civil law questions: David Ormerod, ‘The Fraud Act 2006—​Criminalizing Lying?’ [2007] Criminal Law Review 193, 206. 31 Hugh Collins, ‘Implied Duty to Give Information During Performance of Contracts’ (1992) 55 Modern Law Review 556. 32 Yam Seng Pte Ltd v International Trade Corporation Ltd [2013] EWHC 111, [2013] All ER (Comm) 1321 (QB). 33 Johnson v Unisys Ltd [2001] ICR 480 [16] (HL); approved in Braganza v BP Shipping Ltd [2015] UKSC 17, [2015] ICR 449 [54] (Lords Hodge and Kerr). Robert C Bird, ‘Employment as a Relational Contract’ (2005) 8 University of Pennsylvania Journal of Labor and Employment Law 148; Matthew Boyle, ‘The Relational Principle of Trust and Confidence’ (2007) 27 Oxford Journal of Legal Studies 633; Mark Freedland, The Personal Contract of Employment (OUP 2003) 88; Douglas Brodie, ‘How Relational is the Employment Contract?’ (2011) 40 Industrial Law Journal 232; Jonathan Levin, ‘Multilateral Contracting and the Employment Relationship’ (2002) 117 Quarterly Journal of Economics 1075.

124  Criminality at Work because the work has to be performed personally by the employee. Sometimes it means that the parties to this contract are expecting it to last for a substantial period of time and, in that sense, they will be forming an ongoing relationship rather than merely making a one-​off discrete exchange. While that may be true in many instances of employment, a long-​term contract is not a necessary incident of relational contracts, and nor does its long-​term nature give rise to any special legal obligations of the kind envisaged in relational contracts.34 The legal relevance of describing contracts of employment as relational contracts concerns primarily the implied obligations that arise under the contract.35 The legal category of relational contract recognizes that certain kinds of contract cannot be set out in a determinate way in advance through the express terms of the contract.36 Either expressly or impliedly the parties to such contracts must undertake open-​textured obligations to cooperate in order to bring the purpose of the contract to fruition. At the same time as being cooperative, however, the parties are entitled to look after their own interests in order to ensure that they receive the expected pay-​offs from the contractual undertaking. A relational contract is thus partly an arm’s-​length contract and partly an integrated system of production in which the parties have to be loyal to the common purpose. The parties must both cooperate and compete.37 This duty to cooperate in order to make the venture successful may require from the parties many different kinds of particular duties according to the circumstances. Because it is impossible to pin down in advance what these obligations of cooperation may entail, it is more convenient to say that there is a duty to perform the contract in good faith. This implied duty to perform a relational contract in good faith can be regarded as a term implied by law into contracts that are properly classified as relational contracts.38 It is a term that is inserted not on grounds of fairness but rather because it is necessary for the business efficacy of these kinds of contracts.39 Relying to some extent on Australian authorities,40 Leggatt J has usefully summarized the implied requirement of good faith in relational contracts in these terms: [A]‌n obligation to act honestly and with fidelity to the bargain; an obligation not to act dishonestly and not to act to undermine the bargain entered or the substance of the contractual benefit bargained for; and an obligation to act reasonably and with fair dealing having regard to the interests of the parties (which will, inevitably, at times conflict) and to the provisions, aims and purposes of the contract, objectively ascertained . . . the obligation of fair dealing is not a demanding one and does no more than require a party to refrain from

34 Hugh Collins, Regulating Contracts (OUP 1999) 142. 35 Hugh Collins, ‘Is a Relational Contract a Legal Concept?’ in Simon Degeling, James Edelman, and James Goudkamp (eds), Contract in Commercial Law (Thomson Reuters 2016) ch 3. 36 Charles J Goetz and Robert E Scott, ‘Principles of Relational Contracts’ (1981) 69 Virginia Law Review 1089. 37 Mark Amstutz and Gunther Teubner (eds), Networks:  Legal Issues of Multilateral Co-​operation (Hart Publishing 2009); Gunther Teubner, Networks as Connected Contracts (ed Hugh Collins, Hart Publishing 2011). 38 Sheikh Tahnoon Bin Saeed Bin Shakhboot Al Nehayan v Kent (aka John Kent) [2018] EWHC 333 (Comm), [2018] 2 WLUK 528. 39 For discussion of how terms are implied by law, see Elizabeth Peden, ‘Policy Concerns behind Implications in Law’ (2001) 117 Law Quarterly Review 459; Hugh Collins, ‘Implied Terms: The Foundation in Good Faith and Fair Dealing’ (2014) 67 Current Legal Problems 297; Lord Neuberger in Marks & Spencer Plc v BNP Paribas Securities Services Trust Co (Jersey) Ltd [2015] UKSC 72 [2016], AC 742. 40 Paciocco v Australia and New Zealand Banking Group Ltd [2015] FCAFC 50 [288] (Allsop CJ).

The Duty of Loyalty and the Law of Fraud  125 conduct which in the relevant context would be regarded as commercially unacceptable by reasonable and honest people.41

In the context of a commercial contract, in identical terms to those used in employment, the implied term has also been described as a duty not to destroy the mutual trust and confidence that underpins the contract.42 Although the duty to perform a relational contract in good faith does not lay down precise requirements and depends on the context of each transaction to some extent, the duty will typically require disclosure of certain information in particular contexts. One party may hold or obtain information that should be shared in order to make the common purpose of the contract more likely to be successful or to prevent the venture from becoming a failure.43 The duty of disclosure may not be strictly speaking necessary for the contract to be performed to some extent, but the duty is necessary if the project is to prosper in the way the parties reasonably expected. In contracts of employment the principal mechanism for achieving cooperation without specifying the performance obligations completely in the express terms of the contract is to grant one party, the employer, the legal right to issue orders to the other party, the employee, who is required normally to comply with those instructions. Implied terms of the contract supplement that basic structure in ways that impose the kinds of good faith obligations that are required in relational contracts. For employers, the good faith obligation is usually described as the implied term of mutual trust and confidence.44 For employees, the good faith obligation is described variously as a duty of performance in good faith, a duty of fidelity, a duty of cooperation, a duty of loyalty, and also the implied term of mutual trust and confidence.45 Although different nuances may be attributed to these various formulations of the implied terms that arise in contracts of employment, they function in the same way to supplement the express terms for the purpose of supporting the requirement of cooperation in a contract that is incomplete by design.

D.  Obligations of Disclosure in the Contract of Employment We can now return to address the question of whether there is a duty of disclosure in the contract of employment that is likely to satisfy the requirement of the Fraud Act, section 3 of a ‘failure to disclose to another person information which he is under a legal duty to disclose’. As predicted by the theory of relational contracts, there is no general duty of disclosure imposed on the parties to a contract of employment. Unlike a contract of insurance, employment is not a contract uberrimae fidei. Nevertheless, obligations to disclose information can arise under statute, the express terms of the contract, and implied terms of the contract. 41 Sheikh Tahnoon (n 38) [176], referring also to Bristol Groundschool Ltd v Intelligent Data Capture Ltd [2014] EWHC 2145 (Ch), [2014] 7 WLUK 58 [295] and Astor Management AG v Atalaya Mining Plc [2017] EWHC 425 (Comm), [2018] 1 All ER (Comm) 547 [98]. 42 D&G Cars Ltd v Essex Police Authority [2015] EWHC 226 (QB). 43 For example, Yam Seng Pte Ltd v International Trade Corporation Ltd [2013] EWHC 111, [2013] All ER (Comm) 1321 (QB). 44 Douglas Brodie, ‘The Heart of the Matter: Mutual Trust and Confidence’ (1996) 25 Industrial Law Journal 121. 45 Hugh Collins, Keith D Ewing, and Aileen McColgan, Labour Law (CUP 2012) 143.

126  Criminality at Work The general statutory duties of disclosure placed upon employers mostly concern the need to ensure transparency with respect to the terms of the contract and how remuneration has been calculated.46 Such obligations of transparency have occasionally been supplemented by implied terms.47 There are also statutory obligations of disclosure placed on employers in connection with collective bargaining and consultation bodies.48 In addition, workers in particular professions may incur special duties of disclosure in connection with such matters as safeguarding children or regulation of financial markets. Apart from those statutory duties, the courts have not in general imposed broader implied duties of disclosure or duties to provide information during performance of the contract.49 They have treated the contract of employment as much like other commercial contracts, so that both parties need to look after their own interests. Exceptionally, however, duties of disclosure may arise in particular circumstances through two terms implied by law into contracts of employment: the implied term of mutual trust and confidence, and the implied term of the duty of loyalty.50

1.  Mutual Trust and Confidence One way in which mutual trust and confidence can be undermined is by the discovery of fraud or deceit.51 Breach of the duty may also be committed by failures to disclose material information that is regarded as crucial to the continuation of the relationship. The leading case, Mahmoud and Malik v BCCI,52 may be regarded as a breach of the implied term of mutual trust and confidence by concealment and a failure to disclose the corrupt dealings of the bank. Bournemouth University v Buckland53 involved a failure to disclose information: the chair of the examination board had the scripts remarked and failed to consult Professor Buckland about the change. In Visa International Service Association v Paul,54 the employer was held to be in breach of the implied term by failing to notify the applicant, who at that time was on maternity leave, of a vacancy for which the applicant regarded herself as suitable. These examples of breaches of the implied term of mutual trust and confidence reveal that it does impose duties on employers to disclose information, though only when the omission is serious and is therefore likely to lead to a breakdown in trust and confidence. It can provide the basis for an obligation of disclosure that would satisfy section 3 of the Fraud Act 2006. In principle, the implied term of mutual trust and confidence applies to both parties, so in theory it might be used to impose duties of disclosure on employees. In practice,

46 Employment Rights Act 1996, ss 1, 8, 13. 47 Scally v Southern Health and Social Services Board [1992] 1 AC 294 (HL). 48 Trade Union and Labour Relations (Consolidation) Act 1992, s 181; Information and Consultation of Employees Regulations 2004, SI 2004/​3426. 49 Reid v Rush & Tompkins Group Plc [1990] ICR 61 (CA); Crossley v Faithful & Gould Holdings Ltd [2004] EWCA Civ 293, [2004] ICR 1615. 50 For present purposes, it is assumed that these implied terms are different, though the implied obligations are certainly not unrelated to each other. 51 For example, Post Office v Roberts [1980] IRLR 347 (EAT). 52 [1998] AC 20 (HL). 53 [2010] EWCA Civ 121, [2010] ICR 908 (CA). 54 [2004] IRLR 42 (EAT).

The Duty of Loyalty and the Law of Fraud  127 however, another implied term, the duty of loyalty or fidelity, has been used for centuries to govern the scope of duties of disclosure imposed on employees.

2.  The Duty of Loyalty or Fidelity Under the common law, the duty of loyalty owed by an employee to an employer arises through a term implied by law. The duty is also described as a duty of fidelity and of performance in good faith. The employee’s duty of loyalty requires both performance of the job in a way that serves the interests of the employer and prohibits conduct outside work that may have an adverse impact on the employer’s business interests including its reputation. Employees who denigrate their employer’s business or products are likely to be justifiably dismissed for misconduct, because it is a breach of the implied duty of loyalty. Whistleblowers usually break the implied duty of loyalty when in support of the public interest they report an employer’s misconduct to an outside body such as a regulator or a newspaper, so they require a special exemption for their protection.55 The implied term also requires employees to respect information acquired in confidence, though this duty also arises in tort on the receipt of confidential information.56 Breach of the implied term by the employee is a ground for an employer to justify summary dismissal and to avoid paying damages for wrongful dismissal at common law. Breach of the term can also provide the basis for a claim for an injunction or for compensatory damages. Through express terms of the contract, employers may impose more onerous legal duties of disclosure, confidentiality, and loyalty. To what extent does the implied duty of loyalty impose a duty on employees to disclose information to their employer? As a relational contract, the contract of employment is only likely to require disclosure when it is required as part of the cooperative endeavour to make the performance of the contract successful or at least not to prevent its success. A duty of disclosure seems likely to arise when the employee possesses information that if disclosed would prevent a loss to the employer, provided that such a disclosure would not harm the employee’s own interests under the contract. The courts have held that a duty to disclose information arises where an employee knows about the wrongdoing of other employees. For instance, an employee should tell the employer that a fellow employee is stealing from the business.57 Similarly, an employee should tell the employer that fellow employees are secretly setting up a rival business to the employer’s business in breach of their obligations of loyalty.58 These cases imposing a duty of disclosure concern harm to the employer’s business caused by other employees, but do not require, according to Bell v Lever Bros Ltd,59 the employee to disclose his or her own misconduct. As Lord Atkin said in that case, The servant owes a duty not to steal, but, having stolen, is there superadded a duty to confess that he has stolen? I am satisfied that to imply such a duty would be a departure from 55 Employment Rights Act 1996, pt IVA. 56 Campbell v MGN Ltd [2004] UKHL 22, [2004] 2 AC 457 [14]. 57 Bell v Lever Bros Ltd [1932] AC 161 (HL). 58 Swain v West (Butchers) Ltd [1936] 1 All ER 224, [1936] 3 All ER 261 (CA); Sybron Corp v Rochem [1984] Ch 112 (CA). 59 [1932] AC 161 (HL).

128  Criminality at Work the well established usage of mankind and would be to create obligations entirely outside the normal contemplation of the parties concerned.

This distinction fits the model of the contract of employment as a relational contract well. Although employees can look after their own interests by not confessing their own misconduct and mistakes, their duty of loyalty requires them to report to the employer the misconduct of other employees that is likely to harm the interests of the employer’s business and thus undermine the success of the employment relationship. But the distinction can lead to some surprising results. Consider whether a worker must disclose to his or her employer that he or she took industrial action on a particular day and that, therefore, following Miles v Wakefield,60 the employer is entitled to reduce wages on a pro-​rata basis? On the assumption that there is no express term of the contract of employment that imposed such an obligation to disclose participation in strike action, is there an implied obligation that requires an employee to volunteer the fact and extent of participation in industrial action? If it is correct that there is no additional duty to report one’s own misconduct to the employer, assuming strike action is misconduct, disclosure of one’s own strike action is not required by the implied duty of loyalty. On the other hand, there may be a duty to report that other employees have taken industrial action and a failure to report that misconduct would appear to be a breach of the duty of loyalty. Given that strike action is hard to take on one’s own, it seems that a duty of disclosure will arise indirectly in every case of industrial action. Thus, subject to the issue of dishonesty, a failure to inform an employer that others are on strike may satisfy the main ingredient of the Fraud Act, section 3, whereas there appears to be no duty to admit to one’s own participation in industrial action. Although it is no longer in general a crime for workers to take industrial action, the surprising result of this analysis is that it is possibly a crime of fraud for employees not to inform their employer that their co-​ workers have taken industrial action. Juries may not agree, however, that a failure to snitch on one’s co-​workers amounts to dishonesty. The suggestion that the implied duty of loyalty imposes a duty to betray co-​workers but not confess one’s own misconduct is inevitably speculative in the absence of any closely relevant legal authorities and clear principles. The extent of the civil obligation appears to depend in Lord Atkin’s judgment on the vague standard of the ‘well established usage of mankind’. That phrase tells us that the scope of the duty of disclosure is determined ultimately on customary morality or orthodox views about right and wrong. Yet those moral views are often ambiguous and contested, especially in examples of white-​collar crime such as the presentation of unduly favourable accounts, where the victims are potentially a large and undetermined class of investors or consumers.61 Furthermore, if the ‘well established usage of mankind’ is the linchpin, judgments involved in determining the incidence of employee’s duties of disclosure under the implied duty of loyalty seem to be analogous to the issue raised in the test of dishonesty under the criminal law of what ordinary decent people would think. If so, an appeal to customary moral standards and orthodox views both serves to determine the scope of the actus reus of the offence of fraud by omission under section 3 of the Fraud Act 2006 by identifying when there is a duty of disclosure under



60 61

Miles v Wakefield Metropolitan Borough Council [1987] AC 539 (HL). Stuart F Green, Lying, Cheating, and Stealing: A Moral Theory of White-​Collar Crime (OUP 2006).

The Duty of Loyalty and the Law of Fraud  129 the contract of employment and also to determine whether the employee satisfies the fault element of dishonesty for a criminal offence.

E.  Duties to Safeguard the Financial Interests of the Other Party The second question raised above about the civil law governing the contract of employment is whether there is a duty placed on either the employer or the employee to safeguard the interests of the other party. If such a duty exists, it seems likely that the duty would normally be satisfied by giving a warning to the other party about a risk to their financial interests. Suppose, for instance, that during performance of the job an employee realizes that an external contractor is about to become insolvent, which creates a risk of loss for his or her employer. The question is whether the employee is under a duty to warn the employer about this risk of loss? In other cases, perhaps action on the part of the employee might be required, such as locking a door or blocking an account. Although these duties may not be onerous, in ordinary commercial transactions one party would not be required to take such steps to safeguard the financial interests of the other party unless that was the purpose of the contract. The suggestion that employers owe such a duty to safeguard the financial interests of employees has so far been rejected. In Crossley v Faithful & Gould Holdings Ltd62 the claimant suffered a nervous breakdown and after a period of sick leave agreed to take early retirement. Under the terms of the employer’s long-​term disability insurance scheme, the claimant was entitled to benefits provided he remained in employment. His entitlement therefore ceased on taking retirement. Mr Crossley claimed that, by failing to alert him to this consequence of taking early retirement, the employer had breached an implied term of the contract of employment. The Court of Appeal rejected the claim, holding that there is no standard implied term in all contracts of employment that an employer will take reasonable care of an employee’s well-​being, nor an implied duty for an employer to give an employee financial advice in relation to benefits accruing from his employment. If the employer voluntarily assumes the responsibility to give advice and to help the employee, however, the employer may be liable in negligence for failing to provide careful advice.63 If there is no general duty placed on employers to safeguard the financial interests of employees, does the law place such a duty on employees? The answer to that question depends on the scope of the implied duty of loyalty or fidelity. The origins and development of this implied duty in the contract of employment are contested.64 Older cases do not make sharp distinctions between types of employee, such as managers and servants, and between agents and employees. Nor is it always clear in the past whether courts were considering implied terms in the contract or the imposition of mandatory equitable duties akin to fiduciary duties. The use of language such as ‘fidelity’ seems to confuse the question of whether the discussion concerns a contract term or the imposition of an equitable duty. It now seems clear, however, that the implied duty of loyalty or fidelity 62 [2004] EWCA Civ 293, [2004] ICR 1615 (CA), following University of Nottingham v Eyett [1999] ICR 721 (ChD). 63 Lennon v Commissioner of Police of the Metropolis [2004] EWCA Civ 130, [2004] IRLR 385 (CA). 64 Robert Flannigan, ‘The [Fiduciary] Duty of Fidelity’ (2008) 124 Law Quarterly Review 274; Andrew Frazer, ‘The Employee’s Contractual Duty of Fidelity’ (2015) 131 Law Quarterly Review 53.

130  Criminality at Work in the contract of employment needs to be distinguished sharply from a fiduciary relationship. Unlike for instance the creation of the position of a trustee under a trust or an appointment as a director of a company, the contract of employment does not make the employee a fiduciary.65 If it did, an employee would always have to place the interests of the employer above his or her own and avoid getting into situations of conflict of interest. The duties under a relational contract are not to put the interests of the other party first, as would be the case in a fiduciary relationship, but rather to be loyal to the purpose of the transaction and not take action that would prevent its purpose from being achieved. The purpose of the contract does include the provision of worthwhile pay-​offs for both parties. Both parties can have regard to their own interests in determining how they will perform the contract. What they cannot do under the duty of loyalty is to subvert the purpose of the contract itself. In other words, there is not a duty placed on the employee to safeguard the financial interests of the employer, but rather one to refrain from subverting the purpose of the contract in a way that harms the interests of the employer. Nevertheless, it is true that sometimes employees incur fiduciary obligations or at least fiduciary-​like obligations.66 How can they have fiduciary obligations if they are not fiduciaries? Apart from well-​established fiduciary relationships such as trusts, the law sometimes imposes fiduciary obligations in order to provide a more effective civil remedy for breach of an obligation such as a breach of contract. As Matthew Conaglen explains, the purpose of the imposition of fiduciary duties or remedies in such case is as a mechanism to provide greater reassurance that non-​fiduciary duties will be performed.67 In other words, in order to ensure that the employee does in fact perform the contract loyally or faithfully, equity may insert a fiduciary obligation or equitable remedy on top of any remedy for damages for breach of contract. These equitable remedies will usually be either an injunction, such as a ‘springboard injunction’ against the unauthorized use of confidential information,68 or a claim to account for profits made by misuse of the employee’s position. The remedy of an account of profits ensures that there is no financial incentive for the breach of duty of loyalty even where the employer has suffered no loss. A-​G v Blake illustrates a particularly dramatic instance of such a remedy, when the UK government succeeded in a claim for all the royalties from the memoirs of Blake, who had been a traitorous secret agent, on the ground that he had broken his express confidentiality agreement (both by publishing the book and of course spying for the Russians).69 In order to combat the temptation of misuse of the employer’s property or money, the law imposes a fiduciary duty on the employee with the effect that the employee has a duty to account for the property and any secret profits. For example, where the employee controls the employer’s property or money as part of the job, the implied duty of loyalty or fidelity requires the employee to use that property or money for the employer’s purposes and not for the employee’s own benefit. The use of another’s property including intellectual property without permission in order to

65 University of Nottingham v Fishel [2000] ICR 1462, 1491 (Elias J); Marathon Asset Management v Seddon [2017] EWHC 300 (Comm), [2017] ICR 791 [170]. 66 Douglas Brodie, ‘The Employment Relationship and Fiduciary Obligations’ (2012) 16 The Edinburgh Law Review 198. 67 Matthew Conaglen, Fiduciary Loyalty:  Protecting the Due Performance of Non-​Fiduciary Duties (Hart Publishing 2010) 68 Roger Bullivant Ltd v Ellis [1987] ICR 464 (CA). 69 [2000] UKHL 45, [2001] 1 AC 268 (HL).

The Duty of Loyalty and the Law of Fraud  131 obtain a personal benefit is likely to result in an order to pay for the use of it at the ordinary market rate,70 or to account for the profits made from its unauthorized use. This use of equitable remedies to add to the deterrence of remedies for breach of contract arises in many employment contexts. The duty of loyalty requires an employee not to set up a business in competition with the employer or in a way that subverts the business interests of the employer. In a recent example, Samsung Semiconductor Europe v Docherty,71 the employee set up a business that supplied services to his employer, and was then in a position in his employment to ensure that his contracting business received favourable contracts. For that breach of loyalty, the employee was required to account for all the payments received by contracting business from his employer. Such a case does not demonstrate that an employee is a fiduciary. Rather it shows that in some instances, to ensure an effective remedy for breach of the contractual duty of loyalty, the courts are prepared to impose the equitable remedy that would have been applied had there been a breach of a fiduciary duty. This analysis of the duty of loyalty casts light on the scope of section 4 of the Fraud Act 2006 in its application to employment. As a relational contract, it is not generally true that the contract of employment places the parties under a contractual or equitable fiduciary duty to safeguard the financial interests of the other. In so far as the Law Commission suggested to the contrary by saying that section 4 applied to employment in general, that contention seems incorrect. With regard to employers, the civil law does not recognize a general duty of loyalty or an expectation that employers should safeguard the financial position of employees. With regard to employees, under the contractual duty of loyalty or fidelity, there is similarly no duty to safeguard the financial interests of the employer in general. Instead, under the logic of relational contracts, an employee is in a position where there is an expectation that the employee should not harm the interests of the employer whenever that harmful conduct will fundamentally undermine the purpose of the employment relation rather than simply benefit the employee. Although the Law Commission rejected the view that the presence of fiduciary duties should be essential under section 4, a more nuanced view may be more helpful. It might be argued that the kind of position required under section 4 is only likely to exist when the civil law provides equitable remedies in addition to the ordinary remedy of damages for breach of contract. We have noted that in particular contexts, employees may be subject to equitable remedies such as the action to account for secret profits that are available to remedy particular instances of breach of the contractual duty of loyalty. Where such equitable remedies are available for breach of the contractual duty of loyalty, there is a stronger case for holding that the condition of an expectation of safeguarding position set in section 4(1)(a) of the Fraud Act 2006 has been met. In the absence of such an equitable remedy, the civil law merely provides compensation for losses, a liability that can arise without any kind of special position or relationship. If that analysis of the civil law and its application to section 4 of the Fraud Act 2006 is correct, the most common instances where criminal fraud may take place under section 4 of the Fraud Act 2006 are likely to involve employees who are taking steps to set up a rival business or are using their employer’s resources to make a secret profit for themselves. They may breach their duty of loyalty under the contract of employment by taking confidential

70 71

Strand Electric and Engineering Co Ltd v Brisford Entertainments Ltd [1952] 2 QB 246 (CA). Samsung Semiconductor Europe v Docherty [2011] CSOH 32, 2011 SLT 806.

132  Criminality at Work information from the employer and passing it to others or by diverting potential contracts to their own business. In such cases of breach of the duty of loyalty, it seems possible to argue that there has been an abuse of a position by harming the financial interests of the employer. The employee’s breach of the implied duty of loyalty would likely suffice for a finding of dishonesty.

F.  Over-​criminalization? The final question is whether the Fraud Act 2006 unnecessarily or disproportionately brings to bear the criminal law to sanction what are already breaches of contract that give rise to liability under the civil law of the contract of employment. Are the additional criminal sanctions provided in sections 3 and 4 of the Fraud Act 2006 necessary and proportionate? The courts have been alert to the need to develop equitable remedies that deprive employees of any gains resulting from their breach of duties of loyalty and fidelity. Indeed, A-​G v Blake may be an instance, when properly analysed, of where there was no breach of contract because the information was no longer confidential, yet the court imposed the duty to account in any case to strip the traitor of the fruits of his labours. Where an employee has made a secret profit by using the employer’s equipment, for instance, not only is the employee liable for damages for breach of contract but also the employee must account for any profits made through the use of the employer’s property. If this claim to account is successful, it removes any benefit that may have arisen from the breach of contract. As a deterrent, the action to account for profits or to pay a fair price for use of property is particularly useful where the employer may have suffered no loss from the unauthorized use of the property by the employee. Given that the civil law has already provided this super-​remedy that involves the stripping of any gain as well as compensation for any loss, why is it necessary to add to the deterrence by making this conduct a criminal offence as well? One common justification for the imposition of a criminal offence is to prevent serious harm. Yet the civil law can use damages to compensate for any harm in the context of employment. In fact, in many of these cases, there is in fact no harm done to the employer at all. Indeed, the employer in a sense may obtain a windfall from any equitable remedy, because the employee has produced wealth from a resource that the employer was not using at that time, which the employer then acquires through the action to account for secret profits. Why is it necessary therefore to criminalize such activities as well? Far from there being serious harm that is not being deterred by the civil law, the civil remedies here can almost be described as extravagant. Furthermore, we should remember that the economy benefits in general from former employees starting up their own businesses, including those in competition with their former employers. The criminalization of actions of employees who take short-​cuts in the process of setting up a competitive business may also deter those who play by the rules of competition and restraint of trade as well. It is unclear, therefore, that the criminalization of breaches of the duty of loyalty can be justified by the need to deter serious harm. Is there another justification for the use of the criminal law in this context? Because an employer is not under the same duty of loyalty, employers are much less likely than employees to fall within sections 3 and 4 of the Fraud Act 2006. Though partly modified by the development of the implied term of mutual trust and confidence, the asymmetry of the

The Duty of Loyalty and the Law of Fraud  133 duty of loyalty harks back to the law of master and servant, in which the master lawfully exercised prerogative powers and the servant was utterly subordinate. As a member of the household, disloyalty by a servant was regarded as a betrayal of trust. A disloyal servant is a cancer in the heart of the home. Infidelity of a servant was more than a breach of contract; it was regarded as a grave moral wrong and a subversive threat to the social order. In the context of employment, the Fraud Act 2006 may unconsciously be responding to that tradition by seeking to criminalize at least some breaches of the contractual duty of loyalty because they are regarded as serious moral wrongs that deserve not only civil sanctions but criminal ones as well. If that is correct, embracing a moral attitude that harks back to a bygone age, criminal law is being used here to express a moral condemnation of disloyal servants.

7

Wage Theft as a Legal Concept Sarah Green*

A.  Introduction This chapter will address the question of whether any of the poor employer payment practices commonly and colloquially referred to as wage theft amount in substance to the current statutory offence of theft, as defined by the Theft Act 1968. Since this offence requires that the defendant appropriate property belonging to another dishonestly and with the intention1 permanently to deprive the other of that property, it is not immediately clear that the practices referred to above would fit this definition. The analysis here first considers, therefore, whether the current offence of theft should be modified so as to accommodate those practices so often given the label of wage theft. Since the challenges of this are not inconsiderable, given the pedigree of the concepts already in use, the analysis concludes by reflecting on whether the current statutory offence of fraud by abuse of position is in fact already an ideal fit for such wrongful employment practices, suggesting that perhaps ‘wage fraud’ might be a more appropriate label to apply.2

B.  What is ‘Wage Theft’? In general public discourse, the label ‘wage theft’ has been applied to various unscrupulous employer practices. Whilst all of these are quite varied in terms of their conduct and execution, their consequences for the affected employees are similar: each deprives the victims of what is lawfully due to them as remuneration for their labour. A recent empirical investigation, funded by Trust for London, has identified several common employer practices, which currently attract the label of wage theft: Key Informant interviews and our examination of Employment Tribunal judgments suggest that unpaid wages may present in several forms simultaneously (combinations of deductions from wages, unpaid overtime, withholding holiday pay, failing to pay statutory * I would like to thank Alan Bogg, Bob Sullivan, Nicholas J McBride, and Paul S Davies for the comments and insights they were willing to share, all of which contributed to the best parts of this chapter. Any errors or oversights remain my own. 1 Since many employers will be corporate in form, the extension of any crime of mens rea to them will inevitably give rise to questions of attribution. This is not an issue which can be accommodated within the parameters of the current discussion. 2 For the significance of fair labelling, see, inter alia, Glanville Williams, ‘Conviction and Fair Labelling’ (1983) 42 Current Legal Problems 85 and James Chalmers and Fiona Leverick, ‘Fair Labelling in Criminal Law’ (2008) 71 Modern Law Review 217. Sarah Green, Wage Theft as a Legal Concept In: Criminality at Work. Edited by Alan Bogg, Jennifer Collins, Mark Freedland QC, and Jonathan Herring, Oxford University Press (2020). © Alan Bogg, Jennifer Collins, Mark Freedland, and Jonathan Herring. DOI: 10.1093/oso/9780198836995.003.007

Wage Theft as a Legal Concept  135 pay, or paying allegedly ‘self-​employed’ workers below the NMW), and be accompanied by other abuses of workers’ rights such as failure to provide payslips or written terms of employment.3

There are, therefore, several forms of employer activity that fall within the remit of the current discussion. Perhaps the most significant of these is the failure to pay monies due either under the contract of employment or under statute (such as the National Minimum Wage Act 1998 and the Equality Act 2010). Another example is the practice of making wage deductions in a way which defeats compliance with statutory requirements, such as using employees’ pay to cover uniform or equipment expenses. In addition to these activities, employers have also been said to commit wage theft by failing to make the requisite pension contributions on an employee’s behalf and by not paying holiday pay which is lawfully due.4 The same term has also been used to describe less direct means of reducing employee pay receipts, such as abuse of tipping, and the mislabelling of employees as being self-​employed, a practice which both denies them the benefit of certain statutory protections, and saves the employer resources. There is little doubt, therefore, that all of these employer activities (and this is not an exhaustive list) are both wrongful and deserving of legal censure. The question remains, however, whether the criminal offence of theft is really the most appropriate vehicle for dealing with these forms of behaviour. Whilst, particularly for non-​lawyers, theft is perhaps the most obvious offence to associate with the taking by one party of something which intuitively belongs to another, there are several technical reasons why such an association is not necessarily an apt one.

C.  Is ‘Wage Theft’ Theft? Section 1 Theft Act 1968 contains the definition of theft: 1. Basic definition of theft. (1) A person is guilty of theft if he dishonestly appropriates property belonging to another with the intention of permanently depriving the other of it; and ‘thief ’ and ‘steal’ shall be construed accordingly.

The definition of what amounts to ‘property’ for this purpose is contained in section 5: ‘Property’ includes any money and all other property, real or personal, including things in action and other intangible property.5

3 Nick Clark and Eva Herman, ‘Unpaid Britain:  Wage Default in the British Labour Market’ (Middlesex University November 2017) 10 accessed 2 August 2019. 4 See David Metcalf, United Kingdom Labour Market Enforcement Strategy 2018/​19 (HM Government May 2018) 15 accessed 2 August  2019. 5 Theft Act 1968, s 4(1).

136  Criminality at Work Thus, theft has a distinctive shape. It is concerned with those situations in which victims are ousted from a pre-​existing relationship with assets over which they have established rights. There are two interdependent issues, therefore, which arise when attempting to apply this offence to the wrongful payment practices identified above. The first arises from that requirement for a pre-​existing relationship between an employee and the funds in the possession of the employer which, whilst due to the employee according to legal rules, have not, at or before the time of the employer’s wrongful action, been transferred to her. The second problem lies in theft’s requirement of appropriation: a requirement only satisfied when a defendant assumes the rights already held by the victim over the property in question. The first question is therefore whether monies due to be paid over under a contract and/​ or a statutory requirement (such as the National Minimum Wage Act or the Equality Act), could be said to be the property of the employee. Such a classification would be necessary before they could be said to ‘belong to another’6 and before the employer could be deemed to have appropriated them. It might, however, be that the monies claimed are not capable of amounting to property for legal purposes: the law does not always equate what someone is owed with something she owns. In other words, there is an important forensic difference between saying ‘that is mine; give it to me’ and ‘I performed services for you; pay me what they are worth’. In some cases, therefore, it is not even clear that the employer has appropriated anything which could be said to be the ‘property’ of the victim. Second, even where what is claimed by the employee is recognized by the law as the proper subject matter of property rights (and, as a consequence of the Theft Act, section 4, the criminal law category is more accommodating than that of the civil law),7 this will not always be sufficient. The appropriation element of the offence means that the temporal order of events will be crucial in distinguishing between those situations which might amount to theft and those which do not. In other words, the questions that will need to be answered are whether the subject matter of the victim’s property rights existed, and whether the victim had the relevant interest in it, before the employer performed any action which could amount to appropriation of that subject matter.

D.  Is What the Employee Has Lost ‘property’ for  the Purposes of the Theft Act 1968? There are two remedies available in principle to an unpaid employee:  an action for the agreed sum, which is a debt claim; or an action in damages. The former, being a claim for the enforcement of the primary obligation of the employment contract and not, therefore, subject to the rules on mitigation, penalties, or remoteness,8 is preferable from the employee’s point of view:

6 As required by s 1(1). 7 See Sarah Green, ‘Theft and Conversion:  Tangibly Different?’ (2012) 128 Law Quarterly Review 564, 565: ‘ “Property” includes money and all other property, real or personal, including things in action and other intangible property’ (Theft Act 1968, s 4(1)). ‘ “Goods” includes all chattels personal other than things in action and money’ (Torts (Interference with Goods) Act 1977, s 14(1)). 8 Jervis v Harris [1996] Ch 195. It is also both procedurally and substantively easier for debt claimants to obtain summary judgment: Civil Procedure Rules Part 24.

Wage Theft as a Legal Concept  137 Debt is perhaps the oldest and arguably the most important example of a thing in action. A debt is an obligation that consists of a duty to pay a specified sum of money, either on demand or at a future date that may either be fixed or dated according to an event that is sure to happen. A contingent obligation to pay a sum of money is not a debt. To be a debt, a money obligation must have accrued due though it may not yet be payable. A debt will not accrue due if it arises only in consequence of work to be performed and that work has not yet been carried out. Nevertheless, apart from the classification of an entitlement to be paid in the future as a debt, the extension of the category of things in action to accommodate contractual rights has this consequence. A right to be paid under a contract for work to be done in the future under that contract, though not a debt, is a thing in action and therefore assignable. This principle, nevertheless, will not apply in bankruptcy or insolvent liquidation where the performance for which the payment is to be made has not yet been rendered at the date of the insolvency proceedings. A debt may be either legal or equitable. An action for damages is not a debt unless and until judgment is signed for the claimant.9

The question of which is available in any given case is determined by the terms of the contract and the facts of any given situation: where a specific sum is due, the primary remedy for debt is available, but where such remuneration has not been specified, or has not yet become due, the employee is generally restricted to a claim in damages, and an enforcement of the employer’s secondary obligation only. An action for damages is distinct from a debt, and does not become a debt unless and until judgment is signed for the claimant.10 There is no doubt that a debt, as a classic example of a chose in action, would come within the category of property capable of being stolen for the purposes of the Theft Act. A damages claim is also a chose in action, and therefore potentially included within the section 4 definition in its own right. This does not, however, mean that a debt claim and a damages claim would have the same implications for any potential property offence. It is the nature of a debt claim that the subject matter of the victim’s property rights exists prior to the wrongful refusal to pay it. Incurring a debt (whilst not of course wrongful eo ipso) does generate property rights in the creditor, which then form the basis of any subsequent action for failure to pay that debt. Thus, in the employment context, the recognition of a debt gives the disappointed employee a pre-​existing right, of which the employer’s wrongful payment practices can deprive her. A damages claim on the other hand does not, by definition, arise as a phenomenon until after the relevant activity of the employer has taken place. Even then, the claim is not something of which the employer’s activity can deprive the victim: on the contrary, the employer’s activity is what generates the claim in the first place. For the purposes of the Theft Act, ‘appropriation’ is: Any assumption by a person of the rights of an owner . . . and this includes where he has come by the property (innocently or not) without stealing it, any later assumption of a right to it by keeping or dealing with it as owner.11



9

Michael Bridge and others (eds), The Law of Personal Property (2nd edn, Sweet & Maxwell 2017) [4-​008]. Ex p Charles (1811) 14 East 197, 104 ER 576; Jones v Thompson (1858) E B & E 63, 120 ER 430. 11 Theft Act 1968, s 3(1). 10

138  Criminality at Work It is, therefore, impossible for the employer to appropriate the chose in action which is the damages claim. This is a highly significant distinction, and one which makes a debt claim, but not a damages claim, susceptible in principle to theft. Since both claims are based on the same harm to the employee, and since assigning them to one or other category can, as we will see, often be a matter of technical nuance, allowing the distinction to have such a potent practical effect on any criminal liability would seem undesirable at the very least. The underlying wrong is the same in both instances, yet the criminal law’s approach diverges without any obvious justification. The limits of this analysis have recently been tested before the Employment Appeal Tribunal (EAT). Under the Employment Rights Act 1996 (ERA), the Secretary of State has an obligation to pay any debts owed by employers to employees constituted by ‘arrears of pay’ in the event of the former’s insolvency. It provides as follows: If, on an application made to him in writing by an employee, the Secretary of State is satisfied that: (a) the employee’s employer has become insolvent, (b) the employee’s employment has been terminated, and (c) on the appropriate date the employee was entitled to be paid the whole or part of any debt to which this part applies, the Secretary of State shall, subject to section 186, pay the employee out of the National Insurance Fund the amount to which, in the opinion of the Secretary of State, the employee is entitled in respect of the debt.

To be granted a payment from the National Insurance Fund, the employee must be entitled to payment of the debt on ‘the appropriate date’. The appropriate date is defined by section 185 of the ERA and varies according to the nature of the debt claimed. For the purposes of arrears of pay, the appropriate date is the date on which the employer became insolvent. In Graysons Restaurants Ltd v Miss C Jones and others v Secretary of State for Business, Energy and Industrial Strategy,12 the EAT considered whether equal pay arrears, claimed under section 1 of the Equal Pay Act 1970, (now section 66 Equality Act 2010) but not yet determined or quantified, could count as arrears of pay, and so as a debt, for the purpose of section 182: I accept that an equal pay claim may result in a damages award as opposed to an ‘arrears of pay’ award. However, that is not to say that an employment tribunal simply has a choice to label what is clearly an award of arrears of pay as one of damages. The power to award damages is there to reflect a different sort of equal pay claim. The Equal Pay Act 1970 and the Equality Act 2010 are not concerned solely with ‘pay’ but with contractual terms. The protection applies to benefits in kind, which must, if there is to be a successful remedy, be converted to a monetary value. A successful claim of this type is likely to result in an award of damages, not an award of ‘arrears of pay’. It seems to me however, that as a matter of contractual analysis, rather than election, these proceedings are not concerned with damages claims but with arrears claims.



12

2017 WL 05892028.

Wage Theft as a Legal Concept  139 In any event, as already indicated section 184 does not require the amount for which the Secretary of State is liable to be classed as a ‘debt’ at common law. For example, the provision in section 184(1)(b) for ‘any amount which the employer is liable to pay the employee . . . for any failure of the employer to give the period of notice required by section 86(1)’ cannot necessarily be regarded as a ‘debt’ on common law principles, and may properly be seen as a damages claim. The Secretary of State is however expressly made liable to pay it in cases of insolvency under Part XII, and it is treated as a debt for these purposes. The focus is not on whether the claim is a ‘debt’ at common law but whether it is one for ‘arrears of pay’. If it is a claim for arrears of pay it falls within one of the categories of debt exhaustively defined by section 184(1) of the ERA and therefore is a debt claim within the meaning of section 182. It does not matter that there may also be the facility within the Equality Act and/​or the Equal Pay Act for a tribunal to make an award of damages in appropriate circumstances.13

This judgment demonstrates the extent to which the classification of any given claim, in terms of debt or damages, is dependent upon the specific factual context in which it happens to arise. This has long been true of the common law, and is a phenomenon entrenched by the statutory ‘attribution’ of the label of debt to claims of a certain nature, regardless of whether they would be regarded as debts by virtue of common law analysis. In Graysons itself, for example, the claimants had already performed work for their employer before the date of insolvency, and their work had been rated as equivalent to work done by their male comparators, although they had been paid less for doing it. Whilst those equal pay claims had yet to be formally quantified by the time of the Graysons hearing, the relevant work had been carried out, comparators applied, and the existence of a shortfall thereby identified. Since all that remained was for the calculations to be carried out (the results of which were already pre-​determined, as it were, albeit not yet known), the EAT deemed the equality pay which would in due course be quantified by those calculations to be ‘arrears of pay’ and therefore debts for the purposes of the ERA. By contrast, the right to be paid under a contract for work to be done in the future, whilst being a chose in action and assignable,14 does not amount to a debt, and remains the subject of an action in damages. Such a claim for damages would amount to ‘property’ susceptible to appropriation for the purposes of the Theft Act 1968, as in fact would a claim under the Equality Act for benefits in kind. As established above, however, the damages claim is one that is created, rather than appropriated, by the wrongful practices of employers, and so is not susceptible to the offence of theft. The precise location of the employer’s wrongful behaviour in this context seems not to align with the form of the current statutory offence of theft. In order for the chronological structure of poor employer payment practices to fit the framework for theft, the employee’s proprietary interest in the monies lawfully due would need to arise prior to those actions of the employer which amounted to the purported appropriation. The only way, therefore, in which the factual matrix of poor payment practices can be made to fit the theft model would be for the law to recognize an anterior property right of the employee in the employer’s funds, so that the latter’s failure to pay lawful wages would amount to an appropriation of those funds.

13

14

2017 WL 05892028 [37]–​[39]. Brice v Bannister (1878) 3 QBD 569.

140  Criminality at Work Such an anterior property right would exist were the unpaid monies in the hands of the employer to be subject to a trust in which the employee or employees had a beneficial interest. This proprietary right would clearly then precede the employer’s failure to transfer that money to the employee, thus potentially constituting an appropriation sufficient for the offence of theft.15 There are, however, a number of conceptual obstacles to such an analysis. A constructive trust would need, for English courts at least, to be institutional rather than remedial in nature: In England, the constructive trust has generally been regarded as a substantive institution rather than a remedy. The distinction between these two explanations of the trust is uncertain, owing to a lack of clear definition of the terms used. The commonly accepted distinction is that an institutional constructive trust arises independently of any court order, once the facts on which the creation of the trust depends have occurred. The function of the court is merely to declare its prior existence. A constructive trust would be remedial if it depended on an order of the court for its creation.16

This means that the proprietary right of the employee in the funds would need to have arisen by operation of law and not depend for its existence on discretionary imposition by the court. The recognition of an institutional constructive trust would require an employer, on receiving any money which might be earmarked as owing to an employee, to be regarded as retaining that money unconscionably, or as having received it as an agent of the employee in breach of fiduciary duty.17 The problem with the first of these is again temporal. In order for the employee’s proprietary right to precede any purported appropriation, the employer’s unconscionable retention would also need to occur in advance of that appropriation. Presuming the employer’s failure lawfully to pay her employees amounts to the appropriation, she would have to be deemed to have unconscionably retained the money she should so have paid before she failed to pay it over. In other words, she would need to be regarded as wrongfully keeping the money lawfully owed to her employees before it became due. This is hardly a tenable account of the situation. An alternative potential analysis, which does not come up against the unconscionability problem, would be to make an analogy with the vendor–​purchaser constructive trust, which arises on the conclusion of a valid, specifically enforceable sales contract,18 at which point the vendor becomes a trustee of the property in which the purchaser acquires a beneficial interest: [T]‌he moment you have a valid contract for sale the vendor becomes in equity a trustee for the purchaser of the estate sold, and the beneficial ownership passes to the purchaser, the vendor having a right to the purchase-​money, a charge or lien on the estate for the

15 Although, even then, there might remain an issue of lack of specificity where the employee’s claim is one for damages rather than for a defined sum. 16 John McGhee (ed), Snell’s Equity (33rd edn, Sweet & Maxwell 2014) [26-​014]. 17 Although there are other wrongs which can lead to the recognition of an institutional constructive trust, none of these would be applicable to the factual framework of poor employment payment practices. See Snell’s Equity (n 16) ch 26. 18 Applicable to all contracts for sale of land, but also applicable to contracts for the sale of personality, if specifically enforceable (such as the sale of shares in a private company). See Jacob Meagher, ‘(Re-​defining) the Trust of the Specifically Enforceable Contract of Sale–​–​the Vendor Purchaser Constructive Trust’ (2018) 24 Trusts & Trustees 266, 266.

Wage Theft as a Legal Concept  141 security of that purchase-​money, and a right to retain possession of the estate until the purchase-​money is paid, in the absence of express contract as to the time of delivering possession.19

There is no requirement of unconscionability here,20 just the formation of a contract with specific characteristics. Were a similar analysis to be applied to the employment context, the employer would, on the making of an employment contract, be deemed to become a trustee of any monies earmarked and segregated for the employee’s remuneration, and the employee would gain a beneficial interest in the same. Not only is such earmarking an unusual practice for employers to adopt, but this form of constructive trust will be recognized only ‘if and so far as’,21 and ‘if and so long as’,22 ‘a Court of Equity would under all the circumstances of the case grant specific performance of the contract’23.24 The obvious challenge to the making of any such analogy would be the fundamental and significant difference between contracts of employment and contracts for the sale of land. It is difficult to imagine two types of contract further removed from one another, particularly in remedial terms, than these two.25 Section 236 of the Trade Union and Labour Relations (Consolidation) Act 1992 states ‘no court shall issue an order compelling an employee to do any work or attend any place for the doing of any work’. 26 Whilst the position with regard to the specific performance of employers’ duties is not statutory, nor quite so definitive, the instances in which such a remedy has been granted are exceptional and uncommon.27 Specifically, in relation to the obligation of an employer to remunerate her employees, Lord Sumption has said: The traditional insistence of the courts that contracts of employment are not specifically enforceable has not, at least in the last half century, been dogmatically applied to every obligation under such contracts. It is important to distinguish between the core obligations which provided the original rationale of the rule, and what for want of a better word I will call collateral obligations . . . The core obligations are those which are fundamental to the continued existence of the employment relationship, essentially the obligation of the employee to work and the concomitant obligation of the employer to continue to employ and pay him. When courts say, as they always have, that a contract of employment is not specifically enforceable, they are almost always talking about obligations of this kind.28

19 Lysaght v Edwards (1876) 2 Ch D 499 MR, 506. Although, for the true nature of this trusteeship, see Sir George Jessel MR, at 506, and Meagher (n 18) 295. 20 Since it applies regardless of whether one, both, or neither of the parties wish to continue with the transaction. 21 Howard v Miller [1915] AC 318, 326 PC (BC). 22 Jerome v Kelly [2004] UKHL 25, [2004] 1 WLR 1409 [29] and [32]. 23 Howard v Miller (n 21) 326. See also Bevin v Smith [1994] 3 NZLR 648 (CA), 659–​60; Chi Liung Holdings Sdn Bhd v Attorney General [1994] 2 SLR 354 (CA) 364; Tanwar Enterprises Pty Ltd v Cauchi [2003] HCA 57, (2003) 217 CLR 315 [53]; First Laser Ltd v Fujian Enterprises (Holdings) Co Ltd (2011) 13 ITELR 599 (HKCA) [73]. 24 Peter Turner, ‘Understanding the Constructive Trust between Vendor and Purchaser’ (2012) 128 Law Quarterly Review 582, 582 (footnotes from original). 25 Trade Union and Labour Relations (Consolidation) Act 1992, s 236. 26 Although see Douglas Brodie, ‘Specific Performance and Employment Contracts’ (1998) 27 Industrial Law Journal 37, for an argument that the remedy of specific performance should not be so limited. 27 See Powell v London Borough of Brent [1988] ICR 176; Wadcock v London Borough of Brent [1990] IRLR 223; Keith Ewing, ‘Remedies for Breach of the Employment Contract’ [1993] CLJ 405; and, more recently, Ashfield v Royal National Theatre [2014] EWHC 1176, [2014] 4 All ER 238. 28 Geys v Société Générale, London Branch [2012] UKSC 63, [2013] 1 AC 523 [120] (emphasis added).

142  Criminality at Work The core obligation of the payment of wages is not an obvious subject for specific enforceability.29 Not only are damages likely to be deemed an adequate remedy for the employer’s wrong in such an instance,30 but the enforceability of the employer’s obligation to pay cannot be balanced by a corresponding enforceability of the employee’s obligation to work.31 The fact, therefore, that specific performance is the rule in relation to contracts for the sale of land, and the exception in relation to employment contracts, means that any purported analogy with the vendor–​purchaser constructive trust is unlikely to succeed. The fiduciary analysis is barely more promising. Although the question of whether an employee is regarded as a fiduciary of her employer’s property is one that has been the subject of a considerable amount of attention, both academic and judicial,32 the nature of the employer’s fiduciary status has not received a comparable assessment. Considering the history of the employment relationship, this is perhaps unsurprising: during the course of employment, there have always been many situations in which an employee has been given some form of responsibility for her employer’s property, so the question was bound to arise. By contrast, there have been far fewer situations in which an employer has had reason to have any sort of relationship with her employees’ property, meaning that this issue has attracted far lesser attention. A fiduciary has been described by Millet J as: [S]‌omeone who has undertaken to act for or on behalf of another in a particular matter in circumstances which give rise to a relationship of trust and confidence. The distinguishing obligation of a fiduciary is the obligation of loyalty. The principal is entitled to the single-​ minded loyalty of his fiduciary.33

Whilst the first part of this description, with its explicit reference to trust and confidence, presents an obvious analogy with the employment relationship, the second part, in referring to ‘single-​minded loyalty’ bears much less resemblance to the position of an employer in relation to her employee. The varied commercial objectives and incentives to which employers are subject are on one view fundamentally incompatible with the existence of such elevated standards of loyalty to those in their employ. The more balanced relationship characterized by the implied duty of trust and confidence is a far more authentic

29 Edwin Peel, Treitel on the Law of Contract (15th edn, Sweet & Maxwell 2015) [21-​016]–​[21-​050]. 30 Although for the countervailing argument, see Brodie, ‘Specific Performance and Employment Contracts’ (n 26) 43. 31 Peel (n 29) [21-​046]. 32 See, eg, Reading v Attorney General [1951] AC 507; Hospital Products v United States Surgical Corporation [1983] HCA 64; Neary v Dean of Westminster [1999] IRLR 288; Nottingham University v Fishel [2000] IRLR 471; David Cabrelli, ‘The Implied Duty of Mutual Trust and Confidence: An Emerging Overarching Principle?’ (2005) 34(4) Industrial Law Journal 284; James Edelman, ‘When do Fiduciary Obligations Arise?’ (2010) 126 Law Quarterly Review 302. See also Douglas Brodie, ‘The Employment Relationship and Fiduciary Obligations’ (2012) 16 Edinburgh Law Review 198, in which he takes the view that the employee subject to the implied duty of trust and confidence cannot thereby be a fiduciary in relation to her employer because the relationship is reciprocal in nature. Given, however, that Brodie is considering the question from this perspective, it is perhaps less surprising that he arrives at this conclusion, particularly given that his argument relies to some extent on the lack of vulnerability on the part of the employer. 33 Bristol & West Building Society v Mothew [1996] EWCA Civ 553.

Wage Theft as a Legal Concept  143 representation of the employment matrix.34 Recently, Murray has given this very question detailed consideration:35 [T]‌he contract of employment does not place an obligation of loyalty on the employer: there is no contractual equivalent for the employer of the employee’s contractual duty of fidelity. The implied contractual term which requires both parties to a contract of employment not to act in a way that is likely to destroy or seriously damage the relationship of trust and confidence between the employer and employee . . . does not perform this function . . . There is a misleading coincidence of language here—​the contractual term’s reference to a relationship of trust and confidence, and the language of equity about relationships of trust and confidence that give rise to fiduciary obligations. In fact, the two concepts are distinct.36

Ultimately, however, Murray concludes that The engagement of labour . . . necessarily involves a relationship of trust and confidence requiring the employing entity to place the interests of the worker above their own within a particular narrow sphere. This sphere has been defined as the bundle of assets inherent in the employee to which the employer necessarily gains complete access . . . Just as equity continues to deter inappropriate misuse of access to the employer’s business interests by the employee . . . existing principles of law require that a fiduciary duty be imputed to the employer in respect of the employee’s capacity to engage in future economic activity.37

It remains to be seen whether the courts would agree with Murray’s argument. Her reference to the particular power dynamic between employer and employee has the most potential persuasive force. It was a point made earlier by Brodie when he was making the rather different argument that employees should not be regarded as fiduciaries of their employers’ interests: Over the last 30  years the judicial view of the nature of the employment contract has changed and, as a result, there have been significant changes to the content of the contract. Traditionally the employment contract would have been viewed as commercial in nature and therefore giving rise to a relationship which was not fiduciary: ‘fiduciary obligations are seldom present in dealings of experienced businessmen of similar bargaining strength acting at arm’s length’. Now we view the employment relationship differently; the common law no longer disregards the imbalance of power that is the hallmark of employment relations. In Slaight Communications v Ron Davidson the Canadian Supreme Court endorsed the following passage from Kahn-​Freund’s seminal work, Labour and the Law: ‘The relation between an employer and an isolated employee or worker is typically a relation between a bearer of power and one who is not a bearer of power’. In a similar vein in Malik v BCCI it was said that: ‘An employment contract creates a close personal relationship, where there is often a disparity of power between the parties. Frequently the employee is vulnerable’. Recognition of the power imbalance and 34 Mark Freedland, The Personal Employment Contract (OUP 2003) 177. 35 Jill Murray, ‘Conceptualizing the Employer as Fiduciary: Mission Impossible?’ in Alan Bogg and others (eds), The Autonomy of Labour Law (Hart 2015). 36 ibid 347. 37 ibid 365.

144  Criminality at Work consequent vulnerability in employment relations is important in the current context and, in my opinion, makes it highly improbable that the employment relationship will move into the world of fiduciary relations. Fiduciary obligations exist to protect the vulnerable. In her influential judgment (though in fact dissenting) in Frame v Smith Wilson J stated that relationships in which a fiduciary obligation have been imposed possess three general characteristics; one of which is that the beneficiary is peculiarly vulnerable to or at the mercy of the fiduciary holding the discretion or power. Vulnerability of this nature will often be present in a fiduciary relationship but viewing the employer as being at the mercy of the employee seems counterintuitive. Can it really be said that the employer is vulnerable?38

This recognition of the power imbalance between employers and employees is of course a reason for recognizing an employer as a fiduciary in relation to her employees’ property, as opposed to the converse situation discussed by Brodie. Even were it to succeed in that context, however, there is a further obstacle to the constructive trust analysis. The likely absence of a defined fund from which such monies are to be paid means that the trust device is unlikely to be of much assistance to many unpaid and underpaid employees. In order for a trust to arise, there needs to be sufficiently certain subject matter on which it can be imposed. In situations involving the poor employment practices outlined here, this would require the monies due to the employees to be identified at least to the extent of being a specific proportion of a particular fund, from which the employer was bound to pay the employee.39 Whilst a particular fund of this nature could come into existence in the form of a ‘tronc’ or pool of tips,40 and the failure of employers properly to distribute such gratuities is another practice often referred to as being an instance of wage theft, it is clear that tips are not classed as wages, at least for the purposes of the National Minimum Wage Regulations 1999.41 In Revenue and Customs Commissioners v Annabel's (Berkeley Square) Ltd,42 Rimer LJ described the proprietary implications of a tronc, administered by a third party troncmaster43 thus: upon payment to the troncmaster, the money becomes held by him upon trust for the employees, because that is to attribute a legal interpretation to the arrangement that reflects a realistic business analysis of it; although I regard it as artificial to add the suggested secondary trust since I can envisage no circumstances in which the money so paid would not be enforceably distributable between the employees. If, however, any remote circumstances were to arise in which there could be no such distribution, there would no doubt then be a resulting trust in favour of the employer. But that remote possibility does not mean that, at the point of payment to the troncmaster, the employer retains any relevant beneficial interest in the money. At that point the money becomes instead held by the troncmaster 38 Brodie, ‘The Employment Relationship and Fiduciary Obligations’ (n 32) 208. 39 See Snell’s Equity (n 16) [22-​016]–​[22-​020]. 40 This is a common means of dealing with gratuities–​–​tips are pooled and segregated into the fund or tronc, and then entrusted to an administrator or troncmaster, who distributes them according to an agreed set or rules as to entitlement. 41 Revenue and Customs Commissioners v Annabel's (Berkeley Square) Ltd and others [2009] EWCA Civ 361, [2009] 4 All ER 55. 42 ibid. 43 This is usually an employee, such as a wages clerk or clerical assistant.

Wage Theft as a Legal Concept  145 upon a discretionary trust for the employees; and each of the latter could, in default of the due execution of the trust, sue for such execution. The beneficial entitlement to the money is, pending distribution, in the employees; but until the money is actually paid to them by the troncmaster, none can claim to have been paid.44

On facts such as these, therefore, the employees for whom the funds in the tronc were intended would have something which could form the subject matter of a theft. By the same arrangement, however, the employer also loses the responsibility to pay the tronc monies out, since that falls to the troncmaster. As a result, any appropriation by the employer would require active removal from the tronc, as opposed to an omission to administer it correctly. The troncmaster, on the other hand, could, by failing to pay out of the tronc according to the agreed rules,45 thereby appropriate the monies in which the employees have a collective beneficial interest. If done dishonestly, and with the intention permanently to deprive the employees of it, a potential theft arises. Whilst these actions of both the troncmaster and the employer are conceivable and not perhaps unlikely, they are not normally referred to as instances of wage theft, since their performance is not necessarily facilitated by the employment relationship, and they do not (at least technically) involve an interference with wages. It could be argued that the exclusion of tips and gratuities from the definition of wages for the purposes of minimum wage legislation is a technical nicety, amounting to an attempt by the courts to safeguard, rather than to reduce, the entitlement of employees. It could then follow from this that such an exclusory device should not be used in a different context if its effect thereby is to deny employees a means of protection. Were such an argument to succeed, any action of an employer which required a troncmaster to make any deductions from the tronc prior to its distribution (such as the imposition of fee to cover its administration) could be regarded as an appropriation of ‘wages’. In R v Darroux,46 however, the Court of Appeal distinguished between a defendant who directly controlled the assets alleged to be the subject of theft and one whose actions were merely causally operative in reducing their value. The former set of facts was held to fit the theft framework, whilst the latter was a ‘paradigm example of fraud by false representation’,47 lacking as it did any assumption of the rights of an owner, and should therefore have been charged as such. This analysis precludes the application of theft to any interference by employers with a tronc, since, as we have seen, they lack the requisite element of direct control. The same analysis also excludes any form of ‘truck’ from the remit of theft, since the effect of that practice, in compelling the expenditure of wages in a particular way, is a causal reduction in the money which employees ultimately have at their disposal, rather than a direct action in relation to it. In a broader sense, the Court’s response to the factual matrix in Darroux suggests that charging employers engaged in wrongful payment practices with theft puts the focus of the action on what the employer has gained from her actions, rather than with the essence of what makes her behaviour wrongful. One of the undesirable consequences of this it that a successful charge of theft depends on the employer having made a gain or profit from such 44 Revenue and Customs Commissioners v Annabel's (Berkeley Square) Ltd (n 41)[39]. 45 In Annabel’s, the allocation was done according to a points-​based system, worked out on the basis of relative lengths of service–​–​ibid [14] (Rimer LJ). 46 [2018] EWCA Crim 1009, [2018] 3 WLR 911. I would like to thank Nicholas J McBride for drawing my attention to this decision. 47 ibid [2]‌(Davis LJ).

146  Criminality at Work wrongdoing, since without this, there is nothing on which a purported appropriation can bite. It is not obvious why a dishonest but commercially successful employer should be punished, whilst an equally dishonest but more hapless or less able defendant should escape censure for the same behaviour. It is clear, therefore, that, from a criminal law perspective at least, the shape and tenor of the offence of theft is not the most natural fit with the wrong of exploitative payment practice, or with its consequent harms. The offence of fraud, as suggested by the Court in Darroux, is a more appropriate paradigm.

E. Wage Fraud Under the Fraud Act 2006, fraud is a single substantive offence, which can be committed in three different ways: by false representation; by failing to disclose information; and by abuse of position. The offence is committed even in the absence of the defendant’s gain, the victim’s loss, or even the victim’s deception; all that is required is that the defendant commit the actus reus of the offence dishonestly and with the intention thereby of causing a loss or of making a gain. This is significant for current purposes because it will often be the case, in situations involving sharp employers and vulnerable individuals, that the latter will not in fact be deceived by anything employers have said to them regarding their prospects of pay and conditions. Rather, the vulnerability of the position of such individuals, and the inequality of bargaining power in the relationship between them and their employers (as well as in the absolute sense) means that they are likely to enter into exploitative contracts, despite being aware of their unlawful nature. This should not dilute the employer’s culpability for creating and maintaining such unlawful arrangements for their own gain. Of the three ways in which fraud can be committed, the means of abuse of position, set out under section 4 of the Fraud Act 2006, would seem best to fit the ‘wage theft’ scenario. Whilst it is not inconceivable that an employer might in some factual contexts have a duty to disclose certain information, or not to make false representations, the description of an employer as being in a position in which she is expected not to act against the financial interests of another person is the most universally applicable of the three statutory categories of fraud to the relationships examined here. Section 4: Fraud by abuse of position (1) A person is in breach of this section if he: (a) occupies a position in which he is expected to safeguard, or not to act against, the financial interests of another person, (b) dishonestly abuses that position, and (c) intends, by means of abuse of that position: (i) to make a gain for himself or another, or (ii) to cause loss to another or to expose another to risk of loss (2) A person may be regarded as having abused his position even though his conduct consisted of an omission rather than an act.48



48

Fraud Act 2006.

Wage Theft as a Legal Concept  147 Some idea of what is required in terms of the ‘position’ occupied by the defendant can be gleaned from the Law Commission report from which the Act derives: The necessary relationship will be present between trustee and beneficiary, director and company, professional person and client, agent and principal, employee and employer, or between partners . . . In nearly all cases where it arises, it will be recognized by the civil law as importing fiduciary duties, and any relationship that is so recognized will suffice. We see no reason, however, why the existence of such duties should be essential.49

This lends weight to the argument that wage fraud might in this context be a more intuitively appropriate term to use than wage theft. The former offence, it might be said, is better at criminalizing the essence of the wrong in terms of unscrupulous employment and payment practices: the exploitative actions of the employer, rather than the removal of something which once belonged to the employee. One reason why the Fraud Act offences might work well here is their inchoate nature, in that it does not matter that the victim was not in fact deceived. A common problem for those in a very weak employment bargaining position is not that they are deceived by employer promises, but that they have no choice about the terms they accept. It is the exploitation of this set of circumstances which characterizes the employer’s wrong in this context, and which would be accommodated by the offence of fraud under the 2006 Act. Section 4(2) also explicitly includes omissions as well as acts, meaning that the failure to pay is far more easily accommodated here than it would be by the requirement of an active appropriation under theft. Fraud requires dishonesty and an intention to gain or to cause loss or risk of loss in money or property. Gain and loss are defined in the following way: Section 5: ‘Gain’ and ‘loss’ (1) The references to gain and loss in sections 2 to 4 are to be read in accordance with this section. (2) ‘Gain’ and ‘loss’: (a) extend only to gain or loss in money or other property; (b) include any such gain or loss whether temporary or permanent; and ‘property’ means any property whether real or personal (including things in action and other intangible property). (3) ‘Gain’ includes a gain by keeping what one has, as well as a gain by getting what one does not have. (4) ‘Loss’ includes a loss by not getting what one might get, as well as a loss by parting with what one has.

Subsections 3 and 4 provide a broad definition; one which fairly easily accommodates both the type of loss and the type of gain seen in a ‘wage theft’ situation. Whilst an employer will often gain in the sense of making greater margins of profit by paying her employees less money than she lawfully should, she will always gain in the sense of keeping for herself the



49

Law Commission, Report on Fraud (Law Comm 276, 2002) [738].

148  Criminality at Work difference between what she should pay those workers and what she does in fact pay them. As far as the worker’s loss is concerned, subsection 4 is comprehensive; covering both the shortfall in lawful monies due, as well as any instances of mandated payment for essential work items, such as uniform or equipment. The intention element of the fraud offences need only be of the non-​core Woollin variety,50 meaning that a defendant will be deemed to have intended an outcome where she is certain or virtually certain that a particular consequence will result from her actions. In employment practice terms, this means that an employer who promises a particular level of payment to a worker, knowing that her ability to pay that sum depends upon a particular set of receipts or stream of income that are very unlikely to materialize, will be guilty of fraud. The inchoate mode of the offence means that she does not have to gain anything so long as she intended (in at least the Woollin sense) to gain by so doing: an employer cannot therefore claim that the failure to make a profit means that she cannot be guilty of wage fraud. Like theft, much of the effect of the fraud offences will be determined by the dishonesty element and, consequently, by the state of the employer’s mind. Since the Supreme Court decision in Ivey v Genting Casinos,51 a jury deciding if a particular course of conduct is dishonest will be asked to consider whether the conduct in question would be regarded as dishonest by the standard of the ordinary honest person, but in the light of the facts known or believed by the defendant. It is no longer necessary for the defendant subjectively to believe that her conduct was dishonest according to those standards.52 As Virgo has pointed out, Ivey has led to a ‘dissonance between the objective test of dishonesty and the subjective test of recklessness’ which he regards as having particularly unfortunate implications for the offence of theft. Virgo’s specific issue with this, however, arises from the already expanded remit of theft, meaning that the offence can be committed in the absence of an adverse interference with civil law property rights.53 This, combined with an objective test of dishonesty, risks criminalizing that which might not properly be regarded as the concern of the criminal law. Although appropriation is not relevant to fraud, the question remains whether an employer who abuses his position in relation to an employee, and does so in a way which an ordinary honest person would regard as dishonest but he himself does not, should be subjected to the scrutiny and possible punishment of the criminal law. The answer to this question depends of course on the relative functions and objectives attributed to each sphere of the law; accounts of which are neither unanimous nor static. Much of the analysis above has been conducted from the perspective of the civil law’s consequential effect on the criminal law. Traditionally, this has been the direction of conceptual travel: [D]‌istributive factors are not the concern of the criminal law, since it is not for the criminal law to decide when property rights exist. Rather, property offences exist primarily to reinforce and protect those rights once recognized by the civil law. Indirectly, therefore, 50 AP Simester and others, Simester & Sullivan’s Criminal Law:  Theory and Doctrine (6th edn, Bloomsbury 2016) 626. 51 [2017] UKSC 67, [2018] AC 391. 52 This had been the second limb of the test in R v Ghosh [1982] QB 1053, overruled by the Supreme Court in Ivey (n 51). 53 In R v Hinks [2001] 2 AC 241, the House of Lords upheld the conviction of a defendant for theft where the appropriation required by the criminal law was constituted by the receipt of a gift of indefeasible title at civil law. In other words, the defendant received valid title to property transferred to her voluntarily by her victim, but was nonetheless convicted of stealing it.

Wage Theft as a Legal Concept  149 property offences are not merely a means to protect individuals’ autonomous rights over property, but also a tool for reinforcing political decisions about the distribution of scarce resources in society. Prima facie rules of property arise in the civil law for a variety of underlying reasons. Yet the criminal law does not look behind the prima facie rules. It takes ownership, and its facets, as an axiom.54

Recently, however, Farmer has suggested that there has been a diversion of such thinking: [T]‌he criminal law is pulling away from definitions of property in the civil law. It is not always concerned with the precise nature of property rights or the formal legality of contracts but with the permissibility of modes of acquisition. As a consequence, civil law does not necessarily define a boundary or limit to the scope of the criminal law, as conduct which is judged to be fraudulent or dishonest or exploitative runs the risk of falling foul of the criminal law irrespective of the status of the transaction in civil law. Second, the aims of the criminal law are no longer about protecting property as such, or protecting property rights as defined in the civil law; they are protecting the interests of the vulnerable against conduct which poses a threat to the security of property and the securing of confidence in institutions such as the market in which property is transferred. This is not necessarily driven by actual harm to actual victims, but by the risk of harm. The point here is not only that such conduct is inchoate—​or even ‘pre-​inchoate’—​it is that harms are conceived in relation to an abstract class of victims (the vulnerable) and to an abstract conception of a system of property rights. Third, it is arguable this area is now shaped by a kind of ‘public responsibility’ consisting in the need to show respect for the property interests of others. This is less a sense of subjective awareness of particular outcomes (and so differs in a strict sense from Fletcher’s idea of subjective liability) than the responsibility not to act towards the property of others in a way which others (or the criminal law) would regard as dishonest or fraudulent or which might undermine trust in the security of property. Overall this contributes to a sense that this area of the law is changing in a way which increasingly pulls it away from its eighteenth century roots, protecting different kinds of interests in different kinds of property. Indeed, the inexorable rise of the concept of dishonesty has led certain commentators to identify the area as that of ‘crimes of dishonesty’ rather than property. Not only does this shift the focus of the law, but it also makes it potentially limitless as there is little or no reason why all dishonest contracts or business dealings should not be subject to the criminal law; and indeed that criminal law remedies may be available even where there is no possibility of a civil remedy.55

The time might well be ripe, therefore, for the criminal law to play a part in protecting vulnerable employees from behaviour that would previously have fallen outside of its remit. If Farmer’s argument is accepted, such an extension would be justified by the need to safeguard the institutional integrity of property, and of the labour market. The presence of criminal sanctions would undoubtedly do much to mitigate the fact that, for many vulnerable employees, civil action is simply not a viable option, in terms of financial resources, knowledge, capacity, or awareness. For those not convinced by this argument, and to whom

54 55

Simester & Sullivan’s Criminal Law (n 50) 493. Lindsay Farmer, Making the Modern Criminal Law: Criminalization and Civil Order (OUP 2016) 223.

150  Criminality at Work the alignment of poor payment practices with the stigmatic statutory offence of fraud would seem to be overstepping the mark in terms of both sanction and reputational consequence, the middle ground could be covered by a bespoke statutory offence which defined its own terms and outcomes, such as the one in force in Canada from 1935 to 1955.56 In evaluating the short lifespan of this particular statutory initiative, Tucker concludes: Perhaps the lesson of history for activists is that the invocation of the symbolic image of criminal law is more powerful than its actual application against employers. As a result, current wage theft campaigns are not focused on criminalizing employer wage violations, but rather on changing public perceptions of the seriousness of the problem in order to pressure provincial governments to better enforce regulatory laws. Whether this strategy is more successful than making wage theft a crime remains to be seen.57

The social, political, and economic backdrop to the early part of the twenty-​first century, however, is far removed from the one against which the Canadian experience occurred. As Farmer has suggested, the inclination to protect individual property rights from interference by those in a position to abuse an imbalance in power is currently in the ascendant. This should bode well for any contemporary attempt to formulate a statutory means of regulating employer payment practices. In the meantime, however, and for the reasons already rehearsed, dealing with wrongs of this nature is a job better suited to the offence of fraud than to the offence of theft.

56 See Eric Tucker, ‘When Wage Theft Was a Crime in Canada, 1935–​1955: The Challenge of Using the Master’s Tools Against the Master’ (2017) 54 Osgoode Hall Law Journal 933. Although, as Tucker notes, only one employer was prosecuted during the twenty years of this offence being in force. 57 ibid 957.

8

The Criminalization of Workplace Harassment and Abuse An Over-​Personalized Wrong? Alan Bogg and Mark Freedland

A.  Introduction The wrong of harassment and bullying in the workplace has attracted intense public attention in the wake of high-​profile revelations of widespread abusive practices in the entertainment and media sectors. This exposure of abusive practices, hitherto shrouded in secrecy and silence, have tended to present this wrong in highly personalized terms. This personalized framing has fitted popular narratives of harassment, which have been constructed around the very famous protagonists alleged to have engaged in multiple forms of egregious abuse. The seriousness of the issue has led to widespread calls for the increased use of criminal law to punish the perpetrators of harassment and abuse severely. There is already a significant body of law regulating workplace harassment, and some of that is based in the criminal law. We focus here on the Protection from Harassment Act 1997 (PHA), which makes provision for criminal and civil liability in respect of harassment. We argue that there has been a ‘dragging’ effect which the present form of criminalization of workplace harassment has had upon the general regulation of workplace harassment and bullying (section B); secondly, we put forward a particular notion of criminalization as over-​personalization which may help to explain that negative or dragging effect (section C); thirdly, we consider some recent reform proposals in the International Labour Organization (ILO) and the House of Commons Select Committee on Women and Equalities. These engagements suggest a different possible approach to the criminalization of workplace harassment and bullying which avoid those negative effects. This is based upon a model of employer safeguarding duties, which adopts a more structural perspective on harassment and so avoids some of the limitations of personalized criminalization under the PHA (section D); finally, we consider the desirable shape of criminalization in the light of human dignity, a value that has an architectonic role in harassment law. We suggest that the principle of human dignity supports: (i) the criminalization of employer safeguarding duties; (ii) maintaining and supporting the opportunities for individual workers to seek redress for rights-​violations in the ordinary courts; (iii) dispensing with enquiries into victim consent in liability for workplace abuse (section E). In the conclusion, we propose a new area for research, ‘bargaining in the shadow of the criminal law’, to examine ways in which trade unions can be integrated more deeply into the criminal enforcement process. Alan Bogg and Mark Freedland, The Criminalization of Workplace Harassment and Abuse In: Criminality at Work. Edited by Alan Bogg, Jennifer Collins, Mark Freedland QC, and Jonathan Herring, Oxford University Press (2020). © Alan Bogg, Jennifer Collins, Mark Freedland, and Jonathan Herring. DOI: 10.1093/oso/9780198836995.003.008

152  Criminality at Work

B.  The Negative Effect of the Criminalization of Workplace Harassment in its Present Form The main part of our argument therefore begins with the claim that the present form of criminalization of workplace harassment has had a negative or dragging effect upon the general regulation of workplace harassment and bullying. We take as our starting point for this contention Lizzie Barmes’ magisterial analysis, both doctrinal and empirical, of the whole body of law concerning workplace harassment and bullying, which is presented in her monograph on Bullying and Behavioural Conflict at Work.1 In that work Barmes surveys the extensive apparatus of individual employment rights law which bears upon workplace harassment and bullying and indeed upon what she designates as a broad category of ‘behavioural conflict at work’. This elaborate juridical apparatus consists of elements of the statute law of unfair dismissal and employment discrimination, and of the common law of contract and tort—​and it includes the provisions of the PHA which are our special concern in this chapter because it is they which bring about the criminalization of workplace harassment and bullying. Although Barmes clearly regards this body of law as impressive in its extent, she is by no means approving of its regulatory effectiveness in controlling workplace harassment and bullying. On the contrary, the ‘central finding’ of her study is that ‘individual labour and equality rights, as currently legislated, implemented, and enforced in the UK, overall enhance the status quo by supporting traditional workplace hierarchy and marginalizing more plural, collective ways of being’.2 For her, the sheer elaborateness of this body of law constrains its practical impact upon workplace harassment and bullying; and in the course of making out that argument she refers quite extensively to the counter-​productive complexities of the PH—​that being the legislation which set up a particular combination of and interplay between civil and criminal liability for harassment.3 Although originally primarily aimed at the particular phenomenon of ‘stalking’,4 it has been interpreted as applying to various kinds of workplace behaviour of an equivalent seriousness and impact.5 1 Lizzie Barmes, Bullying and Behavioural Conflict at Work: the Duality of Individual Rights (OUP 2015), reviewed by Alan Bogg (2017) ‘Lizzie Barmes: Bullying and Behavioural Conflict at Work: The Duality of Individual Rights’ 44 Journal of Law and Society 702–​08. 2 ibid 1. 3 The crucial provisions of the Act are: Section 1: Prohibition of harassment. (1) A person must not pursue a course of conduct—​ (a) which amounts to harassment of another, and (b) which he knows or ought to know amounts to harassment of the other.  . . .  Section 2: Offence of harassment. (1) A person who pursues a course of conduct in breach of section 1(1) . . . is guilty of an offence. (2) A person guilty of an offence under this section is liable on summary conviction to imprisonment for a term not exceeding six months, or a fine, . . . or both.  . . .  Section 3: Civil remedy. (1) An actual or apprehended breach of section 1(1) may be the subject of a claim in civil proceedings by the person who is or may be the victim of the course of conduct in question. (2) On such a claim, damages may be awarded for (among other things) any anxiety caused by the harassment and any financial loss resulting from the harassment. 4 Lord Mackay, introducing the Bill on its Second Reading in the House of Lords, HL Deb 24 January 1997, vol 577, cols 917–​43, at col 917, referred in particular to ‘stalking, racial harassment, or anti-​social behaviour between neighbours’. 5 See Barmes (n 1) 89–​92.

The Criminalization of Harassment and Abuse  153 Our argument, specifically focusing upon the effect of the criminalization of workplace harassment by the PHA, suggests that there is a particular problem here which goes beyond the general concerns about the complexity of this legislation. It exemplifies Barmes’ worries about the working of an individual rights system and culture in the employment sphere. Her survey of the working of the PHA in the employment sphere suggests that the criminal offence of harassment has not been frequently prosecuted in that sphere, and that the Act’s provision for civil liability has been of greater practical significance. However, in the development and interpretation of the civil liability provisions, the courts have been keenly aware that these civil and criminal liabilities are tightly yoked together, forming a composite enforcement mechanism for the Act’s prohibition of harassment. In that situation, Barmes’ survey discloses various instances in which the courts’ awareness of the criminal dimension of workplace harassment has tended to constrain their approach to the more frequently invoked civil liability provisions. This dragging effect of the criminalization of workplace harassment upon the civil deployment of the PHA can be seen in a set of cases starting with and stemming from the leading case concerning employers’ vicarious civil liability for harassment of workers by members of their own staff, that of Majrowski v Guy’s and St Thomas’s NHS Trust.6 The House of Lords judges were decidedly reluctant to concede the principle of vicarious civil liability for harassment in the employment context; they clearly felt that they were opening the floodgates to a spate of possibly unmeritorious claims, and they were anxious to impose a restrictive approach upon the adjudication of such claims. In order to sustain and justify this restrictive approach, Lord Nicholls famously invoked the need for caution arising from the parallel co-​existence of civil and criminal liability: Courts are well able to recognise the boundary between conduct which is unattractive, even unreasonable, and conduct which is oppressive and unacceptable. To cross the boundary from the regrettable to the unacceptable the gravity of the misconduct must be of an order which would sustain criminal liability under section 2.7

Barmes’ analysis goes on to show how the idea of this dragging effect of the criminal liability upon the civil liability was to find considerable favour with the courts, in an ascent which reached its zenith in the Court of Appeal in Sunderland CC v Conn,8 where all the three judges were agreed that workplace harassment should attract civil liability only where it was ‘of such gravity as to justify the sanctions of the criminal law’9 or ‘of an order that would sustain criminal liability and not merely civil liability on some other register’10 and where one of them expostulated: [W]‌hat on earth is the world coming to if conduct of the kind that occurred in the third incident can be thought to be an act of harassment, potentially liable to giving rise to criminal proceedings punishable with imprisonment for a term not exceeding six months, and to a claim for damages for anxiety and financial loss?11

6

[2006] UKHL 34, [2007] 1 AC 224. ibid [30]. [2007] EWCA Civ 1492, [2007] 11 WLUK 121. 9 ibid [12] (Gage LJ). 10 ibid [18] (Buxton LJ). 11 ibid [19] (Ward LJ) (emphasis added). 7 8

154  Criminality at Work Although, as she demonstrates,12 this dragging effect was at least somewhat mitigated by the Court of Appeal in Veakins v Keir Islington Ltd,13 it has nevertheless retained significant traction upon adjudications of this kind. That, we wish to argue, is because the courts were responding to an underlying intuition about the way in which criminal liability and civil liability had been interrelated by the PHA, and the specific role of criminalization in that legislative scheme. In fact, the central issue in Majrowski, the employer’s vicarious liability for the employee’s statutory tort of harassment, exposed in an acute form the underlying judicial concern with the personal nature of the wrong. Lord Nicholls described the wrong of harassment as comprising ‘conduct of an intensely personal character between two individuals’.14 He did not regard this as an insuperable difficulty for the employer’s vicarious liability. Indeed, he identified other civil wrongs such as assault and battery which also attracted vicarious liability.15 Each of these civil wrongs could attract criminal liability, and each could also be described as ‘intensely personal’ in nature. Indeed, their personal character may have made them particularly apt for criminalization. By contrast, although Lady Hale shared Lord Nicholls’ view of the ‘personal character’ of the harassment wrong, she was rather more reluctant to reach the conclusion that the Act included vicarious liability: The promoters of the Act might well have thought that this intense focus on the perpetrator and getting him to stop would not be helped, and might even be hindered, by making the employer vicariously liable. Vicarious liability . . . does not depend upon the employer having done anything wrong or even having broken any legal duty imposed upon him.16

Under the discrimination law framework, where there is no criminal liability for discrimination and harassment, the judicial approach to vicarious liability has been more strongly purposive, in contrast to the reluctant tenor of some of the judgments in Majrowski.17 There has been a significant expansion of vicarious liability in recent years, with some of those cases arising in the context of institutional sexual abuse.18 The specific reservations in Majrowksi about vicarious liability for statutory torts that are also crimes would likely be more muted today. Nevertheless, there are deeper puzzles about the interactions between criminal and civil liability that underlie Majrowski. These puzzles, which concern the personal nature of the wrong, remain highly pertinent. We identify three such issues. First, the criminal law is anchored in the fundamental value of individual autonomy and personalized responsibility. As Simester and Sullivan account for it, ‘in a liberal community, the concern with autonomy is fundamental. If the law is to acknowledge and respect individuals as independent members of society, then it must judge them according to their own actions, and not those of others.’19 This has significant ramifications for the structures of liability in crime and tort. In criminal law, vicarious liability and other forms of third-​party 12 Barmes (n 1) 155–​56. 13 [2009] EWCA Civ 1288, [2009] 12 WLUK 33. 14 Majrowski (n 6) [25] (Lord Nicholls). 15 ibid [26]. 16 ibid [68] (Lady Hale) (emphasis added). See also Lord Brown [81] and Lord Hope [43] on the personalized character of the harassment wrong and the resulting tensions with vicarious liability. 17 See, eg, Jones v Tower Boot Co Ltd [1997] 2 All ER 406, [1997] ICR 254. See now Equality Act 2010, s 109. 18 See, eg, Various Claimants v Catholic Child Welfare Society [2012] UKSC 56, [2013] 2 AC 1. 19 Andrew P Simester and others (eds), Simester and Sullivan’s Criminal Law (6th edn, Bloomsbury 2016) 15.

The Criminalization of Harassment and Abuse  155 liability (eg the liability of employers for the foreseeable harassment perpetrated by customers and service-​users against employees) should have no role. The paradigm forms of criminal liability would be either as principal or as secondary party in assisting or encouraging the primary wrong. In each case, criminal liability attaches to the individual’s own culpable acts rather than the attribution of liability for the wrongs of another. By contrast, vicarious liability and third-​party liability may have an extensive role in formulating legal responses to harassment as a statutory tort. Some of the difficulties in Majrowski may have stemmed from the integration of criminal and civil liability into a single statute based upon a common behavioural standard, and a normative spill-​over from the criminal law into the law of torts. There may well be advantages, particularly in terms of clarity, in reforming the law on workplace harassment through a single statute. However, even if reform is implemented in this way, there may be good reasons to develop different behavioural standards for the different purposes of criminal and civil liability.20 Furthermore, the principle of individual autonomy would justify distinctive structures of liability for the criminal and statutory tort variants of workplace harassment. The difficulty in Majrowski was that the existence of criminal liability cast a shadow over the development of vicarious liability for the tort. Secondly, the existence of concurrent criminal and civil liability for the same wrong raise complex issues, particularly with respect to the dignity of the victim. We shall explain this specific meaning of dignity in section E. The initiative in prosecuting crimes lies with the state and the victim occupies a more passive role in the criminal process.21 By contrast, in private law the normative power to seek redress in a public court lies with the right-​ holder.22 The overlap of criminal law and private law can create procedural difficulties, such as whether to suspend civil proceedings in circumstances where a criminal prosecution is being pursued in relation to the same wrong.23 Different regulatory regimes have adopted different approaches to this interaction. In the health and safety context, criminal liability now operates as the exclusive mechanism for addressing statutory health and safety breaches following the repeal of the civil right of action in section 61 of the Enterprise and Regulatory Reform Act 2013 (ERRA).24 In relation to assault and battery, section 45 of the Offences against the Person Act 1861 provides for civil proceedings to be barred for the same cause following a summary hearing in the magistrates’ courts. Any reform of harassment law must ensure that criminal and civil liability are aligned so as to ensure respect for the victim’s dignity. The structure of private law remedies, based on the right-​holder’s normative power to seek redress, is more sensitive to those dignity-​based concerns than the criminal law. This would rule out regulatory approaches that extinguish liability in private law, such as that recently adopted in health and safety law by ERRA, section 61.

20 For example, there appear to be differences between the definitional elements of battery as tort and battery as crime: see Paula Giliker, Tort (6th edn, Sweet & Maxwell 2017) 427. Such divergence may be justified by the different goals of criminal law and tort law. Divergences might also reflect the need to insist upon a more stringent normative threshold for criminalization, reserving the criminal law for the more serious forms of wrong. 21 Matthew Dyson and John Randall, ‘England’s Splendid Isolation’ in Matthew Dyson (ed), Comparing Tort and Crime: Learning from across and within Legal Systems (CUP 2015) 18, 58 22 Arthur Ripstein, Private Wrongs (Harvard University Press 2016) 271–​72. 23 Matthew Dyson, ‘The Timing of Tortious and Criminal Actions for the Same Wrong’ (2012) 71 Current Legal Problems 85. 24 Michael Ford, ‘The Criminalization of Health and Safety at Work’ in Alan Bogg, Jennifer Collins, Mark Freedland, and Jonathan Herring (eds), Criminality at Work (OUP 2020) ­chapter 21.

156  Criminality at Work Thirdly, the preoccupation with the personal character of the harassment wrong obscures the possibility of other non-​personalized forms of liability for employers. There is some recognition of this dimension by Lord Nicholls in Majrowski when he acknowledged that ‘although the victim must be an individual, the perpetrator may be a corporate body’.25 However, this represents only a very modest departure from the personalized framing of the harassment wrong: the ‘corporate body’ is still being conceptualized as a ‘perpetrator’ of interpersonal violence on an individual victim. We think that this model of liability is too restrictive. It might be better to frame the employer’s duty as a safeguarding duty to maintain a healthy and dignified place of work. In this respect, it would have some similarities with the statutory duties that currently exist under health and safety legislation. This would also give a more structural perspective on the phenomenon of harassment. Acts of personalized harassment generally occur within a wider context of toxic organizational structures, norms and practices that create unacceptable psychosocial risks for workers. These risks are intensified in contexts of workplace precarity and insecurity, such as employment on zero-​ hours contracts.26 In its 2018 report Turning the tables: Ending Sexual Harassment at Work, the Equality and Human Rights Commission proposed the introduction of a mandatory duty on employers to take reasonable steps to protect workers from harassment and victimization in the workplace, as a basis for triggering changes in workplace culture.27 This form of corporate liability operates independently of the personal liability of individual harassers. It is thus distinct from vicarious, third-​party, and accessory liability, where the employer’s liability depends upon an act of interpersonal harassment. Liability is focused on the employer’s freestanding responsibility to maintain a workplace where workers are secure from violence and harassment and can work in circumstances of freedom and dignity. This statutory duty could form the basis for a new criminal offence, alongside the existing structures of criminal and civil liability focused on harassment as a personalized wrong.

C.  The Over-​personalization of Workplace Harassment as a Criminal Offence The argument of the previous section exposed a deep underlying issue about the way in which workplace harassment had been criminalized by the PHA. In that previous section, we were concerned with the incidental effect of that particular form of criminalization upon the interpretation and adjudication of parallel civil liability which that Act created. It is now necessary to consider the nature and aspect of this criminalization in and of itself. We contend that the PHA designed the offence of harassment as an essentially personal one, in a way which was not entirely suitable to its functioning in the workplace context. While this personalized conception was suitable for the social problem of stalking, a problem the

25 Majrowski (n 6) [19] (Lord Nicholls). 26 Laura Bates, ‘ “I Wasn’t Protected at All”: Why Women are Made Vulnerable by Zero-​hours Contracts’ The Guardian (19 March 2018) accessed 30 April 2019. 27 Equality and Human Rights Commission, ‘Turning the Tables:  Ending Sexual Harassment at Work’ (27 March 2018) 13 accessed 30 April 2019.

The Criminalization of Harassment and Abuse  157 legislation was designed to address, its application to the distinctive dynamics of workplace harassment has been highly problematic. In order thus to develop our argument, it will be helpful to look at the PHA from a regulatory perspective, following the analytical path taken by Nicola Lacey in her essay of 2004 on ‘Criminalization as Regulation’.28 That chapter sought to break down a divide which in her view had grown up between criminal lawyers and regulatory scholars, in that criminal lawyers were focusing on . . .  the traditional sphere of ‘real crime’—​roughly equating to those offences requiring proof of ‘mens rea’ or ‘fault’ while regulation scholars . . . have tended to focus their attention on . . . regulatory norms . . . ; they , conversely, have paid relatively little attention to the terrain of ‘real crime’, such as serious offences against property or the person . . . 29

She thereby invokes, in a modern guise, the traditional criminal lawyers’ distinction between ‘mala in se’ and ‘mala prohibita’, and justly argues that this distinction should not form a frontier between the territory of criminal law scholars and that of regulatory scholars. This analysis provides a vantage point from which to re-​examine section 2 of the PHA. On the one hand, the Act seems to fit within Nicola Lacey’s stereotype of ‘regulatory criminalization’; we might thus imagine that the 1997 Act was introducing new regulatory offences rather than creating ‘real crimes’. Indeed, at one point in her essay she almost seems to say as much: [T]‌here is a . . . vivid . . . sense in which regulatory criminal law, at least in Britain, is on the increase . . . In the last decade, substantial pieces of criminal legislation have averaged more than one a year in England and Wales . . . most of [this] legislation includes a scattergun array of new offences—​covering . . . newly perceived social problems ranging from overloud ‘raves’ through joy-​riding to harassment, paedophile offences, and international terrorism.30

Yet on closer examination, the legislator in fact created what has been understood and constructed by the courts as a new ‘real crime’. In the previous section of this chapter, we saw how strongly the courts had imputed that character of ‘real crime’ to the section 2 offence, and how that imputation had indirectly impacted on their approach to the construction of the civil liability for the statutory tort. One very important facet of this construction of the offence of harassment as a ‘real crime’ has been to understand it as a particularly personal offence—​truly, a new kind of offence against the person, the infliction of an unacceptable course of conduct upon one person by another person as a form of interpersonal violence. The interpretation of section 2 as it has operated in the context of the workplace, which we reviewed in the previous section, reveals a perception of it as being intensely and archetypally bound up with the personalities of a human perpetrator and a human victim and with the essentially interpersonal relations

28 Nicola Lacey, ‘Criminalization as Regulation: The Role of Criminal Law’ in Christine Parker and others (eds), Regulating Law (OUP 2004). 29 ibid 144–​45. 30 ibid 163 (emphasis added).

158  Criminality at Work between them. This might seem entirely obvious and uncontroversial within the paradigm of ‘stalking’ around which the offence was primarily constructed. How could the practice of ‘stalking’ be legally understood and legally encapsulated in any less personal way? However, in the different context of the criminalization of bullying and abusive behaviour in the workplace, the framing of harassment as an entirely personal offence might be regarded as not being a necessary choice, even as being an egregious one. This will become more apparent when the criminalization of workplace bullying and harassment is subjected to a regulatory rather than a doctrinal scrutiny—​when, in other words, we pursue the question of how well this ultimately personal characterization of the offence of harassment fulfils the regulatory purpose of the PHA and the regulatory needs of the workplace. Engaging in the that kind of scrutiny, we might well conclude the criminalization of workplace harassment by the 1997 Act has turned out to be in significant respects an over-​personalized one, disabling the Act from fulfilling its proper function in the workplace context. We can identify this over-​personalization more precisely by recognizing that it occurs in three aspects of the formulation of the offence of harassment under the 1997 Act: firstly in the conception of the perpetrator as an individual person, secondly in the conception of the victim as an equally individual person, and thirdly in the conception of the infliction of the harassment as a correspondingly individual course of conduct between them. As we saw in the previous section, the actual deployment of the criminal offence of harassment in the workplace context has apparently been very limited, but its conceptualization in this highly personalized way has clearly shaped and influenced the framing and handling of the associated civil liability; and it is very apparent from Barmes’ survey of the adjudication of the civil liability cases that they are handled in exactly that individualized fashion.31 There is a further and even more significant concern about the way in which the 1997 Act has personalized its conception of the perpetrator of the harassment, and hence in turn its conception of the conduct which amounts to harassment. In a way which might be necessary and appropriate to pin down the paradigm of ‘stalking’, around which the Act was originally constructed, the Act understands the perpetrator of harassment to be the human actor who is immediately engaged in the conduct which fulfils the definition of the offence of harassment. The problem is that, in the workplace context, this serves to marginalize or even to suppress an awareness of the institutional role of the employing organization in supporting organizational structures, norms, and practices that facilitate and enable the bullying or abusive behaviour. All this seems to us to indicate the existence of a deep-​seated regulatory problem about the over-​personalized criminalization of workplace harassment by the PHA. In the next section, we turn to consider how criminalization has been framed as a regulatory response to workplace harassment in recent reform proposals at the international and national levels. These incipient reform efforts are promising in moving beyond the limitations of the personalized wrong model in the PHA.



31

Barmes (n 1) 157.

The Criminalization of Harassment and Abuse  159

D. Reform Proposals and the Role of the Criminal Law 1.  Regulatory Activity at the International Labour Organization In 2015 the governing body of the ILO placed a standard-​setting item on the agenda of the 107th Session of the International Labour Conference (May–​June 2018) on ‘Violence against women and men in the world of work’. There was also an extensive report authored by Dr Jane Pilinger in 2017 on Violence and Harassment against Women and Men in the World of Work, which provided a comprehensive assessment of the law and trade union practices on violence and harassment at work across the world.32 The deliberations around a draft instrument revealed the reluctance of some member states to support a strong role for criminalization.33 Furthermore, the International Labour Office identified a diverse mix of current regulatory practices in responding to harassment and violence at work, ranging across labour law, discrimination law, criminal law, and health and safety law.34 While it took the view that criminal prohibitions ‘can be useful for extreme cases of violence and harassment’,35 the criminal law was unlikely to be sufficient. The procedural burdens of the criminal process, such as rules on burden of proof and witnesses, limited the effectiveness of the criminal law as a regulatory tool. It was better to ‘allow victims to seek a variety of remedies depending on the nature of the act, and the type of work relationship and conditions they are working under’.36 Following a consultation process and discussion at the International Labour Conference, there have now been enacted, at the ILO Centenary Conference in June 2019, a Convention and Recommendation Concerning the Elimination of Violence and Harassment in the World of Work.37 These instruments provide some interesting insights into the anticipated role of criminalization. While criminalization is accorded a potential regulatory role, the instruments are somewhat tentative on the utilization of criminal law as a form of labour regulation. This may reflect the caution towards criminalization expressed by some member states during the consultation process.38 Article 1 of the Convention sets out a broad and inclusive definition of ‘violence and harassment’ which . . . refers to a range of unacceptable behaviours and practices, or threats thereof, whether a single occurrence or repeated, that aim at, result in, or are likely to result in physical, psychological, sexual or economic harm, and includes gender-​based violence and harassment . . . 32 Jane Pillinger, ‘Violence and Harassment Against Women and Men in the World of Work’ (International Labour Organization 2017) www.ilo.org/​actrav/​info/​pubs/​WCMS_​546645/​lang-​-​en/​index.htm accessed 30 April 2019. 33 International Labour Office, Ending Violence and Harassment in the World of Work: Report V (2) (ILC 107th Session 2018) (hereafter ILO, Report V(2)) 59  accessed 3 August 2019. 34 International Labour Office, Ending Violence and Harassment against Women and Men in the World of Work:  Report V (1) (ILC 107th Session 2018) (hereafter ILO, Report V(1)) 47  accessed 3 August 2019. 35 ibid 48. 36 ibid 48. 37 Violence and Harassment Convention, 2019 (No 190) and Violence and Harassment Recommendation, 2019 (No 206). 38 ILO, Report V(2) (n 33) 58–​59

160  Criminality at Work Furthermore, ‘worker’ includes ‘persons working irrespective of their contractual status, persons in training, including interns and apprentices, workers whose employment has been terminated, volunteers, jobseekers, and job applicants.’ The ‘core principles’ in Article 4 propose ‘an inclusive, integrated and gender-​responsive approach’ to regulation. This envisages a ‘comprehensive strategy’ which includes ‘sanctions’ and ‘ensuring access to remedies and support for victims’. The Recommendation provides further, in Point 2, that In adopting and implementing this integrated approach, Members should address violence and harassment in the world of work in labour and employment, occupational safety and health, equality and non-​discrimination law, and in criminal law, where appropriate.

Moreover, the two instruments accord a particular emphasis to a health and safety approach to harassment and violence. Article 9 of the Convention requires the adoption of laws ‘requiring employers to take appropriate steps commensurate with their degree of control to prevent violence and harassment in the world of work’. This includes, in particular taking steps to ensure, so far as is reasonably practicable: consultation with workers and their representatives, the adoption and implementation of a workplace policy on violence and harassment, and the integration of psychosocial risks into the management of workplace hazards. The Recommendation at Point 8 further stipulates that Particular attention should be paid to the hazards and risks that: (a) arise from working conditions and arrangements, work organization and human resource management, as appropriate; (b) involve third parties such as clients, customers, service providers, users, patients and members of the public; and (c) arise from discrimination, abuse of power relations, and gender, cultural and social norms that support violence and harassment.

Overall, the ILO instruments reflect a regulatory approach to criminalization. This is based on a concern with outcomes, such that the criminal law should be deployed where it makes a positive contribution to the elimination of harassment and violence in the workplace. In this respect, the instruments do not seem to differentiate between regulatory ­typologies. There is nothing distinctive about the criminal law as a regulatory approach, that would warrant special caution in its utilization. This may reflect a ‘labour law’ perspective on the criminal law, which focuses more on its effectiveness in securing decent work than concerns about whether criminalization is normatively justified. In the view of the International Labour Office, ‘the obligations placed on member States under the proposed Conclusions are broad enough to allow for flexibility in how various forms of violence and harassment are addressed, as long as this is effective’.39 Part IV of the Convention is concerned with ‘protection and prevention’ and part V of the Convention ‘enforcement and remedies’. The instruments are strongly influenced by a ‘health and safety’ model of labour rights. In our view, this would be accommodating of ‘corporate’ liability for employing entities, corresponding to a duty to mitigate the risks and hazards associated with ‘unequal power relations’ and ‘gender, cultural and social norms’. In this respect, the ILO approach could transcend the limitations of a personalized framing of harassment wrongs under the

39 ILO, Report V(2) (n 33) 24 (emphasis added).

The Criminalization of Harassment and Abuse  161 PHA, and lead to wider organizational and structural changes in the workplace. However, the ILO would appear to be agnostic on the question of whether this corporate duty should be enforced through criminal or civil law: that will depend upon an assessment of regulatory effectiveness There is also a strong emphasis on empowering the agency of victims to seek redress through the civil law and the ordinary courts. The ILO approach seems to preclude the displacement of civil law remedies by criminalization, and this would rule out the strategy of ‘deregulatory criminalization’ as seen in the domestic health and safety law context through ERRA, section 61. This is reflected particularly in Article 10 of the Convention, which requires ‘easy access to appropriate and effective remedies and safe, fair and effective reporting mechanisms’. The Recommendation further provides in Points 14–​17 for a range of effective remedies such as reinstatement, compensation for material and non-​material losses, and provisions on legal fees, specialized support and counselling, and legal advice. The recognition that harassment can be perpetrated by co-​workers, customers, and service-​users would also support different bases of liability, such as vicarious, third-​party, and accessory liability. We think that respect for personal autonomy would require that forms of imputed liability, such as vicarious and third-​party liability, be limited to civil wrongs. By contrast, accessory liability as an independent form of wrongdoing could apply to both civil and criminal wrongs. This complex framework of liability would necessitate both civil and criminal law norms in an integrated structure.

2.  Regulatory Activity in the House of Commons Women and Equalities Committee In 2018 the House of Commons Women and Equalities Committee published its report, ‘Sexual Harassment in the Workplace’.40 The report was prompted by the powerful mobilization through the #MeToo movement and the widespread publicity around harassment in the entertainment industry. In the words of the report, the inquiry was ‘a way to harness the momentum of #MeToo to produce practical recommendations for change’.41 The Select Committee adopted a narrower focus than the ILO, framing its proposals specifically around sexual harassment rather than a more general concern with harassment and violence at work. The report identified the range of existing legal measures to address sexual harassment in the workplace. This includes specific provisions under the Equality Act 2010.42 The Committee also noted the existence of criminal offences for more serious forms of sexual harassment, such as the PHA, various offences under the Sexual Offences Act 2003 and the Criminal Justice and Courts Act 2015.43 Despite these prohibitions in civil and criminal law, the Committee nevertheless identified sexual harassment as an endemic problem in UK

40 Women and Equalities Committee, Sexual Harassment in the Workplace:  Fifth Report of Session 2017–​19 (HC 2017–​19 725)  https://​publications.parliament.uk/​pa/​cm201719/​cmselect/​cmwomeq/​725/​725.pdf accessed 3 August 2019. This Committee is currently engaged in an inquiry into the use of non-​disclosure agreements (NDAs) in discrimination cases. 41 ibid 5. 42 Equality Act 2010, s 26. 43 WEC, Sexual Harassment (n 40) 5–​6.

162  Criminality at Work workplaces which required an urgent regulatory response.44 In contrast to the current ILO approach, which adopts a parsimonious approach to criminal sanctions, the criminal law occupies a more prominent role in the Select Committee’s reform proposals. It is configured around a regulatory approach to criminalization, assessing the utility of the criminal sanction as an instrument for securing behavioural change. This is reflected in two main features of its approach. First, the centerpiece of the Select Committee’s proposals is a mandatory duty on employers to take reasonable steps to protect workers from harassment and victimization in the workplace. This builds upon an earlier proposal to that effect by the Equality and Human Rights Commission. The Committee identified appropriate regulatory models for such a duty in the money-​laundering and personal data contexts, where regulators are able to impose a range of powerful civil and criminal sanctions for breaches, including unlimited fines and professional disqualification for individual offenders.45 According to the Committee, such a duty would alleviate the burden on individuals in tackling workplace harassment. It would also respond to the wider problems of organizational culture, social norms, and power structures that cannot be addressed effectively by personalized perpetrator liability. This duty would be supported by a code of practice providing clear guidance to employers on the steps needed to avoid liability. Like the ILO approach, this avoids the limitations of the personalized approach of the PHA. In contrast to the ILO approach, the Committee draws a much stronger connection between the employer duty and criminal sanctions. However, this link is justified from within the domain of regulation theory rather than the domain of criminal law and ‘real crime’. That is to say, the Committee is more preoccupied with the criminal sanction’s likely effectiveness in prompting organizational change, rather than the principled attribution of blame and censure to culpable wrongdoing. Secondly, the Committee’s proposed approach to the criminal–​civil law interface reveals a complex isomorphic interaction between these two bodies of law, with the criminal law reshaping the civil law in potentially productive ways. This provides an interesting contrast with the negative ‘drag’ effect of the criminal law on the civil law in the PHA. The Committee adopts a critical view of the individual enforcement paradigm. The legal process can be arduous, expensive, and traumatizing for individual complainants. These problems are exacerbated for the most precarious workers, who are also more likely to be exposed to sexual harassment. Concerns about legal costs, low compensation awards, and time limits on complaints compound these difficulties.46 We have already encountered a potential regulatory response to such problems, notably the technique of ‘deregulatory criminalization’ under health and safety law, channelling enforcement through an exclusive criminal law mechanism. The Committee avoids this crude and opportunistic strategy, and instead adopts what we term an isomorphic reform strategy. It is isomorphic because the individual civil enforcement paradigm is improved through the imitation of criminal law techniques. For example, the Committee recommends the use of special measures in court, such as a prohibition on defendant cross-​examination or complainant anonymity, imitating measures in the criminal justice system.47 At the remedial

44

ibid ch 1. ibid 12. 46 ibid  28–​33. 47 ibid 33. 45

The Criminalization of Harassment and Abuse  163 level, the Committee recommends the introduction of punitive damages for individual claims.48 It also proposes the reintroduction of powers to make wider recommendations to the employer in discrimination cases, so that individual litigation could prompt wider organizational changes to address ‘toxic organizational culture’.49 This would be convergent with the regulatory focus of the proposed employer duty to take reasonable steps to prevent harassment, enforced through the criminal law. Finally, there is extensive discussion of the misuse of non-​disclosure agreements (NDAs) in harassment cases, particularly in settlement agreements. In particular, the report addresses the inclusion of gagging clauses in NDAs which, though unenforceable, are intended to deter the complainant’s cooperation with public criminal law enforcement.50 To ensure that individual civil litigation does not subvert the criminal process and obstruct public efforts to address systemic organizational problems, the Committee also proposed the introduction of a new criminal offence where an employer or professional adviser propose ‘a confidentiality clause designed or intended to prevent or limit the making of a protected disclosure or disclosure of a criminal offence’.51 We think that the Committee’s isomorphic framing of the criminal–​civil interaction is extremely promising, although the prospects for extensive legislative action appear to be slim at the current time.52

E.  Justifying a Role for Criminalization in Harassment Wrongs: The Role of Human Dignity Dignity has an architectonic role in the law on harassment. The violation of dignity is identified as a core element of the civil wrong of harassment in section 26 of the Equality Act 2010. Often, dignity is used as a basis for framing the character of the wrongs and harms of workplace harassment. For example, Rosa Ehrenreich Brooks introduces the idea of sexual harassment as a ‘dignitary harm’: [B]‌y humiliating, intimidating, tormenting, pressuring, or mocking individuals in their places of work, sexual harassment is an insult to the dignity, autonomy, and personhood of each victim; such harassment violates each individual’s right to be treated with the respect and concern that is due to her as a full and equally valuable human being.53

Or consider Elizabeth Anderson’s account of harassment as a ‘dignitary injury’: ‘On the dignity account, what is wrong about sexual harassment is that it coerces, threatens, torments, intimidates, insults, humiliates, and degrades its victims. These are dignitary injuries, harms to an individual’s standing as a person.’54 48 ibid 30. 49 ibid 36. 50 ibid 41. 51 ibid 45. 52 See Women and Equalities Committee, Sexual Harassment in the Workplace:  Government Response to the Committee’s Fifth Report of Session 2017–​19 (HC 2017-​19 1801)  accessed 3 August  2019. 53 Rosa Ehrenreich Brooks, ‘Dignity and Discrimination: Toward a Pluralistic Understanding of Workplace Harassment’ (1999) 88 Georgetown Law Journal 64, 16. 54 Elizabeth Anderson, ‘Recent Thinking about Sexual Harassment: A Review Essay’ (2006) 34 Philosophy and Public Affairs 284, 292.

164  Criminality at Work The value of dignity’s role in harassment law is contested. For example, the individualized focus on dignity might obscure the structural inequalities that pervade the practices of harassment.55 Anderson has also criticized the way in which dignity deflects attention from ‘the sexist content of nearly all harassment, it individualizes and depoliticizes the wrong’.56 For the purposes of this chapter, we propose to bracket these disagreements about dignity as a basis for a normative account of the wrong of harassment. Instead, we are interested in exploring the specific role of dignity in understanding the overall shape of the criminal law as a regulatory response to harassment in the workplace. We suggest that the principle of human dignity supports: (i) the criminalization of employer safeguarding duties; (ii) maintaining and supporting the opportunities for individual workers to seek redress for rights violations in the ordinary courts; (iii) dispensing with enquiries into victim consent in liability for workplace abuse.

1.  Employer Duties and the Criminal Law Both the ILO and the Women and Equalities Select Committee have identified employer duties as central to progressive legal reform. In so doing, they are recognizing and applying a particular modern understanding of the very idea of ‘the employer’ which it is most important to recognize. This modern understanding of the idea of the employer stands in contrast to an earlier conceptualization of the employer as a single (stereotypically male) human being—​as the ‘master’ who famously figures as the player of the employing role in the ‘Law of Master and Servant’. The contrasting modern understanding, conditioned by historical change in the patterns of productive activity and work relations, envisages the employing role as being archetypally played by an organization or an enterprise. The organization or enterprise may be large or small, but even if it is very small the placing of an employer duty upon it assigns an essentially institutional and structural character to that duty and indeed to the wrong or the harm which it is the object of that duty to avoid or to control. This marks an important departure from the personalized model of duties and of wrongs which underlies the PHA. We suggest that this personalized formulation of the wrong is aligned with the understanding of abuse that is perpetrated in family relations. Indeed, we would regard the PHA as an important precursor to the public recognition of domestic abuse as a distinctive criminal wrong.57 This non-​personalized formulation of the duty marks a regulatory divergence between domestic abuse and workplace abuse. Is this divergence desirable? The spheres of work and family are related in important ways. The articulation of those linkages has been central to feminist work on sexual harassment. As Reva Siegel has argued, ‘the woman’s rights movement had begun to analyze the political economy of heterosexuality in a way that took as structurally interconnected the institutions of marriage and market’.58 In the hands of feminist legal theorists such as Catharine MacKinnon, this structural perspective on sexual harassment exposed the deep continuities between domination 55 Susanne Baer, ‘Dignity or Equality? Responses to Workplace Harassment in European, German, and US Law’ in Catharine A MacKinnon and Reva B Siegel (eds), Directions in Sexual Harassment Law (Yale University Press 2003). 56 Anderson (n 54) 310. 57 See Victor Tadros, ‘The Distinctiveness of Domestic Abuse: A Freedom Based Account’ (2005) 65 Louisiana Law Review 989. 58 MacKinnon and Siegel, Directions (n 55) 8.

The Criminalization of Harassment and Abuse  165 in the family and work relations: ‘men extracting sex from economically dependent women reiterate a coercive relationship that organizes heterosexual relations in marriage and the market both’.59 From a different perspective, recent work in republican political theory has further reinforced the salience of treating the spheres of work and family as engaging common normative concerns.60 Work relations and marital relations are both sites of domination in civil society, and the eradication of domination from work life and family life is a focus of republican public policies. These continuities may have provided an implicit normative justification for the extension of the PHA model, from the regulation of domestic abuse (in the form of stalking) to the regulation of workplace harassment and abuse. Is this recent move towards employer duties a mistake? We do not think so. In contrast to domestic abuse, which is an interpersonal wrong perpetrated in intimate relationships, the paradigm case of workplace harassment is triadic. It certainly involves interpersonal abuse by a harasser of a victim of harassment. It also encompasses the employer. The employer enjoys ‘practical authority’ to direct the running of the enterprise through the organizational rules constituted by the employment contract.61 These organizational rules can provide a context where interpersonal abuse either propagates or is eradicated. This marks a critical difference between work relations and family relations. It would be inconceivable to describe the marriage relation of husband and wife as ‘a relation of practical authority, which is constructed and confined by rules’.62 The employer is an appropriate object of regulation in this triad because of its extensive labour market power, and its capacity to eradicate the structural conditions of ­harassment. This dimension of authority and power means that the case for employer criminalization is particularly strong. There are two aspects to this argument. First, and from a more regulatory perspective, the employer’s authority in work relations means that it is better placed to secure positive behavioural changes in the workplace. It can exercise its authority in positive ways to induce cultural, organizational, and structural changes in its employment arrangements. There is no party that obviously corresponds to this in the domestic abuse context, which in its paradigm case is a bilateral wrong between two individual persons. Secondly, and from a more normative dignitarian perspective, the employer’s authority means that its actions and omissions poses a far greater risk to the worker’s dignity interests than those of other private actors, such as co-​workers. The potential of the employer to cause significant dignitarian injuries makes it a stronger candidate for coercive regulation than private individuals. Let us begin with the regulatory argument. Lacey has emphasized the importance of designing criminalization so as to concentrate upon the actors most capable of positive behavioural change, and so elicit and draw upon their best capacities to bring about such a positive change.63 In this sense, it seems to us to be potentially profitable to use regulatory criminalization to impose institutional safeguarding duties upon employing enterprises rather than using the criminal law purely to punish or deter individual offenders. She also emphasizes the importance of designing the interventions of criminal law so that they form the 59 MacKinnon and Siegel, Directions (n 55) 17. 60 See, eg, the approach in Robert S Taylor, Exit Left (OUP 2017), which has chapters on ‘family’ and ‘market’. 61 Hugh Collins, ‘Is the Contract of Employment Illiberal?’ in Hugh Collins, Gillian Lester, and Virginia Mantouvalou (eds), Philosophical Foundations of Labour Law (OUP 2019) 54–​55. 62 ibid 67. 63 Compare Lacey (n 28) 151, where she refers to ‘the regulatory system’s approach to its subjects’ capacities’, saying that ‘[t]‌his is, perhaps, yet more important than the issue of motivational assumptions for modern criminal law, which proclaims as one of its central doctrines that the subjects of criminalization are . . . self-​determining agents capable not only of rational choice but of taking responsibility for their choices . . .’.

166  Criminality at Work apex of the pyramid of institutional regulation and social regulation of a diffused and pluralistic character.64 In this respect also, the use of the criminal law to impose institutional safeguarding responsibilities might seem a more promising approach than the pursuit of individual perpetrators for ‘real crimes’ of bullying in the workplace, as being more likely to bring about change in institutional practice and culture. This marks an important difference from the case of domestic abuse. Now let us consider the dignity argument. From a dignitarian perspective, Khaitan has endorsed the moral importance of the distinction between vertical and horizontal relationships in enforcing duties to respect dignity.65 In so doing, he builds on a distinction drawn by Margalit between the ‘decent society’ (‘whose institutions do not humiliate people’) and the ‘civilized society’ (‘whose members do not humiliate one another’).66 According to Margalit, there is a much stronger case for using coercion to enforce the ‘decent society’ over the ‘civilized society’. Why might this be so? Margalit provides two reasons. First, restricting humiliating expression by natural persons acting in a private capacity will engage the fundamental human freedoms of those persons, such as their freedom of expression. By contrast, institutions would have a much weaker claim to fundamental human rights’ protections because they are not persons. Secondly, the harm of institutional humiliation is more damaging and pervasive than the harms caused by personal acts of humiliation.67 Institutional humiliation by public institutions has a symbolic dimension that personal acts of humiliation lack. To be demeaned in this way is more corrosive of the public standing of citizens than acts of humiliation by private citizens. It is therefore a stronger candidate for coercive restriction. In sum, there is a stronger case for legislating for a ‘decent society’, because the harms of institutional humiliation are more corrosive and the countervailing rights-​based arguments against criminalization are far weaker. Where should employers be positioned in this mapping? Labour law has always challenged the distinction between vertical and horizontal relations. The distinction between the ‘decent society’ and the ‘civilized society’ is a difference of degree, and we would locate employers towards the former end of the spectrum. Employers wield significant economic power, and consequently the dereliction of safeguarding duties can have catastrophic effects on the dignity of workers. The personification of employers, as in the old terminology of ‘master and servant’, is a simulacrum that obscures the reality of employers as complex institutional entities. A decent labour market, offering a wide range of employment opportunities free from abuse and harassment, should be regarded as an institutional element of the ‘decent society’. Employers occupy a different position in the basic institutional structure, compared with the individual workers who engage in the humiliating treatment of co-​workers. The humiliation caused by derelict employers is more corrosive, akin to institutional humiliation in the public sphere. Employers also have a more attenuated claim that their human rights have been infringed by criminalization, because they are organisations rather than real persons. In sum, the institutional harms of employers are far more serious, and the 64 Compare Lacey (n 28) 150 where she refers to John Braithwaite’s theory of the regulatory pyramid and comments that ‘in terms of this analysis, one could say that the most important regulatory work in relation to criminalization in fact goes on below the bottom of the pyramid, consisting in the personal internalization of norms or general receptivity to peer pressure which secures voluntary compliance . . . The pyramid, perhaps, is focused on the tasks of monitoring and enforcement, rather than that of standard-​setting.’ 65 Tarunabh Khaitan, ‘Dignity as an Expressive Norm:  Neither Vacuous Nor a Panacea’ (2012) 32 Oxford Journal of Legal Studies 1, 16. 66 Avishai Margalit, The Decent Society (Harvard University Press 1996) 1. 67 ibid 171–​72.

The Criminalization of Harassment and Abuse  167 countervailing human rights arguments far weaker, when considering criminal sanctions for employers. For individual harassers, by contrast, the arguments for and against criminalization are more finely balanced. Accordingly, we think that dignity considerations provide a normative prop to the regulatory arguments supporting the use of criminal law to enforce the employer’s safeguarding duties. This provides something of a response to Anderson’s and Baer’s concerns about dignity ‘individualizing’ or ‘depoliticizing’ the wrong of harassment. On Margalit’s approach, attacks on dignity can be readily understood as harms possessing a public dimension, rooted in unjust political and economic structures. In this way, dignity can provide support to the general approach of the Women and Equalities Select Committee in linking employer duties to a more punitive approach. Criminalization need not be aligned with an over-​personalized framing of criminal wrongs.

2.  Dignity and the Relationship between Criminal and Civil Law Harassment wrongs are often presented as assaults on human dignity. Much of the reported case-​law on harassment presents narratives where a worker is demeaned, degraded, or humiliated so that her social standing amongst her peers is attacked and undermined. Given the particular character of this wrong, the law’s response to dignity violations must be particularly sensitive to considerations of human dignity. The legal response must not compound the original wrong by further undermining the victim’s dignity. Ideally, the law should provide ways of reaffirming the dignity of the victim through its own processes and remedies. Jeremy Waldron’s work on human dignity provides some important insights into the connections between dignity, social rank, and the structure of the legal process.68 According to Waldron, dignity should be understood as denoting the high and noble rank of all human beings. In this way, there is a ‘levelling up’ to aristocratic rank so that all human beings are assigned the same high social standing. The legal system performs a critical role in the constituting of noble rank on equal terms to everyone. There are three dimensions that are particularly relevant to our discussion here. First, Waldron identifies the ‘dignity of being a right-​bearer’:  ‘Right-​bearers stand up for themselves; they make unapologetic claims on their own behalf; they control the pursuit and prosecution of their own grievances.’69 Secondly, the requirement of a public hearing and due process is another way in which the legal system respects dignity. Again, as Waldron explains, the legal process contributes to dignity in the following ways: [Litigants] have an opportunity to make submissions and present evidence, and confront, examine, and respond to evidence and submissions presented from the other side. Not only that, but both sides are listened to by a tribunal that is bound to respond to the arguments put forward in the reasons that is eventually gives for its decision.70

The forum of the public court also enables the participants to shape the development of the law itself, by presenting arguments based upon the best interpretation of the law’s demands.71 Finally, Waldron identifies legal representation as an integral element of the legal

68

Jeremy Waldron, Dignity, Rank, & Rights (OUP 2015). ibid  49–​50. 70 ibid 54. 71 ibid 55. 69

168  Criminality at Work system’s respect for dignity.72 Access to representation gives a voice to those who might otherwise be bewildered, cowed, or confused by the technicalities of legal doctrine and the legal process. It is particularly important for those in a situation of social precariousness, so that they can regain their dignified bearing in the courtroom even where that bearing has been trampled upon in the workplace. How can legal responses to dignitarian wrongs ensure that dignity is at least upheld in the ensuing legal response? We suggest that Waldron’s dignity argument indicates a need for caution around criminalization. Criminalization undoubtedly has an important role in securing respect for human dignity. However, criminalization should not displace the ‘dignity of the right-​bearer’ in exercising normative power over the vindication of her own rights. The criminal process can disempower the victim in various ways. The decision whether to pursue a prosecution is in the hands of a state agency rather than the victim herself. The difficulties with this are especially acute when enforcement agencies experience significant resource constraints on their work.73 In the criminal trial, the victim loses control of the legal process. She becomes a witness rather than a right-​bearer. It is important, therefore, that the criminal law operates in auxiliary mode, punishing and deterring culpable behaviour that would infringe the worker’s right not to be treated in a degrading or humiliating fashion. It should nevertheless be possible for the worker as a dignified bearer of rights to seek redress in the civil courts for violations of her individual rights. This would exclude the strategy of ‘deregulatory criminalization’ as under UK health and safety law. It is not sufficient for the legal system to provide formal access to those rights. As the Select Committee observed, the burden of individual litigation has been oppressive for the most vulnerable workers. The discourse of aristocratic dignity will sound rather hollow to workers on zero-​hours contracts who have been sexually harassed and where they do not have public support to pursue their legal claims. Substantive respect for the dignity of the precarious necessitates an array of procedural reforms to support them in their dignity as right-​bearers. This should include at a minimum: (i) legal aid and access to legal representation in civil litigation; (ii) ensuring that NDAs are strictly regulated so that they are only enforceable where there is genuine consent, the worker has received advice from an independent legal representative, and the scope of confidentiality does not obstruct the worker’s cooperation with any criminal investigation of wrongdoing; (iii) finally, the use of mandatory private arbitration clauses should be prohibited in harassment cases. In the recent US case of Epic Sys Corp v Lewis, a majority of the US Supreme Court upheld the preclusion of group claims by mandatory arbitration clauses.74 The system of mandatory private arbitration is antithetical to respect for dignity. The proceedings and judgments are private, so that ‘the public has no way of knowing whether the laws are being faithfully enforced. The law will have disappeared from public view.’75 This undermines the ability of the right-​ holder to enforce or shape the law in an authoritative court. It also blocks an important opportunity for her to regain her public standing through public institutional recognition. In conjunction with this secrecy, the preclusion of group claims impedes the ability of workers to achieve systemic reforms through class actions. This imposes severe constraints on the 72 ibid  61–​62. 73 WEC, Sexual Harassment (n 40) 19–​20. 74 138 S Ct 1612 (2018). 75 Matthew Finkin, American Labor and the Law: Dormant, Resurgent, and Emergent Problems (Wolters Kluwer 2019) 59.

The Criminalization of Harassment and Abuse  169 solidaristic potential of civil litigation, which we would regard as being of overwhelming importance in the harassment context.76 In our view, respect for dignity entails that mandatory arbitration should be prohibited in harassment cases.77

3.  Dignity and the Role of Consent Individual perpetrators of harassment may be liable under a complex web of civil and criminal provisions. In addition to the PHA, there may be personal liability under the Equality Act 2010, section 26 which addresses the wrong of harassment separately from direct discrimination. The most serious instances of harassment may be addressed through the general criminal law. Where there is sexual touching, this could fall within the scope of sexual assault under section 3 of the Sexual Offences Act 2003. Serious cases of violence causing injury could fall within the scope of various offences under the Offences Against the Person Act 1861, such as assault occasioning actual bodily harm (section 47), malicious wounding or infliction of grievous bodily harm (section 20), or wounding or causing grievous bodily harm with intent to do some grievous bodily harm (section 18). Questions of consent are central to many of these different forms of liability. For example, harassment under section 26 of the Equality Act 2010 specifies that the relevant conduct must be ‘unwanted’. For the offence of sexual assault, V must not consent to the touching and D must lack a reasonable belief in V’s consent. Non-​consent is integral to the criminal wrong. For offences against the person involving the infliction of injury, consent to injury sometimes operates as a defence to criminal liability where the situation falls within a legally recognized category. One such category which will sometimes be relevant to workplace violence is ‘horseplay’. Accordingly, legal liability will often turn upon a determination of the victim’s consent. What is the appropriate role of consent in a reformed law of workplace harassment? From an enforcement perspective, unsympathetic personal evaluation is often apt to be the fate of those claiming to have been the victims of workplace bullying or harassment where questions of consent are central to liability.78 In Munchkins Restaurant Ltd v Karmazyn, for example, waitresses were exposed to intrusive sexualized questioning by their employer, required to wear short skirts at work, and pornographic material had also been shown to the waitresses.79 The employer had argued that the sexualized badinage had been initiated or welcomed by the workers, and that their acquiescence in the behaviour provided evidence that this was the case. Although the employer lost the appeal, issues of consent were central to the case. Should consent be retained as an element in the liability test for harassment? Dan-​Cohen has argued in favour of a dignity principle, that ‘the main goal of the criminal law is to defend the unique moral worth of every human being’, as an alternative to the 76 Protests in the wake of #MeToo at Google and Facebook has led to the retraction of mandatory arbitration clauses in sexual harassment cases at those companies: see Moira Donegan, ‘Why Can Companies Still Silence us with Mandatory Arbitration?’ The Guardian (8 January 2019) accessed 30 April 2019. 77 We would favour a system of anonymity for harassment complainants in civil litigation, along the lines suggested by the Select Committee. 78 Indeed, this kind of normative evaluation has already been a feature of the PHA where consent is not an element of the legal definition: see Barmes (n 1) 156–​60. Consent may be expected to intensify this feature of adjudication. For a critical discussion of ‘unwelcomeness’ in harassment law, see Jane Larson, ‘Sexual Labor’ in MacKinnon and Siegel, Directions (n 55). 79 Munchkins Restaurant Ltd v Karmazyn [2010] EWCA Civ 1163.

170  Criminality at Work liberal harm principle in determining the scope of the criminal law.80 Under the harm principle, consent is central to determining whether there should be criminalization.81 Dan-​ Cohen invites us to consider the judgment in State v Brown to test our intuitions about the role of consent.82 In Brown, the defendant beat his alcoholic wife, when she drank alcohol, in order to deter her drinking. This was alleged to be pursuant to an agreement between the parties. The court rejected the defence of consent.83 From the perspective of the harm principle, the decision would seem to be puzzling: volenti non fit injuria is a liberal principle. Criminalizing this consensual violence seems to impede the victim’s autonomy interests and her welfare interests, because she is prevented from participating in an arrangement that she willingly assented to and which she regards as welfare-​enhancing. However, Dan-​ Cohen invites us to consider Brown from the perspective of the dignity principle:  ‘The fact that physical violence does ordinarily hinder both welfare and autonomy is reason enough to render it a blatant manifestation of disrespect.’84 This ‘expressive significance’ would be communicated even in situations, like Brown, where the typical markers of violence (violation of autonomy and welfare) are not present. The treatment of the wife is demeaning and degrading, and it is uncivilized for the criminal law to defer to consent in these circumstances. For our own purposes, we think that R v Aitken might require reconsideration in the light of Dan-​Cohen’s dignity principle.85 In Aitken, an RAF officer suffered serious burns after he had been doused with white spirit and ignited. According to the court, the fellow officers ‘took hold of, and grabbed him. He resisted, but because of the amount which he had drunk he struggled only weakly . . . Aitken that night said that Gibson had gone up “like a torch”.’86 Gibson suffered 35 per cent burns of a life-​threatening nature. According to the Court of Appeal, the issue of consent should have been put to the jury by the trial judge and the convictions under section 20 for malicious infliction of GBH were quashed. ‘Rough and undisciplined horseplay’ was an exceptional category where a valid consent could be operative as a defence to criminal liability under section 20. Further, Gibson’s continued presence at the social event might be construed as evidence of consent to the injuries that resulted. We are sceptical about the defensibility of this ‘horseplay’ category in offences of violence, and the requirement under section 26 of the Equality Act 2010 that conduct should be ‘unwanted’. The very idea of ‘horseplay’ as an exceptional exculpatory category in the criminal law serves to reinforce the norms of toxic masculinity that operate in exclusionary ways in many workplaces. Where a worker has sustained serious injuries in testosterone-​ and alcohol-​fuelled japes in the workplace, or conduct has the purpose or effect of ‘violating B’s dignity’ or ‘creating an intimidating, hostile, degrading, humiliating or offensive environment’ as under section 26 of the Equality Act 2010, there should be liability regardless of consent. The dignity principle provides a strong argument for not deferring to the victim’s consent. In many cases, apparent consent is likely to be vitiated by the existence of power asymmetries in the labour market. In rare cases where such treatment might be

80

Meir Dan-​Cohen, Harmful Thoughts (Princeton University Press 2002) 150. See, eg, Joel Feinberg, The Moral Limits of the Criminal Law Volume 1: Harm to Others (OUP 1987) 115–​17. 82 364 A 2d 27 (NK Super Ct Law Div 1976), aff ’d, 381 A2d 1231 (NJ Super Ct App Div 1977). 83 Dan-​Cohen (n 80) 154. 84 Dan-​Cohen (n 80) 162. 85 R v Aitken [1992] 1 WLR 1006. 86 ibid 1009–​10. 81

The Criminalization of Harassment and Abuse  171 welcomed by victims, this compounds rather than ameliorates the degradation. In Dan-​ Cohen’s terms, a case like Aitken represents an instance of expressive disrespect, and public institutions and laws should have no part in enabling that. The law should never facilitate human degradation.

F.  Conclusion: Bargaining in the Shadow of the (Criminal) Law Workplace harassment is one of the most challenging regulatory issues of our time. We have argued that the criminal law has a legitimate and important role in eradicating all forms of abuse and harassment at work. Comparative studies should examine the existing models of criminal regulation to identify some of the virtues and vices of specific models of criminalization.87 Recently, progressive reform strategies have positioned the employer’s safeguarding duties, on a health and safety model, as central to this next wave of harassment law. This would transcend the limitations on effective enforcement where harassment is conceptualized as a personalized wrong. This has beset the existing law on harassment, especially under the PHA. Drawing upon the dignity literature, we have identified three areas for reform attention: (i) criminalization of employer safeguarding duties; (ii) ensuring that criminalization does not impede the opportunities for individual workers to seek redress for rights violations in the ordinary courts; (iii) dispensing with enquiries into victim consent where a victim experiences serious injury from workplace ‘mobbing’ or where there is harassment under section 26 of the EA. Whatever the shape of legal reform on paper, it is vital that this is translated into real cultural and organizational change in the workplace. As Lauren Edelman has demonstrated, workplace harassment policies can easily become an exercise in defensive ‘window dressing’, designed to block legal liability in harassment claims rather than to prompt deep and enduring change in the workplace.88 Genuine change must be led by the inclusion of audible and diverse voices in the management of workplaces at all levels.89 Lizzie Barmes has also emphasized the need for ‘widening participation in implementation and enforcement as an emancipatory legal device’.90 This will require new forms of collective bargaining and union organizational strategies, and the forging of alliances with social movements that have grown out of the #MeToo mobilization. We suggest that the next wave of activism should explore the possibilities of ‘bargaining in the shadow of the criminal law’. In a recent report, the NASUWT, the Teachers’ Union sought judicial review of a decision not to prosecute pupils who had engaged in ‘upskirting’ of female teachers.91 This led to a successful conviction of a former pupil. Unions must explore the potential of litigation as a point of 87 See, eg, the chapters on criminal harassment laws in Israel and France: Orit Kamir, ‘Dignity, Respect, and Equality in Israel’s Sexual Harassment Law’ (561) and Abigail Saguy, ‘French and American Lawyers Define Sexual Harassment’ (602) in MacKinnon and Siegel, Directions (n 55). 88 Lauren B Edelman, ‘How HR and Judges Made It Almost Impossible for Victims of Sexual Harassment to Win in Court’ Harvard Business Review (22 August 2018) accessed 30 April  2019. 89 ibid. See also Marion Crain and Ken Matheny, ‘Sexual Harassment and Solidarity’ (2019) 87 The George Washington Law Review 56. 90 Barmes (n 1) 261. 91 Richard Adams, ‘Female Teachers Need Protection from Sexual Harassment, Says Union’ The Guardian (21 April 2019) accessed 30 April 2019.

172  Criminality at Work leverage in the criminal process, and to reflect on the possibilities for a new role as ‘private prosecutors’ in appropriate cases. This would give workers a stronger independent voice in the criminal process. This would enhance their dignity. In turn, public agencies should explore ways of working with unions collaboratively on enforcement, so that legal reform leads to enduring cultural change. In the next wave of harassment law, the criminal law has an important part to play in securing dignity at work for all workers.

9

Sex, Work, and Criminalization Michelle Madden Dempsey

A.  Introduction In addressing the topic of sex, work, and criminalization, a preliminary question presents itself: is commercial sex properly regarded as work?1 That is, is the commercial sex trade simply another labour market in which some people seek to earn money in exchange for providing a service—​just like accountancy?2 A secondary question follows: if there are salient distinctions between commercial sex and work, what implications might these distinctions have regarding the proper role of criminal law in regulating or prohibiting commercial sex? This chapter examines the role of criminalization in the context of commercial sex. The examination proceeds in three parts. First, as a preliminary matter, the chapter sets out a framework for conceptualizing various kinds of commercial sex, in hopes of bringing some clarity to the discussion that follows. Second, the chapter explores three key philosophical issues that arise when considering sex as a form of labour. Third, the chapter provides a normative account of criminalization that illuminates four perspectives on the proper role of the criminal law when it comes to commercial sex.

B.  Conceptualizing Commercial Sex: From Sex Work to Modern Slavery This section offers a taxonomy of various ways in which commercial sex is often conceptualized in modern political and legal discourse. Note that this section does not attempt to offer an original analysis of commercial sex or criminal offences involving commercialized sex acts.3 Rather, it merely sets out the various ways in which commercial sex has been understood by others, without defending any conclusions as to which account best reflects the relevant phenomena. The hope is that by surveying and stipulating the meaning of various labels used to refer to commercial sex, this section will clarify some relevant debates, albeit without resolving them.

1 For a strongly negative answer to this question, see Janice G Raymond, Not a Choice, Not a Job: Exposing the Myths about Prostitution and the Global Sex Trade (Potomac Books 2013). 2 For a suggestion that it is, see ‘Cheltenham Lib Dem Chair Quits after Sex Worker Debate’ BBC News (19 September 2016) accessed 24 August 2018. 3 For an illuminating analysis along these lines, see Stuart Green, ‘What Counts as Prostitution?’ (2016) 4 Bergen Journal of Criminal Law and Criminal Justice 65. Michelle Madden Dempsey, Sex, Work, and Criminalization In: Criminality at Work. Edited by Alan Bogg, Jennifer Collins, Mark Freedland QC, and Jonathan Herring, Oxford University Press (2020). © Alan Bogg, Jennifer Collins, Mark Freedland, and Jonathan Herring. DOI: 10.1093/oso/9780198836995.003.009

174  Criminality at Work

1. Commercial Sex As an umbrella term, this chapter will use the term commercial sex to refer to the exchange of sexual services for valuable consideration.4 For simplicity, the understanding of sexual services will centre on paradigmatic acts such as vaginal, anal, or oral sexual intercourse, rather than more indirect services such as phone sex, webcamming, etc.5 Similarly, the exchange of valuable consideration will be understood to include typical methods of commercial exchange (eg money), rather than jewellery, holidays, expensive dining, etc.6 Given the politically charged debates surrounding this topic, and the resulting impossibility of finding universally acceptable labels in this context, it is hoped that commercial sex will serve as an adequately neutral descriptor of the phenomena, without presupposing anything about the circumstances under which any given commercial sex exchange occurs.7

2. Sex Work The term sex work will be used to refer to commercial sex occurring under circumstances in which the persons providing sexual services: (1) regard the commercial sex as a form of labour; and (2)  prefer to engage in commercial sex over alternative options, despite (3) having an adequate range of valuable alternatives for engaging in forms of labour that are consistent with human flourishing. Sex work may involve independent contractors and/​ or people working with a manager or engaging in commercial sex in brothels. The first two criteria used to define sex work above incorporate subjective criteria, while the third depends on objective considerations. First, persons providing sexual services must subjectively regard themselves as engaging in a form of labour.8 Second, persons providing sexual services must have a subjective preference to engage in commercial sex as labour. Third, this subjective preference must be formed under conditions in which the person has an objectively adequate range of objectively valuable options to engage in alternate forms of labour that are consistent with an objective account of human flourishing. Thus, according to this understanding, whether one is engaged in sex work is not merely a matter of one’s subjective interpretation; rather, it incorporates objective criteria consistent with what Andrea Veltman calls meaningful work.9 Whether work is meaningful depends not merely on whether it is consistent with the exercise of personal autonomy. Rather, the key dimensions of meaningful work are multifaceted and include: 4 This definition roughly tracks the ‘Commodification of Sexual Activity’ in Canada’s recently revised statutory framework, Protection of Communities and Exploited Persons Act 2014. The word prostitution has often been used to refer to the umbrella category which is, here, referred to as commercial sex. Given the negative connotations historically associated with the term prostitution and its variants (eg ‘prostitute’), this chapter will avoid its use, aside from when quoting it in the work of others. 5 Teela Sanders and others, Internet Sex Work: Beyond the Gaze (Palgrave MacMillan 2018) 17–​18. 6 As such, phenomena such as ‘sugar dating’ may either involve marginal cases of commercial sex, or more central cases, such as those which include an explicit exchange of sex for money or tuition payments. Natalie Ktena, ‘I’m a Sugar Baby to Pay My Way Through Uni’ BBC Three (5 October 2018) accessed 10 October 2018. 7 Emily St Denny, ‘“The Personal is Political Science”: Epistemological and Methodological Issues in Feminist Social Science Research on Prostitution’ (2014) 16 Journal of International Women’s Studies 76. 8 As such, those who exchange sexual favours in the context of a financially beneficial arrangement would not be regarded here as engaged in sex work. See (n 6). 9 Andrea Veltman, Meaningful Work (OUP 2016) 105–​41.

Sex, Work, and Criminalization  175 [D]‌eveloping or expressing talents and capabilities, earning recognition, acquiring or reinforcing virtues, . . . [and] integrating elements of a life by reflecting [one’s] deeply held values and commitments. . . thus helping to render the patchwork of life more coherent or consistent.10

No doubt, different forms of work, carried out in various social contexts, can be more or less meaningful according to both subjective and objective criteria. Thus, it is best to avoid drawing too sharp a distinction between sex work, understood here as meaningful work, and commercial sexual exploitation (discussed below). That said, however, in determining whether any particular commercial sex act constitutes sex work, it is important to keep in mind the extent to which meaningfulness as work is constrained by the patriarchal and economic structural inequalities which shape the actual experiences in the commercial sex industry in any given society.11

3.  Commercial Sexual Exploitation The term commercial sexual exploitation will be used to refer to commercial sex occurring under circumstances in which the persons providing sexual services do so under conditions in which they lack an adequate range of valuable alternatives for engaging in other forms of labour that are consistent with human flourishing. As with sex work, commercial sexual exploitation may involve independent contractors and/​ or people working with a manager (here, a role better labelled ‘pimp’) or in a brothel. Importantly, this definition is entirely objective: the subjective preferences of the person providing sexual services does not determine whether the activity counts as commercial sexual exploitation. As such, a person might choose commercial sex as a preferred form of labour, yet still be understood as one who is involved in commercial sexual exploitation rather than sex work. The key to distinguishing sex work from commercial sexual exploitation is wholly objective: does the person providing sexual services have an adequate range of valuable alternatives for engaging in meaningful work? Notably, these criteria are objective in the sense of establishing normative baselines, not merely empirical baselines.12 That is, the appropriate question for determining whether commercial sex counts as sex work or commercial sexual exploitation is whether the person providing sexual services is made worse off relative to where she has a right to be, not relative to what her status quo would be without recourse to commercial sex, or her subjective views about her own entitlements.13 A particularly exploitative form of commercial sexual exploitation is sex trafficking. As defined in the UN Protocol to Prevent, Suppress, and Punish Trafficking in Persons 10 ibid. 11 See discussion at nn (29–​35) below. 12 On normative (moralized) and empirical baselines, see Alan Wertheimer, Exploitation (CUP 1996) 22–​23; Alan Wertheimer, Consent to Sexual Relations (Princeton UP 2003) 166–​67. 13 ‘Exploitation therefore does not necessarily harm its victim in the sense of making her worse off than she would have been, had the exploiter never interacted with her at all. Rather, it makes its victim worse off than she should have been, had she been treated fairly.’ Matt Zwolinski and Alan Wertheimer, ‘Exploitation’ Stanford Encyclopedia of Philosophy (2016) accessed 24 July 2018.

176  Criminality at Work (Palermo Protocol), commercial sexual exploitation of children (under 18 years of age), or of adults which occurs by any of the following means constitutes sex trafficking: [T]‌hreat or use of force or other forms of coercion, of abduction, of fraud, of deception, of the abuse of power or of a position of vulnerability or of the giving or receiving of payments or benefits to achieve the consent of a person having control over another person, for the purpose of exploitation. 14

Importantly, according to the Palermo Protocol definition, there is no requirement that the trafficking victim be transported either across borders or within a jurisdiction—​which marks a distinction from UK law.15 Rather, the key to distinguishing sex trafficking from other forms of commercial sexual exploitation rests on one of two considerations:  (1) whether the victim is a child; or (2) (where the victim is an adult) whether a third party (or parties) has employed any of the illicit means listed for the purpose of exploiting the victim. As such, unlike sex work and other forms of commercial sexual exploitation, sex trafficking (as understood herein) does not include independent contracting for commercial sex, but rather assumes the involvement of a trafficker (eg pimp, brothel-​keeper, etc).16

4. Modern Slavery The term modern slavery, often used in political advocacy, has gained new prominence in legal discourse in virtue of its adoption into UK law in the Modern Slavery Act 2015. As the Act’s explanatory notes proclaim, ‘[m]‌odern slavery is a brutal form of organized crime in which people are treated as commodities and exploited for criminal gain’.17 Three features of this definition of modern slavery bear noting. First, it seems to presuppose that modern slavery will involve an organized group of exploiters, rather than, say, an individual pimp exploiting a sex trafficking victim.18 Second, the person providing sexual services is conceptualized as entirely without agency: there is no choice amongst adequate valuable options (as in sex work), nor is there any choice amongst inadequate or low-​value options (as in commercial sexual exploitation and some forms of sex trafficking). The victim of modern slavery is conceptualized as a commodity—​a thing being bought and sold, not a person engaging in a commercial sex act.19 Third, modern slavery is integrally related to broader 14 United Nations Protocol to Prevent, Suppress, and Punish Trafficking in Persons, Especially Women and Children, Supplementing the United Nations Convention Against Transnational Organized Crime [2000] UN Doc A/​55/​383, art 3(a). 15 As noted below, UK law regarding sex trafficking emphasizes the salience of transportation and border in conceptualizing sex trafficking. This is so, despite the UK’s ratification of the Palermo Protocol in February 2006. Michelle Madden Dempsey, Carolyn Hoyle, and Mary Bosworth, ‘Defining Sex Trafficking in International and Domestic Law: Mind the Gaps’ (2006) 26 Emory International Law Review 137. 16 In this sense, the definition of sex trafficking adopted here is narrower than some legal definitions which do not require third-​party involvement. Mary Graw Leary, ‘Dear John, You are a Human Trafficker’ (2017) 68 South Carolina Law Review 415. 17 Explanatory Notes to the Modern Slavery Act 2015 [4]‌. 18 While the legal definition of trafficking in section 2 of the Act does not require that exploiters act in a group, the rhetoric surrounding modern slavery and law enforcement priorities suggest the offence is understood as a matter of group criminality. Thank you to Alan Bogg for raising this point. 19 Note that as a matter of legal definitions, the lines between trafficking and slavery are less clear than the framework set out here. In Rantsev v Cyprus and Russia App no 25965/​04 (ECHR, 7 January 2010) 282, the Court found that ‘trafficking in human beings, by its very nature and aim . . . treats human beings as commodities to be bought and sold’, and as such considered it unnecessary to identify whether trafficking constitutes ‘slavery’, ‘servitude’ or ‘forced and compulsory labour’ in order to find that ‘trafficking itself . . . falls within the scope of Article 4 of the Convention’.

Sex, Work, and Criminalization  177 concerns about border crossing and undocumented immigration. Again, according to the explanatory notes, ‘[m]odern slavery . . . is an international problem and victims may have entered the United Kingdom legally, on forged documentation or clandestinely . . .’.20 Addressing modern slavery thus seems more focused on regulating borders and preventing or punishing irregular migration, rather than creating conditions in which all persons have an adequate range of valuable options from which to choose labour that is conducive to their human flourishing.21 Drawing sharp distinctions between these different forms of commercial sex is at once illuminating and dangerous. One benefit of distinguishing these conceptions is that it allows us to consider how commercial sex operates in a given society and thus develop laws and policies that better respond to these realities. For example, if the overwhelming amount of commercial sex involves commercial sexual exploitation, sex trafficking, or modern slavery, then it makes little sense to develop a legal scheme premised on the false assumption that, in fact, most commercial sex is simply sex work. Yet, there are significant risks associated with drawing these kinds of distinctions. Not least of which, there is a danger that drawing sharp lines between the categories may suggest a lack of fluidity in any given person’s experience of commercial sex. For example, someone who engages in sex work at one time might, at another, lack an adequate range of valuable alternatives (and thus fall into commercial sexual exploitation), and may eventually come to be exploited by a third party (thus falling into sex trafficking). Whatever the benefits or drawbacks of such distinctions, however, it is hoped that this survey adds some clarity to the exploration of sex, work, and criminalization that follows.

C.  Commercial Sex as Labour: Three Philosophical Issues This section examines three philosophical issues that arise when viewing sex as a form of labour: commodification of sex; structural inequalities; and adaptive preferences. By philosophical issues, here, I mean to highlight that the examination below is not stipulative (as was the case in the previous section), nor does it directly respond to questions regarding the best legal scheme to adopt, all-​things-​considered, when it comes to commercial sex (as will be the case in the final section). Rather, this section merely unpacks three considerations that might prove useful in undertaking such an all-​things-​considered evaluation.

1.  Commodification of Sex The first philosophical issue centres on the worry that commercial sex—​in whatever conceptualization, from sex work to modern slavery—​involves the commodification of sex.

20 Seemingly as a bit of an afterthought, the passage continues, ‘. . . or they may be British citizens living in the United Kingdom’. Explanatory Notes (n 17) [4]‌. 21 Moreover, the broader legislative framework suggests not only a lack of political will to remedy conditions that create vulnerability to exploitation, but actually creates new vulnerabilities, such as the creation of an offence of ‘illegal working’ in Immigration Act 2016. Virginia Mantouvalou, ‘The UK Modern Slavery Act 2015: A “Splendid Piece of Legislation”?’ UK Labour Law Blog (6 September 2018) accessed 15 October 2018.

178  Criminality at Work This concern is on display most clearly when it comes to modern slavery, which as noted above, conceptualizes the victim as a literal commodity, being bought and sold ‘for criminal gain’.22 But what of sex engaged in for the commercial gain of the person who provides the sexual service? Do such cases still raise concerns regarding commodification? According to Elizabeth Anderson, the commodification argument grounds a strong objection to commercial sex, insofar as commodification ‘destroys the kind of reciprocity required to realize human sexuality as a shared good’.23 Such reciprocity, she argues, is only realized ‘when each partner reciprocates the other’s gift in kind, offering her own sexuality in the same spirit in which she received the other’s—​as a genuine offering of the self ’.24 Hallie Liberto counters these concerns by arguing that commercial sex need not include genuine commodification insofar as it involves merely ‘alienation in a weak sense’, in which the person providing sexual services retains the right to opt out at any time.25 Similarly, Cecile Fabre argues that persons providing sexual services in a commercial sex sell neither their bodies nor themselves—​and while the commercialization of sex might be regarded by some as an intimate experience that would give rise to the kinds of concerns Anderson raises, Fabre argues there is little reason to suppose that commercial sex is universally experienced as such.26

2.  Commercial Sex and Structural Inequalities A second philosophical issue that can helpfully contribute to an all-​things-​considered evaluation of how the law should respond to commercial sex concerns the relationship between commercial sex and structural inequalities. As typically practised, commercial sex is both a cause and consequence of structural inequalities, especially at the intersection of patriarchy and economic inequality. As Debra Satz observes, ‘[p]‌rostitution shapes and is itself shaped by custom and culture, by cultural meanings about the importance of sex and about the nature of women’s sexuality and male desire’.27 Moreover, as the former UN Special Rapporteur on Trafficking observed: Some prostitute-​users actively seek prostituted women and children of different nationalities, races or ethnic groups for the purpose of exploiting these power disparities, engaging in a ‘highly sexualized form of racism’, by which they rationalize to themselves that woman and children of different races, nationalities or ethnic groups are not harmed by sexual exploitation.28 22 Explanatory Notes MSA (n 17) [4]‌. 23 Elizabeth Anderson, Value in Ethics and Economics (Harvard UP 1993) 153–​54. See also, Margaret Radin, Contested Commodities (Harvard UP 1996). 24 Anderson (n 23). 25 Hallie R Liberto, ‘Normalizing Prostitution Versus Normalizing the Alienability of Sexual Rights: A Response to Scott A Anderson’ (2009) 120 Ethics 138. 26 Cécile Fabre, Whose Body is It, Anyway? Justice and the Integrity of the Person (OUP 2006) 154–​85. Similar issues arise in the context of commercialized care work. See Hayes (this volume). 27 Debra Satz, Why Some Things Should Not Be for Sale: The Moral Limits of Markets (OUP 2010) 146. 28 Sigma Huda, ‘Integration of The Human Rights of Women and a Gender Perspective’ UN Doc E/​CN 4/​2006/​ 62 (20 February 2006), quoting Julia O’Connell Davidson and Julia Sanchez Taylor, Child Prostitution and Sex Tourism: Dominican Republic (ECPAT International, Bangkok, Thailand 1996) 16–​17.

Sex, Work, and Criminalization  179 It is worth noting an important distinction between concerns regarding commodification, on the one hand, and structural inequalities, on the other. Commodification arguments focus on the risk of either treating bodies as commodified objects, or the dehumanizing effects of treating an activity so central to our humanity (sexual intimacy) as something to be bought and sold. In both instances, the concern is primarily about the dehumanizing effects of commercial sex. The structural inequality concern, conversely, focuses on the ways in which persons who sell sex—​fully human persons (most often female/​femme persons)—​are subjugated to the will, wishes, desires, and whims of other human beings (most often male persons). As Kate Manne observes, the problem is not so much that patriarchal inequality denies the humanity of women, but rather places them structurally in the role of ‘human giver’ as opposed to ‘human being’.29 ‘A giver’, she explains: [I]‌s . . . obligated to offer love, sex, attention, affection, and admiration, as well as other forms of emotional, social . . . and caregiving labor, in accordance with social norms that govern and structure the relevant roles and relations.30

The structural inequality concern is clearly not limited to commercial sex: it pervades social life. It both constitutes and perpetuates a range of precious work conditions.31 The point here, however, is that commercial sex is both constitutive of and perpetuates structural inequalities, especially those grounded in patriarchal social forms (sex discrimination, sexism, misogyny). In that sense, exploitation in the commercial sexual industry is a result of patriarchal-​precarity that forms the social context for commercial sex work generally. 32 Commercial sex can be perceived as empowering for individual women, given the poor alternatives that are realistically available. (For, at least when sex is commercialized, women get compensated—​and they do so in a way that does not tie them to any individual husband, etc.33) Yet, as Overall observes, ‘engaging in sex work is “buying into the patriarchal version of independence” ’, and thus does nothing to challenge patriarchal social norms according to which women are structurally positioned as human-​givers, in service of men.34 Ultimately, then, the structural inequality concern illuminates the sense in which commercial sex serves to perpetuate, rather than subvert, such inequalities.35

29 Kate Manne, Down Girl: The Logic of Misogyny (OUP 2018). 30 ibid 301. 31 See Jennifer Collins, ‘Exploitation At Work:  Beyond a “Criminalization” or “Regulatory Alternatives” Dichotomy’ (Chapter 5 of this volume) for discussion of various worker-​precarities. 32 See text at (nn 27–​35). 33 As Engels observed, a married woman ‘only differs from the ordinary courtesan in that she does not let out her body . . . as a wage worker, but sells it once and for all into slavery’. Friedrich Engels, The Origin of the Family, Private Property and the State (International Publishers 1970) 134. 34 Christine Overall, ‘What’s Wrong with Prostitution? Evaluating Sex Work’ (1992) 17 Signs: Journal of Women in Culture and Society 705, 724, quoting Judy Helfand, ‘Silence Again’ in Frederique Delacoste and Priscilla Alexander (eds), Sex Work: Writings by Women in the Sex Industry (Cleis Press 1987) 99–​103. 35 Laurie Shrage, ‘Should Feminists Oppose Prostitution?’ (1989) 99 Ethics 347, 352.

180  Criminality at Work

3.  Commercial Sex and Adaptive Preferences A third philosophical issue that arises when viewing sex as a form of labour is a concern regarding what is referred to as adaptive preferences.36 On one level or another, most everyone’s preferences adapt to the range of options that are realistically available, and these adaptive preferences often shape our choices when it comes to our labour. If one is initially inclined to have preferences that are practically out of reach, or that will result in the loss of other valuable options, one’s preferences will often adapt to the realities one faces. Consider a girl with a natural curiosity regarding the sciences and a preference to become an astrophysicist, who is unable to pursue higher education (for reasons of gender, economics, etc). Add to this the frequent reminders of the importance of marriage and motherhood, the recognition that men in her culture are less likely to prefer marriage to a highly educated woman, and the lack of structural support for women who wish to have a demanding professional career and raise a family. Eventually the girl chooses to abandon her preference for astrophysics and instead become a school science teacher. This way, she can continue to cultivate her interest in science, marry, and have time to spend raising her children. There is no doubt that her choice to become a science teacher, wife, and mother are choices that are genuine and should be respected. Yet, her ultimate preference to pursue those goals was shaped (adapted) by the range of realistic opportunities available to her.37 Oftentimes the influence of social structures, clearly defined social roles, and the limited range of realistic options associated with those roles, will result in adaptive preferences originating from the outset: one’s initial preferences adapt as they come into being, and so map onto the range of options available. In those cases, there is no phenomenological experience of having given up the path one truly preferred—​there is simply the development of a genuine preference for a life that is within reach and will secure a range of goods that are genuinely worth having. Considered in the context of commercial sex, it seems entirely possible that some people (especially female/​femme people) might come to have adaptive preferences to engage in sex work. Particularly in light of patriarchal social scripts that value women and girls for their sexuality and structural inequalities that limit the range of valuable alternatives, it makes sense that some people prefer to engage in commercial sex as their labour.38 Moreover, as Khader explains, adaptive preferences do not (necessarily) undermine people’s capacity for autonomy. ‘People can form adaptive preferences and remain autonomous agents with reflective capacities and senses of what matters to them’.39 As we will see, viewing adaptive 36 John Elster, Sour Grapes: Studies in the Subversion of Rationality (CUP 1983) 2; Serene J Khader, Adaptive Preferences and Women’s Empowerment (OUP 2011) 17. On Khader’s account, adaptive preferences are defined as ‘preferences inconsistent with basic [human] flourishing that a person developed under conditions nonconductive to basic flourishing and that we expect her to change under conditions conducive to basic flourishing’. The term as used above tracks Khader’s account more closely than Elster, insofar as it endorses Khader’s view that adaptive preferences are neither ‘subversions of rationality’ nor autonomy deficits. 37 Martha Nussbaum, Women and Human Development: The Capabilities Approach (CUP 2001). 38 As burlesque entertainer, the Incredible, Edible Akynos, puts it: ‘In a misogynist, sexually repressive culture, where men still earn more than women, jobs are scarce, college education is proving to be a farce, and far too many men emotionally and physically take advantage of females without repercussion, how can a woman redeem herself? She can take a stand and refuse to be played like a mindless twit. Despite the misogynist nature of our society, the fact remains that sex still sells. The good girls aren’t the only ones that finish on top. Selling sex can be the way men learn how to treat women. And how women can learn their proverbial worth.’ Ravishly, ‘Is Sex Work Empowering or Enslaving? 12 Experts Weigh in’ Huffington Post (18 October 2014) accessed 28 October 2018. 39 Khader (n 36) 17–​18, 74–​106.

Sex, Work, and Criminalization  181 preferences as consistent with the exercise of autonomy is important for what follows in our discussion of commercial sex and criminalization, for it suggests at least the possibility that engaging in sex work can be genuinely consensual.

D.  Criminalization and Commercial Sex Historically, the criminal law has taken an interest in commercial sex primarily for the purpose of controlling and punishing persons (mostly female/​femmes) who provide sexual services.40 While commercial sex was not illegal in England, it was ‘surrounded by so many legal restrictions as to be illegal in all but name’.41 Indeed, [P]‌rostitutes were penalized by the legal system simply because they were prostitutes. Actions which would not constitute an infringement of the law by ‘respectable’ women [such a loitering] were illegal if committed . . . by women thought to be prostitutes.42

Views regarding the proper response of the criminal law to commercial sex had begun to shift by the time the Wolfenden Committee Report on Homosexual Offences and Prostitution was published in 1957.43 Taking the view that commercial sex was largely consensual, the Wolfenden Committee endorsed an approach to commercial sex which employed the criminal law as a deterrence for the sake of preventing public nuisance.44 Despite this significant shift, the focus of the criminal law’s interventions remained primarily on the persons (mostly female/​femmes) providing sexual services. Then, as now, the criminal law is less frequently used to target abusive pimps or those (mostly men) who buy sex.45 As a result, ‘a significantly higher number of women than men are convicted [of commercial sex offenses], [and] [o]‌f those convicted, a significantly higher number of women have multiple convictions’ for these offences.46 Alongside these domestic criminal laws, the broader legal regimes addressing commercial sex have long made a sharp distinction between commercial sex that was regarded as freely chosen versus that which was understood as forced. When commercial sex was regarded as freely chosen, those selling sex were targeted by criminal law (as noted above).47 40 For example, the Contagious Diseases Acts of the 1860s authorized the forced medical examination and detention of women suspected of being involved in commercial sex. Judith Walkowitz, Prostitution in Victorian Society (CUP 1980). 41 Paula Bartley, Prostitution: Prevention and Reform in England, 1860–​1914 (Routledge 2000) 4. 42 ibid. 43 Wolfenden Committee Report on Homosexual Offences and Prostitution (HMSO 1957). 44 For critique, see Michelle Madden Dempsey, ‘Rethinking Wolfenden:  Prostitute-​Use, Criminal Law, and Remote Harm’ [2005] Criminal Law Review 444. 45 Although, several recent high-​profiles of abusive trafficking rings have been well-​publicized. ‘Oxford Gang Found Guilty of Grooming and Sexually Exploiting Girls’ The Guardian (14 May 2013) accessed 3 August 2019. As for prosecuting sex buyers, while the offence of ‘Paying for Sexual Services of a Prostitute Subjected to Force etc’ was enacted in 2010, it remains more a matter of ‘law on the books’ rather than ‘law in action’. Sexual Offences Act 2003, s 53A, inserted (1 April 2010) in the Policing and Crime Act 2009. Tellingly, despite the fact that other ‘commercial sex + force’ type offences were moved to the Modern Slavery Act 2015, this offence (targeting buyers) remained in the Sexual Offences Act. 46 The Queen on The Application Of (1) QSA, (2) Fiona Broadfoot (3) ARB v (1) Secretary of State for the Home Department, (2) Secretary of State for Justice [2018] EWHC 407 (Admin) (QB), [2018] 1 WLR 4279 [78]. 47 Bartley (n 41).

182  Criminality at Work Yet, when commercial sex was regarded as forced, and the victims were deemed worthy, it was then framed as a problem of ‘white slavery’.48 While the Modern Slavery Act 2015, discussed above dropped the recognizably racist restriction to ‘white slavery’, this modern criminal legal scheme nonetheless endorses the sharp distinctions between modern slavery and other forms of commercial sex.49 The effect of doing so is to place some commercial sex beyond the scope of criminal law’s concern. For example, by focusing only on the most extreme cases, such as those involving kidnapping, threats of death, and serious violence—​and limiting the scope of concern to modern slavery and trafficking involving undocumented immigrants—​the law fails to address exploitation and harms in the domestic commercial sex industry. With this brief descriptive outline of the relationship between commercial sex and criminal law in view, this chapter now sets out a normative account of criminalization and puts it to use in considering various arguments regarding the proper role of criminalization when it comes to commercial sex.

1.  Criminalization and Commercial Sex: A Normative Account A normative account of criminalization aims to set out ‘the principles and values that should guide decisions about what to criminalize and about how to define offenses’.50 As Duff, et al have rightly observed: [N]‌o philosophical account of the criminal law can be expected by itself to yield determinate conclusions about what ought to be criminalized in a particular social, political, and historical context—​though such accounts should have something to say about the kinds of considerations, both practical and principled, that should be relevant, and how they should be evaluated.51

This section aims to set out some of the principled considerations that should guide our thinking about criminalization and commercial sex (more practical considerations will be examined in the following sections below). Three key principles will guide our discussion regarding criminalization and commercial sex: the principle of minimalism; modest legal moralism; and the harm principle. First, the principle of minimalism provides that criminalization ‘should be reserved for the most serious invasions of interests’.52 Endorsing a principle of minimalism entails recognition that law should typically be regarded as a ‘last resort’, given that there are oftentimes less restrictive and punitive means available. As Andrew Ashworth observes, ‘[public] 48 International Convention for the Suppression of the White Slave Traffic 1910. 49 This sharp distinction is almost universally rejected in feminist political discourse, but advocates are divided as to which category should be abandoned: modern slavery, or sex work? Jo Doezema, ‘Loose Women or Lost Women? The Re-​emergence of the Myth of White Slavery in Contemporary Discourses of Trafficking in Women’ (2000) 18 Gender Issues 23; Julie Bindel, ‘Most “Sex Workers” are Modern-​Day Slaves: Prostitution is Rarely, if Ever, a Choice’ The Spectator (19 August 2017) accessed 15 May 2018. 50 RA Duff and others, ‘Introduction:  Towards a Theory of Criminalization?’ in RA Duff and others (eds), Criminalization: The Political Morality of the Criminal Law (OUP 2014) 1. 51 ibid 19. 52 Andrew Ashworth, Principles of Criminal Law (OUP 2006) 33.

Sex, Work, and Criminalization  183 morality, social convention, and peer pressure are three informal sources of control, and in many spheres it seems preferable to leave the regulation of certain unwelcome behaviour to those forces’.53 This principle counsels in favour of public education campaigns, civil legal remedies, and other non-​punitive approaches to transforming social norms that sustain and perpetuate inequalities and exploitation, rather than using the blunt instrument of criminalization to eliminate the causes and consequences of the commercial sex industry. Second, modest legal moralism is a principle of criminalization ‘according to which only certain kinds of wrong are even in principle candidates for criminalization’.54 The central idea here is that the criminal law ought not be used to address any and all kinds of wrongs—​ but only public wrongs—​that is, wrongs which the public (here, the state) would be justified in condemning and punishing.55 As such, even if there is something wrong about commercial sex, that fact provides no reason in favour of criminalization unless it is a public wrong.56 The basic idea here is simply that some wrongs are simply ‘not the criminal law’s business’.57 For example, recall Elizabeth Anderson’s argument that commercial sex is wrong insofar as it ‘destroys the kind of reciprocity required to realize human sexuality as a shared good’.58 Consistent with modest legal moralism (although not relying on that terminology) Anderson recognizes, this fact alone does not provide a reason to criminalize commercial sex.59 The third key principle that should guide criminalization regarding commercial sex is the presumption of non-​interference, grounded in the harm principle. According to its most recognizable formulation, this principle provides that ‘the only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others’.60 Importantly for our discussion, the harm principle must be read alongside a modifying maxim: volenti non fit injuria (to one who consents, no wrong is done).61 The scope and meaning of the harm principle and volenti maxim, as a unified whole, is complicated and subject to much debate.62 One point of widespread agreement, however, is that where people choose to engage in some activity based on a genuine preference to do so, under conditions where they enjoy an adequate range of valuable alternatives consistent with human flourishing, there is a strong case to be made against the criminal law interfering with that choice. Let us refer to this as the presumption of non-​interference. That said, in applying this presumption, one must be attentive not only to direct harms, but to indirect harms as well. Direct harms include those core criminal offences such as battery,

53 ibid. 54 Antony Duff, The Realm of Criminal Law (OUP 2018) 75. 55 Here, I leave aside my standing debate with Duff regarding what such justification would entail. Michelle Madden Dempsey, ‘Public Wrongs and the “Criminal Law’s Business”:  When Victims Won’t Share’ in Rowan Cruft, Matthew Kramer, and Mark Reiff (eds), Crime, Punishment and Responsibility: The Jurisprudence of Antony Duff (OUP 2011); Duff, Realm (n 54) 94–​101. 56 For example, recall Elizabeth Anderson’s argument that commercial sex is wrong insofar as it ‘destroys the kind of reciprocity required to realize human sexuality as a shared good’. Anderson (n 23) 153–​54. Consistent with modest legal moralism (although not relying on that terminology) Anderson recognizes, this fact does not provide a reason to criminalize commercial sex. Anderson (n 23) 150–​58. 57 Duff, Realm (n 54) 76, quoting ‘Wolfenden’ (n 43) [61]. 58 Anderson (n 23) 153–​54. 59 Anderson (n 23) 150–​58. 60 John Stuart Mill, On Liberty (first published 1859, Dover Publications 2002) 8. 61 Joel Feinberg, Harm to Others: The Moral Limits of the Criminal Law (vol. 1) (OUP 1984) 35–​36. 62 Michelle Madden Dempsey, ‘The Volenti Maxim’ in Peter Schaber and Andreas Müller (eds), The Routledge Handbook of the Ethics of Consent (Routledge 2018).

184  Criminality at Work rape, murder, etc, (malum in se offences) in which the conduct targeted for criminalization bears a direct causal link to the harm sought to be prevented by the criminal law.63 Indirect harms, however, involve conduct that contributes to causing the harm sought to be prevented by the criminal law. In what follows, this chapter will apply these normative principles of criminalization to four different accounts of commercial sex:  commercial-​sex-​as-​work; commercial-​sex-​ as-​exploitation; commercial-​sex-​as-​modern-​slavery; and commercial-​sex-​as-​varied. Each section below explains how the particular account under examination draws upon one or more of the conceptualizations of commercial sex outlined in the first section of this chapter, and takes into consideration the philosophical issues discussed in section B. As we will see, each account differs rather dramatically as to the appropriate relationship between commercial sex and the criminal law.

2.  Criminalization and ‘Commercial-​Sex-​as-​Work’ As its name suggests, the ‘commercial-​sex-​as-​work’ account views commercial sex as a legitimate labour market. It draws upon the conceptualization of commercial sex as sex work—​the exchange of sexual services for valuable consideration, under circumstances in which the persons providing sexual services: (1) regard the commercial sex as a form of labour; and (2) prefer to engage in commercial sex over these alternative options, despite (3) having an adequate range of valuable alternatives for engaging in forms of labour that are consistent with human flourishing. On this account, the appropriate role for criminalization is limited to egregious abuses (malum in se criminal offences), along with basic health and safety regulations (mala prohibita criminal offences). Since commercial sex is understood as sex work, and sex work is understood as a choice based on a genuine preference from amongst an adequate range of valuable alternatives consistent with human flourishing, the harm principle (along with volenti maxim) grounds a strong presumption of non-​interference by the criminal law. That point holds even if the preference to engage in sex work is an adaptive preference (for, as explained above, adaptive preferences are not inconsistent with autonomous choice).64 Pursuant to the principle of modest legal moralism, arguments in favour of criminalizing commercial sex on grounds of commodification are deemed insufficient to overcome this presumption of non-​interference. Perhaps the strongest argument in favour of criminalizing commercial sex according to a sex-​as-​work account is based on the notion of indirect harm, grounded in the claim that commercial sex perpetuates structural inequalities of gender, economics, etc. Still, pursuant to the principle of minimalism, before such concerns could overcome the presumption of non-​interference, it must be established that alternative methods are insufficient to address this problem, such that criminalization would, in fact, be a last resort. Given that this account assumes that people providing sexual services are choosing to do so

63 A malum in se offence is deemed wrong in and of itself and thus forms the core criminal offences; whereas mala prohibita offences are deemed wrong because they are prohibited. The distinction is well recognized but, admittedly, unstable and perhaps more obscuring than clarifying. Duff, Realm (n 54) 20–​22. See below at text to (n 69) for questions regarding how this distinction applies in health and safety regulations regarding condoms. 64 Text and notes at (n 36–​39).

Sex, Work, and Criminalization  185 from amongst an adequate range of valuable options, any concerns regarding the tendency of commercial sex to perpetuate structural inequality in general are deemed inadequate to overcome the presumption of non-​interference. A sex-​as-​work account does, however, take on board the fact that sex workers face violence and abuse (that would clearly count as malum in se criminal offences) at rates far higher than typical labourers.65 Yet, the preferred response to this fact under this account is to demand that the criminal law respond to such violence by directly holding accountable those who perpetrate it, rather than enacting a blanket prohibition on sex work. That is, the criminal law should investigate, arrest, and prosecute sex buyers and others who commit rape, battery, murder, etc. Moreover, in so doing, police and prosecutors should treat sex workers with respect and dignity, affording them the full protections of criminal laws designed to prevent such malum in se violence.66 When it comes to the health and safety of sex workers, the role of the criminal law becomes more complicated. First, it is somewhat difficult to determine if the conduct at issue would best be understood as mala prohibita (as most labour health and safety laws would generally be understood), or whether the conduct should be regarded as malum in se. Take the example of condom use. Clearly, criminal regulation designed to protect the health and safety of sex workers would include a requirement that sex buyers use condoms. What, then, if a sex buyer slips off the condom mid-​coitus? Does this conduct merely violate a mala prohibita regulation designed for the promotion of health and safety—​or does it negate the consent of the sex workers and thus constitute the malum in se offence of rape?67 Another complication regarding health and safety laws in the context of sex work arises by virtue of the fact that such laws generally apply to employers for the protection of employees.68 Yet, where one engages in sex work as an independent contractor, it is not clear how the sex-​as-​work account would wish to design laws to provide health and safety protections. Moreover, as Adrienne Davis observes, even in contexts where there is a clear employer–​employee relationship, ‘there are unique characteristics of sex work that make it much harder to assimilate . . . current [health and safety] regulatory regimes’ into this context.69 For example, ‘monitoring and enforcing health and safety violations [for sex work occurring] in streets, private homes, hotels, and the like will be highly challenging’ due to the lack of a fixed worksite.70 So, too, in fixed worksites such as brothels and strip clubs, commercial sex often takes place in private rooms, and ‘it is during these proximate interactions, while isolated from others, that clients are most likely to’ engage in conduct that jeopardizes the health and safety of sex workers.71 Thus, even with the best intentions of doing so, designing effective health and safety regulations may prove especially difficult in the context of sex work. 65 Hilary Kinnell, Violence and Sex Work in Britain (Willan Publishing 2008). 66 Rebecca Campbell and Hilary Kinnell, ‘ “We Shouldn't Have to Put Up with This”:  Street Sex Work and Violence’ (2000) 42 Criminal Justice Matters 12. 67 Assange v Swedish Prosecution Authority [2011] EWHC 2849 (Admin), [2011] 11 WLUK 63; AP Simester and others, Simester and Sullivan’s Criminal Law: Theory and Doctrine (4th edn, Hart 2010) 475. 68 Ford (this volume). 69 Adrienne D Davis, ‘Regulating Sex Work: Erotic Assimilationism, Erotic Exceptionalism, and the Challenge of Intimate Labor’ (2015) 103 California Law Review 1195. 70 ibid 1201. 71 ibid 1252.

186  Criminality at Work

3.  Criminalization and ‘Commercial-​Sex-​as-​Exploitation’ A commercial-​sexual-​exploitation account (CSE account) recognizes a strong distinction between commercial sex and labour, due to the structural inequalities, abuse, and ­relative lack of autonomy experienced by people who sell sex. That is, while the sex-​as-​work ­perspective views people who provide sexual services as freely choosing from amongst an adequate range of valuable options, a CSE account accuses the sex-​as-​work account of being naïve to the practical realities of commercial sex. According to the CSE account, people rarely if ever choose to sell sex in this robustly free and empowered sense—​which is to say, very little if any commercial sex actually constitutes sex work. Since commercial sex is understood as exploitative under the CSE account, the harm principle (along with volenti maxim) grounds a considerably weaker presumption of non-​interference by the criminal law.72 Indeed, the tendency of commercial sex to perpetuate structural inequalities of gender, economics, etc, and thus increase the prevalence of commercial sexual exploitation, is typically viewed as sufficient to overcome this presumption. (Although, consistent with the principle of minimalization, the CSE account does not look exclusively to punitive criminalization to address these concerns. Instead, public education campaigns and diversion from criminal courts feature prominently in the CSE account’s toolbox.73) Moreover, while concerns regarding commodification do not, on their own, establish a case in favour of criminalization, they are thought to lend weight insofar as the commodification is particularly gendered. For, according to a CSE account, the gendered commodification of sex reinforces the structural inequalities that make commercial sex a proper target of criminalization. That said, the case in favour of criminalizing commercial sex according to a CSE account is directed only towards those actors who benefit from the exploitation. Specifically, the targets of criminalization on the CSE account include only pimps/​traffickers and sex buyers—​not those providing sexual services.74 According to a CSE account of commercial sex, the conduct of pimps/​traffickers are viewed as causing direct harm insofar as they are engaged in conduct of abusing structural and individual vulnerabilities, such as patriarchal-​precarity, of those whom they exploit.75 As such, these actors are deemed to be committing malum in se offences and are proper targets of criminalization and punishment proportionate to the seriousness of their offences. When it comes to sex buyers, however, matters are somewhat more complicated. It is not clear on this account that sex buyers cause direct harm.76 Rather, on a CSE account, sex

72 As Jennifer Collins observes, ‘there is a need for a multi-​faceted legal response to serious exploitation in work relations, which may include, but must extend beyond, criminalization’. Collins (this volume). 73 For examples of the range of non-​criminal approaches consistent with a CSE account of commercial sex, see the initiatives undertaken by the Ending Exploitation Collaborative accessed 3 November 2018. 74 A CSE account of commercial sex, and the kind of criminalization policies outlined above, inform what is referred to as the ‘Nordic Model’—​which, presently, has been adopted in Sweden, Norway, Iceland, Canada, France, Northern Ireland, the Republic of Ireland, and Israel. Michelle Madden Dempsey, ‘Sex Trafficking and Criminalization: In Defense of Feminist Abolitionism’ (2010) 158 University of Pennsylvania Law Review 1729. 75 See text at (nn 27–​35) regarding patriarchal-​precarity. 76 Although, of course, a CSE account also takes on board the facts discussed above regarding malum in se ­offences (rape, battery, etc) often committed by sex buyers. Text and notes (nn 66–​67). The discussion here ­concerns sex buyers who do not commit such offences, but merely purchase sex.

Sex, Work, and Criminalization  187 buyers are understood as committing indirect harms by creating the market demand that drives the commercial sex trade in which sex sellers are exploited.77

4.  Criminalization and ‘Commercial-​Sex-​as-​Modern-​Slavery’ The ‘commercial-​sex-​as-​modern-​slavery’ account takes a significantly different view of the practical realities of the commercial sex trade and, not surprisingly, presents its own recommendations when it comes to the proper scope of criminalization. We can begin by identifying two distinct perspectives which inform this account. The first is a socially conservative, morally puritanical perspective which assumes that anyone providing sexual services for compensation must be doing so under threat of force, or other form(s) of extreme third-​ party coercion. On this view, the paradigmatic victim is a young, innocent, helpless girl who has been tricked, kidnapped, or otherwise forced into sex-​trafficking/​modern slavery.78 The second is a socially progressive, feminist perspective, which views the commercial sex industry as fundamentally grounded in structural inequalities of gender, race, sexuality, class, and economic injustice. On this view, the commercial sex trade is neither ‘just work’ like any other labour market, nor is it merely a matter of exploitation. It is, rather, a matter of what Kathleen Barry famously coined, ‘female sexual slavery’.79 As its name suggests, this account conceptualizes commercial sex as consisting mostly of modern slavery. Accordingly, as noted above, the person providing sexual services is understood to be entirely without agency, and the volenti maxim thus does nothing to ground a presumption of non-​interference from the criminal law. Rather, according to this account, the harm principle is clearly satisfied on grounds that the victims of modern slavery are subjected to direct non-​consensual harms. Moreover, the wrongful conduct and harms inflicted on its victims are of the worst kind—​malum in se offences involving violence and threats resulting in literal enslavement. As such, the wrongs committed easily satisfy the threshold of constituting public wrongs (thus satisfying the principle of modest legal moralism). Finally, on the commercial-​sex-​as-​modern-​slavery account, the wrongdoers are understood to be working in coordinated groups (‘organized crime’80) which require the full investigatory resources and punitive force of the criminal justice system. Thus, while the principle of minimalization plays a restraining role in the other accounts surveyed above, it bears little force when it comes to criminalizing those who perpetrate modern slavery. When it comes to criminalizing sex buyers, however, matters again become slightly more complicated. Where someone buys sex from a victim of modern slavery, the sexual act is understood as tantamount to rape, in light of the utter lack of autonomy exercised by the victim.81 For recall that on this account, the victim is understood as having been subjected to ‘a brutal form of organized crime in which people are treated as commodities and

77 Dempsey, Feminist Abolitionism (n 74) 1752–​62. 78 Janie A Chuang, ‘Rescuing Trafficking from Ideological Capture: Prostitution Reform and Anti-​Trafficking Law and Policy’ (2010) 158 University of Pennsylvania Law Review 1655. 79 Kathleen Barry, Female Sexual Slavery (NYU Press 1984). 80 Text and notes (n 17). 81 Evelina Giobbe, ‘Prostitution:  Buying the Right to Rape’ in Ann Wolbert Burgess (ed), Rape and Sexual Assault III: A Research Handbook (Garland Science 1991).

188  Criminality at Work exploited for criminal gain’.82 Not surprisingly, then, criminalization of buyers is deemed appropriate in such cases.83 As noted above, criminalization under the ‘commercial-​sex-​as-​modern-​slavery’ account focuses primarily on organized criminal gangs recruiting or kidnapping victims and subjecting them to modern sexual slavery. While neither of the two perspectives that motivate this account (conservative religious and progressive feminist) incorporate a particular focus on border security or irregular migration, there is little doubt that recent political and legal discourses have adopted the modern slavery discourse to motivate concerns regarding immigration and justify harshly punitive state action against refugees and other irregular migrants.84 In so doing, this account: [C]‌ondemn[s] as ‘modern slavery’ the application of coercive pressures on migrants [by non-​state actors], but simultaneously  . . .  endorse[s] the application of ever more coercive pressures on migrants by states, often in the name of protecting them from ‘modern slavery’.85

5.  Criminalization and ‘Commercial-​Sex-​as-​Varied’ The three accounts outlined above embrace starkly divergent understandings of commercial sex and recommend correspondingly different roles for criminalization. This final account, however, adopts a more complex understanding of commercial sex, as potentially encompassing each of the other three accounts, to greater or lesser extents. On the ‘commercial-​sex-​as-​varied’ account, there is room to accommodate an understanding of commercial sex as either sex work, commercial sexual exploitation, and/​or modern slavery. This account recognizes that multiple variables intersect in any given society, and in any individual person’s experience, to shape and inform whether they are engaging in sex work, being subjected to commercial sexual exploitation, or are a victim of modern slavery. Recognizing the explanatory power of each of these different accounts . . .  has the potential to move discourse away from the master narratives currently employed in existing debates about the commercial exchange of sexual acts . . . and how the law should and may respond to them.86

In addition to acknowledging the potential diversity of commercial sex, the commercial-​ sex-​as-​varied account embraces many of the criminalization recommendations on offer from each of the three previous accounts. For example, this account counsels in favour of using the full investigatory and punitive force of the criminal law to address malum in se 82 Explanatory Notes MSA (n 17) [4]‌. 83 The more controversial view is whether such criminal penalties should apply even in cases where the sex buyer is not aware that the person providing sexual services is a victim of modern slavery, as in Sexual Offences Act 2003, s 53A (‘Paying for Sexual Services of a Prostitute Subjected to Force etc’). 84 Text and notes (nn 20–​21). 85 Julia O’Connell Davidson, ‘New Slavery, Old Binaries: Human Trafficking and The Borders of “Freedom” ’ (2010) 10 Global Networks 244, 255. 86 Debra M Haak, ‘Re(de)fining Prostitution and Sex Work:  Conceptual Clarity for Legal Thinking’ (2019) Windsor Review of Legal and Social Issues 40.

Sex, Work, and Criminalization  189 violence committed against people engaged in commercial sex, no matter what conceptualization of commercial sex is at issue. Additionally, it views both violent and non-​violent exploitation by pimps and/​or traffickers as serious public wrongs that inflict direct harms which cannot adequately be addressed through non-​criminal means. As Collins observes with respect to seriously wrongful exploitation in other employment contexts, criminalization is appropriate in such cases because ‘we require what is distinctive from the criminal law—​a strong condemnatory state response which penalizes serious wrongdoing’. Thus, consistent with the normative principles of criminalization discussed above, this account endorses criminalization of anyone who exploits persons providing sexual services.87 Once again, however, when it comes to criminalization of sex buyers, matters become rather more complicated. Given the potential diversity of commercial sex in any given society, this account acknowledges that some of it may constitute sex work. If so, then (as discussed above) there is a strong presumption in favour of non-​interference by the criminal law.88 This presumption, however, is not irrebuttable. The question then remains: what considerations might be sufficient to justify criminalizing sex buyers in a context where (at least some of) the commercial sex counts as genuine sex work, rather than constituting commercial sexual exploitation or modern slavery? In answering this question, it is helpful to disaggregate two versions of the commercial-​ sex-​as-​varied account. One version of (let us call it the delineated-​market-​segments version) holds that distinctions between sex work, commercial sexual exploitation, and modern slavery reflect sharply delineated market segments, such that the legalization of sex buying (market demand) in one segment will not impact the number or frequency of sex buying activity in other segments. If this version is correct, then it may be possible to permit market demand to flourish unrestricted when it comes to sex work, while still prohibiting the purchase of sex in segments that involve commercial sexual exploitation and/​or modern slavery. The other version of the commercial-​sex-​as-​varied account (let us call it the non-​ delineated-​market-​segments version) cautions that permitting market demand to flourish unrestricted when it comes to sex work is likely to increase market demand across all forms of commercial sex, including commercial sexual exploitation and modern slavery. The most plausible version remains a matter for further investigation, but recent empirical research suggest that the non-​delineated market version reflects current conditions.89 If permitting market demand to flourish unrestricted with regard to sex work results in an increase in sex buying activity when it comes to commercial sexual exploitation and/​ or modern slavery, then the normative principles of criminalization outlined above may favour criminalizing the purchase of sex per se. For, while commercial sex occurring in the context of sex work would still be entitled to a strong presumption of non-​interference, that presumption may be overcome insofar as increased market demand increases commercial sexual exploitation and modern slavery. To better understand the gist of this argument, let us start with the plausible assumption that where buying sex is not criminalized, there will be a larger number of people seeking to buy sex (and perhaps buying sex at greater frequencies) than there otherwise would be. 87 Text and note (n 53). We require what is distinctive from the criminal law—​a strong condemnatory state response which penalizes serious wrongdoing. 88 Text and notes (nn 60–​62). 89 Seo-​Young Cho, Axel Dreher, and Eric Neumayer, ‘Does Legalized Prostitution Increase Human Trafficking?’ (2013) 41 World Development 67.

190  Criminality at Work That is, market demand for commercial sex will increase as compared to the level of demand that would exist if buying sex were prohibited by criminal law. The question then presents itself: how will this increased market demand be met? Consider two alternatives. First, the increased demand will be met by a large enough supply of people who voluntarily choose to perform sex acts for money, and they do so under non-​exploitative and non-​ abusive conditions. If this happens, then the increased demand for commercial sex will lead to an increase only with respect to sex work, not commercial sexual exploitation or modern slavery. Second, there will not be a large enough supply of people who choose to perform sex acts for money, or they will choose to do so under exploitative or abusive conditions. If this happens, then the increased demand for commercial sex will be met by people providing commercial sex under conditions that constitute commercial sexual exploitation and/​ or modern slavery.90 According to an analysis conducted by Cho, Dreher, and Neumayer, the second answer reflects the reality of what has happened in multiple jurisdictions: even though buyers would prefer to buy sex from sex workers and not from people subjected to commercial sexual exploitation or modern slavery, the scale effect of increased market demand nonetheless increases exploitative and abusive forms of commercial sex, because there are not enough people on the supply side who are engaging in commercial sex under conditions that are non-​exploitative.91 All of this suggests that the criminal law may have a justifiable role to play in prohibiting the purchase of sex per se, if legalization would increase demand and thus increase commercial sexual exploitation and/​or modern slavery. Of course, the principle of minimalism would still counsel in favour of non-​criminal solutions (eg education) to reduce demand and thus prevent an increase in commercial sexual exploitation and/​or modern slavery. However, if non-​criminal alternatives proved inadequate, then criminalizing the purchase of sex per se could be justified.92

E.  Conclusion This chapter has offered a taxonomy of ways in which commercial sex is often conceptualized in modern political and legal discourse, addressed various philosophical issues that arise when viewing sex as a form of labour, and set out a normative account of the proper role of criminalization when it comes to commercial sex. Let us end by returning to the first question raised at the outset of this chapter: Is commercial sex properly regarded as work? The account offered here concedes that while it is possible to conceptualize commercial sex as sex work, there is no universal answer to the question of whether commercial sex actually is, in practice, properly regarded as work. Rather, most or all instances of commercial sex might just as well be best regarded as commercial sexual exploitation or even modern slavery. That is to say, merely acknowledging that commercial sex can be conceptualized as work does not justify the conclusion that commercial sex is, in fact, properly regarded as work. Rather, the answer to our question depends on the extent to which the actual practice 90 In terms of economic theory, the market demand for commercial sex will experience a substitution effect: since the supply of non-​exploitative/​non-​abusive commercial sex will be insufficient to meet the demand for commercial sex, those seeking to buy sex will substitute with exploitative and/​or abusive forms of commercial sex. 91 Cho, Dreher, and Neumayer (n 89). 92 Dempsey, Rethinking Wolfenden (n 44); Dempsey, Feminist Abolitionism (n 74).

Sex, Work, and Criminalization  191 of commercial sex in a given society is shaped by precarities such as patriarchal, economic, and other structural inequalities. When placed into dialogue with the normative account of criminalization set out herein, the implications are two-​fold. First, rather uncontroversially, criminalizing pimping, trafficking, and buying sex is permissible in circumstances where, in practice, commercial sex constitutes commercial sexual exploitation and/​or modern slavery. Second, perhaps more controversially, where the commercial sex industry is varied (that is, where some of the commercial sex is sex work, some is commercial sexual exploitation, and some is modern slavery), then the criminal law may have a justifiable role to play in prohibiting the purchase of sex per se, in order to reduce market demand that would otherwise increase commercial sexual exploitation and/​or modern slavery.

10

The Work of Sex Work Prostitution, Unfreedom, and Criminality at Work Katie Cruz*

A.  Introduction The central and uniting demand of the sex worker rights movement around the world is the decriminalization of consensual adult sex work. This is based on the recognition that criminal law intervention makes sex workers less, rather than more, safe and that sex workers are engaged in a legitimate form of work, not the commission of a crime. A core task of activists and academics has been amassing evidence that criminalization of the seller, buyer, and associated activities impacts negatively on the safety and well-​being of sex workers. The core focus of this chapter is the claim that sex workers, and specifically prostitutes, are engaged in a legitimate form of work that should not be criminalized.1 In activism and scholarship there are two perspectives that contest criminalization on the basis that sex work, including prostitution, is a legitimate form of work. The first focus is on the sex of sex work. The second focus is on the work of sex work. Those who take the latter perspective charge those who take the former with paying insufficient attention to the working conditions and lives of sex workers as workers. Academics and activists who focus on the work of sex work therefore argue for full decriminalization of consensual sex work and the extension of labour law protections. If achieved, such changes would necessitate engagement with the actual and desirable application of criminal law to personal work relations. However, those who focus on the work of sex work have not thought through the implications of their demands for labour rights and criminalization in the (sex)workplace. This chapter has four parts. The first part is contextual. I outline the case for decriminalization of adult consensual sex work, and more specifically prostitution, on the grounds of safety and well-​being. I then move on to contesting criminalization on the basis that prostitution is a legitimate form of work, drawing attention to a focus on either the sex or work of prostitution. In previous research, I have argued that the work position often conflates de jure with de facto access to labour rights. In the second part, I extend this argument and propose that the work perspective pay greater attention to the continuing role of criminalization in the (sex)workplace in decriminalized jurisdictions. Third, I outline a Marxian * I would like to thank Manuel Cruz, Alan Bogg, Mark Freedland, Julia O’Connell Davidson, Jacqueline Sanchez Taylor, Katie Bales, and Michael Ford QC who have all generously engaged with the ideas in this paper. Special thanks goes to Manuel for his top-​notch editorial skills. 1 My use of the term prostitution may seem jarring and confusing to some, particularly given its strong association with abolitionist and radical feminisms. I use the term to refer to a specific form of sex work that involves, but is not limited to, payment for a particular service (intercourse) that is immediately consumed by the customer and not filmed and turned into content for further consumption. I use the term sex worker to refer to those who perform this work. Katie Cruz, The Work of Sex Work In: Criminality at Work. Edited by Alan Bogg, Jennifer Collins, Mark Freedland QC, and Jonathan Herring, Oxford University Press (2020). © Alan Bogg, Jennifer Collins, Mark Freedland, and Jonathan Herring. DOI: 10.1093/oso/9780198836995.003.010

Prostitution, Unfreedom, and Criminality at Work  193 methodology for describing and evaluating criminalization of personal work relations. Fourth, I apply this methodology and offer the work position some provisional remarks about the potential of worker-​protective criminal and civil laws for addressing the unfreedom of sex worker employees of brothels in England and Wales (if prostitution related activities were decriminalized). I focus on criminal and civil laws that relate to health and safety and coerced or forced labour because these are laws that those who view prostitution as work support. However, I spend a relatively short amount of time applying my methodology to these laws. Instead, I spend a good chunk of this chapter focusing on why criminality at work is an important issue for those who view prostitution as work and how it can be engaged with from a Marxian perspective. Overall, I hope to incite future lines of enquiry, debate, and discussion with the sex worker rights movement, union activists, labour and criminal lawyers, about the relationship between criminality at work and worker freedom.

B.  The Case for Decriminalization: Safety, Sex, and Work Sociologists and socio-​legal scholars argue that we are in the midst of an ‘unprecedented rise in, and increased visibility of, sexual commerce and consumption and a corresponding growth in associated forms of regulation’.2 It may be impossible to cite reliable statistics on differentiation (eg the social and economic positioning of the sex workforce) and size of the sex industry (eg the number of brothel-​based workers compared to street workers). Sexual commerce has, however, diversified as a result of the shift to post-​industrial economies (in which service work and consumption dominate, leading to the expansion of leisure and night-​time economies); changing patterns of female migration as women move across the world to secure futures for themselves and their families (no doubt related to the swelling global reserve army of labour following new rounds of primitive accumulation and mass dispossession); postmodern sexual attitudes (wherein the marketization of sex is increasingly acceptable, leading to increased consumption and visibility); and technological transformations (the proliferation of online pornography, web-​camming, and online advertising).3 There is also diversity within each type of sex work. The focus of this chapter, prostitution, ‘is extremely differentiated’. For example, female prostitution: [R]‌anges from well-​paid self-​employed workers, working in private apartments with a high degree of control over their work and providing services beyond ‘intercourse’ (the girl friend experience, companionship, attending events, conversation) to women working on

2 Jane Scoular and Teela Sanders, ‘Introduction:  The Changing Social and Legal Context of Sexual Commerce: Why Regulation Matters’ (2010) 37(1) Journal of Law & Society 1. 3 See, eg, Mechthild Hart, ‘Women, Migration, and the Body-​Less Spirit of Capitalist Patriarchy’ (2005) 7(2) Journal of International Women’s Studies 1; Barbara Ehrenreich and Arlie Hochschild (eds), Global Woman:  Nannies, Maids, and Sex Workers in the New Economy (Metropolitan Books 2002); Jane Scoular and others, ‘Away from Prying Eyes? The Urban Geographies of “Adult Entertainment” ’ (2008) 32(3) Progress in Human Geography 363; Elizabeth Bernstein, Temporarily Yours:  Intimacy, Authenticity, and the Commerce of Sex (University of Chicago Press 2007); Katie Cruz, ‘Beyond Liberalism: Marxist Feminism, Migrant Sex Work, and Labour Unfreedom’ (2018) 26(1) Feminist Legal Studies 65; Sue Ferguson and David McNally, ‘Precarious Migrants: Gender, Race and the Social Reproduction of a Global Working Class’ in Leo Panitch and Greg Albo (eds), Transforming Classes: Socialist Register (Merlin Press 2015) 1–​23.

194  Criminality at Work the streets or in brothels, in assembly line conditions, tightly supervised and often in fairly risky situations.4

Sex worker rights activists and scholars argue that prostitution is a form of sex work and must be decriminalized. In this part, I outline two perspectives that argue for decriminalization on the basis that prostitution is a legitimate form of work. The first focus is on the sex of prostitution and criminalization is understood to stem from the illegitimate and irrational stigmatization and ‘othering’ of sex workers’ private choice of sexual practice or sexual profession. The second focus is on the work of prostitution and criminalization is understood to label as a sexual offence what is in fact a highly gendered and racialized economic exchange, and to act as a barrier to accessing labour rights and protections. This distinction is important because the latter proponents charge the former with paying insufficient attention to the working conditions and lives of sex workers as workers. The prostitution as work perspective also disrupts the claim that those who do not equate prostitution with sexual exploitation are necessarily liberal or ‘sex work’ feminists, narrowly focused on defending it as a meaningful vocation, or an expression of individual choice and sexual agency.5 I make two observations about the work position. First, it contains a nuanced sociological understanding of the work of prostitution, which is persuasively connected to calls for decriminalization. However, those who view prostitution as work tend to (incorrectly) assume that sex workers in prostitution have labour rights and that criminalization ends when labour regulation begins.6 I pause, however, to briefly outline the rigorously evidenced harm reduction and safety case for the decriminalization of adult consensual sex work being made by sex-​worker-​rights activists, academics, and non-​governmental organizations.

1.  Safety and (De)criminalization Sex worker rights activists and their allies and are united on the need for decriminalization of prostitution-​related activities. Criminalization makes sex work less safe for those selling sex and normalizes their abuse by clients and third parties, including bosses and the police. For ten years I have been organizing with sex worker rights and anti-​capitalist feminist groups in London, Jamaica, and with European-​based organizations that support decriminalization. I have conducted empirical research with sex workers and sex worker rights activists in the UK and Jamaica, and I have co-​written an article based on empirical research with the Argentinian sex worker rights organization AMMAR.7 As a result, 4 Camille Barbagallo and Silvia Federici, ‘Introduction’ (2012) 15 (Winter) The Commoner, Care Work and the Commons 1. 5 Catharine MacKinnon, ‘Trafficking, Prostitution, and Inequality’ (2011) Harvard Civil Rights-​Civil Liberties Law Review 271; Michelle Madden Dempsey, ‘Sex Trafficking and Criminalization:  In Defense of Feminist Abolitionism’ (2010) 158 University of Pennsylvania Law Review 1729; Sheila Jeffreys, The Industrial Vagina: The Political Economy of the Global Sex Trade (Routledge 2008). 6 This relative lack of attention to legal details when compared to the careful sociological explanation of the work of sex work has been commented upon in relation to academic research: Jane Scoular, The Subject of Prostitution: Sex Work, Law and Social Theory (Routledge 2015). My argument is that activists are also prone to this asymmetry of attention. 7 Katie Cruz, ‘Unmanageable Work, (Un)liveable Lives: The UK Sex Industry, Labour Rights and the Welfare State’ (2013) 22 Social Legal Studies 465; Katie Cruz, Within and Against the Law: UK Sex Worker Rights Activism and the Politics of Rights (PhD thesis, University of Nottingham 2015); Cruz, ‘Beyond Liberalism’ (n 3); Katie Cruz, Julia O’Connell Davidson, and Jacqueline Sanchez Taylor, ‘The Tourism Industry in Jamaica: Work, Slavery, Freedom?’

Prostitution, Unfreedom, and Criminality at Work  195 I agree with this consensus, which has found a foothold in some countries as well as national and international organizations. The countries that have decriminalized (some, not all) prostitution-​related activities include New Zealand, Australia (New South Wales, Queensland), Germany, Netherlands, and USA (Nevada). Reviews of existing research and consultation with sex workers and activists have been conducted by organizations that have no necessary investment in favouring decriminalization. After two years of primary and secondary research in Papua New Guinea, Hong Kong, Norway, and Argentina, in 2015 Amnesty International passed a resolution supporting decriminalization of all aspects of consensual adult sex work. The policy advocates the repeal, and commitment not to introduce, laws that criminalize the sale, purchase, and organization of sex-​work-​related activities, including brothel-​keeping. This is defended as a human rights position and harm-​reduction strategy to protect sex workers from abuse and violence. Amnesty International’s research has confirmed that laws criminalizing buyers and the organization of sex work related activities (such as brothel-​keeping) force sex workers to work in conditions that compromise their safety and discourage them from approaching the police for protection. It therefore recognizes the harm of criminal laws beyond those that directly target individual sex workers.8 Laws that criminalize buyers are framed as tackling demand but ‘in practice they can lead to sex workers having to take risks to protect their clients from detection by law enforcement, such as visiting locations determined only by their clients’.9 And even when not directly criminalized, sex workers are reluctant to turn to the police when crimes against them are committed. This is because the existence of laws that treat sex-​work-​related activities as a crime effectively allow the police to treat sex workers as criminals and sexual objects. The police are known to respond by arresting sex workers for prostitution-​related activities, closing or monitoring brothels with the attendant actual or potential fear of loss of earnings, denying that a crime has occurred or failing to investigate, practising extortion and enforcing sexual violence, including rape.10 In December 2018, a synthesis of all existing qualitative and quantitative data on the effects of criminalization on the health and well-​being of sex workers was published. The authors concluded ‘that, collectively, lawful or unlawful repressive policing practices linked to sex work criminalization (partial or full) are associated with increased risk of infection with HIV or STIs, sexual or physical violence from clients or intimate partners, and condom-​less sex’.11 Like Amnesty International’s policy, this report demonstrates that criminal laws that impact negatively on sex workers include those that criminalize the seller, buyer, and associated activities. Support for decriminalization on the basis that it makes the sale of sex safer has been voiced for some time. Feminist historian Judith Walkowitz describes a ‘shared assessment across generations of feminist historians in the US and UK’ about the modern history of 7(s1) Journal of the British Academy 191; Kate Hardy and Katie Cruz, ‘Affective Organizing:  Collectivizing Informal Sex Workers in an Intimate Union’ (2018) 63(2) American Behavioral Scientist 244. 8 Amnesty International, ‘Global Movement Votes to Adopt Policy to Protect Human Rights of Sex Workers’ (11 August 2015) www.amnesty.org/​en/​documents/​pol30/​4062/​2016/​en/​ accessed 23 April 2019. 9 ibid 13. 10 ibid 13; Cruz and others, ‘The Tourism Industry in Jamaica’ (n 7). 11 Lucy Platt and others, ‘Associations between Sex Work Laws and Sex Workers’ Health: A Systematic Review and Meta-​analysis of Quantitative and Qualitative Studies’ (2018) 15(12) PLOS Medicine 1, 42.

196  Criminality at Work prostitution that ‘positioned prostitution at the intersection of class and gender’. This perspective dates to the 1970s and ‘stands in contrast to the fierce disputes around sex work raging in other fields of study today’. Three key observations make up this consensus: (1) that prostitution is sexual labour, ‘an integral part of the survivalist strategy of the poor over many centuries’, (2) that ‘intensified policing had negative effects on women in the trade’, and (3) scepticism about ‘political campaigns, including feminist campaigns that repeatedly ended in legislation and other state actions that marked sex work from other forms of labour’.12 Longstanding sex worker rights organizations share this consensus. The English Collective of Prostitutes (ECP) is a UK-​based grassroots organization that also dates to the 1970s. For decades the ECP have provided legal casework support for women prosecuted with prostitution-​related offences, including brothel-​keeping charges faced by women working collectively.13 Like Walkowitz and other feminist historians, the sex worker rights movement and academics also insist that prostitution is a legitimate form of work and should not be criminalized. I now turn to the substance of this demand.

2.  Prostitution as Sex, Prostitution as Work ‘My Body, My Business’ is a key slogan of the sex worker rights movement. It usefully encapsulates the movement’s focus on the sexual nature of prostitution to ground the claim that it is work that should not be criminalized. Activists and academics who take this approach draw comparisons between sex workers and other marginalized groups who experience stigmatization and policing based on their perceived sexuality and/​or gender. Locating sex work within the rubric of consensual sexual behaviour, and calling for the end of criminal laws that punish non-​conforming sexual practices, resonates with gay, queer, and sex-​ positive theory and practice.14 Sex work is perceived to be an issue of sexual practice and sexual diversity, or, in other words, involves respect for ‘My Body’. Another way that the sex of prostitution grounds demands for decriminalization is the view that it is an important vocation. Sex workers and movement activists Juno Mac and Molly Smith have coined the term the ‘Erotic Professional’ to refer to sex workers who defend their work as a socially necessary celebration of ‘pleasure and desire’ against more puritanical social formations. The Erotic Professional, Mac and Smith argue, views herself ‘as answering a vocational “calling” that seems barely to have anything to do with being paid’.15 She freely chooses this sexual profession and insists it is ‘My Business’. Those who focus on the sex of prostitution to argue for decriminalization are therefore opposed to the stigmatization and ‘othering’ of sex workers’ private choice of sexual practice or sexual profession.16 12 Judith Walkowitz, ‘The Politics of Prostitution and Sexual Labour’ (2016) 82(1) History Workshop Journal 188, 189. 13 Cruz, Within and Against the Law (n 7); English Collective of Prostitutes website accessed 23 April 2019. 14 Gayle Rubin, ‘Thinking Sex: Notes for a Radical Theory of the Politics of Sexuality’ in Carole S Vance (ed), Pleasure and Danger: Exploring Female Sexuality (Routledge & Kegan 1984) 275; Cruz, Within and Against the Law (n 7); Andrew Bainham and Belinda Brooks-​Gordon, ‘Reforming the Law on Sexual Offences’ in Belinda Brooks-​ Gordon and others (eds), Sexuality Repositioned: Diversity and the Law (Hart Publishing 2004) 261. 15 Juno Mac and Molly Smith, Revolting Prostitutes: The Fight for Sex Workers’ Rights (Verso 2018) 31. 16 Cruz, Within and Against the Law (n 7).

Prostitution, Unfreedom, and Criminality at Work  197 It is encouraging that in countries where some measure of decriminalization has occurred, or crimes against sex workers are treated as hate crimes, sex workers feel less stigmatized and more humanized.17 Yet continued discrimination against LGBTQ people, despite decriminalization and anti-​discrimination legislation, provides us with ample evidence of the persistence of stigma. A different critique of the focus on the sex of prostitution comes from within the sex worker rights movement. Addressing the ‘Erotic Professional’, Mac and Smith ‘sympathize with the wish to over-​emphasize pleasure, freedom, or power’.18 It can, they argue, feel better than being stigmatized, and can serve as a useful counter to those who consider sex workers to be victims. And Mac and Smith are not denying that some sex workers take sexual pleasure in their work.19 However, they argue that focusing on the sex of prostitution can gloss ‘over the material conditions of sex workers’ workplaces’20 and distract from the ultimate goal of securing labour rights and protections. For these activist critics, the material conditions of sex work—​as a form of gendered and racialized labour that subjects sex workers to varying levels of economic exploitation, abuse, and violence—​ are made less visible through a sole discursive association with minority sexual practices or as an idealized sexual profession. The work position elaborated by Mac and Smith is in line with the views of UK-​based sex worker rights activists, including the ECP and the x:talk project, and Argentinian activists AMMAR.21 It also chimes with the Marxist Feminist understanding of prostitution as a form of sexual labour.22 While prostitution is work, it is not necessarily good or socially valuable work. Sex work is on a par with the many other gendered, racialized, unrewarding, often degrading and highly exploitative service-​based jobs that are available to workers. Other relevant examples are domestic and care work. An important criterion that makes sex work the least bad option is that it often pays better than the alternatives. Take the example of Jamaica. The minimum wage, which is what tourism workers are paid and where most workers who are employed work, is a mere £40 a week. This can be made each day selling sex.23 The economic reality in the UK is not so different.24 Sex work, then, does not have to be valuable or decent (to the self or society) for it to be work and it is possible for workers to choose to engage in exploitative and degrading work. Choice and exploitation or degradation are not mutually exclusive. Those who focus on the work of prostitution therefore insist that decriminalization is necessary to address stigma and for recognition that sex workers are part of the working class.25 But it does not follow that sex workers in decriminalized jurisdictions have been recognized as workers in the eyes of the law. 17 Rose Campbell, ‘Not Getting Away with it:  Linking Sex Work and Hate Crime in Merseyside’ in Neil Chakraborti and Jon Garland (eds), Responding to Hate Crime: The Case for Connecting Policy and Research (Policy Press 2014); Cruz, Within and Against the Law (n 7); Mac and Smith (n 15); Gillian Abel and others, Taking the Crime Out of Sex Work: New Zealand Sex Workers’ Fight for Decriminalization (Policy Press 2010). 18 Mac and Smith (n 15) 33. 19 ibid  33–​37. 20 ibid 35. 21 Cruz, Within and Against the Law (n 7); Hardy and Cruz (n 7). 22 Cruz, ‘Beyond Liberalism’ (n 7). 23 Cruz and others, ‘The Tourism Industry in Jamaica’ (n 7). 24 English Collective of Prostitutes, ‘What’s a Nice Girl Like You Doing in a Job Like This? Comparing Sex Work with other Jobs Traditionally Done by Women’ (2019) accessed 23 April 2019. 25 Hardy and Cruz (n 7); Mac and Smith (n 15); Juno Mac, ‘Selling Sex Is a Working Class Job. It’s Time for Labour to Stand with Sex Workers’ (24 September 2018)  accessed 23 April 2019.

198  Criminality at Work In the next section, I unpack the work position’s assumption that sex workers in brothels have labour rights and that criminal law regulation ends when labour rights begin.

C.  Prostitution as Work: Labour and Criminal Law The prostitution as work position’s support for decriminalization is seldom linked to a clear vision of labour rights or criminal law regulation of personal work relations. What explains the gulf between the claim that sex workers are workers and the total lack of labour rights enjoyed in decriminalized jurisdictions? And if brothels employed sex workers what might be the ongoing role of criminal law in the workplace? I have engaged with the first question in my existing research. I offer a summary of this research to contextualize the second question, which I then pursue in the remainder of this chapter. The prostitution as work approach argues that prostitution should be viewed as a ‘job like any other’ and demands labour and human rights.26 These rights generally include: the right of refusal, health and safety protections including a safe working environment, anti-​ discrimination protections, including sexual harassment, adequate rest breaks, freedom of association, right to a private life, freedom from forced labour.27 Sometimes it is assumed that these rights are already enjoyed. Mac and Smith argue that in New Zealand brothels are now ‘governed by the same labour laws and employment protections that apply to many other workplaces’.28 But in all countries where prostitution has been decriminalized and/​or legalized sex workers are not enjoying the rights of employees or workers.29 I have previously argued that sex workers in the UK—​specifically erotic dancers—​are working in conditions of false self-​employment. The similarity in the organization of UK-​ based erotic dancing clubs and brothels make it reasonable to suggest that were prostitution viewed as legitimate labour those selling sex would also be falsely self-​employed.30 The lack of rights enjoyed therefore conflicts with the reality of working conditions in brothels. Taking UK labour law as an example, brothel-​based workers appear to fulfil the criteria for worker, and arguably employee, status.31 The ‘right of refusal’ that is demanded by sex workers and activists might be thought to prohibit a finding of employee status. A legislative example is provided by section 17 of the New Zealand Prostitution Reform Act 2003.32 26 Cruz, Within and Against the Law (n 7); Mac and Smith (n 15). 27 Mac and Smith (n 15); Scoular (n 6). 28 Mac and Smith (n 15) 194. 29 Cruz, ‘Unmanageable Work’ (n 7) and ‘Beyond Liberalism’ (n 3); Alice Orchiston, ‘Precarious or Protected? Evaluating Work Quality in the Legal Sex Industry’ (2016) 21(4) Sociological Research Online 1. 30 Cruz, ‘Unmanageable Work’ and Within and Against the Law (n 7); AL Daalder, Prostitution in the Netherlands Since the Lifting of the Brothel Ban (Boom Juridische Uitgevers 2007) argues that brothel-​based workers in the Netherlands are falsely self-​employed. 31 Cruz, ‘Unmanageable Work’ and Within and Against the Law (n 7); Ready Mixed Concrete (South East) Ltd v Minister of Pensions and National Insurance [1968] 2 QB 497. For a recent discussion of the elements of a contract of employment and worker status see Autoclenz Ltd (Appellant) v Belcher and others (Respondents) [2011] UKSC 41, [2011] 4 All ER 745. On ‘worker’ status see Pimlico Plumbers Ltd and another (Appellants) v Smith (Respondent) [2018] UKSC 29, [2018] 4 All ER 641. 32 Refusal to provide commercial sexual services: (1) Despite anything in a contract for the provision of commercial sexual services, a person may, at any time, ­refuse to provide, or to continue to provide, a commercial sexual service to any other person. (2) The fact that a person has entered into a contract to provide commercial sexual services does not of itself ­constitute consent for the purposes of the criminal law if he or she does not consent, or withdraws his or her consent, to providing a commercial sexual service.

Prostitution, Unfreedom, and Criminality at Work  199 Employee status requires demonstration of ‘control’ on the part of the employer, ‘mutuality of obligation’ to offer (employer) and perform work (employee), and an obligation to personally perform the work. However, the level of control exercised by an employer is only one factor to be considered and no longer holds the prominent place it once did in determining employee status. In any case, brothel-​based sex workers are subject to levels of control and discipline that would likely fulfil this requirement.33 It is also not clear that refusing a particular client would be viewed as a refusal to perform work per se, and many brothel workers are allocated hours on a rota system, face a fine or dismissal for non-​attendance, and are prohibited from substituting their labour.34 It may be that the toughest, but not insurmountable, legal hurdle to employee status is the byzantine payment system used in much of the sex industry.35 The following description is typical: A manager of the venue will typically set the prices for sexual services. Each worker will pay a percentage to the manager of the amount paid directly to her by the customer. The payment of a percentage of earnings accrued each shift or for each service is often characterized as a ‘commission fee’. The worker will be paid in cash or in vouchers by each customer, which must be turned in at the end of each shift or service, in order for deductions to be made. There are, however, social hurdles standing in the way of worker or employee status claims. Many sex workers would prefer to be ‘unmanaged’ for several reasons, including identification with the self-​employed ideal and desire for autonomy, migration status, and the transitory nature of the industry.36 At the same time, the last few years have been marked by an important judicial shift in the UK to looking beyond the contractual document to the reality of the working relationship. This has led to an increasing recognition that ‘gig economy’ workers are being falsely labelled as self-​employed.37 And while recent case law suggests that ‘substitution clauses’ might militate against a finding of employee or worker status, contractual documents and those who manage sex workers insist on personal service and prohibit the use of a substitute.38 It therefore seems likely that judicial opinion would be that brothel managers are falsely characterizing the sex-​work relationship and that, at least as brothels are currently organized, sex workers could access far greater protection (if, of course, prostitution related activities were decriminalized). The prostitution as work position is partially correct that brothels in decriminalized jurisdictions are ‘governed’ by labour law. However, de jure application of labour rights to brothels has not, to date, been matched by de facto application.39 Those who understand prostitution as work also make conflicting statements about what the continuing role of criminal law is, or ought to be, in personal work relations between sex workers and brothel owners. Sometimes they assume that labour law starts where criminal law ends.40 At the same time, criminal laws, including the prevention of offences against the person, such as 33 Cruz, ‘Unmanageable Work’ and Within and Against the Law (n 7). 34 ibid. 35 ibid; Einat Albin, ‘The Case of Quashie: Between the Legalization of Sex Work and the Precariousness of Personal Service Work’ (2013) 42(2) Industrial Law Journal 180. 36 Cruz, ‘Unmanageable Work’ and Within and Against the Law (n 7); Katie Cruz, Kate Hardy, and Teela Sanders, ‘False Self-​Employment, Autonomy and Regulating for Decent Work: Improving Working Conditions in the UK Stripping Industry’ (2017) 55 Industrial Relations Journal 274. 37 Autoclenz (n 31); Pimlico (n 31). 38 The Independent Workers Union of Great Britain, R (On the Application Of) v Central Arbitration Committee: Re: Deliveroo [2018] EWHC 3342, [2018] 12 WLUK 17. 39 Cruz, ‘Unmanageable Work’ and Within and Against the Law (n 7); Cruz, Hardy, and Sanders (n 36). 40 Mac and Smith (n 15) 202–​03.

200  Criminality at Work assault and rape, and labour abuses, including trafficking and forced labour, are promoted as securing the protection of highly vulnerable sex workers. Adding another layer of confusion, the prostitution as work position is often highly critical of the ‘carceral feminist’ agenda, which attempts to solve the ‘problems of prostitution’, including violence against women and trafficking, using a law and order style response.41 How, then, might activists and academics supportive of prostitution as work differentiate their approach to criminal law intervention from carceral feminism? In the next section of this chapter, I outline a Marxian methodology for evaluating the value of criminal law in personal work relations. I hope that academics and activists who take the work position will find this approach useful in future engagement with the role and limits of criminalization of personal sex work relations.

D.  Capitalism, Unfreedom, and Criminal Law The value of criminalization of personal work relations needs to be assessed in the following descriptive and normative context. Unfreedom—​exploitation and alienation—​are structural features of capitalist societies. This lived reality of unfreedom is, however, masked by the appearance of capitalist social relations, the ideology of voluntary and equal exchange of our alienable property in the person (labour) for payment. These structural unfreedoms exist on a continuum, increase or decrease in severity according to the balance of class forces, but cannot be completely eliminated within class-​structured societies. Unfreedom, then, is not natural or inevitable and can be ameliorated by the conscious collective organization of workers. ‘Freedom’ is defined as workers’ conscious control over their labour and lives (the quotes referencing the fact that human emancipation of our labour is impossible within capitalist social relations). My approach draws from Marxist Feminism. First, unfreedom is gendered and racialized. Second, control over our labour and lives cannot be limited to freedoms in the workplace. ‘Free’ labour exists where waged and unwaged labour are embedded in a system of labour and social rights and protections. This section proceeds in three subsections: exploitation and alienation; the continuum of unfreedom and class struggle; history, unfreedom, and criminality at work.

1.  Exploitation and Alienation Marx’s early work exposed the contradictions of the Rights of Man and Citizen. The universal rights of freedom to practice religion, freedom to own property, and other Rights of Man mask unfreedom.42 Similarly, in his later critique of political economy, Marx argued that the freedom to be a free wage labourer masks unfreedom.43 What, then, did Marx see below the surface appearance of the freedom to be a wage labourer? 41 Mac and Smith (n 15); Melissa Gira Grant, Playing the Whore: The Work of Sex Work (Verso 2014); Elizabeth Bernstein, ‘Militarized Humanitarianism Meets Carceral Feminism: The Politics of Sex, Rights, and Freedom in Contemporary Anti-​trafficking Campaigns’ (2010) 36 Signs: Journal of Women in Culture and Society 45. 42 Karl Marx, ‘On the Jewish Question’ in David McLellan (ed), Karl Marx: Selected Writings (first published 1844, OUP 1977). 43 Karl Marx, Capital: Volume 1 (first published 1867) ch 6, www.marxists.org/​archive/​marx/​works/​1867-​c1/​ ch06.htm accessed 23 April 2019.

Prostitution, Unfreedom, and Criminality at Work  201 Marx’s mature writings reveal the irony that, once endowed with the legal right to private property, individuals were in fact dispossessed and alienated from the land and the means of production. Bourgeois political economy and liberal philosophy tell us that it was the fault of ‘lazy rascals, spending their substance, and more, in riotous living’, who ensured their dispossession, having ‘nothing to sell except their own skins’.44 Marx rejects this foundational myth and replaces it with the ‘actual history’ of primitive accumulation, in which ‘conquest, enslavement, robbery, murder, briefly force, play the great part’.45 Once dispossessed through means of force and enclosures, the landless had to sell their labour power in order to access necessities. Individuals were ‘freed up’ to sell their labour power to any buyer, a process represented as equality and freedom for both capitalist and worker. Marx is clear that the lived reality of this freedom is double-​edged. In many ways it is an improvement upon feudal relations and more direct forms of domination. But he was also clear that new, and more opaque, forms of unfreedom characterized capitalism. Exploitation can be objectively described in the following (simplified) way. The worker sells her labour power to the capitalist for a wage or payment. The capitalist also buys the essentials for production, including rent, machinery, and technologies. During her time at work she produces commodities, which can be a product or a service. This commodity must be greater in value than what the capitalist spends on her labour power and other essentials. Without receiving this greater value the capitalist would not make a profit. It is the value of the product (created by the labourer) for which she is not paid that is an ‘exact expression’ of her exploitation by capital.46 Exploitation of the working class is therefore structural; it is a calculation that all capitalists must perform and act upon in order for the system to function.47 The wage and investment in working conditions is relevant to how we calculate the rate of exploitation and there is an incentive for capitalists to keep these costs as low and ‘competitive’ as possible. Capitalists will use status-​based differences to reduce costs. Female and racialized workers are often found in poorly paid service-​based occupations. Presumed biological and cultural features of non-​white, Muslim, gypsy, and dispossessed and displaced Europeans (white supremacy, Islamophobia, xeno-​racism) and the cultural valuation of traits that are linked to masculinity (sexism), serve to structure divisions in the workplace between low-​paid, menial, and domestic occupations and higher paid and valued professions.48 The exploitation faced by all workers in the workplace, and so the profits generated by capitalists, are only possible because of the domestic and care work in the home, which is primarily performed by women. Without this work there would be no labourers ‘produced’, raised, and made ready for future exploitation. This is not to suggest that women are naturally suited for domestic and care work, in the home or the workplace. Rather, women’s presumed knack for domesticity and care represents a powerful ideology that has material origins in the transition to capitalism, and the emergence of a gendered division between paid/​productive and unpaid/​reproductive labour.49 Exploitation, then, is not simply 44 ibid ch 26. 45 ibid ch 26. 46 ibid ch 9. 47 ibid ch 9; Nancy Holmstrom, ‘Exploitation’ (1977) 7(2) Canadian Journal of Philosophy 353. 48 Nancy Fraser, ‘From Redistribution to Recognition? Dilemmas of Justice in a “Post-​Socialist” Age’ (1995) 212 New Left Review 68. 49 Joanne Conaghan, ‘Gender and the Labour of Law’ in Hugh Collins, Gillian Lester, and Virginia Mantouvalou (eds), Philosophical Foundations of Labour Law (OUP 2019); Silvia Federici, Re-​enchanting the World: Feminism

202  Criminality at Work illegitimate or severely inadequate working conditions, mistreatment, and use of ‘vulnerabilities’, such as racism, sexism, employment or immigrations status, although these factors are relevant. And exploitation cannot be eliminated because it is a necessary feature of the relationship between capital and labour. Our capacity to create is the essence of being human. Being ‘freed up’ for wage labour, Marx argued, alienates workers from the land, but it also alienates and estranges workers from their attributes, skills, knowledge, and emotions.50 Bourgeois political economy and liberal philosophy tells us that workers sell their detachable bodily property or labour for a temporary period of time. But labour is not an alienable ‘thing’. The view that labour can be detached from us and sold like any other commodity is a legal fiction. For Marxists, it is an objective fact that our labour power is embodied. As Holmstrom puts it, ‘labor power is unlike other commodities; unlike other things I “own,” it can’t be stolen or left on the bus!’.51 When labour power is commodified capitalists are buying our body and mind, albeit for a temporary period of time. The commodity produced by the workers’ embodied labour is taken away from the worker and placed under the control of the capitalist. Commodification estranges us from our ‘true’selves, others, and what we produce, and the very planet on which life we know has evolved and called home for millennia. An extensive literature documents how alienation in the labour process is gendered and racialized. As O’Connell Davidson points out, ‘employers can have an interest in workers whose bodies are imagined as unmarked (ie white, middle class, male), especially when those workers will occupy positions of power and authority’. On the other hand, workers ‘whose bodies are read as being “naturally” subordinate or caring are assumed to have a more authentic capacity to perform deference, servility and care, and employers thus have an interest in their bodies as such’.52 Hochschild’s classic study of the sale of airline hostesses’ ‘emotional labour’ is illustrative. It is the personality, the human capacity to care, nurture, and celebrate others that is being bought and commodified, hence controlled for the purposes of exploitation and so profit.53 Again, women’s presumed knack for care is relevant. It helps explain women’s over representation, and men’s ‘feminization’, in jobs that involve the sale of emotional labour. Like exploitation, alienation is therefore not restricted to severe or illegitimate working conditions and the abuse of ‘vulnerabilities’, including gender. Capitalism is the purchase and control of embodied labour (this could be emotional or manual labour) and the control of the commodities we produce when at work. Existing academic literature demonstrates how sex workers in prostitution are exploited and alienated in the Marxian sense of these concepts.54 For the purposes of this chapter and

and the Politics of the Commons (PM Press 2019); Nancy Fraser, ‘Behind Marx’s Hidden Abode: For an Expanded Conception of Capitalism’ (2014) 86 New Left Review 55. 50 Karl Marx, ‘Estranged Labour’ in his Economic and Philosophical Manuscripts (first published 1932) accessed 23 April 2019. 51 Nancy Holmstrom, ‘Sex, Work and Capitalism’ (2014) 13(3–​4) Logos: A Journal of Modern Society & Culture accessed 23 April  2019. 52 Julia O’Connell Davidson, ‘Let’s Go Outside:  Bodies, Prostitutes, Slaves and Worker Citizens’ (2014) 18 Citizenship Studies 516, 521. 53 Arlie Hochschild, The Managed Heart: Commercialization of Human Feeling (University of California Press 1983); see also Kathi Weeks, ‘Life Within and Against Work: Affective Labor, Feminist Critique, and Post-​Fordist Politics’ (2007) 7 Ephemera Theory and Politics in Organization 233. 54 Cruz, ‘Unmanageable Work’ (n 7) and ‘Beyond Liberalism’ (n 3); Holmstrom, ‘Sex, Work and Capitalism’ (n 51); O’Connell Davidson, ‘Let’s Go Outside’ (n 52).

Prostitution, Unfreedom, and Criminality at Work  203 my focus on prostitution that takes place in brothels, the following is most relevant: in its endless quest for new markets, hence profit, capitalism tends not to discriminate or take a moral stance on what should or should not be commodified. Sex sells and sex workers in brothels who work for a boss will be exploited and alienated. Gender, sexism, racism, immigration status, as well as poor pay and working conditions are clearly relevant factors that will exacerbate the exploitation and alienation of workers in brothels. However, it must be remembered that all workers who produce commodities (whether a product or a service) are exploited and alienated. While some workers may experience fulfilment and feel justly rewarded, exploitation and alienation are structural features of the wage labour contract. Without which profits would be impossible. The difference between workers, including sex workers, is not whether they are exploited or alienated, but the degree to which they are.

2.  The Continuum of Unfreedom and Class Struggle From a Marxian perspective, capitalism is not a natural state of affairs nor is it synonymous with work. After capitalism and the wage labour contract there will still be ‘socially necessary’ and ‘really-​free’ work to do or choose from.55 But workers would control this work and what is produced. However, exploitation and alienation and the effects of these processes can be mitigated through the collective organization of workers. And it is this organization that would lead workers to demand new forms of association and production. It is therefore useful to think about exploitation and alienation as unfreedoms that exist on a continuum. At one end of the continuum ‘freedom’ signifies control of labour in favourable working and living conditions. This is not true freedom because capital, left to its own devices, tends towards greater degrees of exploitation and alienation. At the opposite end, unfreedom signifies the lack of control of labour in favourable working and living conditions.56 ‘Free’ labour exists where waged and unwaged labour is embedded in a system of labour and social rights and protections. A number of protections for labour would ease the extent and effects of exploitation and alienation. Marx, of course, wrote favourably about successful collective organization to limit the working day as improving the wellbeing of workers.57 An obvious and important protection for workers today is the (re)establishment of a regulatory collective bargaining system.58 Other statutory rights and reforms should include, but are not limited to, a living wage or a basic income for all workers, the creation of a single and simplified employment status test for all workers that would capture all but the truly self-​employed, and increased protection of zero hour contract workers, including payment for accepted shifts of work that are cancelled.59 Unwaged labour protections should include access to childcare, eldercare, abortion, healthcare, and affordable housing and food. Such struggles to limit the commodification of our labour are the substance of class struggle. These are struggles that sex workers are engaged in and demand to be included as 55 Karl Marx, Capital: Volume III (first published 1894) ch 48  accessed 23 April 2019. 56 Cruz, ‘Beyond Liberalism’ (n 3). 57 Marx, Capital: Volume 1 (n 43) ch 10. 58 Keith Ewing and John Hendy, ‘New Perspectives on Collective Labour Law: Trade Union Recognition and Collective Bargaining’ (2017) 1(1) Industrial Law Journal 23. 59 Cruz, ‘Unmanageable Work’ (n 7) and ‘Beyond Liberalism’ (n 3); Workers (Definition and Rights) Bill 2017–​ 19  accessed 23 April  2019.

204  Criminality at Work part of the working class. In the UK, however, and many other countries, sex workers are effectively barred from formal worker organizing by the doctrine of illegality. The level of decommodification that I am proposing would ensure maximum freedom within capitalist social relations. The need to accumulate profit that defines capitalism does determine, ‘in the sense of exerting real pressures and placing real limits on’,60 the freedoms that can be won within capitalist social relations. In other words, maximum freedoms within capitalist social relations would be met with significant resistance that no amount of ‘renorming’ of labour law is likely to solve.61 In the next subsection I pursue the following question: Has criminal law, in its design and delivery, tended towards more, or less, freedom for workers?

3.  History, Unfreedom, and Criminality at Work Histories of the transition to capitalism in England, and its ongoing transformation in the nineteenth, twentieth, and twenty-​first centuries, have widely acknowledged the use of criminalization to facilitate the (continuing) accumulation of land and labour necessary to turn a profit. Criminalization is a constant feature of global capitalism and has been integral to nation building, colonization, and settlement projects.62 In this subsection, I argue that criminalization of personal work relations reveals a tendency, in design and/​or effect, and in combination with labour and immigration law, to facilitate the exploitation and alienation of formally free workers by penalizing their mobility and ability to exit or enter into work.63 Statutory provision for the compulsion of labour under threat of criminal punishment in England dates to the fourteenth century Statute of Labourers. Forms of compulsion into work that were contained in this statute fell into disuse by the eighteenth century. However, between 1720 and 1843 new statutes emerged, including the Master and Servant Act 1823. These laws allowed penal sanctions for breaches of contract at work, from quitting to disobedience at work, first in relation to specific trades and then more generally. Those convicted could find themselves confined in a house of correction for up to three months and forced to perform hard labour. 64 This was a time of rapid industrialization and there is little doubt that these ‘statutes were enacted in direct response to the heightened demand for labor created by expanding markets’.65 The eighteenth and into the nineteenth century also saw a proliferation of Vagrancy Laws. The array of offences covered by Vagrancy Laws grew with successive legislative amendments and judicial interpretation. Individuals could

60 Sue Ferguson, ‘Intersectionality and Social-​Reproduction Feminisms Toward an Integrative Ontology’ (2016) 24 Historical Materialism 38, 52. 61 Eric Tucker, ‘Renorming Labour Law:  Can We Escape Labour Law’s Recurring Regulatory Dilemmas?’ (2010) 39(2) Industrial Law Journal 99. 62 Robert Steinfeld, Coercion, Contract, and Free Labor in the Nineteenth Century (CUP 2011); Aziz Rana, ‘Settlers and Immigrants in the Formation of American Law’ (Cornell Law Faculty Publications Paper 1075, 2011) accessed 4 February 2019. 63 Julia O’Connell Davidson, ‘New Slavery, Old Binaries: Human Trafficking and the Borders of “Freedom” ’ (2010) 10(2) Global Networks 244. 64 Sidney Webb, Beatrice Webb, and Robert Alexander Peddie, The History of Trade Unionism (BiblioLife 2009); Steinfeld (n 62); O’Connell Davidson, ‘New Slavery’ (n 63); Paul Craven and Douglas Hay, ‘The Criminalization of “Free” Labour: Master and Servant in Comparative Perspective’ in Paul Lovejoy and Nicholas Rogers (eds), Unfree Labour in the Development of the Atlantic World (Routledge 1994). 65 Steinfeld (n 62) 43.

Prostitution, Unfreedom, and Criminality at Work  205 be detained, prosecuted, and housed in correction facilities for offences that were punishment for refusing, or failing, to respect private property interests in land and to integrate into waged labour; the unemployed, surplus labour, sex workers, the ‘idle and disorderly’, and those who took subsistence from the land, including wood and food.66 Organized labour fought back against the use of criminal law sanctions to limit the mobility of workers. Workplace protections were won, including regulation of the working day and the legalization of trade unions, and ‘regulatory’ criminalization of employers were put in place to uphold labour standards.67 This history of class struggle from below is one in which women and people of colour have played an important role.68 Partial decommodification to deregulation of the labour market describes the post-​war twentieth-​ century history of the UK. The fight for decommodification, while never complete for women and people of colour, or indeed sex workers, did return levels of labour and social welfare protections never seen before (or since) in the UK.69 The past forty years have been marked by deregulation as these rights and protections have been eroded. The demand for casual and numerically flexible labour has skyrocketed, rounds of austerity have stripped workers of necessary social protections and basic services, unionization has plummeted, no doubt hastened by successive curtailment of individual and collective and labour rights. Western nation states have, with increasing speed, introduced forms of criminalization at work, which are frequently framed as worker protective. Employers, traffickers, and other parties that subject workers to unfreedom in the labour process, including forcing workers to enter or remain in exploitative conditions, and that employ, or have employed, illegal workers can be penalized.70 These worker protective criminal laws intersect heavily with immigration law and increase the unfreedom of migrant workers in (at least) two ways. The Immigration Act 2016 aims to protect migrant workers from exploitation and British workers from having their wages undercut. One of the means that the Act prescribes is the criminalization of illegal migrant workers. It is, as Fudge stresses, undeniable that this tactic will backfire.71 The criminalization of illegal migrant workers will exacerbate rather than ameliorate unfreedom. The Modern Slavery Act (MSA) 2015 is the latest piece of legislation that aims to protect workers that are trafficked into sexual exploitation and Article 4 abuse. This Act implicitly criminalizes migrant workers. First, the doctrine of illegality bars trafficked workers from accessing remedies for work done and any poor treatment suffered by the employer.72

66 Steinfeld (n 62); O’Connell Davidson, ‘New Slavery’ (n 63); Nicolas Rogers, ‘Vagrancy, Impressments and the Regulation of Labour in Eighteenth Century Britain’ in Lovejoy and Rogers (n 64). 67 Marx, Capital:  Volume 1 (n 43); Christopher Frank, Master and Servant Law:  Chartists, Trade Unions, Radical Lawyers and the Magistracy in England, 1840–​1865 (Routledge 2016); Andrew Miles and Mike Savage, The Remaking of the British Working Class (Routledge 1994); David Gordon Wright, Popular Radicalism: The Working Class Experience 1780–​1880 (Routledge 2013); Steinfeld (n 62). 68 See, eg, Laura Schwartz, ‘ “What We Feel is Needed is a Union for Domestics Such as the Miners Have”: The Domestic Workers’ Union of Great Britain and Ireland 1908–​1914’ (2014) 25(2) Twentieth-​Century British History 173; Angela Davis, Women, Race and Class (Random House 1981). 69 O’Connell Davidson, ‘Let’s Go Outside’ (n 52). 70 The Modern Slavery Act 2015 consolidated legislation relating to trafficking for sexual and labour exploitation (s 2), punishable by imprisonment for twelve months or life (s 5); The Immigration Act 2016 introduces new offences of ‘illegal working’ (s 34) and ‘employing illegal worker’ (s 35), punishable by imprisonment for a maximum of fifty-​one weeks (s 34) or five years (s 35). 71 Judy Fudge, ‘Illegal Working, Migrants and Labour Exploitation in the UK’ (2018) 3(1) Oxford Journal of Legal Studies 557. 72 Hounga v Allan [2014] UKSC 27, [2014] 1 WLR 2889; Taiwo v Olaigbe [2016] UKSC 31, [2016] 1 WLR 2653.

206  Criminality at Work Second, sociologists, socio-​legal scholars, and activists have stressed that the enforcement of anti-​trafficking laws is resulting in a number of negative outcomes for migrant workers in practice, including deportation, prosecution, and deterioration of working conditions. The use of these laws in the sex industry is a case in point.73 The recent history of the criminalization of personal work relations differs from the eighteenth-​and nineteenth-​century focus on penalizing workers who refused or quit work. The end of the twentieth century, and into the twenty-​first, has been marked by criminalization of entry to work. Fudge makes the important observation that the overwhelming concern today is with ‘abusers’ of the system; ‘the “illegal worker” and “unscrupulous employer” figure as the objects of legal opprobrium’.74 O’Connell Davidson has highlighted similar contempt for ‘traffickers’ and ‘smugglers’.75 Criminalization mystifies the real sources of abuse and exploitation, including deregulation, restrictive immigration law, and the hostile environment, and it exacerbates, or does little to alleviate, the unfreedom of migrant workers. Fudge convincingly argues that there is logic at play here. Illegal working laws in the UK are being used to respond to the ‘governance crisis concerning the deteriorating outcomes for workers labouring in the bottom half of the UK’s neo-​liberal labour market’.76 British workers are being conned into thinking that the government’s light touch agenda for labour is the best approach and that the real problem is individual bad apples that ‘abuse’ the system. The recent criminalization of entry to work tells us that we cannot necessarily equate worker protective criminal laws with successful class struggle from below. These criminal laws may not be (migrant) worker protective, but they are (British) worker appeasing. In this section, I have argued that worker unfreedom exists on a continuum. Unfreedom is contested by class struggle from below and limited by paid and unpaid labour rights and protections. The history and present of criminalization reveals a tendency to limit the freedom of workers by restricting their ability to exit or enter work. In the next and final section I apply this methodology and respond to the question: If brothels in the UK were viewed as employing sex workers what might be the ongoing role of criminal law in the workplace?

E.  Prostitution as Work and Criminality at Work Analytical legal philosophy takes a conceptual approach to the rights and wrongs of criminal law, guided by respect for personal autonomy and a concern with serious public wrongdoing and prevention of the harm to others. Capitalist social relations are treated as merely politically contingent factors that have no necessary relationship to, or impact upon, criminal law. My view is that once we situate criminalization of personal work relations within the larger landscape of capitalist class struggle, it becomes clear that its proper use ought to 73 See, eg, Nick Mai, Migrant Sex Workers in the UK Sex Industry: Final Policy Relevant Report (ESRC 2009); The x:talk project, Human Rights, Sex Work and the Challenge of Trafficking (Creative Commons 2010); Julia O’Connell Davidson, ‘Will the Real Sex Slave Please Stand Up’ (2006) 83 Feminist Review 4; Angela Hill, “This Modern Day Slavery”: Sex Trafficking and Moral Panic in the United Kingdom (PhD thesis, University of California Berkeley 2011). 74 Fudge (n 71) 559. 75 O’Connell Davidson, ‘New Slavery’ (n 63). 76 Fudge (n 71) 560.

Prostitution, Unfreedom, and Criminality at Work  207 be mitigation of worker unfreedom. However, the dialectical other of class struggle from below is class struggle from above. We therefore have to pay equal attention to the capitalist interests that shape, and the effects of, criminality at work. As discussed in section B, there is now ample evidence demonstrating that treating prostitution-​related activities as sexual offences does not simply protect victims and persecute villains, if it does so at all. The effect of these laws is to construct sex workers as criminals and objects to be abused through their direct criminalization, or through a close association with illegal activities. This makes prostitution less safe and normalizes abuse on the basis that sex workers are engaging in immoral, harmful, and illegal activities. The effect of the illegality doctrine is to simultaneously bar access to labour and social rights and increase exploitation and alienation. But what might be the role of criminalization in the workplace if prostitution-​related activities were decriminalized? In previous research and in section C of this chapter, I have argued that if prostitution were decriminalized, sex workers in UK brothels would likely be characterized as falsely self-​employed. Would criminal law regulation of brothels then ensure more or less freedom for sex workers (as employees or ‘workers’)? Sex workers in brothels would benefit from exercising the ‘right of refusal’ and this right would likely apply regardless of employment status. Instances of coercion and violence from both bosses and clients would be dealt with as offences against the person. Tort and labour law could also deal with the crimes and mistreatment to which sex workers would likely be subject. It is possible that sex worker employees in brothels (and potentially their families in cases of homicide or death) could claim damages if shown that the employer breached their common law duty of care. This tort-​based duty requires that employers provide a ‘safe system of work’. This includes protection against physical and psychological injury and the employer is responsible for any failure to protect against common dangers, even when the worker is knowledgeable and experienced.77 This common law duty would arguably extend to any failure to adequately protect sex workers from the common danger of client violence.78 Criminal sanctions for the health and safety breaches of bosses would also most comprehensively cover sex worker employees.79 As employees, sex workers would have some health and safety responsibilities, but the employer would bear the brunt of them. There is, then, an important discussion to be had about the interaction of labour, criminal, and tort law for protecting sex workers as employees (or workers) from abuse and violence at work. The other area of criminal law that regulates poor treatment of workers by employers and third parties is contained in the MSA 2015. If prostitution-​related activities were no longer viewed as sexual offences, the Act would need to be amended to remove the distinction between sexual and labour exploitation. Prosecution would then be a possibility for those that subject sex workers in brothels to Article 4 abuses, including when these abuses constitute the ‘purpose’ of trafficking. This criminal law protection would apply regardless of employment status (although, as noted above, trafficked sex workers that are in the country illegally would have no contractual rights). Activists and academics who view prostitution as work argue for health and safety protections and protection against coerced or forced labour. Both strategies could, in theory, help

77

General Cleaning Contractors v Christmas [1954] AC 180. Mcleod v Aberdeen City Council 1999 CSOH 595. 79 Health and Safety at Work Act 1974, s 2. 78

208  Criminality at Work deliver freedom for sex workers. As employees, prostitutes in brothels would also enjoy the workplace freedoms they currently demand, as well as others. These should be combined with broader freedoms that relate to unpaid labour. Indeed, employers and the state ought to deliver these paid and unpaid labour freedoms for all sex workers, regardless of, for example, immigration status. However, the material reality of structural unfreedom and limits to what employers can offer without going out of business cannot go unacknowledged. Our current historical conjuncture is characterized by widespread unfreedom. There are powerful material interests favouring deregulation of the labour market, contraction of the welfare state, and the scapegoating of ‘abusers’ of the system. In the case of the two strategies considered in this section, then, sex worker employees will continually need to fight for these, and it is not at all likely that they will be accompanied by unwaged labour protections, including adequate childcare and housing. Increasingly, labour law scholars argue that anti-​trafficking and modern slavery legislation should, and could, be reoriented towards a worker-​protective model and labour protections for all workers, regardless of immigration status.80 However, it seems highly unlikely that concern with trafficking and modern slavery will be the vehicle in which protections for all workers will be delivered. This is because the raison d’etre of anti-​trafficking law is criminal law and border enforcement. It is also, as I have argued in this chapter, a strategic ‘worker protective’ concern being deployed by governments that in fact want to limit waged and unwaged labour protections. The usefulness of the MSA for sex worker employees would therefore be limited to the potential criminalization of their exploiter. At the same time, ‘illegal’ migrant sex workers will continue to experience extreme levels of unfreedom.81 They will be excluded from employee protections and could be penalized for illegal working. Their only real freedom will be the potential criminalization of their exploiter under the MSA. There are also limits to what many employers can offer service-​based workers within capitalist social relations. Take our example of prostitution in brothels. What is produced is not a tangible commodity that can be sold on to generate further profit. The commodity is the sexual service, which is extinguished in its single use. This is why bosses in the sex industry do not want their workers to be employees. Many would struggle to make a profit and would go out of business. It is not too dystopian to presume monopolies would emerge and sexbots would replace sex workers.82 This does not mean that nothing can be done, or that labour and criminal law lawyers should cease attempting to improve legal protections and access to justice for (sex) workers. But I do want to stress the contradictions and ambivalence of worker protective criminal law-​based strategies for sex worker employees or workers in brothels. Not all worker-​ protective criminal law intervention will contribute to the freedom of sex worker employees or workers, and degrees of unfreedom will persist. And any net gain in the freedom of sex worker employees and workers will not be extended to ‘illegal’ migrant sex workers.

80 Hila Shamir, ‘A Labor Paradigm for Human Trafficking’ (2012) 60 UCLA Law Review 76; Inga K Thiemann, ‘Beyond Victimhood and Beyond Employment? Exploring Avenues for Labour Law to Empower Women Trafficked into the Sex Industry’ (2018) 48(2) Industrial Law Journal 199. 81 Cruz, ‘Beyond Liberalism’ (n 3). 82 Tom Dart, ‘ “Keep Robot Brothels out of Houston”:  Sex Doll Company Faces Pushback’ The Guardian (2 October 2018) accessed 23 April 2019.

Prostitution, Unfreedom, and Criminality at Work  209

F.  Conclusion: Prostitution as Work, Work and Freedom The demand for decriminalization of prostitution is gaining momentum. It is increasingly accepted that criminal law intervention via a sexual offence framework makes workers less, rather than more, safe. The movement, allies, and academics also argue that sex work is a form of work that should not be criminalized. In this chapter, I have argued that the prostitution-​as-​work position is unclear about the content of their demands for labour rights and workplace protections, and what the continuing role of criminalization should, or could, be in the (sex)workplace. I have reiterated my existing position on the labour rights question, which is that if prostitution were viewed as legitimate employment, sex workers in UK brothels would likely be employees, at least in the eyes of the law. I then explored the potential role of criminalization of personal (sex)work relations. Understanding the possibilities and limits of criminality at work, while imagining brothel-​based sex workers as employees/​workers, requires attention to the relationship between capitalist social relations and class struggle. Unfreedom (exploitation and alienation) is the gendered and racialized reality of capitalism that ebbs and flows according to the balance of class forces. The potential and limits of criminalization of personal work relations therefore need to be understood and assessed according to whether it ensures more or less freedom for sex worker employees/​workers. I stressed, however, the contradictions that would accompany the extension of worker-​protective criminal and civil laws to sex worker employees. It is in the living of these contradictions that all (sex) workers will determine the usefulness of criminal law in achieving freedom within, and from, capitalist social relations. Ultimately, however, as some within the sex worker rights movement recognize, real freedom will entail moving beyond capital–​labour relations. This is because capitalism is premised on exploitation, alienation, and the destruction of the environment in the ceaseless quest for new markets. Meaningful collective control over work is anathema to capitalist societies. Work should not be reduced to wage labour, but rather recognized as a vital part of human existence. The goal is to minimize the need to work in order to survive and to maximize time for ‘really free working’.83 Whether, in the absence of unfreedom, people would choose to engage in ‘sex work’ as a ‘really free’ activity is a debate for another time.

83 Karl Marx, Grundrisse: Notebook VI. The Chapter on Capital (first published 1939-​41) accessed 7 February 2019.

11

Human Rights, Labour Rights, and Criminal Wrongs Virginia Mantouvalou*

A.  Introduction When does human rights law impose obligations to criminalize breaches of labour rights? This is the question that this chapter considers with a focus on the European Convention on Human Rights (ECHR or Convention). The European Court of Human Rights (ECtHR) has often ruled that the Convention imposes positive obligations to criminalize conduct, as section B explains. The requirement to criminalize certain violations of human and labour rights is grounded on an understanding that criminalization is an effective form of regulation, primarily because of its deterrent effect. As the Court has frequently emphasized that Convention rights have to be practical and effective, a duty to enact criminal laws aims to achieve exactly that by deterring people from engaging in the conduct in question.1 Another crucial factor to which the Court pays attention when considering whether certain conduct needs to be criminalized involves the nature of the wrongdoing. Some kinds of wrongdoing are such that the condemnation of criminal law is needed so as to influence behaviour, a purpose that civil law may not meet to the same extent. Arguments that conduct should be criminalized should not be made too quickly. They have to meet criteria of serious wrongfulness and harm. The imposition of obligations on states to criminalize violations of labour rights is a relatively new development in the case law of the ECtHR. The Convention has two provisions that explicitly protect labour rights: first, an individual labour right in Article 4 that prohibits slavery, servitude, forced and compulsory labour; second, a right to form and join trade unions in Article 11. The Court has ruled that Article 4 imposes an obligation to criminalize severe labour exploitation, which I examine in subsection C.1 of the chapter. Therein I discuss the case law of the Court, the type of wrong and harm experienced by the workers, and the kind of positive obligations revolving around criminalization, which the Convention imposes. I argue that criminalization is justified in these circumstances, though it is important to appreciate that criminal law has to be accompanied by effective enforcement, as well as other labour protective rules, without which criminal legislation is of limited value. The UK Modern Slavery Act, which I briefly discuss, exemplifies this point. * Professor of Human Rights and Labour Law, UCL, Faculty of Laws. Many thanks are due to Alan Bogg, Hugh Collins, Jennifer Collins, Keith Ewing, Michael Ford, Mark Freedland, John Hendy, Jonathan Herring, George Letsas, and all conference participants for comments and suggestions. 1 On occasion, ensuring compliance with international obligations, including the ECHR, also requires the authorities to amend existing criminal law legislation. Virginia Mantouvalou, Human Rights, Labour Rights, and Criminal Wrongs In: Criminality at Work. Edited by Alan Bogg, Jennifer Collins, Mark Freedland QC, and Jonathan Herring, Oxford University Press (2020). © Alan Bogg, Jennifer Collins, Mark Freedland, and Jonathan Herring. DOI: 10.1093/oso/9780198836995.003.011

Human Rights, Labour Rights, and Criminal Wrongs  211 Article 11 of the Convention has not been found to impose positive obligations to criminalize thus far. Nevertheless, some violations of labour rights under this provision, together with Article 8 (the right to private life), and Article 14 (the prohibition of discrimination) are sufficiently serious and wrongful that they justify and require criminalization, as subsection C.2 argues. Sometimes a duty to criminalize may be the only way to guarantee the rights in question. Civil law may not be sufficient to communicate the gravity of the wrong and deter it. To illustrate this point, I use the example of ‘blacklisting’. Blacklisting is a widespread practice in the UK construction industry, whereby workers who are members of a trade union, and who may therefore be perceived as ‘troublemakers’, are included in a secret list that employers consult. These workers lose their jobs and are unable to find work for years or decades, with grave implications for their well-​being, affecting their income, health, and personal life.2 Both the practice and the effects of blacklisting on workers’ lives are devastating. I suggest that Articles 8 and 11, standing alone or together with Article 14 of the ECHR, should form the basis of a positive obligation to criminalize blacklisting. The cumulative convergence of these rights provides in this case a strong normative argument for criminalization. The section D discusses two objections to introducing new criminal offences: first, the concern of overcriminalization; second, the concern of a possible exclusive focus on criminalization (rather than employing other forms of regulation). On the former objection, even though I  suggest that blacklisting meets the requirements of an especially serious wrong that causes especially serious harm that it ought to be criminalized, I do not of course suggest that all violations of collective labour rights should constitute criminal offences. On the latter objection, I do not suggest that other approaches to regulation, such as through the provision of civil remedies, are not important tools, which can co-​exist with criminalization. However, blacklisting is such a wrong that has caused such harm that the machinery of criminal law is appropriate in order to communicate the wrongfulness of the practice, to punish the perpetrators of this particular wrong, and to deter such systemic violations of trade union rights in the future.

B.  Human Rights and Positive Duties When the Convention was initially conceived, its aim was to place limitations on state action. Over the years, the Court developed a range of positive obligations on states with a view to ensuring the protection of rights in relations between private individuals.3 The Court has repeatedly ruled that human rights have horizontal effect in the employment relation. The extension of human rights in the workplace through the imposition of duties to protect workers against employer intrusions is justified. Human rights violations are moral wrongs that can be committed not only by state actors, but also by private individuals. Employers can exercise power on employees at times greater than state power, and impose 2 This has been documented extensively in publications of the Institute of Employment Rights. The first publication was by Keith Ewing: Keith D Ewing, Ruined Lives: Blacklisting in the UK Construction Industry (Institute of Employment Rights 2009). 3 See Dean Spielmann, ‘The European Convention on Human Rights—​The European Court of Human Rights’ in Dawn Oliver and Jorg Fedtke (eds), Human Rights and the Private Sphere (Routledge 2007) 427. On this, see also the landmark Marckx v Belgium (1979) 2 EHRR 330.

212  Criminality at Work restrictions on their freedom that are incompatible with liberal values:4 the protection of human rights at work can serve to address that. Many ECHR rights have been ruled to give rise to positive obligations applicable in the employment relation, such as the right to private life and freedom of religion,5 but the central Convention provisions that are relevant to labour law are Articles 4 and 11. The obligation to criminalize conduct that violates human rights has been developed in the context of the application of human rights law in relations between private individuals. The legal basis of positive obligations under the ECHR is Article 1, which provides that ‘[t]‌he High Contracting Parties shall secure to everyone within their jurisdiction the rights and freedoms defined in Section I of this Convention’. It is well-​established that Articles 2 (the right to life) and 3 (prohibition of torture, inhuman and degrading treatment), two non-​derogable provisions in times of war and other public emergencies under Article 15 of the ECHR,6 impose extensive positive obligations to deploy the criminal law.7 However, it is important to appreciate that a leading judgment of the Court on positive obligations to criminalize was X and Y v Netherlands,8 which involved the right to respect for private life under Article 8, a derogable and qualified provision. By saying that a right is qualified we mean that it can be restricted if the limitation pursues a legitimate aim, and is proportionate to the aim pursued. In this case, the complaint was that no criminal proceedings could be initiated against a person who raped a young girl with learning disabilities, because the sixteen-​year old lacked legal capacity to appeal against the decision of the prosecutor not to bring criminal charges against the alleged rapist. In examining the claim that criminal law was essential in order to address the wrong that the girl suffered, the Court noted that states have a margin of appreciation as to the means by which they will secure protection of Article 8, and that the type of state obligations that the Convention imposes depends on the aspect of private life that has been affected. ‘Recourse to the criminal law is not necessarily the only answer.’9 But it continued: [T]‌he protection afforded by the civil law in the case of wrongdoing of the kind inflicted on Miss Y is insufficient. This is a case where fundamental values and essential aspects of private life are at stake. Effective deterrence is indispensable in this area and it can be achieved only by criminal-​law provisions; indeed, it is by such provisions that the matter is normally regulated.10

The fact that Article 8 has been ruled to impose positive obligations to criminalize suggests that it is not the absolute or non-​derogable nature of a Convention provision that is crucial when considering the type of positive obligation that exists. What seems to have been crucial in X and Y was the nature of the wrongdoing that the applicant suffered, namely the 4 Hugh Collins, ‘Is the Contract of Employment Illiberal?’ in Hugh Collins, Gillian Lester, and Virginia Mantouvalou (eds), Philosophical Foundations of Labour Law (OUP 2018); Elizabeth Anderson, Private Government: How Employers Rule Our Lives (and Why We Don’t Talk about It) (Princeton University Press 2017). 5 On art 8, see Barbulescu v Romania App no 61496/​08 (Grand Chamber, 5 September 2017); IB v Greece App no 552/​10 (3 October 2013); on art 9 see Eweida and Others v UK App nos 48420/​10, 59842/​10, 51671/​10, and 36516/​10 (15 January 2013). 6 Art 15 provides that derogation from art 2 is permitted only for deaths resulting from lawful acts of war. 7 See further Andrew Ashworth, Positive Obligations in Criminal Law (Hart 2015) 200–​06. 8 X and Y v Netherlands App no 8978/​80 (26 March 1985). 9 ibid [24]. 10 ibid [27].

Human Rights, Labour Rights, and Criminal Wrongs  213 violation of her sexual integrity that constitutes a central aspect of the right to private life, and the practical and effective protection of which requires criminalization of the relevant conduct.11 The positive obligations to criminalize under the ECHR were categorized by Ashworth as follows:12 first, an obligation to secure a right by enacting effective criminal laws, supported by law enforcement mechanisms to prevent, suppress, and sanction; second, a duty to adopt preventive operational measures to protect someone who is at risk of a crime by another person; and third, an obligation to have an effective mechanism for the investigation of allegations of violations of Convention rights, as well as an obligation to have an effective investigation that can lead to the identification and punishment of the perpetrators. Discussing this same issue, Tulkens divided positive obligations to criminalize into two categories: substantive and procedural. The substantive obligation involves a duty to enact effective criminal laws to deter the commission of an offence.13 The procedural obligation contains an obligation to investigate effectively, including a duty to have an effective judicial system that may not only award damages, but which should also lead to the identification and punishment of the perpetrators.14 The investigation has to be prompt in order to maintain public confidence in the state response to a crime.15 Finally, these procedural obligations contain a line of requirements, such as criminalization in primary legislation, interpretation of the criminal legislation, prosecution policy and form, sentencing and execution of sentence.16 These obligations seek to ensure that rights of ‘fundamental importance’,17 as the Court puts it, are practical and effective rather than theoretical and illusory.18 By characterizing Articles 2 and 3 as rights of fundamental importance, the Court suggests that it is their nature as non-​derogable provisions under Article 15 that gives them a status possibly higher than other rights that are derogable. However, the fact that Article 8 was the first provision that gave rise to a positive duty to criminalize shows that the crucial factor is the nature of the wrongdoing and the harm that it generates, rather than the nature of the Convention provision at stake. Against this background, it is possible to imagine positive duties to criminalize stemming from any provision of the Convention, if the wrongdoing and the harm are sufficiently serious.

C.  Criminalization of Breaches of Labour Rights What are the concrete positive obligations to criminalize violations of labour rights? The section that follows examines individual labour rights, and then turns to collective labour rights. 11 ibid [30]. 12 Ashworth, Positive Obligations (n 7) 198. 13 Mahmut Kaya v Turkey App no 22535/​93 (28 March 2000), on art 2; A v UK App no 25599/​94 (23 September 1998), on art 3. 14 McKerr v UK App no 28835/​95 (4 May 2001) [121]. 15 Isayeva v Russia App no 57959/​00 (24 February 2005) [213], on art 2; Khashiyev and Akayeva v Russia App nos 57942/​00 and 57945/​00 (24 February 2005) [177], on art 3. 16 Francoise Tulkens, ‘The Paradoxical Relationship between Criminal Law and Human Rights’ (2011) 9 Journal of International Criminal Justice 577, 586–​87. 17 Khasiyev (n 15) [183]. 18 On the principle that rights have to be practical and effective, see Airey v Ireland App no 6289/​73 (9 October 1979).

214  Criminality at Work

1.  Severe Labour Exploitation The first time that the Court ruled that the Convention imposes positive obligations to criminalize in the labour law context was under Article 4, a non-​derogable provision under Article 15 of the ECHR, in the landmark case Siliadin v France.19 The facts of the case illustrate the severity of the moral wrong of exploitation with which the Court was faced, which led to the imposition of a positive duty to criminalize. Siliadin involved the ill-​treatment and abuse experienced by a migrant domestic worker, who was a minor at the time. The applicant arrived to France from Togo, with someone who had agreed with her father that she would be sent to school, that her status would be regularized, and that she would work until her plane ticket was paid off. In reality she worked for him for a period of time, and then was lent to a family for whom she worked as a domestic worker, sleeping in the children’s room, and working for fifteen hours a day, with no day off, without being paid or being sent to school, with no ID documents, and with undocumented immigration status. Before the Strasbourg Court, Siliadin claimed that the failure of France to have in place criminal legislation to penalize the exploitation that she suffered amounted to a breach of Article 4, because there was no effective way to prevent the crime and punish the perpetrators of the ill-​treatment. In addition, she argued that the existence of civil proceedings and remedies were inadequate for they could not afford the necessary protection to her integrity. France claimed that existing legislation, including some criminal offences, met the requirements of the Convention. The ECtHR referred to its case law on Articles 8, 2, and 3 that emphasized that protection against rape requires criminal legislation, and that children and other vulnerable people require particularly effective protection against breaches of their integrity. It explained that ‘together with Articles 2 and 3, Article 4 of the Convention enshrines one of the basic values of the democratic societies making up the Council of Europe’.20 Having discussed a list of international documents involving forced labour and exploitation of children, the Court said that: limiting compliance with Article 4 of the Convention only to direct action by the State authorities would be inconsistent with the international instruments specifically concerned with this issue and would amount to rendering it ineffective. Accordingly, it necessarily follows from this provision that States have positive obligations, in the same way as under Article 3 for example, to adopt criminal-​law provisions which penalize the practices referred to in Article 4 and to apply them in practice.21

The Court emphasized that Article 4 contains one of the most fundamental provisions in the Convention, in the sense that it is an absolute prohibition which does not permit derogations in times of emergency, and imposes an obligation to both criminalize and prosecute effectively.22



19

Siliadin v France App no 73316/​01 (26 July 2005). ibid [82]. 21 ibid [89]. 22 ibid [112]. 20

Human Rights, Labour Rights, and Criminal Wrongs  215 The Siliadin judgment also considered the definition of the prohibited conduct. It said that slavery should be defined in light of the Slavery Convention of 1927 as ‘the status or condition of a person over whom any or all of the powers attaching to the right of ownership are exercised’. Servitude, on the other hand, involves an obligation to provide services under coercion, and includes an obligation to live in another person’s property and the impossibility of changing the person’s condition. On forced and compulsory labour, the Court relied on the International Labour Organization Forced Labour Convention No 29, which defines forced labour as ‘all work or service which is exacted from any person under the menace of any penalty and for which the said person has not offered himself voluntarily’. Siliadin raised awareness on the vulnerability and ill-​treatment of migrant domestic workers,23 and has been influential at national and international level.24 The severe exploitation of migrant domestic workers was rightly presented as a systemic wrong in France and elsewhere,25 which criminal law could deter.26 The serious violation of physical and mental integrity suffered by migrant domestic workers was ruled to meet the conditions for a crime to be enacted. The positive obligation to criminalize Article 4 violations was reiterated in subsequent case law,27 and the Court developed further extensive positive obligations in relation to criminalization, which mirror the case law on Articles 2 and 3.28 Special mention should be made of Rantsev v Cyprus and Russia,29 the first case that examined trafficking for sexual exploitation under the ECHR. This involved a young woman from Russia who went to Cyprus to work in a cabaret under an ‘artiste visa’, but soon left her job and tried to return to her country of origin. Her employer found her and led her to the police station, but the police returned her to the custody of the cabaret owner. The young woman was taken to a flat and was found dead a few hours later, in an apparent suicide. The case was taken to Strasbourg by the father of the victim. Trafficking for sexual exploitation is not explicitly covered in Article 4. However, in line with its case law on the interpretation of the Convention as a living instrument, the Court ruled that human trafficking ‘by its very nature and aim of exploitation, is based on the exercise of powers attaching to the right of ownership’,30 treating people as commodities, often for little or no pay, involving close surveillance, violence, life and work in poor conditions. It went on to say that trafficking is contrary to human dignity and other fundamental values. As a result, there is no need to identify whether it should be classified as slavery, servitude,

23 Virginia Mantouvalou, ‘Servitude and Forced Labour in the 21st Century: The Human Rights of Domestic Workers’ (2006) 35 Industrial Law Journal 395. See also the discussion by Holly Cullen, ‘Siliadin v France: Positive Obligations Under Article 4 of the European Convention on Human Rights’ (2006) 6 Human Rights Law Review 585. 24 For detailed discussion of the case in documents of international organizations, see Forced Labour and Human Trafficking:  Casebook of Court Decisions (International Labour Organization 2009). See also Maria Fernanda Perrez Solla, ‘Slavery and Human Trafficking: International Law and the Role of the World Bank’ (2009) Social Protection and Labour Paper No 0904, 52. For discussion of the judgment by non-​governmental organizations, see Human Rights Watch Report, ‘As If I Am Not Human’ Chapter V (7 July 2008). For references to the case in parliamentary documents and debates, see UK HL Debs, 5 November 2009, col 400. 25 Siliadin (n 19) [49]ff. 26 ibid [143]–​[144]. 27 CN v UK App no 4239/​08 (13 November 2012); CN and V v France App no 4239/​08 (13 November 2012). 28 For detailed discussion, see Vladislava Stoyanova, Human Trafficking and Slavery Reconsidered (CUP 2017) ch 8. 29 Rantsev v Cyprus and Russia App no 25965/​04 (7 January 2010). 30 ibid [281].

216  Criminality at Work forced and compulsory labour, but it is sufficient to say that it is contrary to Article 4.31 This move was criticized for not examining in depth the particularities of human trafficking under the Palermo Protocol, and its relationship to labour exploitation.32 What is pertinent for present purposes is that Rantsev ruled that the victim suffered a serious moral wrong that violates Article 4 and gives rise to positive duties under the Convention. It is important to emphasize that criminalization is not sufficient for a state to comply with Article 4 obligations. In Rantsev, the Court ruled that the state also has to take positive operational measures to protect victims or potential victims from violations.33 This positive obligation is triggered when the authorities know or ought to have known that someone is a victim or at a real risk to be a victim of treatment contrary to Article 4, and they take no action to protect the individual. This obligation was not met in Rantsev where the authorities failed to examine whether the young woman was a victim, but instead returned her to the custody of her exploiter, or more recently in Chowdury v Greece,34 where Greek authorities knew of the exploitation of undocumented strawberry pickers, but did not take adequate measures to protect them from violations of Article 4.35 In addition, the Convention also imposes an obligation to investigate effectively when there is a suspicion that someone is a victim of treatment contrary to Article 4.36 For an investigation to be effective, it has to be independent and to have potential to lead to the identification and punishment of those who are responsible for the crime. Investigation has to be prompt, and the victim or next-​of-​kin have to be involved in the investigation to better safeguard their interests.37 When there is a case of human trafficking, the authorities have to cooperate with the relevant authorities abroad if there is a cross-​border element.38 That Article 4, along with Articles 2 and 3 of the ECHR, has given rise to positive obligations to criminalize should not come as a surprise. It is worth noting that criminalization of severe labour exploitation, and particularly human trafficking, is not a novel regulatory response to this social problem. It was partly inspired by the international law of human trafficking, which is extensively discussed in Siliadin, Rantsev, and other subsequent case law.39 The European Social Charter, which is the counterpart of the Convention in the area of social and labour rights, also imposes an obligation to criminalize all acts of sexual exploitation against children under Article 7 paragraph 10, which involves the protection of children and young persons from labour exploitation.40 Severe labour exploitation is a grave wrong consisting in taking advantage of workers’ vulnerability, which is often due to their immigration status, and imposing on them appalling working and living conditions.41 The harm suffered by workers involves their mental, 31 ibid [282]. 32 Ashworth, Positive Obligations (n 7) 208; Jean Allain, ‘Rantsev v Cyprus and Russia: The European Court of Human Rights and Trafficking as Slavery’ (2010) 10 Human Rights Law Review 546. 33 Rantsev (n 29) [286]. 34 Chowdury and others v Greece App no 21884/​15 (30 March 2017). 35 ibid [111]–​[115]. 36 Rantsev (n 29) [288]. 37 LE v Greece App no 71545/​12 (21 January 2016) [68]. 38 Rantsev (n 29) [89]. 39 On a discussion of a criminal law approach to human trafficking, see Bridget Anderson and Rutvica Andrijasevic, ‘Sex, Slaves and Citizens: The Politics of Anti-​Trafficking’ (2008) 40 Soundings 135; Hila Shamir, ‘A Labor Paradigm for Human Trafficking’ (2012) 60 UCLA Law Review 76. 40 See for instance European Committee of Social Rights: Conclusions 2011 (Bosnia Herzegovina) (Council of Europe January 2012). 41 In Rantsev it resulted in her death.

Human Rights, Labour Rights, and Criminal Wrongs  217 physical, or sexual integrity. Such violations of personal integrity have in the past given rise to obligations to criminalize under the Convention, and Article 4 case law reiterates the principle that the state response to such wrongs may require the machinery of criminal law. To the extent that criminal law is viewed as the best way to protect fundamental values that human rights embody, criminalization demonstrates societal disapproval, while criminal penalties can also have a preventive or deterrent function.42

2. The Modern Slavery Act 201543 Against the background of the landmark developments in the field and in order to respond to Article 4 case law, the UK enacted the Modern Slavery Act 2015 (MSA), which codified criminal legislation on slavery, servitude, forced and compulsory labour, and human trafficking.44 It can be said that the MSA meets ECHR obligations to criminalize under Article 4, but it is very questionable whether it meets procedural obligations to enforce the law effectively. The stated purpose of the MSA consisted in facilitating the work of prosecutors and the police with regard to modern slavery,45 and increasing the rates of prosecutions, which were viewed as low.46 However, the Act has not led to a significant increase in the number of identifications of victims and prosecutions, which led to criticisms in a National Audit Office Report.47 Existing data suggests that even though there has been a small increase in prosecutions, the numbers are still strikingly low if assessed against National Referral Mechanism (NRM) referrals.48 Between 2015 and 2016, there were 3,146 NRM referrals. In 2016 there were just eighty prosecutions under the MSA, rising from twenty-​six prosecutions in 2015. 49 In 2017, there were 5,145 referrals, while in 2017–​18, there were just 239 prosecutions.50 Even though the number of prosecutions has increased, it is still very low, and there is a striking discrepancy between the number of prosecutions and the number of referrals of victims of human trafficking through the NRM. The underenforcement of the MSA is partly due to the police response to the crime. In October 2017, a police watchdog, Her Majesty’s Inspectorate of Constabulary and Fire & Rescue Services (HMICFRS), produced a report which was highly critical of the police on this matter for a number of reasons, including a lack of understanding of the issue, ineffective investigations, and the frequency with which victims were referred to immigration 42 See analysis of these in Tulkens (n 16) 590–​91. More generally on the criminalization of exploitation, see Jennifer Collins, ‘Exploitation of Persons and the Limits of the Criminal Law’ (2017) Criminal Law Review 169. 43 I discuss the MSA in much more detail in Virginia Mantouvalou, ‘The Modern Slavery Act Three Years On’ (2018) 81 Modern Law Review 1017. 44 Sexual Offences Act 2003, ss 57–​59; Asylum and Immigration (Treatment of Claimants) Act 2004, s 4, as amended by Protection of Freedoms Act 2012, ss 109 and 110, and Coroners and Justice Act 2009, s 71. 45 HC Deb 8 July 2014, vol 584, col 171. 46 For instance, in 2013 there were only sixty-​eight convictions: HC Deb 8 July 2014, vol 584, col 175. The MSA also contains a number of preventive measures, backed by criminal sanction upon breach, discussed in Ashworth and Collins, this volume. 47 Report by Comptroller and Auditor General, Reducing Modern Slavery (HC 630, National Audit Office 15 December 2017). 48 The NRM is the main administrative mechanism dealing with modern slavery allegations and identifying victims. 49 National Audit Office Report (n 47) [4.8]. 50 Crown Prosecution Service, Modern Slavery Report 2017–​18 (CPS August 2018).

218  Criminality at Work authorities instead of being provided with support.51 When it comes to both the identification of victims and the prosecution of individual perpetrators of the crime, in other words, the MSA has thus far failed to meet its stated purpose, as well as the positive obligations under the ECHR to investigate and prosecute effectively any violations of Article 4. I will return to positive obligations and the MSA and to how it falls short of Article 4 requirements in section D of this chapter.

3.  Collective Labour Rights and Blacklisting Do violations of collective labour rights justify the imposition on states of positive duties to criminalize? This has not occurred so far in ECtHR case law, but the remainder of this chapter argues that such duties should be recognized when there is a sufficiently serious breach of the Convention. This analysis inevitably gives rise to the following question: can obligations to criminalize be triggered under qualified and derogable provisions of the ECHR? We saw earlier that Article 8 has been found to trigger such an obligation in a case that involved rape. I argue that we should also recognize such duties in response to breaches of collective labour rights, and develop my argument using blacklisting as an example. Blacklisting serves as one instance of a serious breach of collective labour rights that requires criminalization, but it should not be viewed as the sole example. Blacklisting is the practice of compiling information on individuals regarding their trade union membership and activities, and sharing this information with employers and employment agencies so as to enable them to discriminate against those blacklisted in obtaining a job or in their treatment at work. The information serves as a reason to systematically deny employment to workers, and in the UK has primarily been used in the construction industry, where it has been revealed to be a widespread practice.52 This emerged in 2009, when the Information Commissioner’s Office published a report that revealed the extent of blacklisting by an organization called the Consulting Association. The predecessor of the Consulting Association was the Economic League that was founded in 1919, and included a ‘Services Group’ that was composed of construction companies that collated data on ‘potentially disruptive individuals’.53 The Consulting Association was formed in 1993. In 2009 it was revealed that it put together a secret blacklist on behalf of forty-​four multinational building contractors. The database contained sensitive personal information on over 3,000 individuals, primarily in relation to trade union membership of workers who had raised health and safety concerns. The information on ‘troublemakers’ was secretly put together by construction firms (many of which were large multinationals) that paid an annual subscription. Some of the workers were covertly monitored. While initially this appeared to be an activity of private companies, it was later revealed that Scotland Yard provided information 51 HMICFRS, ‘Stolen Freedom: The Policing Response to Modern Slavery and Human Trafficking’ (HMICFRS October 2017). On this, see further Candice Morgan, ‘Modern Slave or Illegal Worker? The Haze around Modern Slavery and its Implications’ LSE Politics and Policy Blog (17 November 2017) accessed 5 August 2019. 52 For an excellent discussion of the problem of blacklisting, see Hugh Collins, Keith Ewing, and Aileen McColgan, Labour Law (CUP 2012) 453ff. 53 See the discussion in the House of Commons Scottish Affairs Committee, Blacklisting in Employment—​ Interim Report (Ninth Report of Session 2012–​2013, 16 April 2013), 7. More generally the reports of the Scottish Affairs Committee on blacklisting provide a very valuable and useful account of the facts.

Human Rights, Labour Rights, and Criminal Wrongs  219 to the Consulting Association, which means that there was direct state involvement.54 The database was used by employers to check on workers who applied for jobs or were already employed. Many blacklisted workers were repeatedly dismissed and not hired to work for long periods of time. These workers’ description of the practice of blacklisting and its effects serves to illustrate the seriousness of the wrong and harm. A backlisted bricklayer described it as follows: The Blacklist is an economic, social and political prison. I have served a life sentence and other workers continue to be imprisoned. In cases like my own, the Blacklist effectively takes the form of house arrest because of its effect on a person’s social life. My wife was also deeply affected and badly scarred. More often than not, she was forced to financially support me, and our two children, on her low wage as a care worker. This has had a devastating effect on our standard of living. To her great credit my wife supported me and our family unstintingly. She held us together when things got really tough—​which it did quite often. We kept our dignity intact and just managed to keep our heads above water by almost completely sacrificing our social life. My wife had to take out loans, which we could not afford, since my credit rating was zero due to very long spells of unemployment. All of this is the direct result of the building employers deliberately using the Blacklist, time and again, to deny me the right to work and to earn a living.55

The wrongfulness of blacklisting consists in, first, abuse of a vulnerability that is due to economic dependence, and second, discriminatory treatment of individuals for the reason that they exercise a right that they have under the Convention (the right to be a trade union member or the right to free speech). The fact that the practice was concealed and coordinated, with dismissals and refusals of employment being inexplicable for years, is a particularly sinister and insidious aspect of the treatment. Blacklisting raises pressing questions under Articles 11 and 8 of the ECHR, alone or together with Article 14 (prohibition of discrimination).56 More precisely, being included in a list of trade union members compiled by employers to be used in hiring or retention decisions interferes with Article 11. This is contrary to the Convention for it cannot be seen as having a legitimate aim, as the aim of blacklisting is to discriminate against workers who exercise a Convention right. Because of the discriminatory element, the inability to obtain a job and dismissals on the basis of the blacklists also violates Article 14 together with Article 11. The harm caused by the wrongful conduct consists in dismissal, inability to obtain a job in their occupational field, poverty, and social isolation.57 This can also be described as a systemic and arbitrary denial, and hence a violation, of the right to work, with all the severe implications that this has for someone’s life.58 The right to work is not guaranteed explicitly 54 Rob Evans, ‘Officers Likely to Have Passed Personal Files to Blacklisters, Says Met’ The Guardian (23 March 2018) accessed 5 August 2019. 55 Collins, Ewing, and McColgan (n 52) 455. 56 Art 14 is not a free-​standing provision. It can only be violated when invoked together with some other Convention right. 57 Because of the harm caused by blacklisting, it is also described as having ruined lives. See Ewing, Ruined Lives (n 2). 58 On the value and legal protection of the right to work, see Virginia Mantouvalou (ed), The Right to Work: Legal and Philosophical Perspectives (Hart Publishing 2015).

220  Criminality at Work in the ECHR, but it has been protected in case law on other provisions.59 Blacklisting raises questions on the protection of the right to private life under Article 8, a provision that is interpreted broadly by the Court.60 The right to private life is implicated not only because the employers held and used workers’ personal information, a practice that interferes with privacy with no legitimate aim.61 There is a broader Article 8 issue. In Sidabras and Dziautas v Lithuania,62 the Court examined the applicants’ dismissal and ban from access to public and various branches of private sector employment for a period of ten years, for the reason that they were former KGB members. Examining the question whether Article 8 was engaged, it said that the ban ‘affected their ability to develop relationships with the outside world to a very significant degree and has created serious difficulties for them in terms of earning their living, with obvious repercussions on the enjoyment of their private lives’.63 The extensive implications of the restriction engaged the right to private life, and led to a violation of Article 14 together with Article 8.64 Blacklisted workers face serious restrictions on their ability to obtain a job, with devastating implications for their private life. Article 8 imposes several positive obligations,65 and has been found to give rise to a duty to criminalize conduct, as discussed earlier, even though it is a qualified and derogable provision. This is because what carries most weight in the relevant case law of the Court is not the nature of the Convention right as absolute, but the nature of the wrongdoing and the harm inflicted.66 The harm of blacklisting is evident in the following statement by a blacklisted bricklayer: You start to see the effects almost immediately on your wife and family. You don’t have money that you should have to be able to afford everyday things. That sometimes affects the kids, sometimes school holidays, eventually we couldn’t afford a social life, we couldn’t go out with friends, we couldn’t have meals, we couldn’t go to the pictures or things that you take for granted. Our social life was immediately cut off. Probably after three months, you start to really notice the economic and social deprivation. And my wife then had to take on another job and she had one then she had to take two and eventually she had to take three and that put almost impossible demands on her, physically and mentally.67

The wife of a blacklisted worker further explained the effects of blacklisting on private life: At first I resented my husband as he went on strike. I blamed him even though it wasn’t his fault. After that he wasn’t able to find work. It was really hard for both of us. I was earning 59 Virginia Mantouvalou, ‘The Protection of the Right to Work Through the European Convention on Human Rights’ (2014) 16 Cambridge Yearbook of European Legal Studies 313; Rory O’Connell, ‘The Right to Work in the ECHR’ (2012) European Human Rights Law Review 176. 60 See particularly the Grand Chamber judgment Denisov v Ukraine App no 76639/​11 (25 September 2018). 61 On this, see Barbulescu (n 5). 62 Sidabras and Dziautas v Lithuania App nos 55480/​00 and 59330/​00 (27 July 2004). 63 ibid [48]. 64 See further Virginia Mantouvalou, ‘Work and Private Life:  Sidabras and Dziautas v Lithuania’ (2005) European Law Review 573. 65 See, eg, IB v Greece (n 5). 66 X and Y (n 8) [27], [30]. 67 Dave Smith, Blacklisting: The Need for a Public Inquiry (Institute of Employment Rights 2017) 10–​11.

Human Rights, Labour Rights, and Criminal Wrongs  221 for the both of us so I had to do any job I could get. It made me feel quite resentful. As I had to be out working, it meant for a long time we couldn’t have a family. I had no choices. Because I had to be working, it put a strain on both of us. We had our daughter in 1986. We had planned to have more children but without my husband being in work we couldn’t afford to. After having my daughter, I had to go back to work so we had a source of income, but this was through necessity not choice. We were living on the breadline. We could never afford any holidays and weren’t able to go out and do things. There were times we couldn’t even afford the train fare to go and see my mother in Edinburgh. It was existence and only that.68

When the blacklisting scandal came to light, a fine of £5,000 was imposed on Ian Kerr, a person who held an important role in the Consulting Association (and its predecessor, the Economic League since 1969) for the reason that he processed personal information without having registered as a data controller with the Information Commissioner on the basis of the Data Protection Act 1998. None of the companies that were members of the Consulting Association were prosecuted. Some blacklisted workers applied to employment tribunals, and in 2010 the Employment Relations Act 1999 (Blacklists) Regulations 2010 came into force, which made blacklisting a civil, but not a criminal wrong. The Regulations include provisions that define blacklisting, make it unlawful for organizations to refuse employment, dismiss or cause detriment to individuals who are included in a blacklist, and make it unlawful to agencies to refuse service to blacklisted workers. The regulations make blacklisting unlawful, but not a criminal offence. Despite the human rights implications, the Court declared inadmissible two cases on blacklisting,69 though it did accept that there had been interference with Convention rights.70 Mr Smith and Mr Brough were blacklisted and suffered significantly because of their inclusion in the lists. Mr Brough claimed that blacklisting violated Article 11 of the ECHR, but the Court dismissed his claim for non-​exhaustion of domestic remedies, which in this case consisted in not having raised an ECHR argument in employment tribunal proceedings. Mr Smith was an agency worker, who worked in the construction industry. He claimed that the collection of his personal data through blacklisting was in breach of Article 8, that English law did not protect him as an agency worker who was blacklisted because of his union activities, and did not provide him a remedy for that, and that he was not protected from discrimination as a union member and agency worker under Article 14. The ECtHR accepted that there had been an interference with Article 8, and explained that national authorities have a margin of appreciation in securing compliance with the provision.71 In this regard, it found that the authorities had made the retention of personal data a criminal offence under the Data Protection Act, and that they also allowed for a civil remedy, through which the applicant was awarded compensation. The Court ruled that the application was incompatible with the Convention ratione personae,72 68 ibid 11. 69 Brough v UK App no 52962/​11 (Admissibility Decision of 30 August 2016); Smith v UK App no 54357/​15 (Admissibility Decision of 28 March 2017). For discussion of the Brough decision, see Keith Ewing and John Hendy, ‘Article 11(3) of the European Convention on Human Rights’ (2017) European Human Rights Law Review 356, 373ff. 70 Smith (n 69) [36]. 71 Smith (n 69) [36]–​[37]. 72 ECHR, art 35(3).

222  Criminality at Work because the state had met its positive obligations under the Convention by criminalizing the unlawful retention of personal data, and by creating a civil remedy, which the applicant pursued and through which he obtained compensation. In reaching the decision, the Court took into account the fact that ‘it was the actions of private companies rather than the national authorities’.73 However, it is important to note that this occurred before the revelations that there was in fact state involvement in blacklisting.74 The Court also examined the admissibility of the case on the basis of Article 35(3)(c),75 and ruled that the applicant had not suffered ‘significant disadvantage’, because domestic courts recognized the injustice that he suffered, and he received compensation. A number of people received compensation in settlement proceedings in the context of class action before the High Court, the legal basis of which included conspiracy and defamation.76 The defendants also offered an apology for the secrecy and consequences of blacklisting. The ECtHR then turned to the question whether respect for human rights requires the examination of the complaint, which it rejected. This was because domestic courts had recognized the injustice of blacklisting, and Parliamentary and other national bodies had scrutinized and condemned the practice. In Smith and Brough the applicants did not argue that blacklisting as a practice should be a criminal offence, despite the fact that its wrongfulness is not limited to the retention of personal data that is a criminal offence under the Data Protection Act.77 However, the approach of the Court to the admissibility question whether respect for human rights compels it to examine the case, despite the fact that Smith enjoyed some legal protection and was awarded compensation domestically, suggests that the Court did not appreciate fully the moral wrong of blacklisting, its systemic nature, or the harm that it caused to the applicants, their work prospects, their relationships with others at work, as well as their private lives away from work. This was disappointing. The existence of a separate criminal offence that relates to personal data, as well as civil remedies, cannot be viewed as sufficient for the purposes of Articles 11, 8, and 14 of the ECHR, for when there is a serious moral wrong and harm, it is important for the criminal law to identify and condemn it as such. This can be grounded on the guiding principle of ‘fair labelling’, which suggests that a crime needs to be correctly labelled for reasons such as the fair treatment of the offenders78 and the public confidence in the law.79 The ECtHR has recognized this point using broad terms. It has emphasized that

73 Smith (n 69) [40]. 74 Evans (n 54). 75 Paragraph 3 of art 35 says that a case will be declared inadmissible if ‘(b) the applicant has not suffered a significant disadvantage, unless respect for human rights as defined in the Convention and the Protocols thereto requires an examination of the application on the merits and provided that no case may be rejected on this ground which has not been duly considered by a domestic tribunal’. On this admissibility criterion, see Dinah Shelton, ‘Significantly Disadvantaged? Shrinking Access to the European Court of Human Rights’ (2016) 16 Human Rights Law Review 303. 76 Further information on the background of the claims is available here:  Construction Industry Vetting Information Group Litigation, Claims listed in the CIVIG Group Register (29 November 2013) accessed 5 August 2019. 77 I am grateful to John Hendy QC for clarifying this issue, and for providing me with copies of the applications in Strasbourg. 78 See, among others, Andrew Ashworth, Principles of Criminal Law (5th edn, OUP 2006) 88–​89. 79 Victor Tadros, ‘Fair Labelling and Social Solidarity’ in Lucia Zedner and Julian Roberts (eds), Principles and Values in Criminal Law and Criminal Justice: Essays in Honour of Andrew Ashworth (OUP 2012) 67.

Human Rights, Labour Rights, and Criminal Wrongs  223 we need to have a criminal offence that captures the wrong, for otherwise the law will not be effective. In Siliadin it ruled that the existing criminal offences in French law were insufficient for they did not capture the wrongfulness and the harm suffered by victims of servitude, and forced and compulsory labour. France suggested that even though this conduct was not criminalized at the time, there were two other offences in the French Criminal Code that addressed the relevant wrong (obtaining from someone performance of services without pay, and subjecting someone to working or living conditions incompatible with human dignity by taking advantage of his or her vulnerability). The ECtHR responded that these provisions did not provide effective penalties. It noted that ‘the increasingly high standard being required in the area of the protection of human rights and fundamental liberties correspondingly and inevitably requires greater firmness in assessing breaches of the fundamental values of democratic societies’.80 Similarly, in X and Y v Netherlands, which was discussed earlier in relation to a positive duty to criminalize under Article 8, the Court rejected the argument of the government that existing criminal offences were sufficient for Article 8 obligations to be met.81 It is important to identify correctly the wrongful conduct both for purposes of effectiveness but also for purposes of fair labelling. The above reasoning on positive obligations to criminalize conduct under Articles 4 and 8 applies to the criminalization of blacklisting. The Data Protection Act 1998 makes the retention of personal data a criminal offence but this is not sufficient. Obtaining or retaining data is a distinct moral wrong that has to do with a person’s privacy, while blacklisting involves the taking advantage of the economic vulnerability of workers, and their discrimination for the reason that they exercise a Convention right. In addition, the harm of blacklisting does not only consist in a violation of privacy, as argued earlier. It consists in a systemic denial of the right to work with grave implications for private life. It should therefore be regulated as a distinct offence, in addition to existing criminal and civil law penalties. The principle of fair labelling supports the position that blacklisting has to constitute a distinct criminal offence, separate to offences under the data protection legislation. Finally, in relation to making the rights of the Convention practical and effective through deterrence, it is important to appreciate that the penalties under the Data Protection Act offence are low. For this reason, it has been argued that it provides no deterrence, while there is evidence that blacklisting continues to occur.82 The ECtHR has ruled that Article 11 requires ‘real and effective protection against anti-​ union discrimination’,83 but the legal framework in the UK does not meet this standard. The Blacklist Regulations make it unlawful to develop, provide, use, or purchase a ‘prohibited list’, namely one that contains people’s details because of their union membership and activities. However, the Regulations have several weaknesses: they do not have retroactive effect, so those who were blacklisted before the Regulations came into force could 80 Siliadin (n 19) [148]. 81 X and Y (n 8) [28]–​[30]. 82 IER, The Blacklisting Scandal: A History of Blacklisting, Recent Updates, and How the Law Should be Changed (Institute of Employment Rights 2013) 17. For instance, it is said that it occurred during the construction works before the Olympic Games. It is also important to note here that Ian Kerr’s fine of £5,000 was paid by McAlpine, a large construction company. Cullum McAlpine, who was Director of several companies in the McAlpine Group, was the first chairman of the Consulting Association. 83 Danilenkov v Russia App no 67336/​01 (30 July 2009) [124]. See also Wilson, National Union of Journalists and others v UK, App nos 30668/​96, 30671/​96, and 30678/​96, Judgment of 2 July 2002.

224  Criminality at Work not rely on them. Claims have to be brought to a tribunal three months after the offence, but many workers do not know that they are blacklisted for many years after they are included in the list.84 The Regulations do not provide for a right to compensation for blacklisting alone: proof of loss or injury to feelings is required. Moreover, an implication of the fact that blacklisting is only dealt with as a civil matter is that the blacklisted worker has to find evidence and take a claim to court. This can be an insurmountable obstacle given the covert nature of the practice.85 That blacklisting should be criminalized has been suggested in several fora. In parliamentary debates on blacklisting in 2017 it was argued that there is a strong case to criminalize it, and provide for severe sanctions, including high fines and possible imprisonment for the worst cases,86 and ban of companies that have engaged in blacklisting from award of public contracts.87 Representatives of trade unions giving evidence to the Scottish Affairs Committee of the House of Commons said that blacklisting would not cease unless legal sanctions against the practice were strengthened: It is happening, and it will continue in the UK until blacklisting is made a criminal offence punishable by imprisonment and unlimited fines. Until that legislation is brought in by whichever Government decides to bring it in, it will continue.88

The Institute of Employment Rights also proposed changes to the law on blacklisting, including criminal sanctions for employers operating blacklists.89 Given that there is evidence that blacklisting still happens, criminalization can serve to condemn employers who engage in the harmful conduct, and to reduce the instances of it.

D.  Overcriminalization/​Exclusive Focus on Criminalization There are two possible objections to the argument that blacklisting should be criminalized. The first one is the question whether we may extend duties to criminalize too far, which is an important concern expressed both in theory90 and in the context of the Convention.91 Tulkens explained in a concurring opinion in MC v Bulgaria:

84 IER, The Blacklisting Scandal (n 82) 14–​15. 85 Smith, Blacklisting (n 67) 25. 86 HC Deb 5 September 2017, vol 628, col 68WH. 87 Chuka Umunna MP, HC Deb 5 September 2017, vol 628, col 69WH. 88 House of Commons, Scottish Affairs Committee, Blacklisting in Employment: Addressing the Crimes of the Past; Moving Towards Best Practice (Sixth Report of Session 2013–​2014, 12 March 2014) [51]. 89 See Alex Just, ‘A Manifesto Against Blacklisting’ in Smith, Blacklisting (n 67) 39. There are also suggestions that EU public procurement rules can be used to exclude from public contracts businesses that have engaged in ‘grave professional misconduct’. See the EU Public Procurement Directive 2014/​24/​EU of 26 February 2014, art 57(4)(c). See UNITE Legal Services, ‘Blacklisting—​The Battle for Justice’ (August 2016) accessed 5 August 2019. 90 See, eg, Victor Tadros, ‘Criminalization and Regulation’ in RA Duff and others (eds), The Boundaries of the Criminal Law (OUP 2010) 163. 91 Ashworth, Positive Obligations (n 7); Tulkens (n 16); Liora Lazarus, ‘Positive Obligations and Criminal Justice: Duties to Protect or Coerce?’ in Lucia Zedner and Julian Roberts (eds), Principles and Values in Criminal Law and Criminal Justice: Essays in Honour of Andrew Ashworth (OUP 2012); Natasa Mavronicola, ‘Taking Life and Liberty Seriously: Reconsidering Criminal Liability Under Article 2 of the ECHR’ (2017) 80 MLR 1026.

Human Rights, Labour Rights, and Criminal Wrongs  225 [C]‌riminal proceedings should remain, in theory and in practice, a last resort or subsidiary remedy and that their use, even in the context of positive obligations, calls for a certain degree of ‘restraint’.92

She also suggested that the use of criminalization has been broad to the extent that ‘it is no longer necessary to justify the use of the criminal option but rather the absence of its use’.93 As mentioned above, the imposition of duties to criminalize in ECtHR case law stems from the interpretive principle employed by the Court according to which Convention rights have to be practical and effective. State coercion through criminalization is viewed as an effective way to deter the relevant conduct. Sanctions imposed through criminal law aim at influencing behaviour,94 and there is indeed evidence that the criminal law has a deterrent effect.95 In relation to deterrence, Ashworth suggested that when arguing for a new positive obligation to criminalize we should carefully consider the phraseology for the justification of the duty.96 He emphasized that even though criminalization has a deterrent effect, it is doubtful whether higher penalties lead to lower rates of offending.97 He therefore suggested that the ECtHR’s focus should be placed on censuring people for committing particular wrongs, and imposing punishment that is proportionate to the commission of the wrong (separate to or in addition to civil liability).98 With either approach to the justification of criminalization (one focusing on deterrence or one focusing on retribution), the criminalization of blacklisting is required both in order to reduce the likelihood of future violations and in order to punish those who are culpable for the specific wrongful act. The argument of this chapter is not of course that all violations of collective labour rights by private individuals should be criminalized. A good example of a case where the Court imposed a positive obligation to secure Article 11 rights, but which did not trigger a duty to criminalize, is Wilson and Palmer v UK.99 Wilson involved the practice of treating employees who sign an individual employment contract and cease to be represented by a trade union more favourably than those who refused to do so. UK law did not protect employees, and was found to breach Article 11.100 What distinguishes Wilson from the blacklisting cases is that the wrong of blacklisting negates not only union membership but the right to work altogether, with devastating implications for the right to private life. The cumulative convergence of Articles 11, 8, and 14 in the case of blacklisting serve as a particularly strong normative basis for criminalization. A second objection that needs to be considered is whether the Court is reliant exclusively on criminalization when faced with serious violations of labour rights, or whether it is sufficiently appreciative of other forms of regulation, such as civil remedies, which may be suitable and effective in addressing the wrong in question.101 The proposition 92 MC v Bulgaria App no 39272/​98 (4 December 2003). 93 Tulkens (n 16) 585. 94 AP Simester and Andrew von Hirsch, Crimes, Harms, and Wrongs (Hart 2014) 5. 95 See, eg, Andrew von Hirsch and others, Criminal Deterrence and Sentence Severity: An Analysis of Recent Research (Hart Publishing 1999) ch 2. 96 Ashworth, Positive Obligations (n 7) 209. 97 Ashworth did question whether higher penalties conduce to deterrence, ibid 210. 98 ibid. 99 Wilson (n 83). 100 Ewing examined the implications of Wilson and suggested how English law should change in order to provide effective protection to trade union members, but did not suggest that the conduct in question should be criminalized. See Keith Ewing, ‘The Implications of Wilson and Palmer’ (2003) 32 Industrial Law Journal 1. 101 See Tadros, ‘Criminalization and Regulation’ (n 90).

226  Criminality at Work that blacklisting should be criminalized does not imply that there should also not be civil sanctions for the victims. The Strasbourg Court itself recognizes the role of different methods of regulation when examining the effective protection of Convention rights. For instance, it explicitly ruled that criminalization of violations of collective labour rights is not sufficient in Danilenkov v Russia.102 The applicants, members of the Dockers’ Union of Russia (DUR), complained that their employer, a private company, subjected them to discriminatory treatment on the basis of their trade union membership, contrary to Article 11 together with Article 14. Following a period of strike action, which was unsuccessful, the applicants were reassigned to special work teams with limited opportunities, transferred to part-​time posts, and finally made redundant. Even though they won their case in domestic courts that ordered that they are paid compensation, their discrimination claims were not upheld, because the domestic legal framework made discrimination a criminal offence, which civil courts had no jurisdiction to examine. The prosecutor declined to open a criminal investigation, as an inquiry failed to establish intent to discriminate. The Court ruled that because discrimination could only be established in criminal proceedings and not through civil action, Russia did not provide effective legal protection to the applicants. Focusing particularly on the question why a criminal remedy is insufficient, the Court explained that criminal proceedings require proof beyond reasonable doubt of intent by the managers to discriminate. On this basis it was decided not to prosecute. For reasons such as this, criminal proceedings alone were viewed as inadequate by the ECtHR. Civil proceedings would have been more suitable for they [W]‌ould have made it possible to perform the far more delicate task of examining all elements of the relationship between the applicants and their employer, including the combined effects of the various techniques used by the latter to induce dockers to relinquish DUR membership, and to afford appropriate redress.

It may be argued at this point that in order to achieve the practical and effective protection of Convention rights, we could simply impose higher penalties under civil law, rather than invoking the ‘distinctively moral voice’103 and coercive power of criminal law. On this matter, though, it is important to appreciate the different functions of criminal law. As Ashworth rightly pointed out, in addition to deterrence, it is also crucial to identify distinctive categories of blameworthy wrongs, and censure those who commit these wrongs. Blacklisting has to be criminalized as such in order to communicate the gravity of the wrong and harm caused to workers. The imposition of penalties under civil law only for certain wrongful conduct may also be unacceptable under the ECHR. This is because the ECtHR might still classify the relevant sanctions as criminal. In Ozturk v Germany, it ruled: By removing certain forms of conduct from the category of criminal offences under domestic law, the law-​maker may be able to serve the interests of the individual . . . as well as the needs of the proper administration of justice, in particular in so far as

102 103

Danilenkov (n 83). Simester and von Hirsch (n 94) 4.

Human Rights, Labour Rights, and Criminal Wrongs  227 the judicial authorities are thereby relieved of the task of prosecuting and punishing contraventions—​ which are numerous but of minor importance—​ of road traffic rules. The Convention is not opposed to the moves towards ‘decriminalization’ which are taking place—​in extremely varied forms—​in the member States of the Council of Europe. The Government quite rightly insisted on this point. Nevertheless, if the Contracting States were able at their discretion, by classifying an offence as ‘regulatory’ instead of criminal, to exclude the operation of the fundamental clauses of Articles 6 and 7 . . . , the application of these provisions would be subordinated to their sovereign will. A latitude extending thus far might lead to results incompatible with the object and purpose of the Convention.

In other words the Court approaches the question of what constitutes a criminal offence as an ‘autonomous concept’,104 in the sense that is not bound by the authorities’ decision to classify a sanction as civil. This is because when state authorities have the power to impose serious sanctions that are typically found in criminal law, these must be viewed as criminal in nature for the ‘defensive role of human rights’105 under Articles 6 (right to a fair trial) and 7 (no punishment without law) to come into play. It should be added, before concluding, that the Court is not overly focused on sanctions when considering protection of rights. It also pays some attention to structural factors that create vulnerability to workers’ exploitation. In Rantsev, for instance, it explained that ‘the duty to penalize and prosecute trafficking is only one aspect of member States’ general undertaking to combat trafficking. The extent of the positive obligations arising under Article 4 must be considered within this broader context.’106 The Cypriot immigration scheme of artiste visa was ruled to violate Article 4 of the Convention because it did not protect individuals from human trafficking.107 To return to the example of the MSA, it is very questionable whether it complies with its ECHR obligations, not only because of the limited number of prosecutions as discussed earlier, but also because of the lack of political will to address structural factors, and particularly legal structures, that create vulnerability to exploitation. This point is exemplified by the UK Overseas Domestic Worker visa, which was a central political issue during the passing of the Act.108 To conclude this section, the objection that there is a tendency to overcriminalize is not pertinent when it comes to labour rights in the ECHR, where the Court has imposed limited obligations to criminalize. At the same time it cannot be said that the Court is not mindful of the role of civil remedies for victims of violations,109 or of the role of structural factors that create vulnerability to exploitation.110

104 See George Letsas, ‘The Truth in Autonomous Concepts: How to Interpret the ECHR’ (2004) 15 European Journal of International Law 279. 105 Tulkens (n 16) 579. 106 Rantsev (n 29) [285]. 107 ibid [293]. 108 See Virginia Mantouvalou, ‘The Modern Slavery Act Three Years On’ (2018) 81 Modern Law Review 1017; Virginia Mantouvalou, ‘ “Am I Free Now?” Overseas Domestic Workers in Slavery’ (2015) 42 Journal of Law and Society 329. 109 Danilenkov (n 83). 110 Rantsev (n 29).

228  Criminality at Work

E.  Conclusion Certain violations of labour rights constitute grave moral wrongs, which can cause serious harm to a person’s life, physical or psychological integrity. Taking advantage of workers’ vulnerability that is due to economic reasons, or other factors, such as someone’s immigration status, and exploiting them by violating their human and other labour rights can give rise to positive obligations under the Convention. In this context, the systemic exploitation of migrant workers has triggered a positive obligation to put in place effective criminal laws that must be backed up by effective procedural protections. The development of such obligations in this context should be viewed as a positive evolution to the extent that it expresses public disapproval for the specific treatment, punishes those who commit the crime, and deters others. The key questions under the ECHR on this issue are whether the relevant criminal law is effectively enforced, on the one hand, and whether the authorities also take steps to remove structural factors that create vulnerability to such exploitation, on the other. In this chapter, I argued that the violation of collective labour rights may also trigger an obligation to criminalize in certain circumstances. I used the example of blacklisting to illustrate the point that a breach of trade union rights can meet the elements of wrongfulness and harm that are central in the inquiry of criminalization. This positive obligation arises on the basis of Articles 11, 8, and 14 of the Convention. The cumulative convergence of these three rights—​the discriminatory treatment of trade union members and the violation of their right to work—​provides a strong normative argument for criminalization. These are not absolute and non-​derogable provisions. However, their nature as qualified and derogable rights does not preclude them from triggering obligations to criminalize when the conduct that breaches the Convention is a serious wrong that causes grave harm to the workers. The Court has to recognize positive obligations to enact criminal laws (together with other remedies) to address such violations of collective labour rights, both in order to communicate that treating workers in this manner is a grave moral wrong that our legal system disapproves, and in order to contribute effectively to the protection of workers from the wrongful conduct in question and its disastrous effects.

PART III

T HE C ON T E MP OR A RY SHA PE OF CRIM INA L IZ AT ION PR ACT IC E S : RISK, STAT U S, A N D CHA R AC T E R I N THE N E OL IBE R A L C R I M I NA L  L AW

12

The Preventive Role of the Criminal Law in Employment Relations Jennifer Collins and Andrew Ashworth

A.  Introduction What preventive role does the criminal law now play in employment relations, in the light of recent legislative changes? In this chapter we examine the effect of three sets of criminal offences on the protection of workers’ rights and the balance of power in labour relations. We also scrutinize the justifications for using the criminal law in these ways. Thus section B considers offences of omission, analysing a new generation of omissions offences that can be committed by companies. In section C we examine the re-​vamped offence of employing an illegal worker, an offence that exists alongside a regime of civil penalties. We note concerns about discrimination in the enforcement of this law, and suggestions of collusion between immigration officials and employers, leading to questions about whether this offence enhances or impedes the achievement of worker-​protective goals. Section D assesses the new preventive orders associated with modern slavery and human trafficking, which involve the imposition of prohibitions and conditions, reinforced by the criminal offence of breaching the order without reasonable excuse. These preventive orders are contrasted with Labour Market Undertakings and Orders, and the range of supporting interventions. In section E we draw some conclusions about the proper role of criminal law in employment relations.

B.  Offences of Omission The first of the three preventive roles played by the criminal law in employment relations is to be found in offences of omission. It quickly becomes apparent that the criminal law’s role here is controversial on its own terms. This is because, as textbooks on English criminal law tend to state in some form or other, criminal liability for failing to act in a given situation should be exceptional.1 One of the strongest expressions of this doctrinal restriction is that of Simester and Sullivan, who state that ‘it is a guiding principle of the [criminal] law that defendants are liable according to what they do, not what others do and they fail to prevent’.2 They go on to maintain that ‘standard legal doctrine stipulates that the behaviour requirement [in the actus reus of a crime] is a requirement of positive action by the defendant. Except occasionally, an omission will not do.’3

1

For example, David Ormerod and Karl Laird, Smith and Hogan’s Criminal Law (14th edn, OUP 2015) 72–​73. AP Simester and others, Simester and Sullivan’s Criminal Law: Theory and Doctrine (Hart 2016) 15. 3 ibid 72. 2

Jennifer Collins and Andrew Ashworth, The Preventive Role of the Criminal Law in Employment Relations In: Criminality at Work. Edited by Alan Bogg, Jennifer Collins, Mark Freedland QC, and Jonathan Herring, Oxford University Press (2020). © Alan Bogg, Jennifer Collins, Mark Freedland, and Jonathan Herring. DOI: 10.1093/oso/9780198836995.003.0012

232  Criminality at Work Even if we accept Simester and Sullivan’s restrictive view of the place of omissions ­ ffences, it must be recognized that major exceptions have been made in offences relevant o to employment relations. Probably the best known of these is the Health and Safety at Work Act 1974. The Act imposes, inter alia, two distinct duties on employers: section 2 imposes a duty on every employer to ensure, so far as is reasonably practicable, the health, safety and welfare at work of all his employees; section 3 imposes a similar duty towards ‘persons not in his employment who may be affected’ by the conduct of the employer’s undertaking. These duties are reinforced by the criminal offence (section 33) of failing to discharge the obligations imposed on employers by the Act. This offence of omission plays a major role in the labour market, since it underpins much of the work of the Health and Safety Executive4 and is the subject of sentencing guidelines for those cases that result in a conviction in court.5 Recent years have seen further duties imposed on employers through omissions offences. The leading example is the offence in section 7 of the Bribery Act 2010. A relevant commercial organization (C) is guilty of the offence of failure to prevent bribery if a person (A) associated with C bribes another person, intending to benefit C; but it is a defence for C to prove that C had in place adequate procedures designed to prevent persons associated with C from undertaking such conduct. In essence, therefore, this is an offence of omission (by failing to put adequate preventive procedures in place), not requiring proof of fault by C but allowing C a form of due diligence defence. The social function of this corporate failure-​ to-​prevent offence is to change corporate culture by giving companies an incentive to put preventive procedures in place, or at least by providing a deterrent against passivity on the part of companies.6 Other forms of corporate failure-​to-​prevent offences have also been enacted.7 Section 45 of the Criminal Finances Act 2017 holds an employer criminally liable if an ‘associated person’ commits a UK tax evasion facilitation offence when acting in the capacity of a person associated with the company. This offence does not require proof that the associated person was intending to benefit the corporation, and the term ‘associated person’ refers to the capacity in which A was acting and not the formal position held. This offence is also more explicitly preventive, since the conduct element is that the associated person facilitates an offence committed by someone else. On the same model as the section 45 offence but with a few small differences is the section 46 offence of failing to prevent the facilitation of a foreign tax evasion offence.8 The government has expressed an interest in creating similar offences for corporate failure to prevent fraud and to prevent money laundering, without stating precisely which model would be followed.9

4 The HSE’s enforcement policy and other relevant guidance may be found at accessed 6 August 2019. 5 Sentencing Council, ‘Health and Safety Offences, Corporate Manslaughter and Food Safety and Hygiene Offences: Definitive Guideline’ (2016) accessed 6 August 2019. 6 See HM Government, ‘Insight into Awareness and Impact of the Bribery Act 2010’ (2015) accessed 6 August 2019. 7 The offence of employing an illegal worker (discussed in section C below) also applies to corporations, see Immigration, Asylum and Nationality Act 2006, s 22. 8 For details, seeKarl Laird, ‘The Criminal Finances Act 2017—​An Introduction’ [2017] Criminal Law Review 915. 9 Ministry of Justice, ‘Corporate Liability for Economic Crime: Call for Evidence’ (January 2017) accessed 6 August 2019.

The Preventive Role of the Criminal Law  233 The previous two paragraphs establish that a new strain of omissions offences has appeared in English criminal law. Not only is this controversial in terms of criminal law’s own doctrine, but it raises questions about the justifications for introducing such offences. Two questions should be distinguished. The first is whether it is fair and just to impose such duties on employers, corporations or companies. Should it not be the task of the police and the regulatory agencies to prevent and detect bribery, fraud, money laundering, and the facilitation of tax evasion? Indeed it should, but the further issue is whether there is good reason for coopting employers into this enforcement process. Lindsay Farmer has recently argued that the rise of preventive offences can be linked to an interest in protecting the vulnerable (such as children, victims of abuse and trafficking).10 On his view, it is ‘the relationship between the scale of the threat and the vulnerability of a particular class of victims that justifies the adoption’ of preventive measures.11 But the cluster of omission offences examined here, imposed on corporations in relation to securing the stability of financial interests, cannot be said to be underpinned by this rationale. More persuasive is a version of what leading criminologists such as Pat O’Malley and David Garland have termed ‘responsibilization’. They used the term to characterize the movement to delegate the responsibility for policing from the state to local organizations.12 We use it here in an extended sense, to cover the imposition of responsibility for aspects of law enforcement on corporations and employers, particular where they are in sufficient proximity to the wrongdoing or wrongdoer to be able to prevent or report an offence. This raises a normative question: to what extent is it justifiable to place such a preventive duty upon employers and corporations? One answer to this is to point out that the profit-​based structure of companies is the essential context for economic crimes of the kind discussed above. As Celia Wells puts it, ‘a corporation benefits from the wrongdoing of associated persons acting in pursuit of contractual or commercial advantage or tax limitation’.13 That benefit supplies a reason for expecting the corporation to play some role in preventing offending within or on behalf of the company. While this role cannot be seen as protecting the rights of workers, limits ought to be placed on an employer, business or corporation achieving competitive advantage over another through turning a blind eye to unlawful practices. However, it is one thing to find good arguments in favour of laying a duty on the corporation. It is a significantly further step to find good arguments in favour of imposing criminal liability on the company for failing to prevent these offences. One key issue is how serious these new omissions offences are: there is no doubt that bribery and facilitating tax evasion are serious offences—​although their impact is financial, and does not involve sex or violence—​but we must recall that the company is not the progenitor of these offences, but rather stands some distance away from the actual commission. The decision whether the criminal law or regulation is the more appropriate response calls for some difficult calculations. It can be argued that criminal law should be the last resort rather than the ‘go-​ to’ response,14 and that it should require the full criminal procedure and fair trial rights.

10 Lindsay Farmer, Making the Modern Criminal Law: Criminalization and Civil Order (OUP 2016) 107. 11 ibid 107. 12 Pat O’Malley, ‘Volatile and Contradictory Punishment’ (1999) 3 Theoretical Criminology 175; David Garland, The Culture of Control (University of Chicago Press 2001) 126. 13 Celia Wells, ‘Corporate Failure to Prevent Economic Crime—​A Proposal’ [2017] Criminal Law Review 423, 430. 14 Douglas Husak, ‘Criminalization as Last Resort’ (2004) 24 Oxford Journal of Legal Studies 207.

234  Criminality at Work Regulatory enforcement is another form of response that should be considered, not least because the fines imposed by regulators may be larger (and therefore perhaps more effective) than those imposed by the criminal courts.15 The government seemed to concede that in some fields, such as financial regulation, ‘strengthening deterrents to misconduct through regulatory reform’ would be appropriate and effective.16 However, the regulatory approach should not be selected in order to avoid granting fair trial rights to corporations accused of transgressing. The crucial issue is whether, ‘intrinsically’, these failure-​to-​prevent offences are at a sufficient level of seriousness to justify criminalization and therefore fair trial rights. Identifying the factors that determine the ‘sufficient level of seriousness’ is a difficult and contested task. Health and safety duties are clearly a settled core, but in the current crisis surrounding enforcement of labour standards it is uncertain how this will be developed.17

C.  The Offence of Employing an Illegal Worker The second of the three preventive roles played by the criminal law in employment relations is to be found in the re-​vamped offence of employing an illegal worker. Part of the background to this offence is the re-​vamped offence of illegal working that is committed by the illegal worker: section 34 of the Immigration Act 2016 inserts into the Immigration Act 1971 a new offence of ‘illegal working’, committed by a person who works when disqualified by reason of immigration status, and who knows or has reasonable cause to believe that he or she is so disqualified. The government’s aim of making ‘it harder for people to live and work illegally in the UK’ is demonstrated in the legal position that for those convicted, their wages are in principle recoverable under the Proceeds of Crime Act 2002, and any outstanding wage bill need not be paid.18 The doctrine of illegality applies as a defence to contract claims on that basis that an individual ought not to benefit from their own wrongdoing in order to preserve the integrity of the legal system.19 Exceptions to the application of this doctrine are drawn around considerations of public policy and proportionality, to be determined on a case-​by-​case basis using a range of factors.20 The background to harsher sanctions upon employers is one of increased vulnerability for those bearing irregular work status.21 Our chief interest is in the provisions of section 35 of the Immigration Act 2016, which amends the offence of knowingly employing an illegal worker contained in section 21 of 15 Wells (n 13) 432–​35. 16 For details, see Laird (n 8) 15–​16. 17 See David Metcalf, ‘United Kingdom Labour Market Enforcement Strategy 2018/​19’ (HM Government May 2018). 18 Announced by Prime Minister David Cameron on 21 May 2015 in his immigration speech. Council Directive 2009/​52/​EC (18 June 2009) (Employer’s Sanctions Directive) has not been enforced in England and Wales, providing for minimum standards on sanctions and measures against employers of third-​country nationals holding irregular migration status. 19 Patel v Mirza [2016] UKSC 42, [2017] AC 467. For discussion, see Sarah Green and Alan Bogg (eds), Illegality After Patel v Mirza (Hart 2018). 20 Patel (n 19) [107]. 21 This sits awkwardly alongside the NRM referral system for victims of trafficking, established in 2009, see ATMG and Human Trafficking Foundation, ‘Submission to the UN Special Rapporteur on Contemporary Forms of Slavery (March 2017)  accessed 6 August 2019. cf the statutory defence in s 45 of the Modern Slavery Act 2015.

The Preventive Role of the Criminal Law  235 the Immigration, Asylum and Nationality Act 2006. Criminal sanctions for employing an illegal worker were first introduced in section 8 of the Asylum and Immigration Act 1996: failure to carry out a right to work check for a prospective worker was a criminal ­offence that could attract a fine of up to £5,000 per worker.22 Between 1998 and 2004 there were twenty-​five convictions.23 There are two principal changes in the amended section 21 offence, extending its scope and seriousness. The first is to broaden the basis for conviction, from ‘knowingly’ to ‘has reasonable cause to believe that the employee is disqualified from employment by reason of the employee’s immigration status’ (the new section 21(1A)(b) of the 2006 Act). The effect of this change is to broaden the offence by replacing a ­subjective test with an objective test, and the government’s expressed intention was to criminalize employers who are wilfully blind to the employee’s immigration status, usually by failing to carry out the appropriate checks.24 However, the Minister also stated that in cases where the employer is simply negligent, the civil penalty scheme (in force since 29 February 2008) will continue to be used, now with a maximum penalty of £15,000 or £20,000 per worker depending on whether it is a first offence.25 Some 2,594 civil penalties were issued to employers in 2015–​16.26 The criminal offence is aimed at employers who ‘actually know that someone has no right to work’ or who deliberately refrain from checking, and the formulation of ‘reasonable cause to believe’ was intended to make the offence easier to prove. The second principal amendment is the more than doubling of the maximum penalty, from two to five years’ imprisonment. This steep increase, combined with the broadening of the basis for criminal liability from ‘knowingly’ to ‘reasonable cause to believe’, renders the law much tougher than it was before 2016. To what extent these powers will be used remains to be seen: the criminal offence in the 2006 Act resulted in sixty-​nine prosecutions in 2008 but only nine in 2013 and, as stated in the previous paragraph, 2,594 civil penalties were imposed in 2015–​16.27 There is no legal obligation to perform right to work checks on employees, but there is strong financial incentive to do so. A right to work check operates as a due diligence defence to the imposition of a civil penalty, so long as an employer does not have knowledge of a worker’s illegal work status.28 Employers are responsible for staff who carry out right to work checks on their behalf.29 Performance of a check also means that an employer can demonstrate that they had reasonable cause to believe that a worker had legal work status, thus ensuring that the mens rea requirement for the section 21 criminal offence is not met.

22 See Bernard Ryan, ‘Employer Enforcement of Immigration Law after Section 8 of the Asylum and Immigration Act 1996’ (1997) 26 Industrial Law Journal 136. 23 Home Office, Control of Immigration: Statistics—​United Kingdom 2000 (Cm 5315, November 2001) Table 7.5; Home Office, Control of Immigration: Statistics—​United Kingdom 2004 (Cc 6690, November 2005) Table 6.5, noted by Bernard Ryan, ‘Employer Checks of Immigration Status and Employment Law’ in Cathryn Costello and Mark Freedland (eds), Migrants at Work: Immigration and Vulnerability in Labour Law (OUP 2014) 240. 24 These details are to be found in Current Law Statutes, Immigration Act 2016,  98–​99. 25 Immigration, Asylum and Nationality Act 2006, s 15. The civil penalty was increased from £10,000 to £20,000 in May 2014. 26 Home Office, ‘Immigration Act 2016:  Factsheet:  Illegal Working’ https://​assets.publishing.service.gov.uk/​ government/​uploads/​system/​uploads/​attachment_​data/​f ile/​537205/​Immigration_​Act_​-​_​Part_​1_​-​_​Illegal_​ Working.pdf accessed 6 August 2019. 27 ibid, setting out the government’s Explanatory Notes to the Bill, 2–​3. 28 Immigration, Asylum and Nationality Act 2006, s 15(3). 29 See Home Office, ‘An Employer’s Guide to Right to Work Checks’ (August 2017) accessed 6 August 2019.

236  Criminality at Work The assumption throughout the Home Office’s official guidance papers and the various government statements is that it is fair and appropriate to impose these duties on employers, rather than leaving the various government immigration services to identify and track down illegal workers. The Home Office justifies placing a preventive obligation upon employers on the basis that: Illegal working often results in abuse and exploitative behaviour, the mistreatment of illegal migrant workers, tax evasion and illegal housing conditions. It can also undercut legitimate businesses and have an adverse impact on the employment of people who are lawfully in the UK.30

The same approach is applied by the Immigration Act 2016 to landlords, who are required to check the immigration status of their tenants and prospective tenants. In respect of both employers and landlords it can be argued that (a) they are in business in order to make a profit, and (b) they are well positioned to check on people wanting either work or lodging, so that (c) so long as the process of checking is straightforward and not especially burdensome, it is fair to expect employers and landlords to act as gatekeepers and law-​enforcers in this way. At the same time, there is vigorous denial that these offences amount to delegating to employers or landlords an immigration enforcement function. This is evident in debate of the Bill in the House of Commons, where it was emphasized that it is not an employer’s job to scrutinize the relevant documents in detail, but simply to have taken reasonable steps to have checked that they exist and appear valid (in the employee’s presence), before making copies of them.31 This argument decoupling the functions of an employer, corporation or landlord from immigration enforcement in performance of this duty can be challenged. There is clear financial incentive to go beyond performance of a basic right to work check. Civil penalties can be reduced if there is cooperation with Home Office officials. Deductions of £5,000 are available for each of the following: (i) for reporting suspicions about an employee’s right to work; (ii) for active cooperation with the Home Office when an employer is under investigation; and (iii) for having effective document-​checking practices in place (only if an employer has not employed illegal workers in the last three years).32 Voluntarily opening up business premises and providing access to employment records are a form of ‘active cooperation’, removing the need for immigration enforcement officers to apply for warrants.33 The financial incentive is considerable. For an employer who has not been found to have employed an illegal worker in the last three years, the maximum penalty 30 Home Office, ‘Code of Practice on Preventing Illegal Working: Civil Penalty Scheme for Employers’ (May 2014) 3 accessed 6 August 2019. Almost identical justifications were given for employer sanctions by President Obama when in office, see Katherine Brandon, ‘Working Together for Immigration Reform’ (25 June 2009) accessed 6 August 2019. 31 Immigration Bill Deb 27 October 2015. 32 Home Office, ‘Code of Practice on Preventing Illegal Working’ (n 30) 11–​12. 33 Corporate Watch, ‘Snitches, Stings, and Leaks: How “Immigration Enforcement” Works’ (30 August 2016) accessed 6 August 2019, notes that 50 per cent of workplace raids are not sanctioned by a court-​ordered warrant; David Bolt, ‘An Inspection of How the Home Office Tackles Illegal Working: October 2014–​March 2015’ (HM Government, December 2015) [4.13].

The Preventive Role of the Criminal Law  237 amount is £15,000. However, this could be reduced to £3,500 in total if (i) and (ii) of the mitigating factors stated above are found and if an employer pays the penalty using the Home Office’s ‘fast payment’ option. It could be reduced to a warning notice of all three mitigating factors are present. Employers will be concerned not to have a civil penalty on record, since this could affect future immigration applications, an employer’s ability to sponsor migrants to enter the UK or to hold a gangmaster’s licence.34 This system of financial incentive turns what might appear a modest legal preventive duty into something more onerous and more closely connected to immigration enforcement. The merits of this may be contested from the perspective of both employers and workers. We argue here that the balance falls disproportionately against vulnerable workers. Focusing first on employers, there is some evidence that employers are not fully aware that immigration authorities cannot enter premises without clear legal permission. An immigration enforcement team requires a warrant to search premises where illegal working is alleged unless ‘informed consent’ is given by an employer or an ‘AD letter’ is authorized by a Home Office Assistant Director.35 Between 2009 and 2014, 36,381 ‘illegal working’ visits were undertaken by the Home Office.36 The Independent Chief Inspector of Borders and Immigration reported in 2015 on the Home Office’s response to illegal working. His report noted that in a sample of 184 cases, Immigration Officers claimed in 102 visits that they had obtained ‘informed consent’ to enter, despite ‘managers at all levels [believing] that warrants were used in the majority of cases’.37 According to the report, in nearly all of the premises visited, ‘English was not always the first language of those encountered, and files rarely documented how officers confirmed that consent was “fully informed” as required’.38 In cases where ‘informed consent’ is provided, it is questionable whether this is free and voluntary. A concern to reduce the quantum of civil penalties, noted above, may mean that employers feel pressure to agree, though it is difficult to obtain empirical evidence on this matter.39 A seemingly modest preventive duty of this kind carries substantial implications for workers. First, the preventive duty must also be reconciled against employers’ other obligations towards workers, particularly in relation to discrimination law. The key provisions are to be found in the Equality Act 2010. As noted above, in order to have a statutory excuse, employers are encouraged to perform right to work checks. Employers must ensure that they do not discriminate directly or indirectly in employment practices in relation to ‘protected characteristics’ under the 2010 Act. The Home Office has produced a Code of Practice for Employers which focuses on the need for fair recruitment practices. For example, if an employer asks for right to work documents from one applicant they ought to ensure that the same is asked of all applicants at the same stage in the recruitment process.40 34 Home Office, ‘An Employer’s Guide to Right to Work Checks’ (n 29) 9. 35 On the ‘AD letter’, see Immigration Act 1971, s 28CA. 36 Corporate Watch (n 33) 5. 37 Bolt (n 33) [5.21]. 38 Bolt (n 33) [5.22]. 39 For arguments focusing on employers’ resistance to sanctions in relation to illegal working, see Yara Evans and others, ‘ “Papers Please”: The Impact of the Civil Penalty Regime on the Employment Rights of Migrants in the UK’ MRN Migration Perspectives Paper (Migrants Rights Network 2008)  accessed 6 August 2019; Sonia McKay and Tessa Wright, ‘Tightening Immigration Policies and Labour Market Impacts’ (2008) 14(4) Transfer: European Review of Labour and Research 653. 40 Home Office, ‘Code of Practice For Employers:  Avoiding Unlawful Discrimination While Preventing Illegal Working’ (May 2014) 7 https://​assets.publishing.service.gov.uk/​government/​uploads/​system/​uploads/​

238  Criminality at Work If a candidate is unable to supply acceptable documents, an employer should aim to keep open a job for as long as possible in order for them to do so, if this is possible according to business needs.41 So far, this appears reasonable. But the issue of discrimination is highly pertinent in relation to immigration raids. Recent research reveals that Home Office enforcement raids have focused on . . .  small, minority ethnic businesses, such as restaurants, takeaways and small retail shops.42 . . . Between July and the end of September 2013, 90 workplaces were raided of which 71 are identifiable by name as Indian, Bengali or Chinese restaurants or takeaways.43

The legal position for irregular migrants uncovered on raids is weak. The Supreme Court in Taiwo v Olaigbe44 held that immigration status falls outside of the protected characteristic of nationality. A worker identified by immigration authorities via a workplace raid with suspected immigration status will not be able to make a claim under the Equality Act 2010. This is of concern, coming, as it does, alongside increased powers for immigration officers to suspend business practices for up to forty-​eight hours where a civil penalty has previously been issued or where an employer has been prosecuted for employing illegal workers.45 Upon application to a court, an immigration officer may apply for a compliance order which can be used to extend closure of premises or otherwise require positive steps from the employer to ensure that they do not employ illegal workers.46 This not only impacts employers’ ability to make a living, but also has a disproportionate effect on vulnerable workers who may be living from one paycheck to another. The imposition of this preventive duty has been accompanied by reports of increasing collaboration/​collusion between employers and immigration enforcement.47 For example, employers may provide initial intelligence which will be scrutinized by intelligence officers. The Home Office’s guidance to staff on ‘Illegal Working Operations’ identifies employers as providing the majority of ‘reports about suspected illegal working’.48 Employers may also be involved in gathering together workers for immigration officers. The sting operation set up by Byron Hamburgers in July 2016 in association with immigration officials attracted attention of the wider media via a Twitter account. A routine staff meeting was called in branches. In fact it was used to gather together workers for immigration authorities, which led to thirty-​five arrests in relation to illegal working. Corporate Watch has identified evidence of similar operations at Deliveroo in June 2016 and cleaning companies Amey and attachment_​data/​file/​311665/​Code_​of_​practice_​on_​avoiding_​unlawful_​discrimination_​while_​preventing_​illegal_​working.pdf accessed 6 August 2019. 41 ibid 7. 42 Alice Bloch, Sonia McKay, and Leena Kumarappan, ‘Employer Sanctions: The Impact of Workplace Raids and Fines on Undocumented Migrants and Ethnic Enclave Employers’ (2015) 35(1) Critical Social Policy 132. 43 ibid 134. 44 [2016] UKSC 31, [2016] 1 WLR 2653. 45 Immigration Act 2016, sch 6. 46 Note there are similarities to closure notices under Anti-​Social Behaviour, Crime and Policing Act 2014, pt 4. 47 Corporate Watch (n 33); John Vine, ‘An Inspection of the Intelligence Management System: February–​April 2014 (HM Government 2014). 48 Home Office, ‘Illegal Working Operations’ (HM Government 12 July 2016) accessed 6 August 2019.

The Preventive Role of the Criminal Law  239 ISS in May 2007 and 2009 respectively.49 It is estimated that the Home Office performs 6,000 raids per year.50 The driver for employer cooperation beyond discharge of legal obligations was noted above. Employer involvement in this way creates a barrier to workers complaining and unionizing about unacceptable working standards (which may include exploitation) because they are afraid to speak out.51 Irregular migrant workers who do not have legal work status are highly vulnerable in immigration raids or on ‘rescue’ raids in relation to modern slavery. A report by two prominent anti-​trafficking organizations notes that on the latter there is a significant risk that ‘any undocumented worker who escapes modern slavery will be detained as an immigration offender rather than identified as a potential victim’, thus producing a barrier to reporting exploitation.52 As a means of eliminating exploitation of workers, a preventive measure which in practice draws stronger links between employers and immigration enforcement seems ill-​suited. In the US context, it has been argued that employer sanctions for illegal working makes workers more vulnerable by having the effect of punishing workers.53 The severity of the legal landscape in England and Wales, where wages are treated as proceeds of crime, is intensified through a preventive offence backed up by an incentivized system of civil penalties. Rather than ensuring that employers who exploit illegal workers are deterred and punished, employers’ cooperation with immigration authorities—​which extends beyond a duty to ensure correct paperwork—​is rewarded through the reduction of civil penalties. This prompts a wider conceptual question as to how far the criminal law furthers or impedes the achievement of labour law’s worker-​protective goals? We begin by observing the uneven balance of power in personal work relations between employer and worker. This uneven balance of power is redressed through labour law’s worker-​protective goals. On occasion the criminal law supplements this function through legislation such as the Modern Slavery Act 2015. The preventive duty examined prioritizes the enforcement of immigration rules over the protective capacities of the criminal law. Strong financial incentives have the effect of heightening the preventive obligation. This system of a preventive criminal law measure, accompanied by a civil penalty scheme, contributes to and perpetuates the government’s ambition to create a ‘hostile environment’ for workers holding illegal work status.54 A 2010 review of the coalition government’s strategy to reduce net migration through the civil penalty scheme revealed that ‘the system [of civil penalties] did not create the hostile environment for those who benefit from illegal working that the Government 49 Corporate Watch (n 33) s 3. 50 ibid s 1. 51 ATMG and Human Trafficking Foundation (n 21) 17: ‘This immigration enforcement approach will also make communities whose members have insecure immigration status and who are vulnerable to exploitation and trafficking less likely to trust, work with authorities or disclose slavery for fear of drawing attention from law enforcement.’ 52 ATMG and Human Trafficking Foundation (n 21) 17. GLAA, The Nature and Scale of Labour Exploitation Across All Sectors Within the United Kingdom (May 2018) 17 accessed 6 August 2019. Section 45 of the Modern Slavery Act 2015 provides only a limited statutory defence for victims of slavery or trafficking. 53 David Bacon and Bill Ong Hing, ‘The Rise and Fall of Employer Sanctions’ (2012) University of San Francisco Law Research Paper No 2011-​12, 10. See also, Philip Martin and Mark Miller, ‘Employer Sanctions:  French, German and US Experiences’ (2000) ILO International Migration Paper 36, 7. 54 On the ‘hostile environment’, see FLEX, ‘Hostile Environment Undermines UK Government’s Modern Slavery Agenda’ (1 May 2018) accessed 28 August 2018; Home Office Delivery of Brexit:  Immigration Contents (9 February 2018)  accessed 6 August 2019.

240  Criminality at Work has intended’.55 The Home Office’s ‘Summary Guide for Employers on Preventing Illegal Working in the UK’ informs employers that:  ‘We carry out regular, intelligence-​led operations . . . to target illegal working, with the aim of removing the most harmful people from the UK first.’ The conflation of illegal work status with an individual being ‘harmful’ is troubling. To what extent ought a wider range of individuals—​employers, corporations, universities, landlords—​be opted in to a political agenda which is unapologetic about its commitment to a ‘hostile environment’?56 It can be doubted whether it is fair and appropriate to impose a preventive duty with strong links to hostile immigration enforcement upon employers. We have seen that a regulatory system, backed by substantial financial penalties, is in force. To take the further step and impose criminal liability, not least because it would be to criminalize an omission (failing to make the required checks), calls for further justifications—​especially if a maximum sentence as high as five years’ imprisonment is involved. So far, enforcement of the civil penalty scheme has been limited.57 Between 2008 and March 2018, 20,315 civil penalties were issued.58 But there is a strong argument from the perspective of liberal criminal law theory that any coercive preventive measures ought to include only the most serious wrongs which cannot be dealt with adequately by civil law measures through administrative penalties.

D.  The New Preventive Orders The third of the three preventive roles played by the criminal law in employment relations relates to a new group of preventive orders. Chief among these are the Slavery and Trafficking Prevention Order (STPO) and the Slavery and Trafficking Risk Order (STRO), under Part 2 of the Modern Slavery Act 2015. Another two-​step model, with some similarities to and some differences from the STPO and STRO, is that introduced by sections 14–​30 of the Immigration Act 2016, consisting of Labour Market Enforcement (LME) Undertakings and LME Orders. It is useful to contrast and compare these two sets of preventive measures in appraising the preventive role of the criminal law in the labour market. A key feature of both models is that they are tools given to state enforcement bodies and inspectorates to enforce 55 John Vine, ‘UK Border Agency’s Operations in the North West of England—​ An Inspection of the CPCT—​Illegal Working’ (HM Government March–​April 2010) 10 accessed 6 August 2019. 56 For example, the recent Windrush scandal has revealed that internal targets were placed in the Home Office as to how many individuals should voluntarily depart from the UK, rather than through forcible removal. Individuals affected did not have the correct paperwork to support their status as second-​generation migrants to the UK, and were threatened with deportation, denied benefits and pensions, removed from their jobs, and denied access to the NHS. See Rob Merrick, ‘Theresa May Vows Her “Hostile Environment” on Illegal Immigration Will Continue, Despite the Windrush Scandal’ The Independent (25 April 2018) accessed 6 August 2019. On Universities and students undertaking paid work, see Peter Scott, ‘How Universities were Swept into the Hostile Environment’ The Guardian (1 May 2018) accessed 6 August 2019. 57 Melanie Gower, ‘Employers’ Duties to Prevent Illegal Working’ (Home Affairs Section, SN06706, February 2015) 9ff. 58 Home Office, ‘Undocumented Workers:  Fines:  Written Question—​137049’ (HM Government 26 April 2018) accessed 6 August 2019.

The Preventive Role of the Criminal Law  241 labour standards. Bodies such as the Gangmasters and Labour Abuse Authority (GLAA) and HMRC (NMW/​NLW) have experienced increased government investment in recent years.59 Here we see coercive preventive measures, enforced by specialist state bodies, prioritized over private enforcement by individuals through Employment Tribunals (ETs). State enforcement is a departure from what Michael Ford describes as the model of ‘privatized social justice’ marked by the Redundancy Payments Act 1965. These preventive measures are an exception to the individual enforcement paradigm.60 A strong basis for justifying this new system of state-​administered preventive measures is its commitment to the fundamental right of access to justice. The context is a legal environment which has recently been bold in its rejection of barriers to the common law right of access to the courts for individuals. Tribunal fees were introduced for ET claims on 29 July 2013.61 This led to a substantial fall in ET claims—​it was estimated that 14,000 claims were not issued per year because of fees.62 The Fees Order was reversed by the Supreme Court in the case of UNISON on the basis of public law principles that (i) fees must not effectively prevent access to justice and the systemic effect of the Fees Order was as such, and (ii) that the Fees Order was disproportionate in relation to its objectives.63 Prior to the UNISON judgment, the government was keen to emphasize the availability of alternative means of enforcement. Ford notes that Perhaps embarrassed at how fees had undermined the individual enforcement paradigm yet politically and ideologically committed to them, instead of abolishing fees the Government created more exceptions to the paradigm.64

The LME Undertakings and Orders, for example, were a valuable political tool for the government in seeking to dispel concerns about the effects of tribunal fees on individual enforcement of labour rights. Political and rule of law drivers are evident. But from a criminal law perspective, are these principled measures? So far limited attention has been given to this question. We turn first to the STPO. This order can be made either on conviction for a slavery or trafficking offence, or on application to a magistrates’ court. The police, the National Crime Agency, an immigration officer or a labour abuse prevention officer (LAPO), can apply to the court for an STPO. While the latter is in the nature of a civil order, it may fairly be regarded as an aspect of the preventive role of the criminal law because breach of its conditions is a criminal offence with a maximum sentence of five years’ imprisonment. The

59 With recent recommendations for increased enforcement resources for the Employment Agency Standards (EAS), see Metcalf (n 17) 54. 60 Michael Ford, ‘Employment Tribunal Fees and the Rule of Law: R (Unison) v Lord Chancellor in the Supreme Court’ (2018) 47 ILJ 1, 4. 61 Fees Order 2013, SI 2013/​1893, made under the power in Tribunals, Courts and Enforcement Act 2007, s 42(1). Justifications for the imposition of fees were set out in BIS, ‘Resolving Workplace Disputes: A Consultation’ (HM Government 2011). 62 Ford (n 60) fn 4: ‘Based on comparing the mean cases received in the quarter to June 2013 with the mean number received in the quarters between October 2013 and June 2017’. For data, see Doug Pyper, Feargal McGuiness, and Jennifer Brown, ‘Employment Tribunal Fees’ (House of Commons Briefing Paper No 7081, 18 December 2017). 63 R (UNISON) v Lord Chancellor (Equality and Human Rights Commission Intervening) [2017] UKSC 51, [2017] 3 WLR 409. 64 Ford (n 60) 22.

242  Criminality at Work STPO on conviction can be imposed if the court finds that there is a risk of the defendant committing a further slavery or trafficking offence. The court may tailor the prohibitions in the order to the specific risk posed by the defendant: common requirements would be to notify change of address to the police or a ban on travelling outside the UK. The STPO as a freestanding order can be applied for at a magistrates’ court by a relevant officer. The order can only be made in respect of a person who has already been convicted or cautioned for a slavery or trafficking offence, and where the court is satisfied that a STPO is ‘necessary to protect against the risk of harm from the defendant committing’ such offences. Part 2 of the Modern Slavery Act 2015 also introduces a Slavery and Trafficking Risk Order (STRO), which is a civil order that may only be made on application on a person who has not been convicted of a slavery or trafficking offence, where a magistrates’ court is satisfied that the STRO is ‘necessary to protect against the risk of harm from the defendant committing’ such offences. Once again, it is for the court to specify conditions and prohibitions, based on an assessment of risk by the applicant that the defendant (who has not yet been convicted of a slavery or trafficking offence) is at risk of committing such an offence. The standard of proof is ‘akin to’ the criminal standard of ‘beyond reasonable doubt’.65 The STRO must be for a minimum of two years, and breach of its conditions is a criminal offence with a maximum sentence of five years’ imprisonment. Like the two STPOs, the STRO is said by the government to be ‘a preventive measure to deter unlawful and harmful activity’. The STPO and the STRO are put forward as essentially preventive, and will often be aimed at employers of the victims of modern slavery or human trafficking. The preventive rationale is pursued in a two-​step arrangement. In the first place the court sets out conditions (usually, prohibitions) that are considered necessary to provide protection against the defendant committing a slavery or trafficking offence. In the second place, any breach of those conditions amounts to a criminal offence (unless there is a ‘reasonable excuse’) that can be punished by up to six months’ imprisonment in a magistrates’ court or up to five years’ imprisonment at the Crown Court. This means that the criminal offence will be committed by any deviation from the conditions or prohibitions, even if that deviation creates no actual risk of slavery or trafficking. The terms of the order amount to a kind of personal criminal code for the defendant. For example, someone subject to a STPO may take a flight to Spain for purposes unconnected with slavery or trafficking: that would be an offence of breaching the order, even though the risk against which the STPO is intended to guard is not proved to arise.66 In effect, the STPO imposes a penalty via a prohibition which is a significant incursion of an individual’s liberty.67 So far there has been limited use of both orders despite the fact that it remains difficult to secure a conviction for a slavery or trafficking offence given the hidden nature of 65 See Home Office, ‘Guidance on Slavery and Trafficking Prevention Orders and Slavery and Trafficking Risk Orders under Part  2 of the Modern Slavery Act 2015’ (April 2017) accessed 6 August 2019. A requirement of ‘beyond reasonable doubt’ follows R v Manchester Crown Court ex parte McCann [2002] UKHL 29, [2002] 1 WLR 2033, see Rory Kelly, ‘The Right to a Fair Trial and the Problem of Pre-​inchoate Offences’ (2017) European Human Rights Law Review 596. 66 An example of a STRO is given in HM Government, ‘2017 UK Annual Report on Modern Slavery’ (October 2017) 30. It was used because Devon and Cornwall Police were unable to prosecute a modern slavery offence because victims did not wish to give evidence in court. No information is given about the length of the prohibitions which were imposed. 67 See AP Simester and Andrew von Hirsch, Crimes, Harms, and Wrongs: On the Principles of Criminalization (Hart 2011) 217.

The Preventive Role of the Criminal Law  243 exploitation and its complexity.68 Between July 2015 and March 2017, fifty-​six STPO and STROs were issued in England and Wales.69 Thirty-​seven of these were Prevention Orders and nineteen were Risk Orders.70 A 2017 report noted ‘limited use’ of Slavery and Trafficking Prevention and Risk Orders by police forces, citing in one force ‘clear misunderstanding’ as to the breadth of the Risk Order ‘and a belief that orders could be obtained only following a conviction’.71 Preventive measures in relation to slavery and trafficking can be compared with the system of LME Undertakings and Orders introduced by the Immigration Act 2016. Two points require emphasis. First, where a trafficking or slavery offence is involved there is potential for overlap between the system of preventive slavery and trafficking measures outlined above, and Labour Market Undertakings and Orders. This begs a question about whether both sets of measures are necessary. Second, unlike the Slavery and Trafficking preventive measures which supplement the core exploitation offences found in the Modern Slavery Act 2015, the Labour Market Undertakings and Orders are supported by a wider range of interventions, which include civil law measures. They are supposed to complement the pre-​existing ‘deterrent’ enforcement regime in relation to a wide spectrum of labour market non-​compliance extending beyond serious exploitation. These include: the risk of inspection, financial (civil law) penalties, reputational penalties, prosecutions under the National Minimum Wage (NMW), Employment Agencies and Gangmaster Licensing legislation, Orders and Undertakings, and prohibitions.72 These new Undertakings and Orders are a new enforcement tool to support the enforcement functions of the National Minimum/​Living Wage Enforcement Teams in HMRC (NMW/​NLW), the GLAA, and Employment Agency Standards Inspectorate (EASI), who have been given increased powers and a wider remit since November 2016.73 This has been accompanied by some funding increases despite the backdrop of government austerity measures. The GLAA was allocated an annual budget of £7.775m in 2017–​18—​an increase of £3.474m from 2015–​ 16.74 The HMRC (NMW/​NLW) budget was supplemented by £5.3m 2017–​18 from the year previous, bringing its budget to £25.3m.75 LME Undertakings and LME Orders ‘are intended to tackle serious and persistent non-​ compliance where existing penalties are failing to stop recidivism’.76 For example, employers who seem to regard a fine as an operational cost of business will be required to take action to prevent continued breaches.77 The usual approach will be to start to deal with 68 In the year to March 2017, 2,255 modern slavery offences were recorded: Jack Dent, ‘Modern Slavery Act 2015: Recent Developments’ (House of Commons Briefing Paper No 07656, 25 October 2017) 21; see GLAA (n 52) 29: while the number of referrals of cases of modern slavery and human trafficking rose in 2016–​17, convictions decreased by 6 per cent in this period. 69 HM Government, 2017 UK Annual Report on Modern Slavery (October 2017) 5 accessed 6 August 2019. 70 ibid 31. On 13 October 2016, the government found that sixteen STPOs had been made and three STROs, PQ HL 2057 (13 October 2016). 71 HMICFRS, ‘Stolen Freedom: The Policing Response to Modern Slavery and Human Trafficking’ (October 2017) 78 accessed 6 August 2019. 72 Metcalf (n 17) 51 ff. 73 The Health and Safety Executive is excluded, see Immigration Act 2016, s 3. 74 Margaret Beels, ‘Preventing Labour Exploitation’ (2017) 107 Warwick Papers in Industrial Relations 9. 75 Metcalf (n 17) 23. 76 Metcalf (n 17) 63. 77 Home Office and Department for Business, Energy and Industrial Strategy, ‘Code of Practice on Labour Market Enforcement Undertakings and Orders’ (November 2016) [4]‌, identifies the purpose of these measures as

244  Criminality at Work non-​compliance by warning letters, then to move on to civil penalties. If these measures do not result in compliance, prosecution should be considered, and this is where LME Undertakings may have a role to play, with a custodial sentence available for breach. 78 If an employer has failed to comply with labour market laws, and has failed to respond to lesser warnings and sanctions, it may be appropriate to require an LME Undertaking. HMRC, the GLAA and the EASI are all able to make use of this power. Under section 14 of the 2016 Act an enforcing authority may request a LME Undertaking from a person if that authority ‘believes that person has committed, or is committing, a trigger offence’.79 The Secretary of State can add to these trigger offences by regulation.80 An LME Undertaking may take the form of a ‘prohibition, restriction or requirement’ in order to reduce the risk of the commission of a further trigger offence or its continuance (section 15(1)(5)). The enforcement authorities should take account of matters such as the number of workers affected by the non-​compliance, any harm to them, the amount of any underpayment, and whether the non-​compliance has been deliberate or careless. An LME Undertaking is ‘a voluntary but binding undertaking to comply with any requirements or restrictions as set out by the enforcement body’.81 The Undertaking should be fulfilled within two years. If the LME Undertaking is not accepted, fulfilled in time, or is breached, the enforcement authority may apply for an LME Order.82 The court may make an Order if it satisfied on the balance of probabilities that a person ‘has committed, or is committing, a trigger offence’, and it is just and reasonable to do so (section 18(1)). An LME Order may also be made on conviction for a trigger offence, such as failing to pay the National Minimum Wage or running an unlicensed employment agency.83 According to the Code of Practice, LME Orders should contain requirements or prohibitions designed to reduce the risk of non-​compliance through taking specified steps.84 Interestingly, the government preferred to rely on the LME Undertakings and Orders regime rather than creating a new offence of aggravated labour market breach, on the grounds that it would be burdensome for prosecutors to prove the requisite intent or motivation for the criminal offence and therefore it would be unlikely to be much used.85 However, there are weaknesses in this justification. One is that for sentencing purposes it will be necessary to make a finding on matters of intent and motivation, particularly if imprisonment is in prospect (two years being the maximum). The second difficulty is that there are references in the government papers to the involvement of organized criminal gangs in some cases ensuring ‘there is a level playing field for legitimate competition between law-​abiding businesses, in which workers are guaranteed the national living wage and are protected from exploitation’. ­accessed 6 August 2019. 78 Metcalf (n 17) 64. 79 Immigration Act 2016, s 14(4). Trigger offences include an offence under the Employment Agencies Act 1973 (except s 9(4)(b)), the National Minimum Wage Act 1998, the Gangmasters (Licensing) Act 2004, and include secondary and inchoate offences. 80 Immigration Act 2016, s 14(4)(d). 81 Metcalf (n 17) 64. 82 Immigration Act 2016, s 18. 83 Immigration Act 2016, s 20. 84 Home Office and Department for Business, Energy and Industrial Strategy (n 77) [19]. 85 Home Office, ‘Tackling Exploitation in the Labour Market:  Government Response’ (January 2016) 12 accessed 6 August 2019.

The Preventive Role of the Criminal Law  245 (which needs to be verified), although it should also be mentioned that a significant number of those responding to the government’s consultation emphasized the non-​serious nature of some breaches, and argued in favour of minor responses to minor infractions.86 No LME Undertakings or Orders have been issued by a relevant enforcement body in their first year of operation.87 There is some indication that Orders on conviction (section 20) will be preferred by enforcement bodies.88 This is because the process of issuing an LME Undertaking is more involved. It requires escalation to an Order where a respondent has failed to give an LME Undertaking, has breached it, or its overall deterrence value is undermined.89 It is not clear how, if at all, this function is significantly different from issuing a formal warning letter.90 It has also been highlighted that there are further barriers to the GLAA utilizing LME Undertakings. An Undertaking or Order can only be issued in relation to an offence which relates to an employer’s status as licensed or unlicensed. They cannot be used in relation to an employer’s application to licensing conditions which set out their labour market obligations.91 This sits alongside an open-​ended selection of trigger offences which the Secretary of State can add to by regulation. It is puzzling that Working Time Regulation offences currently fall outside of the recognized list of offences. There appears to be no clear rationale for their exclusion. The Director of Labour Market Enforcement (DLME) has also highlighted limitations of LME Orders having effect for two years’ maximum. The concern is that an Order becomes ‘redundant’ where a sentence in excess of two years is imposed. Once again, we see custodial sentences being introduced in the context of measures that are put forward as preventive, and where the essence of the offence is breach of the conditions or prohibitions in a civil order (and not necessarily the commission of a trigger offence). LME Undertakings and Orders are ‘intended for more serious and persistent offenders where this type of intervention is judged appropriate to prevent further offending’.92 One significant change is the maximum penalty: trigger offences tend to have only fines as sanctions, whereas the maximum for breach of an LME order is two years’ imprisonment. The government states that this is to avoid employers treating fines as ‘business overheads’,93 but this is hardly a justification for depriving someone of their liberty by means of a sentence of imprisonment. Unless it is clear that the offence is so serious that neither a fine nor a community sanction is a sufficient response,94 other means of enforcement ought surely to be adopted. How strong are the justifications for the STPO and the STRO as compared with LME Undertakings and Orders? First, the standard of proof is higher for deciding whether the actions giving rise to an application for an STPO or STRO took place. It requires a court to be satisfied ‘that it is sure 86 Home Office, ‘Tackling Exploitation in the Labour Market: Consultation’ (October 2015) 8 accessed 6 August 2019. 87 Metcalf (n 17) 64. 88 The GLAA was in the process of securing the first order in February 2018, see Metcalf (n 17) 64. 89 Metcalf (n 17) 64. 90 ibid 64. 91 ibid 64. 92 Home Office and Department for Business, Energy and Industrial Strategy (n 77) [6]‌. 93 ibid  [5]‌. 94 This is the statutory threshold that has to be met before a court imposes a custodial sentence: Criminal Justice Act 2003, s 152(2).

246  Criminality at Work that behaviour or actions giving rise to the application took place’.95 This is an enhanced civil standard of proof, which is similar in effect to the criminal standard. A court must then go on to decide whether the STPO or STRO is necessary to protect an individual or individuals from harm. By contrast, a court may make an LME Order if it is satisfied on the balance of probabilities that a person ‘has committed, or is committing, a trigger offence’, and it is just and reasonable to do so (section 18(1)). The evidential standard set by the STPO and STRO is strongly preferable. The STPO and STRO are narrower in scope in that they relate to slavery and trafficking ­offences specifically. LME Undertakings and Orders relate to a wider range of offences which are set out as trigger offences. This is not an argument against the latter set of measures. Secondly, we could argue that there is a weaker case for imposing LME Undertakings and Orders given that there is already a strong deterrent regime in place in relation to the trigger offences covered by these preventive measures. Are there clear reasons for enacting LME Undertakings and Orders alongside a wide range of other deterrence measures, such as the risk of inspection and civil law penalties? This point is pertinent given the DLME’s proposal in May 2018 to significantly increase civil penalties for non-​compliance.96 ‘Turnover taxes’ have been proposed, with the value of a fine directly linked to a company’s turnover.97 Significant revenue streams could then be used to fund the enforcement system.98 Is there still a place for these criminal law measures too, beyond the various criminal offences supporting labour market standards? Why has there been little consideration or evaluation of the relationship between civil and criminal measures? It is not proven that state enforcement via coercive preventive measures offers more effective deterrence than civil law measures carrying substantial financial penalties. A third issue is whether an individual could otherwise enforce their rights. 99 It is useful to set out the current complex framework for labour market enforcement. Unfair pay deductions, rest break entitlements, non-​payment of wages, unfair dismissal, and discrimination claims, among others, can only be pursued by an individual through employment tribunals. Modern slavery, trafficking, non-​payment of the NMW/​NLW, operating without a licence, maximum limits on weekly hours, among others, are regulated by a number of state enforcement bodies (HMRC, EASI, GLAA, HSE, the police), which often overlap in their remit. State agencies such as the GLAA have been endowed with greater powers, resources, and remit ‘as a response to an immediate problem’, such as exploitation and trafficking, rather than developed with regard to the powers of other enforcement agencies and to the system of labour enforcement as a whole.100 This system of individual action via ETs 95 Home Office, ‘Guidance on Slavery and Trafficking Prevention Orders and Slavery and Trafficking Risk Orders Under Part 2 of the Modern Slavery Act 2015’ (April 2017) [3.3.2] accessed 6 August 2019. 96 Metcalf (n 17) 55. 97 ibid 55. 98 ibid 55. 99 Ford (n 60) 23, notes ‘[a]‌bsent from the Government’s celebration in Tackling Exploitation of the UK’s “strong statutory framework” of employment rights was any recognition of the problems for delivery of labour standards caused by the decline in claims resulting from fees.’ 100 See Linda Dickens, ‘Another “A” for the GLA in Enforcing Labour Rights and Preventing Exploitation’ in Linda Dickens and Guglielmo Meardi (eds), Warwick Papers in Industrial Relations (September 2017) 5 accessed 6 August 2019. Proposals for a single public enforcement organizations were rejected as inappropriate by the DLME in his UK Labour Market Enforcement Strategy 2018/​19, see Metcalf (n 17) 20.

The Preventive Role of the Criminal Law  247 or by state enforcement bodies is supplemented by the Acas Helpline, which seeks to provide advice, mediation, and referral to enforcement bodies or to an ET. This complicated patchwork creates difficulties for individuals in enforcing their rights and entitlements, and for enforcement bodies in ‘coordinating their activities and intelligence to deliver effective and efficient enforcement’.101 It is also of concern that there are gaps in enforcement.102 Victims of domestic servitude and forced labour can pursue a claim in an ET for abuses related to their employment. These might include discrimination claims and claims regarding withheld wages.103 But barriers to enforcing labour rights are high for slavery and trafficking victims. The Deduction from Wages (Limitation) Regulations 2014 limit the back-​payment of wages to a period of two years before a claim to an ET.104 The rest will remain unpaid. We have highlighted above the procedural limitations posed by tribunal fees prior to the Supreme Court’s decision in UNISON. The Queen’s Bench Division of the High Court held in April 2018 that victims of modern slavery and trafficking can access legal aid for immigration advice on their right to remain in the UK.105 The legal position prior to this was that an applicant such as LL, who had been brought to the UK and sexually exploited as a child, would be unable to access free immigration advice while the government considered whether she was a victim of trafficking. Arguably the STPO and STRO are preventive measures introduced in response to a legal environment which has created barriers for individual enforcement of employment rights, through complexity, through ET fees, and through lack of legal aid. While a number of these barriers have now been removed certain difficulties remain. First, it is difficult to secure conviction of modern slavery offences given the complexity of exploitation.106 Second, there is the concern that victims of trafficking and slavery, especially individuals holding illegal work status, will not pursue individual enforcement through ETs.107 This does not provide justification for slavery and trafficking preventive measures. But it does provide some strength to a narrowly defined and proportionate preventive measure where the regulatory backdrop is not otherwise regulated by civil law measures.

E.  Conclusions A number of themes emerge from our discussion in this chapter. First, on the corporate omission offences, there is an important question about criminalization of an organization rather than an individual employer, which is in line with labour lawyers’ concern to move beyond a ‘master and servant’ relation in personal work relations. We have queried to what extent current measures to ring-​fence the security of financial interests ought to be supported 101 Metcalf (n 17) 20. 102 ibid 20. 103 ATMG and Human Trafficking Foundation (n 21) 6. 104 ibid 6. 105 LL v The Lord Chancellor CO/​3581/​2017 (16 April 2018)  accessed 6 August 2019. Considering the Legal Aid, Sentencing and Punishment of Offenders Act 2012 (LASPO). 106 See GLAA (n 52) 29. 107 Dickens (n 100) 6: ‘Pro-​active state funded agencies and inspectorates offer advantages in rights enforcement over the individualized private law model which leaves individuals to enforce their rights—​requiring an individual to be in a position to complain, to have knowledge of rights and capacity and willingness to enforce them and which offers redress to only the individual complainant rather than effect wider improvement.’

248  Criminality at Work by coercive preventive measures. Second, whether it is justifiable to ‘responsibilize’ employers by imposing preventive duties on them remains controversial, with principled and pragmatic considerations for and against. Moreover, a regulatory duty is not so difficult to justify as one backed by the criminal sanction. Third, and consequentially, there is concern about the severity of the preventive offence of knowingly employing an illegal worker, both in how the offence is drafted and in its effects when combined with the system of civil penalties. Both the civil and criminal law penalties operate within a ‘hostile environment’ which seems to prioritize the enforcement of immigration offences over worker-​protective measures. Fourth, on slavery and trafficking preventive orders, we see a welcome concern to ensure state-​administered measures to protect vulnerable workers where previously there have been barriers to accessing justice (such as through ET fees). However, the LME Undertaking and Orders scheme in particular seems heavy handed when set alongside the current civil penalties scheme, with proposals now on the table for the civil scheme to carry significantly increased penalties.

13

Licensing of Employing Entities and Criminalization ACL Davies

A.  Introduction In this chapter, I consider the use of licensing as a regulatory technique in employment law.1 In simple terms, licensing regimes involve a requirement to obtain permission from a regulatory agency before engaging in a particular activity. Licensing is usually enforced via the criminal law, by making it an offence to engage in the regulated activity without holding a valid licence. The main use of licensing in employment law is for agencies (sometimes known as ‘gangmasters’) supplying workers in the agricultural and food processing industries, which are regulated by the Gangmasters and Labour Abuse Authority (GLAA).2 It is an offence to supply workers in the regulated industries without holding a licence, and it is also an offence to obtain workers from an unlicensed agency.3 Licensing is controversial as a regulatory technique, and not just because of its reliance on criminal sanctions. It is regarded by its critics as unduly bureaucratic and insufficiently effective at securing compliance with the required standards. The licensing scheme operated by the GLAA has been under a great deal of scrutiny and some threat of ­abolition,4 though at the time of writing, the Director of Labour Market Enforcement has proposed a pilot scheme extending it to other sectors, with a view to gathering data about its effectiveness.5 I begin the chapter by giving an overview of licensing in general, and its relationship with the criminal law. I then explore uses of licensing in employment law settings, before revisiting the discussion about licensing and the criminal law in the employment context. I conclude that, on balance, there are good justifications both for the use of the criminal law to enforce licensing regimes and, if certain conditions are met, for the use of licensing regimes in employment law. 1 This chapter considers developments up to 10 April 2019. I am grateful to the editors and workshop participants for comments on an earlier draft. Responsibility for errors and omissions remains my own. 2 See, generally, Gangmasters (Licensing) Act 2004 (as amended) and accessed 10 April 2019. 3 Gangmasters (Licensing) Act 2004, ss 12 and 13. 4 See, eg, Adrian Beecroft, Report on Employment Law (October 2011) 15 accessed 5 August 2019. 5 The proposal is for regional pilots of licensing for nail bars and hand car washes:  David Metcalf, United Kingdom Labour Market Enforcement Strategy 2018/​19 (HM Government May 2018) 9 accessed 5 August 2019. ACL Davies, Licensing of Employing Entities and Criminalization In: Criminality at Work. Edited by Alan Bogg, Jennifer Collins, Mark Freedland QC, and Jonathan Herring, Oxford University Press (2020). © Alan Bogg, Jennifer Collins, Mark Freedland, and Jonathan Herring. DOI: 10.1093/oso/9780198836995.003.0013

250  Criminality at Work

B.  Licensing Ogus identifies four main characteristics of licensing regimes: the licence must be obtained before the regulated activity can be undertaken; the regulator must make a judgement about the competence of all applicants for a licence; the licence sets minimum standards which are the same for all licence-​holders; and those who do not comply (either with the requirement to obtain a licence or with the licence conditions themselves) face being barred from operating in the relevant sector.6 These elements contrast with a classic ‘command and control’ regulatory regime in which the regulator sets and publishes standards for those wishing to engage in the regulated activity, with sanctions for non-​compliance. In this kind of regime, operators in the relevant sector may have no contact with the regulator unless an infringement comes to light, and even then, they may be given an opportunity to put things right, or a lesser sanction such as a fine, with the possibility of being banned from engaging in the activity being reserved for the most serious (and often deliberate) breaches.7 Ogus notes that licensing regimes are costly to operate, because of the need for the regulator to engage with and assess everyone seeking to work in the relevant sector, and argues that strong reasons are required to justify the use of a licensing regime.8 He identifies two such reasons: cases in which the potential harm from non-​compliance is so great that it is worth making every effort to prevent it, such as an accident in the nuclear industry, and cases in which it is too difficult to set out workable standards to enforce via command and control or other less intrusive regulatory techniques because of the complexity of the regulated activity.9 For example, a licensing system for doctors might be justified both because of the potential for an unqualified practitioner to cause death or serious injury, and because of the wide variety of standards doctors have to meet.10 Ogus also points out that licensing may be an appropriate policy choice where consumers of the services in question would not be well-​placed to form their own judgements about the qualifications of practitioners, something which is again relevant to the regulation of doctors, lawyers, and other similar occupations.11 The problem for licensing regimes, in my view, is their role in granting what Ogus refers to as ‘prior approval’.12 By checking the suitability of firms or individuals before they start engaging in the regulated activity, licensing regimes might seem as if they are intended to prevent harm from occurring, with the result that every time a licensed operator infringes the applicable standards, there is a nagging sense in which the licensing system might have failed. In some cases, this has led to changes to licensing regimes to ensure that licences must be renewed on a regular basis, giving the regulator an additional opportunity to scrutinize those it regulates. In the UK, for example, the General Medical Council (GMC), which regulates doctors, was forced to introduce a ‘revalidation’ system after a series of scandals 6 Anthony Ogus, Regulation: Legal Form and Economic Theory (Hart 2004) 214–​15. 7 See, generally, Ian Ayres and Robert Braithwaite, Responsive Regulation:  Transcending the Deregulation Debate (OUP 1992). 8 Ogus (n 6) 215. 9 ibid 215. 10 It should be noted that licensing regimes may sometimes be used to limit the number of operators in a particular sector, which gives rise to criticism about restricting competition. This is not an objective (or an effect) of the GLAA regime. 11 Ogus (n 6) 216–​20. 12 This is the title given to his discussion of licensing: Ogus (n 6) ch 10.

Licensing of Employing Entities  251 involving harm caused negligently or even deliberately by doctors who had been licensed at the start of their careers but had not had any subsequent contact with the regulator.13 More generally, though, it is important to note that any inspection of someone’s qualifications and basic competences, however careful and however often repeated, cannot guarantee that that individual will never infringe standards whilst engaged in the regulated activity. For example, the fact that a person has passed his or her driving test does not mean that he or she will never lose concentration or get angry or take an unnecessary risk while driving. But it has some value in ensuring that people meet minimum standards of competence before they get onto the roads on their own. It is not realistic to suppose that any regulatory regime—​even licensing—​can prevent harm from occurring, though licensing may be better than some other techniques at reducing the risk of harm occurring. The roads would clearly be less safe if the driving test were abolished, for example.

C.  Licensing and the Criminal Law It was noted above that the criminal law is routinely used to enforce licensing regimes. It is usually a criminal offence to carry on the relevant activity without having first obtained a valid licence from the regulator. This is intended to serve two purposes: to deter non-​ engagement with the regulatory regime by people wishing to engage in the regulated activity, and to deter non-​compliance with regulatory standards by licence-​holders because of the threat of losing the licence and thus being unable legally to continue with the activity. There may also be supplementary offences such as providing false information or obstructing the regulator. A key question is whether it is appropriate to use the criminal law in these ways. Creating a new criminal offence is the most serious step Parliament can take, so it is generally accepted that this should only be done where there is a strong justification for criminalization as opposed to some other strategy, such as making the conduct a civil wrong or using taxation or education to discourage it.14 In recent years, a number of eminent theorists have expressed concern about a tendency on the part of legislators to ‘over-​criminalize’: to create new offences in order to be seen to be tackling a problem, without paying due regard to the question of justification.15 A useful basic starting-​point for justifying the creation of a new criminal offence is Feinberg’s positive reformulation of Mill’s well-​known ‘harm principle’:16 It is always a good reason in support of penal legislation that it would be effective in preventing (eliminating, reducing) harm to persons other than the actor (the one

13 See, generally, accessed 10 April 2019. 14 See, generally, AP Simester and others, Simester and Sullivan’s Criminal Law: Theory and Doctrine (6th edn, Hart 2016) ch 16. 15 Some of the leading contributions to a large literature are Andrew Ashworth, ‘Is the Criminal Law a Lost Cause?’ (2000) 116 Law Quarterly Review 225; RA Duff, Answering for Crime (Hart 2007); Douglas Husak, Over-​ Criminalization (OUP 2008); AP Simester and Andrew von Hirsch, Crimes, Harms and Wrongs: On the Principles of Criminalization (Hart 2011). 16 John M Robson (ed), The Collected Works of JS Mill, Vol XVIII—​Essays on Politics and Society Part I (On Liberty) (Routledge 1977) 223.

252  Criminality at Work prohibited from acting) and there is no other means that is equally effective at no greater cost to other values.17

As the quotation makes clear, the focus is on the prevention or reduction of harm, which may be done by criminalizing not only acts that cause harm but also acts that create an (unjustified) risk of harm. This explains why it is justified to criminalize, for example, attempted murder as well as murder itself. The harm principle taken alone would be both under-​and over-​inclusive, so it is necessary to add two qualifications. First, the harm in question must also be wrongful: many acts cause harm but it would not be justified to criminalize them.18 For example, it is harmful to deprive a competitor of trade by undercutting its prices, but this behaviour would not normally be regarded as wrongful in the context of a market economy. Second, as the quotation makes clear, even when a wrongful harm has been identified, it is still necessary to weigh up the advantages and disadvantages of criminalization.19 These might include the seriousness of the harm and the likelihood of it occurring, set against any value attaching to the harm-​ causing conduct. For example, where a particular type of speech causes wrongful harm, it is necessary to consider whether the fundamental right to freedom of expression, especially in political contexts, tips the balance against criminalization. These basic principles present obvious, but not insurmountable, challenges in relation to regulatory offences such as those used to support licensing regimes.20 A first problem is that practising without a licence may not necessarily cause harm or even any risk of harm. For example, a doctor who is qualified in another jurisdiction and has not obtained the right to practise medicine in the UK is engaged in an unlicensed activity, but may be perfectly capable of safe and effective practice. However, we can treat the harm principle as satisfied if we step back and consider at a more general level the harm likely to occur in the absence of a system in which doctors’ qualifications have to be checked, which would obviously include death, serious injury, and outbreaks of contagious disease. From this perspective, the harm principle may be satisfied in relation to licensing offences if the licensing system itself reduces the risk of harm.21 The second requirement is that the harm must be wrongful. Again, engaging in a licensed activity without a licence might not be regarded as especially morally wrong. But this can be addressed on two levels. One involves examining the moral dimension of the licensing system itself, using a similar approach to that taken to the harm principle, above. If the licensing system is designed to reduce the risk of harmful wrongdoing which is deserving of moral censure, for example, because it would involve causing physical injury or exploiting vulnerable people, we might be prepared to extend that censure to people who disregard the licensing regime. The other is to examine the morality of disregarding the licensing regime itself. As Duff puts it, even if the unlicensed practitioner is perfectly safe, ‘they display a culpable arrogance in refusing to subject themselves to regulations which are properly designed to protect public safety’.22 17 Joel Feinberg, The Moral Limits of the Criminal Law: Harm to Others (OUP 1984) 26. 18 Simester and Sullivan (n 14) 665. 19 ibid 665. 20 See, generally, Jeremy Horder, ‘Bureaucratic “Criminal” Law: Too Much of a Bad Thing?’ in RA Duff and others (eds), Criminalization: The Political Morality of the Criminal Law (OUP 2014). 21 Simester and Sullivan (n 14) 677. 22 RA Duff, ‘Crime, Prohibition and Punishment’ (2002) 19 Journal of Applied Philosophy 97, 105.

Licensing of Employing Entities  253 The third requirement is to ensure that the advantages of criminalizing the harmful and wrongful conduct outweigh the disadvantages. In many cases, this will involve considering whether a lesser measure than creating a new criminal offence would be equally effective in practice. As the Law Commission explains: ‘[i]‌n a regulatory context, it is not enough to justify making engagement in conduct criminal, that the conduct must be deterred or punished in some way.’23 In relation to licensing, a key question that often arises is whether it would it be possible to achieve the same effect using some form of civil enforcement. To a large extent, this is an empirical question about how individuals and firms might react to different types of penalty. On the one hand, it might be thought that civil penalties would be less effective because they lack the element of condemnation and stigma associated with the criminal law and might simply be seen as a ‘tax’ on doing business. Regulators may need to have a criminal offence at the top of what Ayres and Braithwaite term the ‘enforcement pyramid’ as their ultimate threat.24 On the other hand, simply criminalizing certain conduct may not be sufficient to stigmatize it, particularly if a majority of people do not regard the conduct as morally wrong. Under these circumstances, a substantial civil penalty might be equally effective. Indeed, if regulated individuals and firms perceive that the regulator is more likely to invoke the civil penalty (eg because there are fewer procedural hurdles and a lower burden of proof), this may increase its deterrent effect.25 However, an important caveat is in order. The more stringent the penalty, the more important it is that the person accused of wrongdoing has proper safeguards in place to protect his or her interests. This is reinforced by Article 6 of the European Convention on Human Rights, which provides a set of protections for those accused of criminal offences that goes beyond the requirements of a fair trial in a civil case. The European Court of Human Rights (ECtHR) is alert to the possibility that states may ‘misclassify’ wrongdoing as administrative or civil in order to avoid having to observe those safeguards, and it does not take their classifications at face value.26 Instead, it considers the nature of the wrongdoing and the associated penalty and determines the matter for itself. This means that it may not be possible to put in place a very severe penalty and to classify it as administrative. Thus, if we consider that non-​compliance with a licensing regime warrants a heavy fine, it may be necessary to enforce this via the criminal law on Article 6 grounds. This will certainly be the case where there is a possibility of imprisonment. Finally, it is worth noting one common criticism of the use of the criminal law in regulatory contexts: that Parliament enacts offences but regulators rarely use them, so that any deterrent effect they may generate is much reduced in practice.27 The deterrent effect of the criminal law in any particular instance is, of course, an empirical question, but it is self-​ evident that—​if individuals think about the criminal law at all—​they consider the likelihood of getting caught and the likelihood of being prosecuted as well as the seriousness 23 Law Commission, Criminal Liability in Regulatory Contexts:  A Consultation Paper (No 195, August 2010) [4.7]. 24 Ayres and Braithwaite (n 7). 25 Compare the finely balanced argument around this issue which is presented by Michael Ford in the Introduction to his Chapter 21 in the present volume with respect to the criminalization of health and safety at work. 26 See, generally, David J Harris and others, Harris, O’Boyle and Warbrick’s Law of the European Convention on Human Rights (4th edn, OUP 2018) ch 9. 27 Law Commission (n 23) [1.25]–​[1.27].

254  Criminality at Work of the offence and the sanction. However, this concern can be addressed both in relation to regulation generally and in relation to licensing regimes in particular. Most regulators employ a staged approach (an enforcement pyramid28) in which lesser sanctions are used for minor or accidental infringements,29 reserving prosecution for egregious cases of deliberate non-​compliance.30 From this perspective, occasional uses of the power to prosecute are both rational and sufficient. In any event, licensing regimes are less prone to the ‘insufficient prosecution’ criticism than other kinds of regulation, because it is likely that for many people, the decision to refuse or withdraw a licence will be a sufficiently serious sanction in itself.31 Prosecution is inherently confined to cases in which an individual has chosen to engage in or continue with the regulated activity despite not being licensed to do so. Of course, prosecution needs to be a sufficiently realistic possibility to deter people from operating without a licence, but a low number of prosecutions is not necessarily a sign of an inactive regulator, as it might be in a classic ‘command and control’ system.

D.  Licensing in Employment Settings Perhaps the most common form of licensing that is of relevance in employment settings is the requirement for many people employed in professional occupations to hold a valid licence in order to do their job.32 This is acknowledged in unfair dismissal law, for example. Under section 98(2)(d) Employment Rights Act 1996, a ‘potentially fair’ reason for dismissal ‘is that the employee could not continue to work in the position which he held without contravention (either on his part or on that of his employer) of a duty or restriction imposed by or under an enactment’. Thus, if an individual is employed in a role that requires him or her to be a qualified solicitor with a current practising certificate, but he or she is ‘struck off ’, the employer is likely to be entitled lawfully to dismiss him or her, provided that a fair procedure is followed. However, the focus of the present chapter will be on licensing obligations for individuals or firms as employers or suppliers of labour, of which the main example is the licensing regime for suppliers of labour in the agriculture and food processing industries, operated by the GLAA. I will also explain the old regime for employment agencies in all sectors put in place by the Employment Agencies Act 1973, and repealed in 1995. An important caveat before I turn to these examples is that my focus on licensing for the purposes of this chapter cuts across a different conceptualization of enforcement in this area pursued by the government. The Immigration Act 2016 created a new office of Director of Labour Market Enforcement (DLME), with the task of overseeing the GLAA, the Employment Agency Standards Inspectorate (EAS), which survived the abolition of 28 Ayres and Braithwaite (n 7). 29 The Regulatory Enforcement and Standards Act 2008 sought to facilitate greater use of ‘intermediate’ civil sanctions. It was based on recommendations made by Richard B Macrory, Regulatory Justice: Making Sanctions Effective (Final Report, November 2006) accessed 5 August 2019. 30 For an empirical case study of this phenomenon, see Keith Hawkins, Law as Last Resort: Prosecution Decision-​ Making in a Regulatory Agency (OUP 2002). 31 Procedural protections for disappointed applicants and licence-​holders are generally a matter of administrative law, enforced by means of judicial review. The classic case is McInnes v Onslow Fane [1978] 1 WLR 1520. In the case of the GLAA, there is a right of appeal to a tribunal: Gangmasters (Appeals) Regulations 2006 (SI 2006/​662). 32 For lawyers, the principal statute is the Legal Services Act 2007, and for doctors, it is the Medical Act 1983.

Licensing of Employing Entities  255 licensing for agencies and continues to enforce the remaining aspects of the regulatory regime, and the National Minimum Wage (NMW) compliance team within Her Majesty’s Revenue and Customs (HMRC), which enforces the NMW. The government sees these three bodies as having a shared responsibility for ‘public enforcement’ of labour law for vulnerable workers and the current role of the DLME is, in large part, to encourage them to work together.33 The Taylor Review34 recommended an increase in the scope of HMRC’s enforcement powers to include sick pay and holiday pay and a possible extension of EAS’s powers to include the Agency Workers Regulations 2010.35 In its response in 2018, the government appeared broadly to accept these proposals.36 Then, in a surprise move in March 2019, as part of an attempt to win the support of Labour MPs for Theresa May’s Brexit deal, it was announced that the three agencies would be merged into a single body with enhanced powers.37 At the time of writing, no further detail about this proposal has yet emerged. Even if a merger goes ahead, it is worth noting there are significant differences between the roles and powers of the three bodies and, in some respects, grouping them together highlights the extent to which ‘public enforcement’ of labour law in the UK does not form a comprehensive or coherent whole, given the historical accidents of its development.

1.  Gangmasters and Labour Abuse Authority In 2006, after the tragic deaths of a number of Chinese migrant workers who were collecting shellfish in Morecambe Bay, the Gangmasters Licensing Authority (GLA) was set up to regulate the agriculture, shellfish, and food processing industries by means of a licensing regime for agencies who supply labour in those sectors.38 The legislation originated in a Private Members’ Bill and was shaped by the very considerable public outrage over the exploitation of the workers who lost their lives in the tragedy.39 In 2016, the GLA was given enhanced investigative and enforcement powers which it can use across the labour market as a whole,40 not just in the regulated sectors, and as a consequence it was renamed the Gangmasters and Labour Abuse Authority (GLAA).41 However, somewhat confusingly, the 33 Immigration Act 2016, s 2. 34 Matthew Taylor, Good Work: the Taylor Review of Modern Working Practices (July 2017) accessed 5 August 2019. 35 SI 2010/​93. 36 See, generally, HM Government, Good Work: A Response to the Taylor Review of Modern Working Practices (February 2018) accessed 5 August 2019, and on the specific issue of enforcement, Department for Business, Energy and Industrial Strategy and Ministry of Justice, Good Work: the Taylor Review of Modern Working Practices: Consultation on Enforcement of Employment Rights Recommendations (February 2018) accessed 5 August 2019. 37 UK Government press release, ‘Workers’ Rights to Be Protected in UK Law’ (6 March 2019) accessed 10 April 2019. 38 See (n 2). 39 Introduced by Jim Sheridan MP. 40 Immigration Act 2016, ss 11–​13. Labour Abuse Prevention Officers can investigate offences relating to gangmasters, agency work, minimum wage and modern slavery, and have at their disposal a new system of civil orders (Immigration Act 2016, ss 14–​30) as well as prosecutions. 41 Immigration Act 2016, s 10. According to the explanatory notes on the Lords amendments, the continued reference to ‘gangmasters’ in the name is intended to retain the organization’s ‘international reputation’. The term

256  Criminality at Work licensing regime was not extended at the same time, and there is an ongoing discussion, as noted above, led by the new DLME, about whether it should apply in any additional sectors.42 The Gangmasters (Licensing) Act 2004 defines its scope of application in section 3(1) as ‘agricultural work’, ‘gathering shellfish’, and ‘processing or packaging’ any agricultural produce, shellfish, or fish. Under section 4(2), the concept of a gangmaster is given a simple definition: ‘A person (“A”) acts as a gangmaster if he supplies a worker to do work to which this Act applies for another person (“B”).’ Section 4(3) makes clear that this definition applies broadly and is not affected by factors such as the precise nature of the contractual relationships between the various parties or the presence of intermediaries. The requirement to hold a licence is set out in section 6. It is an offence under section 12(1) to act as a gangmaster without holding a licence. This is a strict liability offence. It is also an offence under section 12(2) to possess a document knowing or believing it to be ‘false’ or ‘improperly obtained’, or belonging to someone else, ‘with the intention of inducing another person to believe’ that the individual in question is a licensed gangmaster. On summary conviction for a section 12 offence, the individual may be imprisoned for up to six months or fined up to the statutory maximum, or both. On indictment, the maximum sentence of imprisonment is ten years. The GLAA now has access to the system of hybrid civil/​criminal Labour Market Enforcement Undertakings (LMEUs) and Labour Market Enforcement Orders (LMEOs) under the Immigration Act 2016, but it is too early to judge the effect of this change.43 Interestingly, the licensing regime is further reinforced in the case of gangmasters by virtue of section 13, which makes it an offence to enter into an arrangement to obtain workers from an unlicensed gangmaster. It is a defence for the end user to show that it took reasonable steps to satisfy itself that the gangmaster had a valid licence. The intention is to encourage end users to check gangmasters’ licences as a way of making it difficult for unlicensed gangmasters to remain in business. This offence carries a maximum sentence of six months’ imprisonment or a fine not exceeding the statutory maximum, or both. The Gangmasters (Licensing Conditions) Rules 2009 set out the conditions that must be fulfilled for a gangmaster to obtain and retain a licence,44 but it is important to read these in conjunction with the GLAA’s own guidance.45 There is substantial overlap between the Rules and the equivalent provisions governing employment agencies more generally.46 For example, there are restrictions on charging fees to workers, on supplying workers during industrial disputes, and withholding payment to workers.47 However, the Rules also contain some provisions that are intended to tackle problems specific to the sector, such as those governing accommodation and travel, or the proper supervision of shellfish gatherers.48 The GLAA’s own guidance replicates these rules in more user-​friendly language, but also has, however, always been somewhat problematic because it conjures up a much narrower image than the GLAA’s licensing remit in fact covers. I use the abbreviation GLAA throughout for convenience. 42 Metcalf Report (n 5). 43 Immigration Act 2016, ss 14–​30. For an overview, see ACL Davies, ‘Immigration Act 2016’ (2016) 45 Industrial Law Journal 431. 44 SI 2009/​307. 45 GLAA, Licensing Standards (October 2018). 46 Employment Agencies Act 1973; Conduct of Employment Agencies and Employment Businesses Regulations 2003 (SI 2003/​3319). 47 Gangmaster Rules 2009 (n 44) sch 1: paras 7, 10, and 13. 48 Gangmaster Rules 2009 (n 44) paras 21 and 26.

Licensing of Employing Entities  257 makes clear that gangmasters are expected to comply with the general law as a condition of holding a licence. Thus, for example, the Rules do not replicate the law on health and safety, but health and safety compliance is one of the eight licensing standards enforced by the GLAA.49 The GLAA deals with the number and complexity of the licensing standards by operating a scoring system.50 Standards regarded as ‘critical’ attract a score of thirty points if they are not met, and most others a score of eight points. A score of thirty points or more will usually lead to refusal of, or immediate revocation of, a licence. This means that it is not generally possible to hold a licence whilst being in breach of a critical standard. A score between zero and thirty will usually lead to the grant of a licence but with Additional Licensing Conditions (ALCs) requiring the licence-​holder to tackle areas of non-​compliance with non-​critical standards, or a non-​immediate revocation. A GLAA licence must be renewed once a year, giving the GLAA regular opportunities to review licence-​holders’ ongoing compliance.51 According to the latest available GLAA annual report, for 2016–​17, the GLAA estimates that there are some 464,000 workers in the sectors subject to licensing.52 During that year, it managed around 1000 licences, refused to grant a licence in sixteen cases, revoked eighteen licences covering 4543 workers, and recovered £93,165 owed to some 1225 workers. Most published evaluations of the GLAA are broadly positive, but it is difficult to state with any precision how effective it is.53 While it is possible to evaluate various aspects of the GLAA’s performance in relation to licensed individuals and firms (such as its speed and effectiveness in investigating complaints), it is much harder to tell how much unlicensed activity is taking place. Since unlicensed labour providers are committing a criminal offence, they operate ‘under the radar’ and are inherently difficult to detect or quantify. It will be interesting to see how the proposed pilot scheme to extend the scope of licensing to new sectors will be evaluated.

2.  Employment Agencies and Businesses The Employment Agencies Act 1973 created a nationwide licensing scheme for employment agencies and businesses,54 which was abolished in 1995.55 This scheme is of interest for two reasons. First, it gives us another example of licensing in the employment context

49 GLAA, Licensing Standards (n 45) Licensing Standard 6, 22–​26. 50 ibid  3–​5. 51 ibid  5–​6. 52 Gangmasters Licensing Authority, Annual Report and Accounts 1 April 2016 to 31 March 2017 (HC 1402, July 2018) accessed 5 August 2019. 53 See, eg, Andrew Geddes and others, Gangmasters Licensing Authority:  Evaluation Study:  Baseline Report (Gangmasters Licensing Authority/​DEFRA 2007); Mick Wilkinson and others, An Evaluation of the Gangmasters Licensing Authority (Oxfam 2009) accessed 5 August 2019. For the most recent government evaluation, see Department for Environment Food & Rural Affairs, Report of the Triennial Review of the Gangmasters Licensing Authority (April 2014) accessed 5 August 2019. 54 For definitions, see Employment Agencies Act 1973, s 13. 55 Deregulation and Contracting Out Act 1994, s 35.

258  Criminality at Work to consider, even if it is now only of historical interest. And second, despite the repeal of the licensing provisions of the Act, some related elements of that regime, including criminal offences, remain in force. This also offers some interesting points of comparison which will be examined later in the chapter. Before 1973, some local authorities regulated agencies through local legislation, and there was sectoral legislation for some occupations, for example, nursing.56 The Employment Agencies Act 1973 was the first attempt to lay down a national regulatory regime for agencies. The scheme of the Act was relatively simple. Under section 1, anyone wishing to run an employment agency or business was obliged to hold a current licence, and it was a criminal offence attracting a fine to run an employment agency or business without a licence. The grounds for refusing to grant or revoking a licence were set out in section 2(3), as follows: (a) that the applicant is an individual under the age of twenty-​one years; (b) that the applicant is a person who on account of misconduct or for any other sufficient reason is unsuitable to hold a licence in respect of an agency or business of the class in question; (c) that any person (other than the applicant) who is or is to be concerned with the carrying on of the employment agency or employment business is a person who on account of misconduct or for any other sufficient reason is unsuitable to be associated with an agency or business of the class in question; (d) that the premises at which the employment agency or employment business is or is to be carried on are unsuitable in respect of an agency or business of the class in question; (e) that the employment agency or employment business has been or is being improperly conducted.

The licensing scheme was initially run by the Department of Employment and provision was made for appeal to a person nominated by the Secretary of State.57 An identifiable agency within the Department, the Employment Agency Licensing Office, appears to have been established in the early 1980s.58 This changed its name to the Employment Agency Standards Inspectorate (EAS) when the licensing regime was abolished in 1995. More detailed standards for agencies were set out in the Conduct of Employment Agencies and Employment Businesses Regulations 1976.59 These were quite strongly procedural in character, focusing on matters such as the information to be provided to end users by agencies about their terms and conditions and records to be kept by agencies.60 Special provision was made in respect of young workers, the recruitment of workers from outside the UK for jobs in the UK, and the recruitment of UK workers for jobs abroad.61 Perhaps most importantly from the perspective of workers themselves, the Regulations required agencies to maintain a separate client account and to pay workers promptly.62 In the

56 For example, the Nurses Agencies Act 1957, which consolidated and amended provisions from earlier Acts. 57 Employment Agencies Act 1973, s 4. 58 See Hansard, Written Question 19954, asked by Tulip Siddiq on 14 December 2015, and answered by Nick Boles on 12 January 2016. 59 SI 1976/​715. 60 ibid regs 2 and 8 (employment agencies); 9 and 12 (employment businesses). 61 ibid regs 5, 6, and 11. 62 ibid reg 7 and sch 2.

Licensing of Employing Entities  259 case of employment businesses in particular, there was an express prohibition on making it a condition of paying a worker that money had been received from the end user.63 It is perhaps worth noting, however, that the regime did not fully address two central problems with agency work. First, in relation to employment businesses in particular, it did not attempt to regulate workers’ employment status, simply requiring that the employment business should notify the hirer and the worker as to whether the worker was its employee or self-​employed.64 Second, although section 6 of the 1973 Act made it a criminal offence for an agency to charge a work-​seeker a fee, this ban was not absolute and exceptions were permitted, principally in relation to modelling and various parts of the entertainment industry.65 As Hepple and Napier pointed out at the time, this meant that the UK government could not ratify International Labour Organization (ILO) Convention 96,66 something which it had undertaken to do in 1951.67 It is evident from the statistics that the licensing regime for employment agencies and businesses was significantly under-​resourced. A mere fifty-​eight licences were refused and nine revoked between 1977 and 1994.68 This was against a background of 5057 licences in force in 1977 and 14,482 licences in force in 1994. Licensing in the sector was abolished as part of the ‘bonfire’ of regulation brought about by the Deregulation and Contracting Out Act 1994.69 Curiously, though, the abolition focused only on licensing itself and left much of the substance, and the small group of officials (renamed EAS, as noted above) charged with enforcing it, untouched. The cynical explanation for this would be that EAS was not seen as worth abolishing because it was not perceived as a threat. The current position is that employment agencies and businesses are subject to the Conduct of Employment Agencies and Employment Businesses Regulations 2003,70 and it is an offence not to comply with the Regulations.71 EAS investigates complaints and conducts targeted ‘operations’ in sectors perceived to be at high risk of non-​compliance, sometimes in collaboration with other agencies such as HMRC.72 It has at its disposal the possibility of prosecuting offenders or even seeking an order prohibiting a person from running an employment agency or business,73 but proceeds principally by persuasion and warning.74 Since the Immigration Act 2016, it has also had access to ‘intermediate’ sanctions in the form of LMEUs and LMEOs.75

63 ibid reg 9(10). 64 ibid reg 9(1), (4), and (6). 65 See the Employment Agencies Act 1973 (Charging Fees to Workers) Regulations 1976 (SI 1976/​714). 66 ILO Convention 96, Fee-​Charging Employment Agencies Convention (Revised) 1949. 67 BA Hepple and BW Napier, ‘Temporary Workers and the Law’ (1978) 7 Industrial Law Journal 84, 90. 68 See Michael Wynn, ‘Regulating Rogues? Employment Agency Enforcement and Sections 15–​18 of the Employment Act 2008’ (2009) 38 Industrial Law Journal 64, 70. 69 Deregulation and Contracting Out Act 1994. 70 SI 2003/​3319. 71 Employment Agencies Act 1973, s 5(2). 72 See Department for Business, Energy and Industrial Strategy, Employment Agency Standards Inspectorate:  Annual Report 2017–​2018 (August 2018) accessed 5 August 2019. 73 Employment Agencies Act 1973, ss 5(2), 6, and 3A respectively. 74 See Department for Business, Energy and Industrial Strategy, Employment Agency Standards Inspectorate: Enforcement Policy Statement (June 2017) accessed 5 August 2019. 75 See (n 43).

260  Criminality at Work EAS remains significantly under-​resourced, though at the time of writing its profile has been raised somewhat now that it is overseen by the DLME and would be included in the proposed new single enforcement agency, discussed above.

3.  Evaluation As I noted above, it is often pointed out by the critics of licensing that the system promises more than it delivers: the fact that someone holds a valid licence does not guarantee that they will not break the rules, and the very existence of a licensing regime does not guarantee that there are no unlicensed operators. However, it is not possible to claim that any regulatory regime is completely successful in preventing harm from occurring. In this subsection, I will attempt to assess some of the advantages and disadvantages of the licensing system operated by the GLAA, and what (if anything) has been lost by the move to a ‘command and control’ approach in relation to agencies. I will also draw some comparisons with the much more elaborate ‘command and control’ regime operated by HMRC’s National Minimum Wage (NMW) compliance team, which is better resourced and has a much wider range of sanctions, organized into a more recognizable ‘enforcement pyramid’.76 One caveat is in order before I begin. There is, of course, an important debate to be had about whether public enforcement is justified either in relation to a particular employment right or group of rights, or in relation to a particular sector of the economy.77 I do not engage with these debates in this chapter for reasons of space, focusing instead on the question of what type of public enforcement is appropriate once the initial decision not to rely solely on workers’ own (collective or individual) enforcement efforts has been taken. A first advantage of licensing is that it puts the regulator on notice as to the identities and locations of those seeking to operate lawfully in the regulated activity or sector, and gives it an opportunity to conduct an inspection before granting a licence. As explained above, the GLAA’s approach to this task contains an in-​built opportunity for firms and individuals that are not fully compliant, but not in breach of a ‘critical’ standard, to improve their level of compliance. Commentators are often dismissive of this aspect of licensing, arguing that it is relatively easy to get the paperwork in order to make a successful application. However, although the GLAA refuses a relatively small number of licences (sixteen in 2016–​17, against an overall total of 1004) this does indicate that the licensing process is not automatic and does involve scrutiny of applicants’ credentials.78 It also means that the GLAA can gather intelligence about the degree of risk of future non-​compliance among licensed operators, and about individuals or firms which may decide to operate without a licence. Of course, it is difficult to say how many unlicensed operators there are who do not put themselves forward for a licence at all, but it does mean that the GLAA starts at a considerable advantage relative to, for example, EAS, which only comes into contact with an agency when a complaint is made.

76 Ayres and Braithwaite (n 7). 77 I discuss some of the arguments in favour of special treatment for the agriculture sector in ACL Davies, ‘Migrant Workers in Agriculture: A Legal Perspective’ in Cathryn Costello and Mark Freedland (eds), Migrants at Work: Immigration and Vulnerability in Labour Law (OUP 2014). 78 Gangmasters Licensing Authority, Annual Report (n 52).

Licensing of Employing Entities  261 A second, related advantage is that a licence may be granted for a fixed period of time, so that there is a renewal process in which the regulator can engage in further scrutiny. In the case of the GLAA, this is annual.79 Again, it is often argued that this is ineffective, because it should be relatively easy for a firm or individual to create the appearance of compliance. However, there is a good example in a recent GLAA annual report of an individual who had set up an agency using a false identity who was caught at the renewal stage, when he was unable to produce up-​to-​date identity documents in the false name he had been using.80 This shows that unscrupulous operators may not be able to sustain an appearance of compliance over time. Again, other methods of regulation are entirely reactive, to complaints or other intelligence, in identifying cases of non-​compliance. The GLAA carries out investigations on this basis in addition to operating the annual renewal process. A third advantage of licensing, and in my view the most significant, is that it facilitates the public enforcement of multiple standards. To understand this, it is useful to compare and contrast the GLAA’s licensing scheme with the current approach to the regulation of agencies. As noted above, it is a criminal offence under section 5(2) Employment Agencies Act 1973 attracting a fine to fail to comply with the Regulations governing the conduct of employment agencies and businesses. Of course, while the courts (including magistrates’ courts) now have power to set fines with no maximum limit, any fine would have to be proportionate to the offence, so less serious offences, if prosecuted at all, would attract smaller fines. In practice, EAS (which has very limited resources) brought a prosecution in just one case in 2016–​17, but it is significant for the fact that the offender was charged with eight violations of the Regulations suggesting, as one might expect, that prosecution is unlikely unless there are multiple infringements.81 Under the gangmasters’ licensing regime, by contrast, minor infringements attract a points score which may, cumulatively, put an operator’s licence in jeopardy. It seems likely that this is a more effective way of promoting compliance with the full range of standards enforced by the GLAA than a threat of prosecution, because it combines a serious consequence—​the loss of the licence—​without the need for the GLAA to go to court to bring it about. One response to this might be to point out that it would not be unreasonable to reserve public enforcement for the most serious cases of exploitation, where workers’ basic rights are being violated. This might be done for a variety of reasons, the most obvious of which is that public enforcement is costly and therefore a scarce resource which should be used with care. However, it is important to think carefully here about the kind of labour law enforcement system we want. Whilst this is a large question for a short chapter, there are two broad alternatives that we might consider, assuming that private enforcement remains the principal method for workers to secure their rights. Public enforcement might supplement private enforcement either because the rights thus enforced are considered to be of particular importance, or because the workers affected are considered to be unlikely to be able to enforce their rights without support. The former is intuitively appealing and would suggest

79 GLAA, Licensing Standards (n 45) 5–​6. 80 GLA, Annual Report and Accounts 1 April 2015 to 31 March 2016 (HC 279, December 2016) 10 accessed 5 August 2019. 81 Department for Business, Energy and Industrial Strategy, Employment Agency Standards Inspectorate Annual Report 2016–​2017 (November 2017) [35]–​[36] accessed 5 August 2019.

262  Criminality at Work that the state should identify a ‘core’ of fundamental rights for workers, such as freedom from forced labour, safe working conditions and a minimum rate of pay, and focus on enforcing these by public means. But the latter is also worth considering. Workers may struggle to enforce their rights either because of the nature of the rights themselves, or because of their personal circumstances. In the former case, the worker may be subjected to repeated small infringements, for example in relation to the NMW, each of which is minor in itself and therefore not worth challenging, but when taken together over time, or across the workplace as a whole, have a significant cumulative effect.82 In the latter case, workers who are vulnerable because of, for example, migration status, temporary working, or working via an intermediary, are likely to face difficulties in enforcing any rights, not just the fundamental ones. From this perspective, it might be argued that we should not settle for the enforcement of just a subset of rights—​even the most fundamental ones—​and should focus instead on ensuring that vulnerable workers can enjoy conditions of ‘decent work’.83 This latter objective is more readily achieved by a licensing regime than by one which provides for prosecution, even where a range of offences are in principle subject to prosecution. A fourth and final advantage of licensing is that it can lead to relatively rapid removal of particular individuals from the relevant sector. Under the gangmasters’ licensing regime, an individual who is refused a licence, or has his or her licence revoked, can no longer lawfully operate as a supplier of workers in the regulated sectors, and faces the threat of prosecution if he or she continues to do so. Under the regime for agencies, it is possible for EAS to obtain a prohibition order under the 1973 Act, which bans an individual from running an agency on threat of prosecution.84 These orders can operate for up to ten years at a time. At the time of writing, eleven people are the subject of prohibition orders.85 However, this requires EAS to obtain evidence that the individual in question is unsuitable and to take active steps to seek the order from a tribunal. Before concluding, however, there are two important caveats that are not unique to licensing but are crucial to a rounded evaluation. First, the effectiveness of any enforcement regime depends on the level of resources devoted to it. There are striking differences between the three enforcement bodies we have been considering in this chapter.86 In 2016–​17, the GLAA had a budget of £4.8 million to deal with around 1000 licence-​holders covering 460,000 workers. HMRC had an enforcement budget of £20 million to monitor roughly two million workers covered by the NMW. EAS had a budget of £500,000 to cover some 18,000 agencies dealing with over a million workers. Thus, while the GLAA and HMRC have a roughly equivalent budget of £10 per worker, EAS had a tiny budget of about 50p per worker. It is easy to see how a government that did not favour licensing in particular 82 Simester and Sullivan (n 14) 678. 83 This is now a key organizing concept for the work of the ILO. See, generally, accessed 10 April 2019. 84 Employment Agencies Act 1973, s 3A. 85 Employment Agency Standards Inspectorate, ‘Guidance People Prohibited from Running an Employment Agency or Business’ accessed 10 April 2019. This list was last updated on 23 February 2018. 86 Director of Labour Market Enforcement:  David Metcalf, United Kingdom Labour Market Enforcement Annual Report 2017/​18 (March 2019) 7–​8 accessed 5 August 2019.

Licensing of Employing Entities  263 or public enforcement in general could simply starve a regulator of the resources needed to do its job effectively. Although all three bodies received dramatic increases in funding in 2017–​18, the differentials remain, and in the case of the GLAA, the new funding was to support its extended remit to investigate abuses across the economy, rather than to support the licensing scheme. Indeed, there is a risk that the GLAA’s extended remit may, over time, divert resources away from licensing. Second, a significant disadvantage of licensing, which it shares with other techniques of public enforcement, is that it is not formally concerned with obtaining redress for affected workers.87 Indeed, if a breach of licensing conditions is detected and an operator’s licence is revoked, it will no longer be able to supply workers in the regulated sectors and any workers for whom it has been responsible may lose their jobs and accommodation. Even if the workers are being exploited, this may not be a particularly good outcome for them. Moreover, if their migration status is not in order, they may be at risk of being deported. There is no guarantee even for individuals identified as victims of modern slavery through the National Referral Mechanism that they will be allowed to remain in the UK should they wish to do so.88 The only public enforcement mechanism we have considered that focuses on affected workers is the process of issuing a notice of underpayment by HMRC, which requires the employer to pay workers and former workers the NMW arrears it owes them, on relatively generous terms, in addition to paying a penalty to the Secretary of State.89 Of course, in practice, the GLAA does not operate the licensing scheme without paying careful regard to the needs of workers and a significant part of its work involves recovering money owed to workers, as noted above.90

E.  Licensing and the Criminal Law in Employment Settings In this section, I will revisit the discussion surrounding the use of the criminal law to enforce licensing regimes, focusing in particular on whether the offence of acting as an unlicensed gangmaster can be justified in the light of concerns about over-​criminalization, particularly in regulatory contexts.91 It is important to begin by reminding ourselves of the context in which the Gangmasters Licensing Act 2004 was passed. The Morecambe Bay tragedy meant that there was considerable public awareness of, and outrage at, the potential for vulnerable workers to be forced into dangerous situations by unscrupulous gangmasters. Indeed, the gangmaster who had been in charge of the cockle-​pickers was ultimately convicted of the manslaughter of at least twenty-​one people. It is not therefore surprising that the full force of the criminal law was brought to bear on the problem. Moreover, the fact that the legislation was initially 87 Interestingly, there is civil liability for breach of licensing conditions under r 7 of the Gangmasters (Licensing Conditions) Rules 2009 (SI 2009/​307). See Antuzis v DJ Houghton Catching Services Ltd [2019] EWHC 843 (QB), [2019] IRLR 629; G v DJ Houghton Catching Services Ltd [2016] EWHC 1376 (QB), [2016] IRLR 859. But this requires that the affected workers are able to bring proceedings. 88 Home Office, Discretionary Leave Considerations for Victims of Modern Slavery: Version 2.0 (September 2018) accessed 5 August 2019. 89 National Minimum Wage Act 1998, ss 17, 19, and 19A. 90 Gangmasters Licensing Authority, Annual Report (n 52). 91 For reasons of space I do not discuss the s 13 end user offence in detail here.

264  Criminality at Work presented as a Private Members’ Bill meant that it did not reflect the concerns of the then government about taking a risk-​based approach to regulation in order to reduce ‘burdens’ on compliant businesses.92 This explains how a licensing scheme came to be enacted at a time when such schemes had fallen out of fashion. In asking whether the criminal offences contained in the 2004 Act can be justified, we are therefore asking an ahistorical question. The offences exist because there was a political imperative to tackle an immediate problem, not a carefully worked-​out strategy for the use of the criminal law. Nevertheless, the question remains relevant, not least because the future of the GLAA and of the licensing system in particular are by no means secure. We saw above that the proper use of the criminal law to enforce a licensing regime turns primarily on whether we are satisfied that the regime itself is a necessary means of reducing the risk of serious and wrongful harm. For a labour lawyer, this is a relatively simple question to answer. The presence of an intermediary is a risk factor for exploitation, because it creates a ‘triangular’ relationship between the intermediary, the end user, and the workers, in which responsibility for protecting the workers’ rights may be obscured. Moreover, there are a number of additional risk factors for exploitation in the agriculture, shellfish, and food processing sectors, including a high proportion of migrant workers (who may be vulnerable because of poor English language skills, a lack of knowledge of their rights, and fears about their migration status), workers who depend on their employer for accommodation as well as a job, and well-​documented health and safety problems. These are borne out by the kinds of problems uncovered by the GLAA’s investigations, including trafficking and forced labour, dangerous working conditions, non-​payment of wages, confiscation of passports, and overcrowded accommodation. From a labour law perspective, there is no real difficulty in saying that these are serious harms that should attract moral censure. There are two considerations which may undermine this position. One is the concern that exploitation exists across the labour market, and not just in the sectors regulated by the GLAA. It is often alleged that the presence of the GLAA may simply have shifted the problem to other sectors, such as construction or hospitality. However, while this may create inequality of treatment between labour exploiters in different sectors, it seems more logical to treat this as a reason for extending, rather than removing, the licensing regime, or at least exploring other avenues for addressing the problem of exploitation. It is perverse to suggest that a mechanism for tackling some exploitation should be abolished because it does not combat all exploitation. The other is the point I made in the previous section about the relationship between licensing and ‘decent work’. Paradoxically, the very ability of licensing regimes to enforce a range of standards may undermine the justification for the use of the criminal law, because not every standard so enforced will be tackling serious harm deserving moral censure. However, this concern can be met by reference to the ‘enforcement pyramid’.93 Given the points system operated by the GLAA, a gangmaster would need to infringe a number of less serious standards in order to be denied or to lose a licence, and would only be at risk of prosecution if they continued to operate despite not having a licence to do so.

92 See the government-​ commissioned Hampton Review:  Philip Hampton, Reducing Administrative Burdens: Effective Inspection and Enforcement (HM Treasury 2004) (can be found at accessed 5 August 2019). 93 Ayres and Braithwaite (n 7).

Licensing of Employing Entities  265 A final point to consider relates to the offence itself, and whether it can be regarded as deserving of moral censure in its own right, with a relatively high maximum prison sentence. This is relatively easy for the section 12(2) offence which involves an element of dishonesty. It deals with the situation in which, for example, an individual is in possession of a document he or she knows to be false (such as a fake GLAA licence) and intends to use it to persuade end users that he or she is a licensed gangmaster. It is more difficult for the strict liability offence in section 12(1), which could cover anything from a (relatively) innocent mistake to a deliberate decision to carry on as a gangmaster after losing a licence, or to avoid engaging with the GLAA altogether in order to run a business that is engaged in the serious exploitation of workers. The benefit of the current formulation of the offence is that the prosecution has a relatively straightforward task: proving that the individual acted as a gangmaster and did not have a licence. It seems likely in practice that a combination of prosecutorial discretion and appropriate sentencing will serve to ensure that this offence is largely confined to cases in which some moral censure for ‘culpable arrogance’ is warranted.94 Indeed, the guidance on sentencing laid down by the courts might be seen as reflecting an assumption that the cases they are likely to see will be at the serious end of the spectrum.95 To sum up, although the use of the criminal law to enforce the licensing regime for gangmasters is explained by the historical and political context in which the 2004 Act was passed, I would argue that it can also be justified in principle. The licensing system is designed to reduce the risk that gangmasters will exploit vulnerable workers, causing them serious physical, mental, and economic harm, and although not all cases will be serious, it does deal with situations in which the moral censure and stigma of the criminal law are warranted.

F.  Conclusion This chapter set out to explore the role of licensing as a regulatory technique within employment law. The principal example of this technique in operation is the licensing regime for firms and individuals who supply labour in the agricultural and food processing industries, as operated by the GLAA, though it has also been used in the past for agencies across the labour market. Licensing is often criticized for being expensive to operate, and burdensome and bureaucratic for compliant firms. In particular, it is often attacked for failing to prevent non-​compliance even though this is an unrealistic objective for any regulatory regime. As I have sought to argue in this chapter, however, licensing has some advantages, including its ability to create opportunities for contact between the regulator and those it regulates and, most importantly, its ability to require operators to comply with a range of labour standards, not just a small selection of fundamental rights. Thus, it has an important role to play in creating the conditions for decent work where it applies. An effective licensing regime must be reinforced by a powerful sanction to deter firms and individuals from seeking to operate without a licence, either because they have not attempted to obtain one, or been refused one, or had one revoked. The use of the criminal law to achieve this objective is regarded as controversial by some theorists because of fears

94

95

Duff, ‘Crime, Prohibition and Punishment’ (n 22). R v Morkunas [2014] EWCA Crim 2750, [2014] 11 WLUK 36 [51].

266  Criminality at Work about over-​criminalization. I have argued that the licensing regime operated by the GLAA should not be regarded as an example of this phenomenon, because it seeks to reduce the risk of exploitation of vulnerable workers which can involve serious wrongdoing by unscrupulous labour intermediaries and significant harm to workers. It is, of course, essential that this sanction can be enforced by a competent and well-​resourced regulator, so that the strength of the sanction is coupled with a reasonable chance of detection and punishment. It is to be hoped that the GLAA’s extended role under the Immigration Act 2016 will not prove to be a distraction from its core licensing functions. Ultimately, many of the questions raised in this chapter are empirical questions about the extent of exploitation in the labour market, the effectiveness of regulators, and the deterrent effect of particular regulatory techniques. These empirical questions are inherently hard to answer, given that labour exploitation by unlicensed operators is a criminal activity that the perpetrators are at pains to conceal. Nevertheless, even if it is hard to make conclusive statements about the efficacy of licensing, the 2004 Act made an important statement about the type of labour market we want to have: one in which the promotion of decent work, not just the prevention of exploitation, is the objective of public enforcement.

14

Criminalizing Care Workers A Critique of Prosecution for Ill-​treatment or Wilful Neglect LJB Hayes

A.  Introduction A recent extension of UK criminal law has carved a new space of public interest in abuse in care settings. It enables the state to intervene in matters of care worker conduct with powers of prosecution for ill-​treatment or wilful neglect. Arguably, this represents a radical advance in the protection of the human rights of vulnerable adults, demonstrating high regard for moral conduct in care-​giving and ensuring the accountability of people in positions of trust. Such laudable objectives could be seen to inform the contours of a normative role for the state in the policing of care-​giving relationships. The espoused purpose of the ‘care worker offence’ at section 20 of the Criminal Justice and Courts Act (CJCA) 2015 is to protect vulnerable adults from abuse.1 Accordingly, paid care workers may be subject to prosecution where their behaviour at work appears to fit judge-​made definitions of ill-​treatment or wilful neglect of a person for whom care is provided. A charge of ill-​treatment will relate to allegedly abusive action(s) whereas a wilful neglect charge will relate to allegedly abusive inaction(s). Conduct that might fall within definitions of ill-​treatment includes aggressive shouting, teasing, and taunting; feeding a vulnerable person in a medically incorrect way; or slapping them.2 Conduct that may give rise to a charge of wilful neglect includes singular or repeated omissions such as failing to take a person to the toilet or to change their incontinence pads, failing to put a person to bed, and failure to assist with the taking of medication.3 The focus of this chapter is the application of ill-​treatment or wilful neglect provisions to workers caring for elderly and disabled people in residential and nursing homes, domiciliary care services, or homecare. It follows concerns expressed by Herring that prosecutions are being launched against low-​paid and overworked care staff while the real causes of neglect may well be found higher up the management chain.4 Public discourse and policy typically regards hands-​on care workers as ‘unskilled’ and there are few formal training or qualification requirements.5 Their duties centre on personal care tasks such as helping vulnerable adults to eat, drink, use the toilet, changing incontinence pads, assisting or prompting with medication, dressing, washing, putting to bed etc. The size of the UK adult 1 Hereafter CJCA 2015, s 20. 2 See, eg, R v Heaney [2011] EWCA Crim 2682, [2011] 11 WLUK 155; R v Pumbien [2015] EWCA Crim 2186, [2015] 12 WLUK 423. 3 See, eg, R v Maghmouj [2016] EWCA Crim 1647, [2016] 10 WLUK 308; R v ESM [2016] EWCA Crim 1496, [2016] 9 WLUK 410; R v Nursing [2012] EWCA Crim 2521, [2013] 1 WLR 1031. 4 Jonathan Herring, Vulnerable Adults and the Law (OUP 2016), 215. 5 For details see discussion later in this chapter, particularly text to (nn 106–​21) below. LJB Hayes, Criminalizing Care Workers In: Criminality at Work. Edited by Alan Bogg, Jennifer Collins, Mark Freedland QC, and Jonathan Herring, Oxford University Press (2020). © Alan Bogg, Jennifer Collins, Mark Freedland, and Jonathan Herring. DOI: 10.1093/oso/9780198836995.003.0014

268  Criminality at Work social care workforce is large, with 2  million workers fragmented across approximately 30,000 employer organizations plus over 100,000 disabled individuals who act as employers to workers whom they personally hire.6 State-​funded care provision has been highly marketized over two decades and most employment is with for-​profit private enterprises with service-​delivery contracts with local authorities. Terms and conditions have been described as ‘the worst of any’ and the weakness of employment relationships is reflected in the sector’s considerable recruitment difficulties and higher labour turnover rates than in any other part of the UK labour market.7 According to section 20(3) and section 20(4) of the CJCA 2015, unpaid carers (including those receiving state-​funded social security allowances as caregivers), are not liable for prosecution under the provisions. Evidently, Parliament did not intend for paid care workers to be held to account as abusive caregivers per se. Rather, care workers are held to account for their misconduct as paid workers. This framing of the provisions is troubling and in this chapter, I suggest that the offences are doing something more than protecting vulnerable adults. That something more is to amplify the gendering of worker subordination and weaken the prospect of safer and sustainable social care by individually targeting the care worker and excising her conduct from the context of her work. With an offence in which ill-​treatment or wilful neglect by a caregiver is actionable only if that caregiver is paid, it seems at first glance that relations of employment are too firmly in the spotlight. Since it is not possible to prosecute unpaid carers under CJCA 2015 provisions, wilful neglect by an unpaid carer is not a specific criminal offence unless circumstances fit with earlier provisions set out in the Mental Capacity Act 2005; these require that the neglected adult lacks capacity or is reasonably believed to lack capacity.8 For example, in R v Kurtz, a daughter was prosecuted and convicted under provisions at section 44 of the Mental Capacity Act 2005 (MCA) for a sustained neglect of her mother, including failure to arrange necessary medical treatment which caused prolonged and unnecessary pain, coincidental with early death.9 However, the Court of Appeal overturned her conviction because the MCA 2005 offence of wilful neglect could only be committed if the neglected adult lacked capacity or the defendant reasonably believed the adult to lack capacity. Had Kurtz been paid to provide care as a worker, a conviction under CJCA 2015 would have been possible. In respect of the care of adults who lack capacity, provisions at section 44 of the MCA 2005 treat paid and unpaid carers equally.10 It is therefore difficult to understand why the recent extension of criminal law to protect all adults in receipt of care, as per CJCA 2015, 6 Information on employer numbers and employment via direct payments in England see Skills for Care, ‘The State of the Adult Social Care Sector and Workforce in England’ (September 2018) accessed 5 August 2019. 7 Denise Kingsmill, ‘The Kingsmill Review: Taking Care. An Independent Report into Working Conditions in the Care Sector’ (2004) accessed 31 May 2019. On labour turnover see text to (n 91) below. 8 Note that wilful neglect is an omissions offence and qualitatively distinct from general provisions such as those set out in Offences Against the Person Act 1861, eg assault occasioning actual bodily harm under s 47 or malicious infliction of grievous bodily harm under s 20. There is potential for prosecution under s 5 of the Domestic Violence, Crime and Victims Act 2004 for allowing a vulnerable adult to die or suffer serious physical harm caused by an unlawful act, but it is not an omissions offence. However, omission may constitute gross negligence manslaughter see R v Stone and Dobinson [1977] QB 354. 9 [2018] EWCA Crim 2743, [2019] 2 WLR 1509. Mental Capacity Act 2005 (MCA 2005). 10  See Herring (n 4), 214.

Criminalizing Care Workers  269 diverges from this. Its prosecuting requirement of payment for care suggests section 20 of the CJCA 2015 sets the criminal law in a distinctive relation to the contract of employment in respect of vulnerable adults who do not lack capacity.

B. A Workers’ Crime In this chapter, I explore section 20 of the CJCA 2015 as a provision by which misconduct at work becomes criminal behaviour. By framing instances of elder abuse by caregivers as specific crimes of ill-​treatment or wilful neglect, occurring only in the presence of pay, CJCA 2015 conceives of ill-​treatment or wilful neglect as a workers’ crime. Applying the label, ‘workers’ crime’, serves to highlight a crime in which care workers exhibit inadequate dedication, poor self-​discipline, and weak attachment to their duties of employment. A useful illustration is provided in the sentencing comments of Judge Stead in R v Logan, in which a six-​month prison sentence for wilful neglect was imposed on two care workers who had tried to cover up the true cause of a serious injury sustained by a care home resident.11 The resident was very frail and needed to be lifted in a hoist operated by two people. Logan attempted to lift her in the hoist while working alone and the resident fell to the floor, breaking her pelvis. Logan and her colleague Gillies recorded the fall in the incident book but falsely recorded that it had happened when the resident was being lifted by them both. In passing sentence, Judge Stead said that their corner-​cutting had been motivated by ‘a combination of idleness and self-​interest’ so as to ‘maximise their rest breaks’.12 Accordingly, Logan and Gillies had failed as workers and the judge narrated their offending conduct as wilful neglect by workers. It is pertinent to note that staff shortages in the UK care sector are stubbornly persistent and services have been chronically underfunded for many years by successive governments.13 Thus, it is highly likely that, in some care settings, occupational cultures exist in which cutting corners (such as operating a hoist singlehandedly) is implicitly accepted as though a necessity of circumstance.14 There is some evidence that severe cost pressures across the care industry can influence the habitual behaviour of some staff, embedding sub-​ standard routines and attitudes which may well persist even in the absence of the immediate stimulus of staff shortage.15 This is a serious problem and vulnerable people can be hurt as a consequence. However, from an employment perspective, important normative questions arise about the extent of management responsibility for staff conduct, about managerial

11 Crown Court hearing reported by Andrew Bardsley, ‘Care Workers who Colluded to Cover Up Fall of an 87 Year Old Woman from her Bed Jailed’ Bolton Times (Bolton, 2 April 2015) www.theboltonnews.co.uk/​news/​ 12867238.care-​workers-​who-​colluded-​to-​cover-​up-​fall-​of-​an-​87-​year-​old-​woman-​from-​her-​bed-​jailed/​ accessed 5 August 2019. 12 ibid. 13 See Association of Directors of Adult Social Services, ‘ADASS Annual Budget Report’ (2018) accessed 1 June 2019. 14 See, eg, report of inquest at Cannock Coroners Court into the death of Mrs Teale, ITV report ‘Care Home Staff Did not Use Hoist to Move Elderly Resident, Inquest Told’ ITV News (2 September 2015) accessed 11 June 2019. 15 See, eg, West Sussex Adults Safeguarding Board ‘Orchard View Serious Case Review’ (June 2014) accessed 31 May 2019.

270  Criminality at Work capacity to shape the conditions in which staff are operating, and the scope of managerial prerogative. Within the contractual context of employment, managerial prerogative can shape worker behaviour via powers of discipline or dismissal, as well as via powers of reward and recognition.16 We might ask for example, if judicial ‘readings’ of ill-​treatment or wilful neglect, such as those of Justice Stead, result in the imprisonment of care workers for failures of performance which should have been headed off at a workplace level by effective management and could have been addressed by day-​to-​day formalization of professional standards through strong and effective relations of employment?17 With a particular focus on the crime of wilful neglect, the case law reviewed in section C of this chapter suggests that the line navigated by courts in the context of social care (between failure of performance issues and criminal conduct) is indeed a fine one. I argue that this is because social care workers are afforded an identity as ‘part-​care-​professional’ and ‘part-​family-​member’ hybrid; meaning that precedent established in the gendered context of family care does not sit comfortably, and neither does precedent established in respect of breach of professional occupational duties. Accordingly, the capacity to prosecute care workers under occupationally specific criminal provisions has the effect of strengthening the power asymmetry that characterizes the employment relationships in which care workers are typically engaged. Managerial prerogative is enhanced by state power, but terms and conditions are not reciprocally improved. This is concerning in light of the academic literature on elder abuse in care settings, much of which highlights connections between poor employment, lack of training, the low status of social care work, and the mistreatment and neglect of vulnerable adults.

C.  A Crime against Caring Further potential problems lie in establishing the guilty mind of an accused care worker. For neglect to be judicially regarded as ‘wilful’ in situations of social care, the prosecution must show offending behaviour was motivated by an attitude of ‘not caring’.18 This mens rea element of the wilful neglect offence is distinctive from that applied to trained medical professionals (being a deliberate decision not to act on medical knowledge). The guilty mental state of ‘not caring’ must be present at the moment of inaction. This reflects normative expectations that while care work is not ‘skilled’, it is more than a ‘mere’ economic exchange of labour for pay and is underpinned by obligations which extend beyond explicit contractual duties. Yet these vocational and moral obligations are not reflected in the labour market status of care work. Rather, regard for hands-​on caring as a moral and relational labour draws on gendered concepts of unpaid care, the family work of mothers, and the bonds of affection that are culturally assumed as adequate motivation in the absence of financial reward.19 Hence in situations of social care, the crime of wilful neglect is constructed as a

16 See Mia Ronnmar ‘The Managerial Prerogative and the Employee’s Obligation to Work:  Comparative Perspectives on Functional Flexibility’ (2006) 35(1) Industrial Law Journal 56. 17 See also Alan Bogg and Mark Freedland (this volume). 18 R v Turbill [2013] EWCA Crim 1422, [2013] 7 WLUK 407. 19 Marian Barnes, Care in Everyday Life: An Ethic of Care in Practice (Policy Press 2012); Neil Cobb, ‘Compulsory Care-​Giving:  Some Thoughts on Relational Feminism, The Ethics of Care and Omissions Liability’ [2008] 39 Cambrian Law Review 11.

Criminalizing Care Workers  271 crime against caring, by which I mean a crime committed when the accused violates gendered expectations that are culturally associated with home and family, rather than those associated with notions of professional duty, workplace, and contract. Judicial sentencing comments provide good evidence that care workers are regarded in the courtroom as replacement (female) family members. For example, Mrs Justice Thirlwall in R v Healey stated: Elderly people have a right to be treated with respect by everyone in the community, when they are ill and living in residential homes, they are entitled and we must demand, they are properly cared for. What this appellant did was the opposite of that.20

Healey was thus held to standards that transferred from the family home into the residential home and were identified by the judge within notions of ‘the community’ rather than the worker–​employer relationship. A further example is the case of R v Karen Fitzpatrick, in which a homecare worker was imprisoned for giving a back-​handed slap across the face to a woman with dementia who spat out her tablets.21 In handing down the maximum sentence available, Mr Justice Irwin described Fitzpatrick’s behaviour as ‘unforgivable in a civilized society’ because ‘families entrust the care of their most vulnerable relatives to people like you’ (my emphasis). In so doing he imparted that women in paid care work were substitute family members and that their behaviour ought to align with normative expectations of familial care. Feminist scholars of care ethics have drawn attention to ways in which such expectations are highly gendered and they have explored how the prevalence of gender norms in matters of caregiving can fuel assumptions that caregivers ought to be ‘caring’ on the basis of attributes assumed ‘natural’ in women.22 Such gender norms lead to a conflating of paid caregiving with unpaid labour, equating the workplace with the home and privileging a discourse of family over a discourse of skill and professionalism.23 It is pertinent to note that abusive actions such as slapping are addressed in general criminal provisions such as common assault and battery or actual bodily harm. We might thereby question whether the capacity for prosecution on grounds of ill-​treatment satisfies perceived disciplinary deficits that are rather more symbolic and cultural, than legal. Questions of gender are particularly salient to wilful neglect provisions, which criminalize care workers for inaction and omission. In a general sense, the concept of criminal liability for neglecting to take action is more contentious than for undertaking an action prohibited in law. Traditional common law principles permit the freedom to choose not to act, to ‘mind one’s own business’, in order to avoid unjustified interference with individual autonomy.24 A classic exposition involves a hypothetical passer-​by who sees a drowning man and knows he can save the man’s life by extending his hand. Despite this knowledge, the passer-​by commits no offence by choosing not to help him, even though it results in the man drowning.25 20 [2011] EWCA Crim 2682, [2011] 11 WLUK 155 [9]‌. 21 R v Karen Fitzpatrick [2013] EWCA Crim 730, [2013] 4 WLUK 590. 22 Berenice Fisher and Joan Tronto ‘Toward a Feminist Theory of Caring’ in Emily K Abel and Margaret K Nelson (eds), Circles of Care: Work and Identity in Women’s Lives (SUNY Press 1990) 35. 23 See Gabrielle Meagher, ‘What Can We Expect from Paid Carers?’ (2006) 34 Politics and Society 1, 33–​54; LJB Hayes, Stories of Care: A Labour of Law. Gender and Class at Work (Palgrave 2017). 24 William Wilson, Central Issues in Criminal Theory (Hart 2002) 91–​94. 25 James Stephens, Digest of the Criminal Law (Macmillan 1877) 135.

272  Criminality at Work Some scholars regard the deference to individual freedom in the drowning man example to be based on a misreading of autonomy.26 They point to the fact that individual autonomy is possible only in the context of social cooperation and argue therefore that liberal jurisprudence ought to balance respect for individual autonomy with an appreciation of social responsibility. However, such a balance would demand agreement on where the law ought to draw the line. If criminal liability were to arise because the law expects us all to care for one another, might a rich man commit a crime for ignoring the pleas for financial help of a hungry man in need of food? Despite the moral case for action to assist, it is hard to imagine that law would ever compel the rich man to care by criminalizing his omission and indifference. Liability for omissions is exceptional: it requires clear justification and is the preserve of a limited number of statutory instruments.27 These address instances in which individuals choose to take on responsibilities that create ‘special relationships’ (including parent/​child relationships) and are therefore considered to owe positive duties to support or assist particular others.28 Neil Cobb has rightly drawn attention to the gendered implications of an assumption that relationships of care and support are chosen.29 Women, as a group, bear most responsibility for care, and conformity with powerful social norms means it is illusory to regard women as volunteer parties to relationships that are ‘special’, in the sense that they warrant exceptional status in law. This line of argument may provide support for the exclusion of unpaid caregivers from the reach of section 20 of the CJCA 2015.30 If so, it also sharpens the case for critique of specific measures in criminal law arising on account of the caregiving functions of an overwhelmingly female workforce. Arguably, it suggests that women, in their ‘natural’ domain of the family, are defined by voluntary relations of care that normatively lie beyond the reach of law, so long as they are unpaid. With pay, the employment status of care workers brings the labour of care into public sight and makes their conduct a matter of public interest, such that they are subject to discipline both as parties to an employment relationship and as parties to a pre-​existing relationship of gender. Additionally, we might consider that where hands-​on care is at the core of paid work, the relation of employment is juridically regarded as a ‘special relationship’ in which gendered social responsibilities, external to the employment contract, are policed via the disciplinary power of the state, as an adjunct to employer powers of managerial prerogative. Cobb also argues that, in matters of criminal law, juridical ‘readings’ of individual conduct in respect of omissions liability are influenced by gendered social norms. Accordingly, responsibility for care is juridically assumed of individual women because women, as a group, are society’s assumed caregivers.31 A gendered critique of section 20 of the CJCA 2015 requires attention to the mens rea of ‘not caring’, which defines the mental attitude necessary for conviction. Through discussion of case law in this chapter, I suggest that the doctrine of ‘not caring’ is an unsatisfactory measure with which to judge, as criminality, the inaction of care workers. It appears that judicial consideration has constructed the mens rea 26 Andrew Ashworth, ‘The Scope of Criminal Liability for Omissions’ (1989) 105 Law Quarterly Review 424. 27 Glanville Williams, ‘Criminal Omissions—​The Conventional View’ (1991) 107 Law Quarterly Review 86. 28 Cobb (n 19). 29 ibid. 30 Such exclusion means that unpaid relations of care are not marked out as ‘special relations’ for the purpose of omissions liability in the form of wilful neglect beyond relations of care for children and relations of care for adults with lack of mental capacity, see Children and Young Persons Act 1933; MCA 2005. 31 See Cobb’s discussion and analysis of R v Stone and Dobinson [1977] QB 354: Cobb (n 19).

Criminalizing Care Workers  273 of ‘not caring’ from gendered norms of unpaid family caregiving and has set it to work in an employment context. I evidence that this approach hides from view important matters pertaining to the employment relationship. Indeed, the process of individual prosecution hinges entirely on individual behaviour and excludes from consideration a raft of matters controlled by management, such as understaffing, lack of training, and poor supervision. Thus, my analysis suggests an ontological confusion about the status and role of a paid care worker, part-​care-​professional and part-​family-​member hybrid. It is reflected in the dual framing of the section 20 of the CJCA offence as a ‘workers’ crime’ (in which the potential for prosecution is too firmly wedded to care as an employment activity) and a ‘crime against caring’ (in which the route to potential conviction is insufficiently attentive to the employment context in which care activities are conducted).

D. Ontological Confusion In sections E and F below, I point to the organizational and political implications of the capacity of the state to criminalize care workers as per section 20 of the CJCA 2015. I show how criminal law at work in the care sector can mis-​frame the problem of elder abuse and by-​ pass the significance of low-​quality jobs, of widespread inadequacies of care provision, of employment deregulation, and of longstanding under-​funding. Conceivably, the criminal law can increase care workers’ subordination to their employers in ways which do not improve the safety and well-​being of vulnerable adults, for example if prosecution can ensue from a failure of contractual performance dressed up as a failure of care. Research about the problem of ill-​treatment and neglect in care settings correlates violence and poor care with low-​quality employment, chronic underfunding, lack of training, and aggressive marketization. The evidence indicates that abuse in care settings is principally a systemic rather than individually driven problem. Indeed, official statistics suggest that the most frequently reported form of abuse is ‘neglect and omission’, a problem arising from a sector-​wide lack of resources in the context of government-​imposed austerity measures.32 Care which is safe, effective, and ethical is a corollary of good quality jobs and strong relations of employment. It is therefore concerning that criminal provisions, by their very existence, isolate the conduct of individuals from their workplace settings. The framing and interpretation of those provisions means that judicial narration about care failure is deflected away from the workplaces and industrial context in which problems arise. Therefore, I argue in conclusion that section 20 of the CJCA 2015 exacerbates a pre-​existing ontological clumsiness about the role and status of care workers. It draws on gendered social norms to protect the ideological validity of a marketized system of care in which employment relationships have been strategically weakened in pursuit of cost-​savings, despite abuse in care-​settings being a known consequential risk.

32 NHS Digital, ‘Safeguarding Adults: Annual Report, England 2015–​2016 Experimental Statistics’ (Department of Health 2016) 20 accessed 5 August 2019; NHS Digital, ‘Safeguarding Adults Collection (SAC), Annual Report, England 2016–​17 Experimental Statistics’ (Department of Health, 15 November 2017) 15–​ 16  accessed 5 August 2019.

274  Criminality at Work

E.  Care Worker Prosecution When care workers are implicated in allegations of abuse, the possible consequences range from the loss of a job, to occupational barring, to imprisonment.33 It is frequently the case that prosecutors draw on general criminal provisions regarding theft, grievous bodily harm, sexual assault etc, when people in receipt of care are victims of offences perpetrated by care workers. Where such general crimes occur in the context of caregiving relationships they are made out in aggravated form and sentencing decisions reflect a recognition that victims are especially vulnerable. Yet, crimes committed in the private and personal context of caregiving are not easily discovered and it can be difficult to obtain reliable evidence where the victims are frail, confused, or have impaired communication.34 This is an important driver for the creation of bespoke provisions in criminal law to enhance the protection of adults considered vulnerable because they lack mental capacity. Ill-​treatment or wilful neglect offences in respect of vulnerable adults are set out in three distinct statutes (Mental Health Act 1983; MCA 2005; CJCA 2015). Offences of ‘ill-​ treatment or wilful neglect’ do not include a requirement to prove harm and do not require an act to have been perpetrated.35 They are either-​way offences carrying a maximum sentence of five years’ imprisonment on conviction on indictment, or a fine, or both and on summary conviction the maximum sentence is twelve months, or a fine, or both.

1.  Ill-​treatment and Wilful Neglect in the Courts Statutory provisions set the scope of each offence by defining who is liable for wrongdoing, yet they offer no prescription about what is meant by ‘ill-​treatment or wilful neglect’. Definitions of the actus reus and mens rea elements of these offences are a matter for the courts. The judicial meaning of ‘wilful neglect’ developed in relation to offences against children according to provisions in the Children and Young Persons Act 1933. Perpetrators, as parties to the special relationship of parenting, were found to have failed to respond to the needs of their child when medical attention was required. In R v Sheppard, parents were prosecuted for ‘wilful neglect’. According to the Children and Young Persons Act 1933, neglect must be such that it is ‘likely to cause [the child] unnecessary suffering or injury to health’ and includes a failure to provide ‘medical aid’.36 For the House of Lords, Lord Diplock interpreted such neglect as ‘a failure, for whatever reason, to provide the child whenever it in fact needs medical aid, with the medical aid it needs’.37 However, for neglect to be wilful, and thus fall within the scope of sanctions, evidence was required that a parent was either aware that ‘the child’s health might be at risk if it were not provided with medical aid or that the parent’s unawareness of this fact was due to not caring whether the child’s health was at risk or not’.38 33 See discussion in Jill Manthorpe and Kritika Samsi, ‘Care Professionals’ Understanding of the New Criminal Offences Created by the Mental Capacity Act 2005’ (2015) 30(4) International Journal of Geriatric Psychiatry 384. 34 Alison Brammer, ‘Carers and the Mental Capacity Act 2005: Angels Permitted and Devils Prosecuted’ (2014) 8 Criminal Law Review 589. 35 Set out in CJCA, s20; MCA 2005, s 44; and Mental Health Act 1983, s 127. 36 Set out in the Childrens’ and Young Persons Act 1933, s 1(1) and s 1(2). 37 R v Sheppard [1981] AC 394, 404 (Lord Diplock). 38 ibid 403 (Lord Diplock).

Criminalizing Care Workers  275 Thus, for the Children and Young Persons Act 1933 offence to be made out, the mens rea element of wilful neglect includes the requirement that the accused’s state of mind is one of ‘not caring’. This was affirmed in the assessment of Lord Keith of Kinkel who noted in Sheppard the applicability of wilful neglect where a parent ‘fails to provide m