Missing Voice?: Worker Voice and Social Dialogue in the Platform Economy 1839105534, 9781839105531

Citation preview

Missing Voice?

THE FUTURE OF WORK AND EMPLOYMENT The future of work and employment has rarely been as opaque as it is now and never as speculated upon. This important series will explore the biggest issues facing the modern workforce, policy-makers and businesses today. Its books will include topics as diverse as the rise of the gig economy to the role of platform companies and an ageing workforce. Book proposals on topics such as employment (in)security, inclusivity, equity, remote teams, wellbeing, AI, digitalization and voice will be welcomed. Books will be theoretically rigorous and empirically grounded but also will light the way for future research and debate in the field of employment relations.

Missing Voice?

Worker Voice and Social Dialogue in the Platform Economy Edited by

Adrian Wilkinson Professor of Employment Relations and Human Resource Management, Griffith University, Australia and University of Sheffield, UK

Tony Dundon Professor of Employment Relations and Human Resource Management, University of Limerick, Ireland and the Work and Equalities Institute, University of Manchester, UK

Paula K. Mowbray Senior Lecturer in Human Resource Management, Griffith University, Australia

Sarah Brooks Lecturer in Organisational Behaviour, University of Sheffield, UK THE FUTURE OF WORK AND EMPLOYMENT

Cheltenham, UK • Northampton, MA, USA

© Adrian Wilkinson, Tony Dundon, Paula K. Mowbray and Sarah Brooks 2022

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, electronic, mechanical or photocopying, recording, or otherwise without the prior permission of the publisher. Published by Edward Elgar Publishing Limited The Lypiatts 15 Lansdown Road Cheltenham Glos GL50 2JA UK Edward Elgar Publishing, Inc. William Pratt House 9 Dewey Court Northampton Massachusetts 01060 USA A catalogue record for this book is available from the British Library Library of Congress Control Number: 2022943075 This book is available electronically in the Business subject collection http://dx.doi.org/10.4337/9781839105548

ISBN 978 1 83910 553 1 (cased) ISBN 978 1 83910 554 8 (eBook)

EEP BoX

Contents List of figuresvii List of editorsviii List of contributorsx PART I 1

INTRODUCTION

Missing voices? Integrating worker voice and social dialogue in the platform economy Adrian Wilkinson, Tony Dundon, Paula K. Mowbray and Sarah Brooks

PART II

2

MACRO CONSIDERATIONS

2

The legal framework of worker voice in the platform economy Alan Bogg

3

Giving platform workers a say: regulating for voice in the gig economy Andrew Stewart and Jim Stanford

4

How do workers gain voice on digital work platforms? Hotspots and blind spots in research on platform worker voice Laura Thäter, Thomas Gegenhuber, Elke Schüßler and Markus Ellmer

20

48

71

PART III MESO FORCES 5

How anger sparks voice: exploring individual and collective voice in the remote gig economy Alex J. Wood, Nick Martindale and Vili Lehdonvirta

6

Work, spatial changes and ‘arm’s-length bargaining’ in the platform economy Miguel Martínez Lucio, Debra Howcroft, Holly Smith, Stephen Mustchin and Stefania Marino v

109

130

vi

7

Missing voice?

Voice in platform-enabled gig work Anne Keegan and Jeroen Meijerink

153

PART IV MICRO EXPERIENCES 8

Is anybody there? An examination of the voice experiences of Uber ride-share drivers Paula K. Mowbray

PART V 9

175

INTEGRATING WORKER VOICE IN THE GIG ECONOMY

When voices from below are heard: the case of a Swiss online food-delivery platform Nicola Cianferoni, Luca Perrig and Jean-Michel Bonvin

195

Index216

Figures 4.1

Hotspots and blind spots in platform worker voice

5.1

‘Anger’ independent variable underlying items

117

5.2

‘Dependence’ independent variable underlying items

118

5.3

‘Communication’ independent variable underlying items

119

9.1

Courier problem-solving examples

200

9.2

In-app chat

201

vii

85

Editors Adrian Wilkinson is Professor at Griffith University, Australia and Visiting Professor at the University of Sheffield. He has been shortlisted by HR magazine for the award of HR (Most Influential International Thinker). Adrian has authored/co-authored/edited 30 books, 190 articles in academic journals and 100 book chapters. Recent books include The Sage Handbook of Human Resource Management (Sage, 2019), The Future of Work and Employment (Edward Elgar Publishing, 2020), The Handbook of Research on Employee Voice (Edward Elgar Publishing, 2020), Human Resource Management at Work (Kogan Page, 2020) and Contemporary Human Resource Management (Sage, 2021). He is a Fellow of the Chartered Institute of Personnel and Development in the UK and a Fellow of the Australian Human Resource Institute. He is an Academician (Fellow) of the Academy of Social Sciences in the UK as well as a Fellow of the Academy of Social Sciences in Australia. Tony Dundon is Professor of HRM and Employment Relations at the Department of Work & Employment Studies, Kemmy Business School, University of Limerick. He is a Visiting Professor at the Work and Equalities Institute (WEI), the University of Manchester, and Honorary Professor at St Andrews University, Scotland. He is a Fellow of the Academy of Social Sciences (FAcSS); a Fellow of the Chartered Institute of Personnel and Development (FCIPD); former Editor-in-Chief of the Human Resource Management Journal (HRMJ); and International Editorial Board Member for the journal Work Employment & Society (WES). His research looks at mechanisms of employee voice and social dialogue; employment regulation; trade union organising; work and technology; and debates about HRM and corporate performance. He has published in a range of journals including the British Journal of Management, Human Relations, British Journal of Industrial Relations and International Journal of Human Resource Management, among others. His books include Routledge Companion to Employment Relations (Routledge, 2018); A Very Short, Fairly Interesting and Reasonably Cheap Book about Employment Relations (Sage, 2017); Theories of Human Resources and Employment Relations (Edward Elgar Publishing, 2019); Power Politics and Influence at Work (MUP, 2020); Handbook of Research on Employee Voice, 2nd edn (Edward Elgar Publishing, 2020); and Human Resource Management: Cases and Texts, 6th edn (Sage, 2021), among others. viii

Editors

ix

Paula K. Mowbray is a Senior Lecturer in Human Resource Management at Griffith University, Australia. Her research interests include employee voice, participation and involvement, and line managers. Her research focuses on how employee voice is managed within diverse settings and contexts, such as manufacturing, universities and hospitals, as well as exploring the voice experiences of workers, such as ride-share drivers, casual workers and paramedics. She has published in journals such as International Journal of Management Reviews, British Journal of Management, Personnel Review and International Journal of Human Resource Management and has book chapters on employee voice in Employee Voice at Work (Springer, 2019), The Sage Handbook of Human Resource Management (Sage, 2019) and Handbook of Research on Employee Voice (Edward Elgar Publishing, 2020). Sarah Brooks is Lecturer in Organisational Behaviour at the University of Sheffield, UK. Prior to joining academia, Sarah was a behavioural change implementation specialist and operations manager working in the private sector. Her research interests build on her industry experience and include voice about well-being, voice about concerns of unethical behaviour and temporal processes of voice. She is also an ICF accredited coach and is currently exploring the ways in which coaching can serve as a route to voice. She has published in the International Journal of Human Resource Management and the Journal of Managerial Psychology. She also has a book chapter with Professor Adrian Wilkinson titled ‘Employee Voice as a Route to Wellbeing’ in the Handbook on Management and Employment Practices (Springer, 2022).

Contributors Alan Bogg, Professor of Labour Law, University of Bristol, UK. Jean-Michel Bonvin, Professor of Sociology and Socioeconomics at the University of Geneva, Switzerland. Nicola Cianferoni, Scientific Officer, Occupational Health Administration, State Secretariat for Economic Affairs SECO and Associate Researcher, Institute for Sociological Research University of Geneva, Switzerland. Markus Ellmer, TeamEcho, Linz, Austria. Thomas Gegenhuber, Assistant Professor Digital Transformation, Leuphana University of Lüneburg and Johannes Kepler University Linz, Austria. Debra Howcroft, Professor of Technology and Organisation, University of Manchester, UK. Anne Keegan, Full Professor, Department of Human Resource Management and Employment Relations, University College Dublin, Republic of Ireland. Vili Lehdonvirta, Professor of Economic Sociology and Digital Social Research, Oxford Internet Institute, University of Oxford, UK. Stefania Marino, Senior Lecturer in Employment Studies, University of Manchester, UK. Nick Martindale, Postdoctoral Research Fellow in Sociology, Nuffield College, University of Oxford, UK. Miguel Martínez Lucio, Professor of International HRM & Comparative Industrial Relations, University of Manchester, UK. Jeroen Meijerink, Associate Professor of Human Resource Management, University of Twente, Netherlands. Stephen Mustchin, Senior Lecturer in Employment Studies, University of Manchester, UK. Luca Perrig, PhD candidate, University of Geneva, Switzerland. Elke Schüßler, Professor of Organization Studies, Johannes Kepler University, Linz, Austria. Holly Smith, Research Associate, Work & Equalities Institute, University of Manchester, UK. Jim Stanford, Economist and Director of the Centre for Future Work, the Australia Institute. Andrew Stewart, John Bray Professor of Law, University of Adelaide, Australia. Laura Thäter, Research Associate, Hertie School, Berlin, Germany. x

Contributors

xi

Alex J. Wood, Lecturer in Human Resource Management and Future of Work, University of Bristol, UK.

PART I

Introduction

1. Missing voices? Integrating worker voice and social dialogue in the platform economy Adrian Wilkinson, Tony Dundon, Paula K. Mowbray and Sarah Brooks INTRODUCTION The gig economy can be defined as a broad range of labour transactions taking place through digital media, which relies on what have become known as gig workers (Kenney & Zysman, 2016; Schor & Attwood-Charles, 2017; Srnicek, 2017). Interest in the platform (or gig) economy has highlighted the importance of technology in shaping customer experiences, policy about the future of work transformation, and societal relationships. These transformations matter to many people and organisations globally: to companies operating in the gig economy who offer goods and services via a digital platform which acts as the intermediary between buyer and seller, to gig workers delivering the goods and services, to clients purchasing the goods and services through a digital platform, and to policy-makers and governments concerned with rights and the economic functioning of society (ILO, 2021). In fact, so prominent is the role of technology in the platform economy that it is common for a gig worker to never meet or speak to another human being in the enterprise for which they work. This book highlights the variation in gig work and the different types of digital labour platforms, to show there are significant societal issues that need to be grappled with to ensure that this growing group of workers is not disadvantaged. Freedom of association, voice and representation are often contextualised as “human rights”, set against transitional labour standards and UN sustainability goals. Yet how those who work through digital labour platforms can articulate their voice and express their rights is uncertain and remains contentious. A lack of voice opportunity may adversely affect workers’ dignity and well-being, whilst the objective of enhancing organisational effectiveness may be missed. Alongside, or separate to the human rights argument, is also 2

Integrating worker voice and social dialogue in the platform economy

3

the idea that effective voice benefits both the organisation and the employee, such as better performance, reduced turnover, enhanced individual well-being, and potentially longer-term sustainable people management outcomes with wider societal relevance (Boxall & Macky, 2014; Guest, 2017). However, such claims are typically drawn from studies of standard employment systems; analyses specifically in relation to gig economy and platform work performance have rarely, if ever, been developed. As Vallas and Schor (2020, p. 278) argue, the structure and design of the platform organisation and the way in which gig workers undertake their tasks, reduces the ability of the worker to “resist, elude, or challenge the rules and expectations that firms establish as conditions of participation”. Hence, greater consideration by platform organisations and regulatory bodies is needed in order to ensure that the voice of platform workers is not neglected or missing from organisational-level decisions, or from wider policy level choices about rights and regulations. In this book we focus specifically on the voices of those who work in the gig economy and, in particular, explore why such workers’ voices are missing and/or neglected in policy debates, and what can be done about it for the future (Wilkinson et al., 2021). We highlight a number of different perspectives from which voice in the platform economy can be considered, bringing to the fore some key areas for further research and questions that need to be answered. Finally, we advocate the benefits of drawing on analyses done at the macro, meso and micro levels to highlight ways in which the extant literature can be strengthened, while recognising that these forces can overlap across space and influence the forms of representative agency at other levels.

THE PLATFORM ECONOMY AND ITS RELATIONSHIP WITH TRADITIONAL WORK Gig work is not straightforward when compared to more traditional forms of work in which most voice and silence research has taken place. Firstly, gig work can be distinguished from more traditional work due to the presence of an intermediary, in the form of a provider using a digital platform that organises and allocates work tasks, and which facilitates commercial interactions between the worker and customer. Secondly, the platforms that serve as intermediaries vary considerably and can be distinguished by definition as being part of the “sharing economy” or “crowdworking” (also known as “cloudworking”). While platform work in the sharing economy can involve the use of digital platforms to sell or lease assets, such as Airbnb and Etsy, our focus in this book is on those individuals who sell their services and perform work tasks by using the platform which connects them with the end-user customer, referred to as crowdworking. The allocation of crowdwork can take place either through websites (e.g. Amazon Mechanical Turk, Fiverr) or

4

Missing voice?

apps (e.g. Uber, Deliveroo). Typically, work organised via a smart phone app suggests the task is restricted to a particular space and locality, referred to as “local” gig work (e.g. Deliveroo, Uber), whereas work organised by a computer suggests the task can be anywhere in the world, referred to as “remote” gig work (e.g. Amazon Mechanical Turk, Upwork, Fiverr) (see Howcroft & Bergvall-Kåreborn, 2019). Finally, the length of time that workers are matched with clients can range from short term to long term (e.g. Uber as a rider-sharing service to individual customers is immediate and a short, one-off transaction). However, other online platforms which source temporary agency labour (e.g. hosco.com or YoungOnes.com) provide more long-term labour services, either as direct recruitment platforms for individuals or as agencies for corporate clients, in areas such as cleaning or hospitality (Meijerink & Arets, 2021). Although online digital platforms and gig work are growing fast, the actual numbers engaged in gig employment may be relatively small and precise data is difficult to ascertain. Across Europe, gig work may account for around 7 per cent of the labour force (Pesole et al., 2018). Figures for the UK also vary, between almost five million gig workers in some studies (Huws & Joyce, 2016; Shenker, 2019), to lower estimates of around 2.8 million in another (Lepanjuuri, Wishart & Cornick, 2018). In the USA, estimates indicate that around 0.5 per cent of people were earning income from work via digital labour platforms, such as Uber and TaskRabbit (Farrell & Greig, 2016). Arguably, while numbers are relatively small compared to other economic sectors, the rate of growth is rapid and the use of digital gig work has huge potential to transform the world of work (Berg et al., 2018; Wilkinson & Barry, 2020).

THE PLATFORM ECONOMY AND EMPLOYMENT STATUS This book contributes to the growing concern and debates surrounding the relationship between individuals working with different platform providers (such as Uber, Deliveroo, Amazon Mechanical Turk, TaskRabbit) and the legal delegation of those workers as so-called “independent contractors” rather than “employees” (Healy, Nicholson & Pekarek, 2017). Of central importance therefore to understanding voice in the platform economy is a consideration of the employment status of those who work in the sector. While contested legally in some jurisdictions, platforms such as Uber and Deliveroo typically place workers outside the typical employer–employee arrangement so that workers are legally recognised as so-called “independent contractors”. While the platform providers claim this arrangement ensures greater workforce flexibility and an opportunity to make money (Rosenblat & Stark, 2016), workers face job precarity and employment insecurity as a result of their non-employee

Integrating worker voice and social dialogue in the platform economy

5

status (Bergvall-Kåreborn & Howcroft, 2013; Wood et al. 2019). To address the relationship between employment status and job precarity, government and international bodies have produced reviews and reports recommending the need to update employment laws to improve conditions (ILO, 2016, 2021). For example, the UK government commissioned a review of modern working practices, known as the Taylor Review, which recommended some change to the employment status of gig workers (Taylor Review, 2017), but others argue that it did not go far enough and left many gig workers without clear rights and limited access to representation (Bales, Bogg & Novitz, 2018). International agencies such as the International Labour Organization (ILO) and Organisation for Economic Co-operation & Development (OECD) have argued that gig workers need to be protected by adopting policies in platform-based labour markets so they align more with those that apply to traditional businesses (Schwellnus et al., 2019; see also Mowbray, Chapter 8). To this end, a ruling in the UK in 2021 stipulates that Uber drivers are “workers” and not “self-employed” because they are assigned ratings by customers and subject to discipline based on those ratings. Consequently, the court stipulated that Uber drivers should be entitled to minimum wage, holiday pay and pensions. Notably, the GMB trade union in the UK has bargained and secured a recognition agreement for Uber drivers, which provides for collective voice and representation rights, although such an arrangement has been a long, hard fought for battle through the courts (see Bogg, Chapter 2). While this improves conditions for UK-based Uber drivers, this is not something that has been applied globally by Uber (Barratt, Veen & Goods, 2021). Uncertainty regarding employment rights including sick pay, holidays, minimum wages and working conditions may link employment patterns in the gig economy with limited or inferior voice opportunity because independent contractors have little motivation or few mechanisms to voice (Healy, Nicholson & Pekarek, 2017; Berg et al., 2018). When considering the literature, there is a paucity of research on this expanding group of so-called independent contractors, both in relation to issues associated with their work, and more specifically their opportunity to speak up about things that concern them. Most of the research that has been conducted in relation to these independent contractors has focused primarily on collective action and self-organisation as a means to voice (Johnston & Land-Kazlauskas, 2018; Tassinari & Maccarrone, 2020). The main aim of voice is to “potentially influence organizational affairs relating to issues that affect people’s work, their interests, and the interests of managers and owners” (Wilkinson et al., 2020). Therefore, considering voice as a means to have a say both individually and/or collectively, formally and/or informally, suggests that a broader understanding is necessary as to how intermediary and digital platforms engage workers, in order to provide us with better insight into worker voice and social dialogue in

6

Missing voice?

emerging gig-economy sectors. As such, we argue that there are still voices at work which need to be heard and this book has been designed to explore those.

THE PLATFORM ECONOMY AND TRADITIONAL MODELS OF POWER The voice literature has identified that power differentials created by the organisational hierarchy underpin voice and silence between employers and employees in traditional workplaces (Morrison & Rothman, 2009). However, the platform economy based on new technologies, including robotics and digital platforms, has reconfigured work and management relationships with attendant implications for regulation, governance and voice (Thornthwaite et al., 2020). The ILO has identified five ways in which the use of digital platforms contributes to worker exploitation: the removal of free agency, reduced bargaining power and rights, domination, dependence, and the unfair allocation of risks and rewards (Choudary, 2018). This could suggest that power may shape voice in the platform economy in different ways. For example, the Internet as a source for activism has been highlighted as having the potential to create solidarities (Greene et al., 2003; Fitzgerald et al., 2012; Frangi et al., 2018) but may also generate counter-mobilisation and reduce effort put into voice, resulting in “clicktivism” (Upchurch & Grassman, 2016). Looking more deeply, worker exploitation and disempowerment are linked to a high level of information asymmetry which is a key feature within the sector: for example, the withholding of specific information concerning jobs, stipulations about the number of jobs workers must accept, reward structures associated with ratings workers must receive, and the encouragement of competition between workers (Vallas & Schor, 2020). The platform economy is not confined to traditional organisational boundaries and, as such, worker voice arrangements across organisational boundaries and international borders are increasingly more complex and uneven. As a result, there have been calls for improved corporate governance to take into account the potential for power asymmetries in new sectors of employment activity, such as the gig economy and globalised supply chain networks across different platforms (e.g. Uber, Deliveroo, professional freelancers). The importance of increased corporate governance can be illustrated by using the notion of a digital cage, which describes an organisation where workers find themselves disempowered because they are controlled by algorithms rather than managers, effectively removing any opportunity for voice (Vallas & Schor, 2020; Kougiannou & Mendonça, 2021). Such conditions have led to a large power imbalance in the platform economy which creates conditions that do not encourage voice (Kwon et al., 2016).

Integrating worker voice and social dialogue in the platform economy

7

THE PLATFORM ECONOMY AND HETEROGENEITY One challenge presented to researchers is that the platform economy is not homogeneous. Notably, platforms differ in the spatial (physical and digital) location of work, the way workers interact with platforms, and the technologies used to motivate, control, remunerate and organise work. Developments at work such as increasing insecurity due to crowdworking “in which digital platforms act as a form of ‘internalised offshoring’” (Findlay & Thompson, 2017, p. 132) will pose further challenges for voice; challenges which have yet to be theorised let alone researched empirically. Adding to this complexity is that in some cases, such as crowdworking tasks channelled via internet platforms, workers comprise highly skilled professionals located anywhere around the globe, in areas such as software programming, but other digital labour platforms may involve low paid and low skilled micro tasks, such as those offered through the like of Amazon Mechanical Turk (Bergvall-Kåreborn & Howcroft, 2014; Wilkinson et al., 2021). Other platform providers, such as Uber and Deliveroo, regulate and distribute work tasks through a digital application, although the nature of labour is executed in a specific (local) space, which has very different implications for voice and employment standards. For example, under these systems, voice may be mediated by the technology itself, with workers rarely, if ever, actually speaking directly with a human manager despite the geographical proximity (Inversi, Buckley & Dundon, 2017). Hence, voice often becomes a written exchange, with algorithmic, template, phone text or email responses serving as the primary or only voice mechanism (Rosenblat & Stark, 2016). The result is a dilution in the influence of these workers and their ability to engage in two-way conversations and express genuine concerns that could improve both their own outcomes and those of the organisations they work for. Taking recent events at Uber as an example, the lack of homogeneity is highlighted. In 2021, the GMB union in the UK formed an agreement with Uber that allows the union to represent Uber drivers (see https://​www​.gmb​.org​ .uk/​news/​uber​-and​-gmb​-strike​-historic​-union​-deal​-70000​-uk​-drivers). In other countries, Uber drivers have set up their own online groups to facilitate social dialogue and communication between themselves as an organising tactic. Interestingly, these workers are not only concerned about traditional Industrial Relations (IR) issues such as working conditions and pay, but are also sharing suggestions useful for the company, such as better placement of markers for airport pickups (Rosenblat, 2018; Walker, 2021). However, given that this voice is between drivers themselves rather than between the drivers and Uber, the platform provider, the extent to which these concerns reach managers at Uber is not transparent. Certainly, what has been demonstrated is that this

8

Missing voice?

type of peer-to-peer voice (Kalfa & Budd, 2020) via social media can convert to solidarity and the development of networks (Kougiannou & Mendonça, 2021), potentially improving voice influence and outcomes for these types of workers. However, as Heiland (2020) notes, it is difficult to generate trust and commitment in online communities, and worker voice based on this avenue alone is unlikely to be successful, with face-to-face communication more likely to generate positive outcomes for more significant issues. The use of union influence, strikes and spontaneous collective action through online platforms and social media have also been used by platform workers globally. Of note, the Independent Workers’ Union of Great Britain (IWGB) is the first organic, bottom-up trade union to be formed since the 19th century, specifically for riders in Deliveroo and to challenge the lack of universal employment rights (Wilkinson et al., 2021).

THE PLATFORM ECONOMY, FLEXIBILITY AND REGULATION It is important to understand the context of the platform economy to appreciate voice. For example, it is unsurprising that voice in the platform economy is mostly generated by poor working conditions, joint experiences of dependence on particular platforms, and solidarity between workers about issues on the same platform. However, variables specific to different countries may not only configure institutional work and voice arrangements in certain ways, but might also affect the effectiveness of informal and formal voice channels differently. For example, the relationship between and potential for informal communications, even when technologically mediated via an app, to build on and complement formal voices structures is untested across digital platforms, and an exploration of whether formal voice mechanisms shaped by digital technologies undermines informal social dialogue among actors is required. Most of the focus on motivation has been directed at the individual and associated with the reasons why these workers are attracted to platform work, with low barriers to entry for workers, the alleged attractiveness of flexible work and enhanced job autonomy as some of the main reasons for joining a platform (Hall & Krueger, 2018; Wood et al., 2019). Opportunities for social interaction and for some occupations, particularly knowledge workers, a validation of professional expertise, have also been cited as motivating factors (Rockman & Ballinger, 2017). However, researchers have noted there is often a “fictitious freedom” associated with gig work (Malhotra, 2020) and platform workers often accept the precarity of work in order to attain the benefits of flexibility. Questions arise as to whether platform workers are willing to forsake their rights, whether uncertain contract status or insecure pay counterbalance minimal rights, and

Integrating worker voice and social dialogue in the platform economy

9

whether long hours and unregulated working time and a lack of voice are other motivating factors.

TOWARDS A THREE-LEVEL ANALYTICAL MODEL FOR VOICE IN THE PLATFORM ECONOMY Unlike much extant research which looks at employees who have access to voice via corporate policy or trade union presence, we are looking at those who may not be employees and do not have easy access to voice structures. While the research on employee voice has generated important findings, much of it is done at the organisational level and emphasises the role of managers and different voice structures. As a result, research has tended to neglect employees and, importantly, much previous research on voice has neglected workers who are not legally defined as employees. Therefore, one of the primary aims of our book is to help contribute to such gaps on the missing voices of those not fully recognised as employees, and who work in the gig economy. We argue that there are still voices at work which need to be heard and research in this area is very important. Taking into account the variables which are likely to influence voice in the platform economy, namely, legal status of employment, heterogeneity of the intermediary platforms, lack of trade union presence and an absence of face-to-face dialogue with an authority figure, we propose that scholars should focus research at and across three levels, namely macro (societal), meso (organisational) and micro (worker), with the potential to examine forces at higher levels and assess possible influences on the ground (Wilkinson, Barry & Morrison, 2019). The chapters in this book explore voice across these three levels in the following ways. The Macro Level The macro level can be characterised by an exploration of regulatory frameworks within and across societies which determine organisational policy. In this book we include three chapters which highlight the importance of legislation, industrial relations, and temporal and spatial boundaries in shaping voice at the macro level of the platform economy. Highlighting the legal statutes which underpin the right to voice, Chapter 2 by Bogg brings to the fore a pressing need for transnational legislation and collective bargaining to adequately support voice in the platform economy. The prevalence of legal debates surrounding the employment status of gig workers is highlighted as one of the main obstacles blocking individuals working in the platform economy from achieving voice. Ranging from “employees” to “self-employed” and the “dependent self-employed” known as workers, each status provides access

10

Missing voice?

to a legal gateway offering a differing suite of statutory protections for gig workers, with employees receiving the most protection. Taking a broad macro regulatory approach, Chapter 3 by Stewart and Stanford comments that the growing individualisation of work, as driven by both technology and geography, represents a major societal problem for platform workers anywhere in the world to express their voice. Indeed, as they argue, the very concept of worker “voice” takes on a surreal nature when there is no one to listen. They suggest a need to develop systems through which platform workers can overcome their technologically and geographically mediated isolation to share common concerns and demand better treatment, whether from platform owners or regulators. They examine the use of regulation to facilitate either collective bargaining or other methods by which platform workers or their representatives can be given a meaningful say over work organisation, compensation and working conditions. Although much of the chapter is concerned with instrumental state regulation at the national, regional or sectoral level, they also examine types of co-regulation between platform businesses, worker representatives and/or the state that have emerged both nationally and even transnationally sometimes leading to codes of practice and other forms of “soft law”, rather than legally binding agreements. They list some of the most promising priorities for regulatory improvement to support the development of genuine and effective mechanisms of collective voice and representation as a possible avenue to provide a wider set of voice rights for marginalised and isolated workers. A further macro consideration is provided in Chapter 4 by Thäter and colleagues, who argue for sector- and societal-type policy recommendations, using four different lenses to show how conditions can be created which support agency for voice: worker-driven voice, platform-driven voice, union-driven voice and regulation-driven voice. They identify that most research has documented attempts to support formal collective and formal individual voice, but there has been very little exploration of informal individual voice, highlighting a gap which might yield important insights. Based on their analyses, they provide four policy recommendations designed to strengthen and embed voice in the platform economy: the creation of a digital space so that workers can speak informally, acting as a proxy for a physical workplace where none exists; a greater involvement by unions to support sustained efforts to augment worker voice; the adaptation of unions to work in the platform economy by embracing digital solutions to revolutionise attempts at organising and funding; and finally, clarity by the state on the employment status of workers in the platform economy to ensure fair, democratic debate.

Integrating worker voice and social dialogue in the platform economy

11

The Meso Level In contrast to the macro level, the meso level can be characterised by the variability in voice systems across different digital platform types and different labour market institutions, as well as attempts to organise in trade unions and across other collective civic society groups and network alliance spaces. In this book, we feature a discussion about conditions underpinning trade union renewal spaces for gig worker representation, individual and collective voice channels, formal and informal voice, and the role of algorithmic management. In Chapter 5, Wood and colleagues investigate the circumstances under which workers in the “remote” gig economy engage in three different voice channel activities: taking action against poor treatment by clients, showing interest in collective organisation to improve pay and working conditions, and supporting the advancement of state regulation. They identify that antagonism, dependence and solidarity are the three variables that underpin worker motivations for voice. Furthermore, despite being treated as a homogeneous group, the study shows that the voice behaviour of platform workers is shaped by very different circumstances, demonstrating an important heterogeneity across type and levels. An investigation of voice among remote gig workers is seen as revealing of voice dynamics in the wider platform economy, most closely resembling self-employment, where freelancers are matched with clients and both parties are free to choose who they contract with, the rate charged to do so, and whether other workers can be brought into a project. In contrast, local gig work perhaps resembles a different non-standard type of arrangement, as these gig workers often experience less discretion in terms of setting fees, nor can they choose their methods of work or even the clients they work for. To investigate the influences over different voice activities across multiple levels or segments of the gig economy, Wood and colleagues situate platform work within industrial relations concepts and highlight that this field should not be limited to legally contracted employees alone, but workers more generally who labour under these platform regimes. Focusing on the changing landscape of industrial relations in response to the growth of the platform economy, Chapter 6 by Martínez Lucio and colleagues highlights the growing fragmentation of representational spaces and new challenges to the way that workers negotiate with employers. They argue that focus has moved on to exploring the newer waves of mobilisation and representation that have emerged with a greater use of organising, mobilising and networked forms of representation. These developments parallel many of the debates we have seen in relation to the way migration has changed and brought forth new innovative and socially oriented forms of worker voice which are not merely part of a set of institutional readaptations but may signify a new system of decentred activism and collective action that represents a more direct partic-

12

Missing voice?

ipative form of worker voice. The authors outline how these disruptions are best seen as the outcomes of a decline in traditional and stable workplaces, the undermining of established spatial boundaries and relations between employers and employees, and the emergence of public space as a focus for worker representation and organising. The chapter notes how we are seeing a form of bargaining from a distance or at arm’s-length and a greater attention to more complex forms of communication between employers and workers, signalling variable meso-type forces and influences across networks and newer emergent collective and union alliances disrupting the spaces for mobilising actions. Furthermore, ruptures in terms of ownership patterns, consumption processes and the very nature of work are also disruptive for both traditional employers and digital platforms as new spaces and reference points within mobilising practices are carved out amongst actors within different spaces and locations. It also seems to suggest that the “locations” of worker voice and representation mechanisms are established at a variety of levels and are directed in terms of their demands towards a broader set of organisational actors and dynamics. Next, focusing on voice arrangements across different types of online labour platforms, Chapter 7 by Keegan and Meijerink reviews the extant literature on worker voice in the gig economy and identifies a range of platform-sponsored voice mechanisms which serve to both enable and disable voice. They also highlight that even where platforms enable voice, it is limited by providing channels that are mainly downward in nature, and largely focus on gathering information to improve how the platforms function. The authors see algorithmic management, at the heart of institutionally complex platform human resource management (HRM) practices, as having negative consequences for worker voice by disabling voice and leaving many gig workers unable to exercise voice, or too anxious to share grievances, express dissatisfaction, and have a say in the conditions that impact on them. They note that responses to this issue underpin new voice initiatives, some led by unions, even though the institutional context for voice in platform work is dynamic and uncertain. Signalling a meso direction, Keegan and Meijerink also ask if changing the legal status of employees will by default lead to better working conditions for all. Using an example of Uber and Deliveroo, it could be that changes result in caps being applied to maximum worker-levels, leaving many of the most vulnerable workers outside of the formal employment relationship and unable to continue working. The risks are that these workers will be forced back into informal and unregulated work through fragmentation of the market and sector, which is perhaps even more precarious and lower quality work with inferior rights than that offered by major platforms. In short, solving issues regarding workers’ rights and vulnerability in platform-enabled gig work must be seen as part of a larger struggle across firms, sectors and levels to tackle low paid, vulnerable work for marginalised workers.

Integrating worker voice and social dialogue in the platform economy

13

The Micro Level The micro level examines how workers experience voice in practice, drawing on informal channels and digital media platforms at the job level, and how alliances and networks build on union and other labour market activist groups on the ground (as also discussed by Martínez Lucio et al. in Chapter 6). Indeed, the Covid-19 pandemic has changed the nature and risks associated with some key platform jobs. For those working for the likes of Uber and Deliveroo during the pandemic, many of these workers effectively became essential front-line support staff, exposed to increased health and other safety hazards working through a global pandemic, with little or no personal protective equipment, while ensuring that food was delivered to people or providing essential transport/ride-sharing services (Inversi, Cefaliello and Dundon, 2020). In Chapter 8, Mowbray discusses some of the key issues associated with the voice of ride-share drivers specifically, by presenting the results of a study of Australian Uber drivers that was conducted during the first year of the Covid-19 pandemic. Mowbray’s research illustrates some of the lived experiences of these ride-share drivers when attempting to voice to their intermediary platform provider, Uber. In addition to this, Mowbray argues that significant legal developments in a number of countries concerning the employment status of Uber drivers, including the United Kingdom and California in the United States, could cause a ripple effect whereby Uber’s apparent reluctance to provide voice opportunities for its drivers may wane on the ground/work level. There are early indications that Uber may be creating new opportunities for its drivers to voice their opinions and concerns, even in other jurisdictions such as Australia, where despite it being a matter of debate, their employment status has not yet been changed, thereby signalling a potential micro-level counter-response by Uber as the platform provider (employer). An Integration of Macro, Meso and Micro Levels Highlighting the value of drawing on all three levels of analysis, we also feature one case study which examines the context that led to the emergence of social dialogue within a bike delivery platform provider in Switzerland. Chapter 9, by Cianferoni and colleagues, takes a broader view of voice in the platform economy by describing and analysing events that took place at macro, meso and micro levels to show how workers were able to raise their voices and successfully negotiate better working conditions. By focusing on capacity for voice at each level, they identify the precise conditions that allowed for this dialogue to emerge in the gig economy. At the macro level, they look at the broader legal and regulatory context in which gig platforms operate in Switzerland. They identify that the independent contractor status was the

14

Missing voice?

main source of concern which did not offer workers protection under labour law provisions, or access to social security. At the meso level, the trade union used the dispute at this delivery platform provider as a vehicle to engage with the government about the platform economy more widely. The chapter shows how workers communicated their grievances collectively to managers, initially using an unofficial works council and, later, a trade union. In the case of gig workers that had no legal recourse for individual dialogue, turning to trade unions to facilitate access to senior managers at the work unit level enhanced their confidence to voice their concerns. At this micro level, the relationship between gig workers and the field managers was mediated by minimalist in-app chats. Furthermore, the authors illustrate how the voice system was designed to capture organizational issues and exclude worker issues, indicating the extent to which organizational attitudes to voice can shape its perceived legitimacy amongst workers. However, even with the support of the trade union, the case study illustrates that the workers had recourse to very few tactics and found themselves disempowered. Nonetheless, and even though the lack of direct access to supervisors and the limited topics for voice caused fragmentation amongst the workers, the frustration created from disjointed and protracted conversations with managers led to worker solidarity, facilitating a process whereby workers collectively searched for a way to make their voices heard.

CONCLUSION Most of the research on employee voice has been focused on workers employed in traditional industries and under standard work arrangements (i.e. full-time employees) (Wilkinson et al., 2018). There is an opportunity for voice scholars across all disciplines to come together with the aim of creating voice opportunities for all workers including individuals who work in the platform economy (Wilkinson et al., 2020). Such non-traditional workers have been largely neglected in the voice literature. So, we need to ask how the motivators and outcomes of voice might be different for those in gig-economy jobs, working across platform types and levels, and interconnected across a triad of “worker”, “digital intermediary platform” and “end-user”. In this book we explore to some extent what this might mean for designing effective voice mechanisms for non-standard employment relationships and occupations. Questions and debates for future researchers may usefully ask what the appropriate structures of voice are for these workers in light of the barriers they face, such as legal contract status uncertainties, trade union mobilisations, business model transformation, and isolation from other co-workers on the same platform or across different platform types in the same market space. When considering the platform economy, it is natural for researchers to want to identify patterns and trends and compare them to more traditional

Integrating worker voice and social dialogue in the platform economy

15

workplaces. However, although there are many similarities in terms of task management, task allocation and fixed payment for services, there is a need to appreciate the differences in the gig economy which make traditional forms of voice difficult for platform workers. What is clear is that the growing supply and demand in the platform economy suggests that research in this area is a priority and offers the potential to inform future policy and practice.

REFERENCES Bales, K., Bogg, A. & Novitz, T. (2018). ‘Voice’ and ‘choice’ in modern working practices: problems with the Taylor Review. Industrial Law Journal, 47(1): 46–75. Barratt, T., Veen, A. & Goods, C. (2021). A new deal for Uber drivers in UK, but Australia’s ‘gig workers’ must wait. The Conversation. https://​theconversation​.com/​ a​-new​-deal​-for​-uber​-drivers​-in​-uk​-but​-australias​-gig​-workers​-must​-wait​-157597. Berg, J., Furrer, M., Harmon, E., Rani, U. & Silberman, M. S. (2018). Digital labour platforms and the future of work: towards decent work in the online world, International Labour Office, Geneva ILO. Bergvall-Kåreborn, B. & Howcroft, D. (2013). ‘The future’s bright, the future’s mobile’: a study of Apple and Google mobile application developers. Work, Employment and Society, 27(6): 964–81. Bergvall-Kåreborn, B. & Howcroft, D. (2014). Amazon Mechanical Turk and the commodification of labour. New Technology, Work and Employment, 29(3): 213–23. Boxall, P. & Macky, K. (2014). High-involvement work processes, work intensification and employee well-being. Work, Employment & Society, 28(6): 963–84. Choudary, S. (2018). The architecture of digital labour platforms: policy recommendations on platform design for worker well-being. ILO Future of Work Research Paper series. https://​www​.ilo​.org/​wcmsp5/​groups/​public/​-​-​-dgreports/​-​-​-cabinet/​ documents/​publication/​wcms​_630603​.pdf. Farrell, D. & Greig, F. (2016). Paychecks, paydays, and the online platform economy: big data on income volatility. JP Morgan & Chase Institute Report. https://​www​ .jpmorganchase​.com/​corporate/​institute/​document/​jpmc​-institute​-volatility​-2​-report​ .pdf. Findlay, P. & Thompson, P. (2017). Contemporary work: its meanings and demands. Journal of Industrial Relations, 59(2): 122–38. Fitzgerald, I., Hardy, J. & Lucio, M.M. (2012). The Internet, employment and Polish migrant workers: communication, activism and competition in the new organisational spaces. New Technology, Work and Employment, 27: 93–105. Frangi, L., Noh, S.-C. & Hebdon, R. (2018). A pacified labor? The transformation of labor conflict. In A. Wilkinson, T. Dundon, J. Donaghey & A. Colvin (eds), The Routledge Companion to Employment Relations, pp. 285–304. London: Routledge. Greene, A.M., Hogan, J. & Greco, M. (2003). E-collectivism and distributed discourse: new opportunities for trade union democracy. Industrial Relations Journal, 34: 282–90. Guest, D. (2017). Human resource management and employee well-being: towards a new analytic framework. Human Resource Management Journal, 27(1): 22–38. Hall, J.V. & Krueger, A.B. (2018). An analysis of the labor market for Uber’s driver-partners in the United States. Industrial and Labor Relations Review, 71(3), 705–32.

16

Missing voice?

Healy, J., Nicholson, D. & Pekarek, A. (2017). Should we take the gig economy seriously? Labour and Industry, 27(3): 232–48. Heiland, H. (2020). Workers’ voice in platform labour: an overview, Düsseldorf: Hans-Böckler Foundation. Howcroft, D. & Bergvall-Kåreborn, B. (2019). A typology of crowdwork platforms. Work, Employment and Society, 33(1): 21–38. Huws, U. & Joyce, S. (2016). Crowd working survey: size of the UK’s “gig economy” revealed for the first time. University of Hertfordshire. http://​englishbulletin​.adapt​.it/​ wp​-content/​uploads/​2016/​02/​crowd​-working​-surveypdf1​.pdf. ILO (2016). Non-standard employment around the world: understanding challenges, shaping prospects. Geneva: ILO. ILO (2021). World employment and social outlook: the role for digital labour platforms in transforming the world of work. Geneva: ILO. Inversi, C., Buckley, L.A. & Dundon, T. (2017). An analytical framework for employment regulation: investigating the regulatory space. Employee Relations, 39(3): 291–307. Inversi, C., Cefaliello, A. & Dundon, T. (2020). #HereToDeliver: valuing food delivery workers in the future. Policy@​Manchester Blog, 20 June. https://​blog​.policy​ .manchester​.ac​.uk/​health/​2020/​06/​heretodeliver​-valuing​-food​-delivery​-workers​-in​ -the​-future/​. Johnston, H. & Land-Kazlauskas, C. (2018). Organizing on-demand: representation, voice, and collective bargaining in the gig economy. Conditions of work and employment series, 94. International Labour Office, Geneva. https://​www​.ilo​.org/​ wcmsp5/​groups/​public/​-​-​-ed​_protect/​-​-​-protrav/​-​-​-travail/​documents/​publication/​ wcms​_624286​.pdf. Kalfa, S. & Budd, J. (2020). The future of employee voice. In A. Wilkinson, J. Donaghey, T. Dundon & R. Freeman (eds), The Handbook of Research on Employee Voice, 477–88. Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing. Kenney, M. & Zysman, J. (2016). The rise of the platform economy. Issues in Science and Technology, 32(3). https://​issues​.org/​rise​-platform​-economy​-big​-data​-work/​. Kougiannou, N.K. & Mendonça, P. (2021). Breaking the managerial silencing of worker voice in platform capitalism: the rise of a food courier network. British Journal of Management, 32(3): 744–59. Kwon, B., Farndale, E. & Park, J.G. (2016). Employee voice and work engagement: macro-, meso- and micro-level drivers of convergence? Human Resource Management Review, 26: 327–37. Lepanjuuri, K., Wishart, R. & Cornick, P. (2018). The characteristics of those in the gig economy. Department for Business, Energy & Industrial Strategy Report, London. https://​assets​.publishing​.service​.gov​.uk/​government/​uploads/​system/​uploads/​ attachment​_data/​file/​687553/​The​_characteristics​_of​_those​_in​_the​_gig​_economy​ .pdf. Malhotra, A. (2020). Making the one-sided gig economy really two-sided: implications for future of work. In S. Nambisan, K. Lyytinen & Y. Yoo (eds), Handbook of Digital Innovation, 228–49. Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing. Meijerink, J. & Arets, M. (2021). Online labor platforms versus temp agencies: what are the differences? Strategic HR Review. http://​doi​.org/​10​.1108/​SHR​-12​-2020​ -0098.

Integrating worker voice and social dialogue in the platform economy

17

Morrison, E.W. & Rothman, N.B. (2009). Silence and the dynamics of power. In J. Greenberg & M.S. Edwards (eds), Voice and Silence in Organizations, 175–202. Bingley: Emerald. Pesole, A., Urzí Brancati, M. C., Fernández-Macías, E., Biagi, F. & González Vázquez, I. (2018). Platform Workers in Europe. Joint Research Centre of the European Union, Luxemburg. http://​publications​.jrc​.ec​.europa​.eu/​repository/​bitstream/​JRC112157/​ jrc112157​_pubsy​_platform​_workers​_in​_europe​_science​_for​_policy​.pdf. Rockmann, K.W. & Ballinger, G.A. (2017). Intrinsic motivation and organizational identification among on-demand workers. Journal of Applied Psychology, 102(9): 1305–16. Rosenblat, A. (September 1, 2018). The network Uber drivers built. Fast Company. https://​www​.fastcompany​.com/​40501439/​the​-network​-uber​-drivers​-built. Rosenblat, A. & Stark, L. (2016). Algorithmic labor and information asymmetries: a case study of Uber’s drivers. International Journal of Communication, 10: 3758–84. Schor, J.B. & Attwood-Charles, W. (2017). The “sharing” economy: labor, inequality, and social connection on for-profit platforms. Sociology Compass, e12493. https://​ doi​.org/​10​.1111/​soc4​.12493. Schwellnus, C., Geva, A., Pak, M. & Veidel, R. (2019). Gig economy platforms. OECD Economics Department, Working Paper 1550, 1–33. Shenker, J. (August 31, 2019). Strike 2.0: how gig economy workers are using tech to fight back. The Guardian. https://​www​.theguardian​.com/​books/​2019/​aug/​31/​the​ -new​-resistance​-how​-gig​-economy​-workers​-are​-fighting​-back. Srnicek, N. (2017). The challenges of platform capitalism: understanding the logic of a new business model. Juncture, 23: 254–7. https://​doi​.org/​10​.1111/​newe​.12023. Tassinari, A. & Maccarrone, V. (2020). Riders on the storm: workplace solidarity among gig economy couriers in Italy and the UK. Work, Employment and Society, 34(3): 0950017019862954. Taylor Review (2017). The Taylor Review of Modern Working Practices, July. London: Department for Business, Energy and Industrial Strategy. Thornthwaite, L., MacMillan, C. & Barnes, A. (2020). The internet, the web and social media: the promise and practice of e-voice. In A. Wilkinson, J. Donaghey, T. Dundon & R.B. Freeman (eds), Handbook of Research on Employee Voice, 2nd edn, pp. 509–52. Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing. Upchurch, M. & Grassman, R. (2016). Striking with social media: the contested (online) terrain of workplace conflict. Organization, 23(5): 639–56. Vallas, S. & Schor, J.B. (2020). What do platforms do? Understanding the gig economy. Annual Review of Sociology, 46: 273–94. Walker, M. (2021). Peer-to-peer online voice as emergent collective action. Journal of Industrial Relations, 63(5): 777–97. https://​doi​.org/​10​.1177/​00221856211031940. Wilkinson, A. & Barry, M. (eds) (2020). The Future of Work and Employment. Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing. Wilkinson, A., Barry, M. & Morrison, E. (2019). Toward an integration of research on employee voice. Human Resource Management Review, 30(1). http://​doi​.org/​10​ .1016/​j​.hrmr​.2018​.12​.001. Wilkinson, A., Donaghey, J., Dundon, T. & Freeman, R.B. (eds) (2020). Handbook of Research on Employee Voice (2nd edn). Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing.

18

Missing voice?

Wilkinson, A., Gollan, P.J., Kalfa, S. & Xu, Y. (2018). Voices unheard: employee voice in the new century. International Journal of Human Resource Management, 29(5): 711–24. Wilkinson, A., Knoll, M., Mowbray, P. & Dundon, T. (2021). New trajectories in worker voice: integration and application to contemporary challenges in the organization of work. British Journal of Management, 32(3): 693–707. Wood, A.J., Graham, M., Lehdonvirta, V. & Hjorth, I. (2019). Good gig, bad gig: autonomy and algorithmic control in the global gig economy. Work, Employment and Society, 33(1): 56–75.

PART II

Macro considerations

2. The legal framework of worker voice in the platform economy Alan Bogg 1. INTRODUCTION This chapter will consider the legal mechanisms for worker voice in the platform economy. There has been significant scholarly, political and trade union attention focused on the problems of ‘gig work’. Rarely a day passes without another employment status case on gig work, or a newspaper report on a work stoppage, or systemic failures to pay minimum wage rates, or the death or serious illness of a frontline gig worker during the Covid-19 pandemic. Gig work is often taken to be the exemplar of work in the platform economy. There is no doubt that this mode of arranging work through platforms has placed significant strains on existing legal institutions and regulatory techniques of voice. In turn, the ingenuity of workers and trade unions has been a source of creative organizational strategies, new forms of collective resistance, and the forging of solidarities in the platform economy. Voice has not been silenced in the platform economy. Rather, workers and unions are exercising their agency through both conventional and novel channels of voice. As new forms of agency and voice emerge, legal norms must be reimagined to address the distinctive regulatory challenges of platform work. This is particularly important as platform technology spreads to new sectors and occupations. In conceptualizing voice and the law, the chapter will track the legal framing of worker voice, rather than a non-legal sociological understanding. The legal institutionalization of voice is based upon a foundation of individual statutory entitlements. These legal rights are allocated to individual workers and employees, and the substantive protections usually depend upon a nexus with formal trade union structures. These rights provide a substrate whereby individuals can exercise their voice, either singly or with others, protected from the most serious forms of anti-union victimization. This basic foundation supports more developed forms of voice institutionalization, for example through collective bargaining, strike action, and information and consultation mechanisms. While the general attitude of the law is neutrality towards different 20

The legal framework of worker voice in the platform economy

21

forms of labour association, closer scrutiny reveals a structural alignment with trade union voice, reflecting the historical development of British labour law. The next section identifies four legal dimensions of voice in the platform economy: casualization and precarity, the growth in self-employment, the spatial decentring of work, and the fragmentation of the bilateral employment model. These dimensions provide a critical contextualization of the operation of legal norms framing voice in the gig economy. Section 3 explores the issue of employment status in the gig economy. In legal terms, employment status provides the gateway into the statutory norms and institutions facilitating voice. It has been the pivotal battleground in gig economy litigation around the world, and it is often regard as the key to legal emancipation. Section 4 explores the dilemmas of juridification in the context of worker voice. Legal norms simultaneously empower and constrain autonomous worker resistance, channelling economic conflicts into institutions of ordered dispute resolution. Sometimes, exclusion from conventional legal mechanisms of voice provides opportunities for radical and agonistic resistance ‘beyond the law’. Alongside these agonistic strategies, unions have also used courts and litigation as a way of shaping the regulatory contexts of work. Litigation in the courts becomes a form of political voice, and a new site to challenge the neoliberal capture of the legislative process. Section 5 explores how existing legal norms and institutions might be reimagined, tailored to the distinctive needs and regulatory challenges of precarious work in the platform economy.

2.

LEGAL DIMENSIONS OF VOICE IN THE PLATFORM ECONOMY: AN OVERVIEW

To what does ‘platform’ refer to in the framing of gig work? According to Woodcock and Graham, the terminology ‘comes from its more traditional usage as a raised surface on which people can stand. In this context, the platform is a digital environment upon which other software can be run’ (Woodcock and Graham, 2020, 42). This provides a mechanism for what Prassl has described as ‘digital work intermediation’ (Prassl, 2018, 14). While there are a wide variety of business models encompassed by this general category, they are unified by the use of platform software to connect work-providers with customers or service-users. The technology provider usually presents itself as an intermediary in the background, providing contracting opportunities for the buying and selling of services (e.g. labour) as a commercial agent rather than as an employer. This is integral to the dominant business model of platform work, which is based upon the reduction of costs through the shedding of employer responsibilities by platform companies. That is why employment status is one of the central legal issues of worker voice in the gig economy. It represents a contested site where workers assert their normative agency and claim their

22

Missing voice?

legal rights against an employer as the responsible duty-bearer. In this chapter, the platform intermediary will be referred to as the employer because in most situations this is the legally accurate descriptor: the intermediary generally performs at least some of the functions of an ‘employer’ (Prassl, 2015). Following the typology of Woodcock and Graham, we can distinguish between ‘geographically tethered work’ and ‘cloudwork’ (Woodcock and Graham, 2020, 41). Typically, geographically tethered work is represented in the visible forms of labour we encounter daily, in private hire transportation, fast food and parcel deliveries, cleaning services, and home care. It is labour that necessitates physical presence at a particular time and place displaying many of the material features of traditional employment. By contrast, cloudwork does not have the same spatial and temporal constraints, and may occur in online freelancing in IT services, transcription, or data collection and processing (Woodcock and Graham, 2020, 55). The challenges for worker voice are particularly acute in the cloudwork context, with work-providers widely dispersed across global labour markets, working in different jurisdictions and time zones, in highly atomized settings. Work in the platform economy displays a rich variety. It spans many types of job and many sectors. The picture is dynamic, too, as employers outside the existing parameters of platform work experiment with new technologies in an unfolding process of ‘Uberization’. Woodcock and Graham cite research indicating that about 1.1 million people in the UK have undertaken work in the gig economy (Woodcock and Graham, 2020, 1). In the Global South, Heeks has estimated that there are approximately 30–40 million gig workers (Woodcock and Graham, 2020, 39, citing the work of Heeks, 2017). Citing the work of Guy Standing, Woodcock and Graham suggest that by 2025, ‘one-third of all labour transactions will be mediated by digital platforms’ (Woodcock and Graham, 2020, 1, citing the work of Standing, 2016). Nor is there a single type of ‘gig worker’ with homogeneous needs and preferences. It encompasses independent and high-earning IT professionals as well as migrants with precarious work status working for multiple employers on different apps doing food deliveries. Given this variation and complexity of circumstances in platform work and the position of different workers, is it possible to say anything general about law and voice in the platform economy? The scholarship indicates four legal dimensions of voice. The first dimension is casualization and the creation of precarious work arrangements through platform intermediation (Woodcock and Graham, 2020, 16–18). Usually, there are no guaranteed offers of work under the contractual arrangements in many gig work contexts. Workers are free to ‘log on’ to the app and accept work to suit their lifestyle preferences, and employers are free to tailor their offers of work to match fluctuations in economic activity. Translated into legal terminology, there is no ‘mutuality of obligation’ in the sense of ongoing

The legal framework of worker voice in the platform economy

23

reciprocal offers to offer and accept work. This is significant because mutuality of obligation is still a key legal element in some employment status tests, particularly for ‘employee’ status in the UK. New technologies can be used to enhance the ‘digital legibility’ of active work time (Woodcock and Graham, 2020, 24), which facilitates what McCann has described as a ‘fractured’ model of working time regulation (McCann, 2021). This means that payment is restricted to periods of active work (rather than payment for the willingness and availability to undertake work, as under more conventional contractual employment arrangements). While the growth of casualization and precarity in labour markets preceded the platform economy, often in the form of ‘zero hours’ contracts, this process has intensified and accelerated through enabling platform technologies. It has led to a kind of hyper-casualization in the gig economy. This has intensified inequalities of power in the platform economy, as platform companies acquire great monopoly powers through the strategic pursuit of network effects in product markets. Gig companies have invested huge subsidies to attract new customers to their networks, augmenting their market power in the process (Prassl, 2018, 22–4). In this way, power disparities are even starker in platform work compared with non-platform employment relationships. This has strengthened the ability of platform companies to promote deregulatory strategies in the political arena, through corporate political voice driving deregulatory legislation. It has also intensified the dominating power of companies in designing and enforcing one-sided written contractual arrangements that exclude employment status for workers, while also ratcheting control over the labour process through surveillance technology. These conditions of precarity obstruct the ability of workers and unions to organize by strengthening the employer’s capacities to suppress autonomous collective mobilization. This is because individual security of employment provides a shield for workers to share their experiences with each other and to begin to articulate dissent. These informal and individualized forms of voice are often a vital stepping stone to more developed forms of collectivism. The second legal dimension is the growth in ‘sham’ or ‘bogus’ self-employment. Contractual arrangements often present a picture of gig workers as independent entrepreneurs, with platforms simply reducing transaction costs for these entrepreneurs to hook up with new customers (Prassl, 2018, 52–5). This has been described as ‘regulatory arbitrage’, in that large corporations wield their market power to organize corporate structures and contractual documentation to opt out of labour market protections for workers (Prassl, 2018, 20–22). The reality is that many platforms create and sustain structures of extensive direct and indirect control, fixing payment rates, specifying delivery routes, utilizing systems of algorithmic discipline based upon customer ratings, and collecting data (De Stefano, 2016). Despite formal

24

Missing voice?

contractual appearances, gig workers experience high levels of dependence, subordination and vulnerability to contractual exploitation. Around the world, there has been extensive litigation to establish ‘employment status’ as a basic legal stepping stone to the protection of voice in legal frameworks. There is an emerging judicial consensus across these jurisdictions that many of those working in geographically tethered work should be regarded as employees or workers, and within the scope of protective labour law statutes (Moyer-Lee and Kountouris, 2021). The third legal dimension is the spatial decentring of work (Woodcock and Graham, 2020, 63–4). Increasingly, there is no physical workplace, and management and discipline is depersonalized as it operates through algorithmic processes. Sometimes this correlates with a high degree of ‘spatial autonomy’, for example for workers engaged in cloudwork; by contrast, the physical locations of those engaged in geographically tethered work are often tracked and subject to a high degree of control (Woodcock and Graham, 2020, 63). This dispersal creates significant challenges in the forging of solidarities and coordinating collective action. For those in geographically tethered work, such as food delivery riders, there is no physical workplace and management is often mediated through an app. The absence of a physical workplace, where workers come together, can create barriers to shared dialogue and geographically situated solidarities. Yet workers have also found new ways to communicate with each other, through WhatsApp groups, web-based forums, and congregating in public spaces proximate to restaurants (Woodcock and Graham, 2020, 94–5). This is an example where creative engagement can transform an obstacle into a resource. Traditionally, the physical workplace augmented the employer’s power through private property and the right to exclude union organizers and recruitment activities. The disappearance of the physical workplace has had the effect of eliminating private property as a locus of sovereign control. Where the ‘workplace’ is the street or in the cloud, the employer loses an important tool of control. These spaces may also be less vulnerable to the intrusive surveillance that can operate in factories, offices and other forms of physical private property. The shifting nature of private property is testament to the continuing importance of strategic choices of unions and workers: new forms of domination may also carry an Achilles’ heel that reveal new opportunities for resistance and evasion of employer control. The fourth legal dimension is the fragmentation of the bilateral bond between a single employer and a single worker in a stable and continuing employment relationship (Prassl, 2015). This fragmentation is occurring in a number of ways. Recent litigation against Uber in the United Kingdom and in Canada revealed the organizational complexity of a transnational corporate entity that operated in multiple jurisdictions. In Uber v Aslam,1 for example, there were different Uber entities: Uber London Limited (ULL) (which held the operating

The legal framework of worker voice in the platform economy

25

licence for London private hire), Uber BV based in the Netherlands which owned the Uber app, and Uber Britannia Ltd which was responsible for private hire services outside the London region. While the written contractual nexus between ULL and the individual drivers was tenuous, the Supreme Court treated ULL as the ‘employer’ in the work arrangement. This fragmentation of the employing entity, and the allocation of distinct functions across different organizational units, makes it more difficult to identify a single unitary employer. In the Canadian case of Uber v Heller, class action litigation was focused on the validity of a private arbitration clause in the employment contract (which was held to be unenforceable on the facts).2 This required a driver based in Canada to pursue a grievance through mediation and arbitration in the Netherlands, and this included the payment of significant upfront filing and administration fees. This arrangement represented a form of juridical mobility, capital flight through choice of law, using contracts to escape from national regulatory systems through mandatory private arbitration clauses (see also Mowbray, Chapter 8). Workers themselves are increasingly engaged in multiple overlapping relationships with different employers. Workers may even have multiple apps open simultaneously in order to maximize the opportunities for paid work, which could mean that they have many employers at a single point in time.3 This presents significant challenges to regulators in allocating employer responsibilities fairly across different entities. Multi-apping can also be used as a way for putative employers to deny employment status, providing evidence that the worker is marketing her services and so in business on her own account. Increasingly, legal voice mechanisms must operate in a world that is significantly more complex than the post-war model of stable bargaining units, based upon a single and stable bilateral relation between a worker and a unitary employer. Given these four legal dimensions, gig work creates particular challenges for the norms and institutions of trade union organization, collective bargaining and formalized strike action. Sometimes, the more established trade unions have not engaged with gig workers, perhaps given the ambiguities around employment status and the high turnover/low wages of those workers. Organizing the hyper-precarious may bring uncertain gains in terms of stable recruitment and membership dues. This has led to the emergence of grassroots ‘indie’ unions such as the Independent Workers’ Union of Great Britain (IWGB) or the App Drivers and Couriers Union (ADCU). Collective action and resistance has also taken more spontaneous forms rooted in more informal social networks (see also Martínez Lucio et al., Chapter 6). The extreme insecurity experienced by many gig workers has accentuated the importance of basic individual legal protections as a minimum legal foundation for all forms of voice, both individual and collective. For example, basic legal protections

26

Missing voice?

from unfair dismissal and arbitrary treatment at work are integral to worker empowerment, even though we might not think of these as legal aspects of ‘freedom of association’ (Bogg and Estlund, 2014). This could have far greater importance for the voices of gig workers than statutory union recognition machinery and other more conventional legal supports for trade union voice. The courts have also been an important location of voice for gig workers. There has been significant litigation around employment status, as a gateway into legal rights relating to dismissal, wages, working time and discrimination. It is also integral to access legal voice institutions such as individual freedom of association protections, legal support for recognition for collective bargaining purposes, and mechanisms such as workplace consultation. As we shall see, unions such as the IWGB have also used judicial review and strategic litigation to reshape regulatory frameworks, for example securing the extension of health and safety protections to gig workers. This provides a way of securing improved terms and conditions, on a generalized basis, in circumstances where conventional bargaining power might be lacking to support effective collective bargaining. We must also attend to the conflictual politics of worker voice. As the Canadian case of Heller demonstrates, platform companies are often constituted by labyrinthine and complex corporate structures that extend across many national boundaries. To match the transnational scale of platform work and platform companies, solidarities must also be developed across jurisdictions as a basis for transnational collective action. As Woodcock and Graham have argued, neoliberal political ideologies have enabled a permissive legal framework that has empowered the capital freedoms of platform companies (Woodcock and Graham, 2020, 33–6). The contestation of neoliberal ideologies in the public sphere, securing worker-protective legislation and resisting deregulation, is a key site for worker voice in achieving structural changes to the distribution of power in labour markets. This political contestation must occur in both national and transnational norm-setting entities (for example, the European Union or the International Labour Organization (ILO)). Theories of worker solidarity must also recognize and address the reality of conflicting interests and identities across gig workers (Davies, 2012). These conflicts may be magnified where work hierarchies reflect and constitute racial and gender inequalities in the labour market and civil society (Woodcock and Graham, 2020, 28–30). The global scale of platform work may also create the conditions for conflicts of interest between groups of workers situated in different geographical regions. The era of homogeneous worker interests has certainly been interred by the platform economy, if it ever existed at all.

The legal framework of worker voice in the platform economy

3.

27

EMPLOYMENT STATUS AND WORKER VOICE

Legal and sociological scholarship on worker voice usually identifies the prevalence of self-employment amongst gig workers as one of the main legal obstacles impeding ‘voice’ in the platform economy (Prassl, 2018, 42–50). This is because ‘employee’ or ‘worker’ status provides the legal gateway into a suite of individual statutory protections and legal institutions of collective voice. This is often contrasted with ‘self-employment’, which is taken to fall outside the scope of statutory employment protection and is instead regulated through commercial law and the general law of contract. From a legal perspective, however, the distinction between ‘worker’ and ‘self-employed’ is not as clear cut as it might appear to the non-lawyer. In the UK system, for example, there is not a single ‘binary divide’ between ‘employees’ (enjoying statutory protections) and the self-employed (excluded from statutory protections). The UK system recognizes an ‘intermediate’ category of employment status, described as ‘worker’ status, and this includes some self-employed within its scope.4 There are a range of comparative legal approaches, with some systems now utilizing an intermediate category while others maintain the traditional binary divide. A brief explanation of the basic distinctions between these different legal categories is useful, taking the UK system as an example. ‘Employee’ is a common law category based upon the ‘contract of employment’. Historically, the courts developed strict criteria for defining this category, based upon elements such as control and subordination, the requirement of personal work, ‘mutuality of obligation’, and a wider consideration of the overall relationship. These strict criteria excluded many individuals who displayed similar features of subordination and dependence to employees but failed to meet the high legal threshold of employee status. These were often individuals in non-standard forms of work, such as casual or temporary work. This pattern of exclusion was dysfunctional because these features of precariousness operated to negate employment status and the resulting statutory protections for employees. In response to this dysfunction, the UK legal system has relied increasingly upon extended employment status categories for certain fundamental statutory entitlements such as the national minimum wage, working time protections, and anti-discrimination rights. The ‘worker’ category is an extended statutory category which includes a wider range of personal work relations otherwise excluded from the narrower ‘employee’ category. The main type of ‘worker’ (known as the ‘limb (b) worker’) is defined in s. 230 (3) (b) of the Employment Rights Act 1996 as ‘any other contract, whether express or implied and (if it is express) whether oral or in writing, whereby

28

Missing voice?

the individual undertakes to do or perform personally any work or services for another party to the contract whose status is not by virtue of the contract that of a client or customer of any profession or business undertaking carried on by the individual’. There are similar extensions in the Equality Act 2010 and Trade Union and Labour Relations (Consolidation) Act 1992 which, while having different statutory definitions, are now regarded as having wide and inclusive boundaries co-extensive with ‘limb (b) worker’. Some self-employed workers are included within these intermediate categories, especially where their work arrangements display the vulnerabilities to exploitation that justify basic statutory protections. For example, where the self-employed worker is subordinate to an employer, or where the worker has no effective influence over the contractual terms, or is integrated into another’s business, she will likely meet the lower threshold for worker status. The courts have adopted a ‘purposive’ approach to characterization of the relationship, ensuring that the relevant statutory rights are applied where the weaker party is in need of – and therefore an intended beneficiary of – the statutory protections. In the recent landmark case of Uber, this led the Supreme Court to treat drivers as ‘workers’ for the purposes of their statutory claims, and to focus on the economic substance of the employment model rather than the formal and elaborate contractual documentation drafted by Uber’s lawyers. As Lord Leggatt explained in Uber, UK law on employment status adopts a tripartite structure: The effect of these definitions … is that employment law distinguishes between three types of people: those employed under a contract of employment; those self-employed people who are in business on their own account and undertake work for their clients or customers; and an intermediate class of workers who are self-employed but who provide their services as part of a profession or business undertaking carried on by someone else. Some statutory rights, such as the right not to be unfairly dismissed, are limited to those employed under a contract of employment; but other rights, including those claimed in these proceedings, apply to all “workers”.5

While the identification of a ‘limb (b) worker’ involves a consideration of many of the elements relevant to the ‘employee’ category (for example, subordination, control, dependence, personal work), ‘limb (b) worker’ should be understood as a wider and more inclusive category encompassing a broader group that includes the dependent self-employed. In the words of the Employment Appeal Tribunal in Byrne Bros (Formwork) Ltd v Baird, the threshold of this intermediate category should be easier to meet and involves a ‘lowering of the passmark’ for putative workers.6 For this reason, and as exemplified in Uber itself, the ‘worker’ category provides an important gateway to legal protections for some self-employed workers in the platform economy. This depends upon

The legal framework of worker voice in the platform economy

29

the legal allocation of voice rights, and the extent to which those rights are extended to ‘workers’ and not restricted to ‘employees’. It is important, therefore, to map the legal rights relevant to worker voice onto the different employment status categories. The more extensively that these rights are extended to ‘workers’, the more likely they will be accessible to some of those who are ‘self-employed’ in the gig economy, and in need of statutory protections. To this end, we should distinguish the following rights: (i) the right to form a trade union; (ii) the right to engage in trade union organizational activity; (iii) the right to recognition for the purposes of collective bargaining; (iv) the right to strike; (v) the right to consultation over workplace decision-making. Starting with the right to form a trade union, there is a basic liberty for individuals to form associations and to act collectively within the constraints of the general law. Such associations can provide mutual support to members, provide legal advice and share information on the enforcement of legal rights, and provide a conduit for political action to seek changes in the legal framework. There are many examples of this informal coordination and spontaneous resistance in the platform work context. The law generally does nothing to prevent this basic form of association unless it causes a civil wrong. It may encompass loose and informal aggregations of workers (Routh, 2016). To constitute a special kind of association in law, a ‘trade union’, there is a specific statutory definition in s. 1 of the Trade Union and Labour Relations (Consolidation) Act (TULRCA) 1992. This defines a ‘trade union’ as an association ‘which consists wholly or mainly of workers of one or more descriptions and whose principal purposes include the regulation of relations between workers of that description or those descriptions and employers or employers’ associations’. In turn, ‘worker’ is defined in s. 296 (1) TULRCA as anyone who works ‘under a contract of employment, or (b) under any other contract whereby he undertakes to do or perform personally any work or services for another party to the contract who is not a professional client of his’. This bears the same wide meaning accorded to ‘limb (b)’ worker, and hence it includes some categories of self-employed worker within it. Where an association does not consist ‘wholly or mainly’ of ‘workers’ so defined, it cannot be ‘listed’ as a trade union by the Certification Officer. While this does not prevent individuals from forming an association, it does mean that the association cannot secure the benefits of ‘listing’ as a ‘trade union’, which include applying for a certificate of ‘independence’. The effect of this exclusion from listing is to restrict the association’s access to some important statutory rights in the legislation, specifically the statutory recognition machinery in TULRCA. So while non-trade union associations may enjoy a liberty to associate (in the sense that they are not prohibited from forming and acting collectively), they and their

30

Missing voice?

members do not necessarily have a right to associate that is correlative to some important legal duties on employers. The importance of worker status is reflected in (ii) the right to engage in trade union organizational activity. The ‘right to organize’ is based upon a perimeter of individual rights that protect workers from anti-union victimization, correlative to duties on employers not to coerce or victimize workers on a range of trade union grounds. The structure of legal protection for freedom of association under TULRCA is focused on ‘workers’ and ‘employees’. ‘Workers’ are protected from being subject to a ‘detriment’ (section 146) and financial inducements by their employer (section 145A) where the sole or main purpose is to prevent or deter them from being (or not being) trade union members, use of union services, or taking part in the activities of an independent trade union at an appropriate time. There are also two statutory rights restricted to ‘employees’ working under a ‘contract of employment’. Refusal of employment (section 137) on the grounds of trade union membership is restricted to employees. Furthermore, and in common with general protections from unfair dismissal, dismissal (section 152) in relation to trade union membership, use of union services, and taking part in the activities of an independent trade union at an appropriate time, is also restricted to ‘employees’. Where the association does not meet the definition of a ‘trade union’, this excludes the basic statutory protections in sections 137, 145A, 146 and 152. Individual claimants must themselves meet the definition of ‘worker’ or ‘employee’, depending upon which right is being claimed. Without these protections, gig workers would be exposed to the risk of serious forms of anti-union victimization by employers.7 This will have a significant chilling effect on even the most basic forms of collective organization and voice. In relation to (iii), the right to recognition for the purposes of collective bargaining, there are three levels of recognition in UK law, corresponding to the extent of legal support. The most extensive legal protections are provided by the statutory recognition procedure in Schedule A1 of TULRCA. This gives a listed trade union with a certificate of independence a right to recognition for collective bargaining where it has demonstrated majority support in an appropriate ‘bargaining unit’. Even where a union meets the s. 1 definition, and is listed and certified as independent, the union’s recognition claim must be on behalf of a group of ‘workers’ in the ‘bargaining unit’. This was a significant exclusionary barrier in a high profile gig economy recognition claim, where the individual riders at Deliveroo failed to meet the statutory threshold of worker status under s. 296 (even though the IWGB counted as a ‘trade union’).8 There is a second tier of recognition, which we can describe as legally supported voluntary recognition, where an independent trade union is accorded ‘voluntary’ recognition by an employer. In this situation, the volun-

The legal framework of worker voice in the platform economy

31

tary recognition of an independent trade may arise out of the ‘shadow effect’ of a potential Schedule A1 claim. It is further supported by important legal rights. Thus, the union has a right to information for collective bargaining purposes under s. 181 TULRCA. Time off for trade union duties is restricted to officials of recognized ‘independent trade unions’ (s. 168), and time off for trade union activities is restricted to members of recognized ‘independent trade unions’ (s. 170). There is also an important right in s. 145B which restricts the making of ‘offers’ to members of an ‘independent trade union’ that is recognized or seeking recognition, where the ‘sole or main purpose’ is that terms of employment will no longer be set by collective agreement. These all provide important indirect legal supports to the ‘voluntary’ collective bargaining process. Finally, a very basic form of voluntary recognition for collective bargaining is possible where the association is not an ‘independent trade union’. This arrangement is purely voluntary, and it attracts no positive legal supports under TULRCA. It is simply recognition that is permitted under the general law, and it is entirely dependent upon the consent of the employer. In terms of (iv), the right to strike in the UK is protected through a legal framework of immunities from common law liability. Where organizers of strike action commit certain listed economic torts, it is possible to claim a ‘trade dispute defence’ under TULRCA 1992. The scope of this defence is limited to ‘a dispute between workers and their employer’ (section 244 TULRCA 1992). If those engaged in gig work are not ‘workers’, their dispute with the employer cannot be a ‘trade dispute’. This would mean that the organizers of strike action would be exposed to common law liabilities in tort. The main form of tort liability in strike action is the economic tort of inducing breach of contract. There are also continuing uncertainties around the torts of conspiracy to injure and the parameters of ‘unlawful means’ sufficient to leave trade unions exposed to injunctions, uncertain common law liabilities, and expensive litigation. Individual strikers may also be unprotected from dismissal. The most robust dismissal protection – ‘protected industrial action’ – is restricted to ‘employees’. For the strike action to count as ‘protected’, the organizers of the strike must not be liable in tort ‘by virtue of s. 219’. In turn, s. 219 requires a ‘trade dispute’ which, as we have seen, must be a ‘dispute between workers and their employer’. Finally, the various consultation rights (v) related to collective redundancies, transfers of undertakings, national level information and consultation, and European works councils, are currently restricted to ‘employees’ in the UK. In this respect, these EU-based consultation rights have a narrower personal scope than most domestic trade union rights, which mostly apply to the wider group of ‘workers’. This account of the legal framework demonstrates the continuing importance of employment status to worker voice in the gig economy. While the allocation

32

Missing voice?

of many of these statutory rights to workers ensures that some self-employed individuals will be legally protected, the recent litigation in Deliveroo reveals ongoing problems in the gig economy. The IWGB did qualify as an ‘independent trade union’ in these circumstances, because most of its members were workers. The Central Arbitration Committee (CAC) nevertheless held that there was a valid and unfettered ‘substitution clause’ in the contract, which meant that the Riders were not ‘workers’ for the purposes of the specific recognition claim. A ‘substitution clause’ means that the individual can designate a ‘substitute’ to do the work. The effect of a wide substitution clause may be to negate employment status, because all the categories depend upon the work being undertaken ‘personally’ under the contract. In Deliveroo, this meant that the statutory recognition procedure in Schedule A1 of TULRCA was inapplicable to them since it is restricted to ‘workers’. The failure to find ‘worker’ status also potentially means that the other legal protections in TULRCA are unavailable, with the result that Riders are unprotected from basic forms of anti-union victimization. Following the landmark victory in Uber, the legal arguments now point decisively in favour of ‘worker’ status for Deliveroo Riders. The ongoing appeal in Deliveroo is focused on the narrow legal point that Article 11 of the European Convention on Human Rights, which protects trade union rights such as the right to collective bargaining, requires ‘worker’ to be interpreted widely in accordance with the criteria in ILO Recommendation No. 198 on Employment Relationship.9 Under Part II of the Recommendation, the determination of an ‘employment relationship’ should ‘be guided primarily by the facts relating to the performance of work … notwithstanding how the relationship is characterized in any contrary arrangement, contractual or otherwise, that may have been agreed between the parties.’ It also recommends ‘allowing a broad range of means for determining the existence of an employment relationship’; and where there is a ‘relevant indicator’ present in the factual arrangements there should be a ‘legal presumption that an employment relationship exists’. These relevant indicators ‘might include: (a) the fact that the work: is carried out according to the instructions and under the control of another party; involves the integration of the worker in the organization of the enterprise; is performed solely or mainly for the benefit of another person; must be carried out personally by the worker … involves the provision of tools, materials and machinery by the party requesting the work.’ In Deliveroo the existence of an apparently unfettered substitution clause was treated as fatal to worker status. This is incompatible with the ILO Recommendation. Its effect was to characterize the Riders as self-employed despite the existence of other indicative factors supporting an ‘employment relationship’. The personal performance of work is simply one indicative factor, to be considered as part of a wider assessment of the contractual

The legal framework of worker voice in the platform economy

33

party’s entrepreneurial independence and the corresponding need for statutory protection. Uber now emphasizes that the scope of statutory rights should be construed ‘purposively’. In Deliveroo, the Riders were presented with written terms on a take-it-or-leave-it basis, and had no effective opportunity to negotiate the contracts. The Riders displayed all the features of vulnerability to contractual exploitation. Since the purpose of Schedule A1 is to implement the right to collective bargaining through a specific statutory scheme, as a response to individual inequality of bargaining power, ‘worker’ should be construed inclusively so that it is aligned with ‘employment relationship’ under Article 11. Surprisingly, the recent Court of Appeal judgment in Deliveroo upheld the earlier determination of the CAC and High Court, treating the domestic law on substitution clauses as compatible with the range of possible approaches permitted by the ILO Recommendation. This may be a situation where the narrow focus on Article 11 was unhelpful, because the Court of Appeal did not have an opportunity to consider the more general significance of the shift in domestic law brought about by Uber. The strong purposive approach in Uber is likely to herald an approach that is less in thrall to substitution clauses, where the focus is on the substantive issue of whether statutory protections are justified by the economic situation of the individual workers. There seems little doubt that the Riders in Deliveroo are vulnerable to contractual exploitation, even despite the existence of a wide substitution clause. On the Uber approach, they would be treated as included within the statutory category of worker. To conclude, it is important not to fetishize ‘worker’ status. This worker litigation represents important legal progress, but it does not resolve all of the difficulties faced by gig workers. The question of ‘employee’ status for many gig workers has been untested at the highest appellate level. Crucially, unfair dismissal protection (which is currently confined to ‘employees’) provides an important general constraint on arbitrary treatment, and it provides a context of secure employment within which a ‘culture of contestation’ can develop (Bogg and Estlund, 2014). While the Uber decision was concerned with ‘limb (b) worker’, the statutory purposive approach should also apply to ‘employee’ claims. It seems very likely that many gig workers would also meet the employee threshold on this ‘purposive’ approach, with protection then depending upon whether there is sufficient ‘continuity’ of two years’ continuous employment. This provides a vital foundation against general arbitrary treatment, and a crucial form of empowerment for individual voice and contestation. This also provides a necessary foundation for other forms of voice, including collective voice through trade unions. We should also recognize the growth of ‘flexible scheduling’, which has been described as a new ‘despotism’ in the workplace (Wood, 2020). This flexible scheduling is a feature of many gig work contexts. This highlights the general importance of stable employment as a precondition of voice. Wood

34

Missing voice?

explores the potential for arbitrary treatment in the distribution of working time, and the gratitude and loyalty that results in a ‘gift’ of scheduling that is sensitive to the worker’s needs. These ‘gifts’ provide a powerful disciplinary tool for managers. It may also engender a pliable and ingratiating attitude in the ‘beneficiary’ of the ‘schedule gift’. The general regularization of scheduling, for example by regulating or even prohibiting ‘zero hours’ contracts, provides another legal technique for empowering workers. A right to stable and predictable working hours would also provide an effective counter to the domination latent in management-led flexible scheduling of working time. It is an important reminder that in evaluating the effectiveness of legal regimes of voice, we need to examine the full range of legal rights available to workers. Effectiveness depends upon an effectively enforced general floor of fair and just working conditions, as much as it depends upon legal rights more narrowly concerned with voice and trade union organization. The distinctive features of gig work also have important implications for the configuration of freedom of association rights. Gig workers are often under no obligation to switch on the app and be available for work, and the employer is under no obligation to provide work. Such arrangements are described as ‘casual’ or ‘zero hours’ contracts (which are by no means confined to gig work contexts). In these settings, the refusal to offer further work can be a powerful retaliatory measure against trade union activity. The chilling effects of these refusals of employment are just as coercive as classic ‘dismissals’ for trade union reasons, particularly in circumstances of casual work. The statutory provision regulating ‘access to employment’, s. 137, is restricted to refusals of employment on grounds of ‘union membership’. It thus excludes trade union activities and use of union services (but the courts have interpreted ‘union membership’ purposively and broadly to ensure compliance with Article 11 of the European Convention on Human Rights). More importantly, the right may only be claimed by ‘employees’, not ‘workers’. Those engaged in casual work have often found it difficult to meet the legal test for ‘employee’, which provided part of the justification for the intermediate worker category. In ‘casual’ gig work contexts, therefore, s. 137 is likely to be a central element in the structure of legal protection. The current restriction of the right to employees is very problematic.

4.

PLATFORM WORKERS FIGHT BACK

Employment status has been a significant obstacle in accessing legal protections for voice. Nevertheless, workers and unions have continued to exercise their voice and agency. Gig work has been a site where relatively small unions, such as the IWGB and the ADCU, have had highly visible impacts on the business practices of large well-resourced gig employers. These public

The legal framework of worker voice in the platform economy

35

victories, such as the litigation on ‘limb (b) worker’ status in Uber, can have a significant mobilizing effect on other forms of voice such as collective bargaining, striking, political lobbying, and media campaigns to expose and shame exploitative working practices. For example, the recent initial public offering (IPO) of shares in Deliveroo coincided with a media expose of poor working practices and high-profile strike action. This led to one of the worst IPOs in the history of the London Stock Exchange, with investors reluctant to invest in the company given the adverse publicity generated by a highly effective media campaign against Deliveroo. Often, resistance has been based upon spontaneous protest and strike action, fostered through informal networks of communication such as WhatsApp groups, and developed outside formal trade union structures. As Woodcock and Graham point out, ‘this resistance is happening within structurally difficult conditions, often in grey areas of legality, or even taking place illegally’ (Woodcock and Graham, 2020, 108). This liminal legality can also provide important openings for workers and trade unions to exercise their agency. Let’s consider two examples. First, the legal status of strike action by gig workers, and secondly the use of strategic litigation by unions such as the IWGB, as a technique of resistance and a voice channel for political activism in the courts. Starting with strike action, some forms of gig work (such as food delivery) provide important examples of strong solidarity and visible coordinated withdrawals of labour. Does the employment situation of gig workers facilitate or impede effective strike action? This is an area of significant legal complexity, and a basic explanation of the law on strikes is warranted. At common law, strikes are usually unlawful both for individual strikers and for those that organize the strike action. Most strikes involve a fundamental breach of contract by the individual striker because there is a continuing obligation to be available for work under the contract. In organizing a strike, the organizer has usually committed a civil wrong, a tort, in inducing this breach of contract. This exposes the strike organizer to tort liabilities and remedies, including damages and injunctive relief by the employer. In the UK, the collective legality of the strike is dependent upon a statutory exclusion of tort liability for the strike organizer. This is conferred in limited circumstances, where there is a ‘trade dispute’ and where various substantive and procedural requirements have been met, such as a lawful ballot and sufficient notice. Crucially, and as we have seen, the ‘trade dispute’ defence is restricted to disputes between employers and their workers. This body of statutory law is highly complex and very restrictive. It warrants emphasis, however, that this statutory framework is only necessary where there is an underlying common law liability for the strike organizer. If there is no tort, and hence no civil wrong, there is no need for its statutory exclusion.

36

Missing voice?

Paradoxically, the precarious nature of the gig employment model may sometimes provide gig workers with a wider freedom to withdraw their labour, both individually and in coordination with others, than ‘standard’ workers. Where there is no obligation to accept offers of work, as in a casual arrangement, the individual decision to not switch on the app does not constitute a breach of contract. This is because there is no contractual obligation to be available for work. Where there is no breach of contract, there is unlikely to be tort liability for the strike organizer where this has been coordinated as a collective action. The absence of tort liability removes the need for the ‘trade dispute’ defence and its tight substantive and procedural qualifications. In fact, this pattern is reflected in the early period of trade union development in the USA, where trade unions supported ‘at will’ contracts terminable with little or no notice as a basis for avoiding common law liabilities in strike disputes (Jacoby, 1982). This provides a vivid demonstration of how casualization can be exploited by trade unions to avoid a highly restrictive statutory regime of strike law based upon a ‘standard employment’ model of contracting. The contractual vulnerability can be flipped into a potent weapon of resistance. The second area in which unions operating in the gig economy have made a significant impact is through strategic litigation and ‘legal mobilization’. With unions like the IWGB, legal victories on employment status have had a galvanizing effect on union mobilization, leading to widespread media coverage, political leverage in seeking legislative and regulatory changes to employment conditions, and providing extra momentum to recognition campaigns for collective bargaining rights. Some of this litigation has originated in individualized disputes enforcing individual statutory claims, which have been resisted by employers on the basis that the claimants lack employment status. Such disputes have generated wider effects beyond the success of the individual statutory claim itself, which may ‘act as a springboard for collective bargaining or may influence policy in the political sphere’ (Ford and Novitz, 2020, 272). In the UK context, employment status litigation has sometimes been akin to public interest litigation, using an individual statutory claim as the vehicle. As an example, Ford and Novitz point to the success of the GMB union in securing a recognition agreement with the delivery firm Hermes, which they attribute to the successful strategic litigation around employment status at the firm (Ford and Novitz, 2020, 271). The IWGB has also supported employment status litigation in the gig economy and other non-standard employment contexts.10 Litigation has provided an important form of leverage for smaller unions, in circumstances where these grassroots organizations have a small membership base and limited resources. In many ways, ‘employment status’ has transformed into a symbolic political device for articulating a critique of work insecurity, precariousness, extreme social and economic inequalities, and political dis-

The legal framework of worker voice in the platform economy

37

empowerment. The remarkable media and political attention generated by the Supreme Court judgment in Uber demonstrates how ‘employment status’ bears a collective and political meaning that transcends the individual nature of the dispute itself. It is as much a political as a legal identity, whereby the claimant demands respect and recognition as a bearer of rights (Bogg, 2020). This litigation around employment status, although rooted in private disputes between workers and employers, has become emblematic of the wider struggle for justice. The IWGB has also been active in using judicial review and public law to secure the health and safety and income security of precarious workers in the gig economy. This public law litigation can be understood as a form of ‘political bargaining’ which is particularly important in those sectors in which the IWGB has been most active, such as private hire transportation. For example, Dubal has traced the growth of precarious work in the San Francisco taxi economy (Dubal, 2017). This was enabled by a breakdown in the model of municipal regulation of transportation which allowed platforms such as Uber to enter the market. In what she describes as a ‘political bargain’ in the San Francisco context, ‘the Chauffeurs’ Union shaped business models, set prices, and effected public policy, establishing strong municipal regulation of a once unregulated industry’ (Dubal, 2017, 79). The restriction of operating licences under this model was pivotal to the maintenance of high labour standards. The erosion of the ‘political bargain’ left a regulatory vacuum which allowed platform-based companies to drive down labour standards. The maintenance of strong regulatory regimes at the local or municipal level is supported by the provision of social welfare provision at the national level, through a universal floor of labour standards and social welfare entitlements. In the context of platform work, this litigation reveals how unions have exercised political voice and influence in creative ways. Judicial review provides a mechanism whereby unions can ensure that their arguments are listened to and addressed by public regulators. Where the judicial review is successful, the impacts can be systemic, improving the welfare of large numbers of workers. This avoids some of the weaknesses of a collective bargaining strategy focused on single employers and smaller bargaining units. Here are three recent examples of political voice through judicial review. The first example is concerned with the regulation of taxis and private hire vehicles in London. In R (Independent Workers’ Union of Great Britain) v London Mayor,11 the IWGB sought judicial review against the Mayor of London following the modification of the ‘congestion charge’ that applied in the ‘central congestion zone’ in London. This involved a change to the scheme of exemption, which had applied to taxis and private hire vehicles. Given the significant growth in private hire vehicles operating in the London area, the exemption was removed for private hire vehicles. This change was introduced to reduce traffic congestion in central London, while maintaining

38

Missing voice?

service provision for disabled service users (taxis were required to have special adaptations to facilitate disabled access). The congestion charge would have a significant impact on the livelihoods of many private hire drivers, the vast majority of whom lived in the most deprived areas of London. There were also very significant racial disparities between taxi drivers and private hire drivers: 94 per cent of private hire drivers were black, Asian and minority ethnic (BAME) compared with only 12 per cent of taxi drivers. Given the disparate impact on different racial groups, the IWGB argued that the change constituted indirect race discrimination under the Equality Act 2010. The Court of Appeal concluded that the scheme was a proportionate means of achieving a legitimate aim. Consequently, it did not constitute unlawful indirect race discrimination. The case reveals the complexities involved in recasting ‘political bargains’ in local regulatory schemes. In essence, this case involved conflicts between different groups of workers providing services in passenger transportation, against a backdrop of racialized disparities in the local labour market. The IWGB’s interests were aligned with private hire employers like Addison Lee, in supporting the economic viability of private hire services. The mechanism of judicial review provided the IWGB with a ‘voice’ in respect of the congestion charge, in circumstances where it had less opportunity for political influence than the privileged incumbent group of (mostly white) taxi drivers. While the direct legal challenge failed, the Court of Appeal stressed that the racial disparities were so stark that it was necessary to subject the policy change to intensive review. This form of legal accountability, based in the intensive judicial scrutiny of public decision-making, could have indirect effects on the future policy development. Though the IWGB did not have a direct voice in shaping regulatory decisions, it was able to participate through court action. The second example, R (Adiatu and another) v H M Treasury,12 was concerned with the national implementation of job support measures in response to the economic impact of the Coronavirus pandemic. This arose out of the Coronavirus Job Retention Scheme (CJRS), which provided financial payments to employers in respect of employees who were ‘furloughed’ as a result of the situation brought about by Covid. The CJRS was restricted to those individuals who received their pay through PAYE. While this might include some intermediate limb (b) workers, many individuals in the limb (b) worker category would be paid on a self-employed basis. This group of non-PAYE limb (b) workers were excluded from the CJRS. While there was also a parallel scheme for self-employed individuals, the qualification thresholds for the self-employed scheme meant that some limb (b) workers would be left without a safety net. The IWGB challenged the exclusion of limb (b) workers from the CJRS, and the government’s decision not to raise the level of Statutory Sick Pay (SSP)

The legal framework of worker voice in the platform economy

39

or remove the lower earnings limit to qualify for SSP. The legal challenge failed. The High Court emphasized the extraordinary political context of the decision-making, which required the rapid implementation of policy in circumstances of emergency. On that basis, the exclusion of some self-employed workers from the CJRS was subject to a wide margin of appreciation and it could not be said to be ‘manifestly without reasonable foundation’. As with the congestion charge case, it is important to evaluate the effectiveness of the IWGB’s intervention beyond the direct outcome in the case itself. This was a high-profile intervention that ensured the political visibility of precarious workers and the absence of a safety net for workers affected by the pandemic. It placed a spotlight on issues of insecurity and precarity, particularly among frontline workers exposed to significant health risks. These issues have been central to the public discourse about post-pandemic economic recovery. Litigation can trigger a wider political dialogue, even where the case itself fails on narrower legal grounds. The final example, R (on the application of The Independent Workers’ Union of Great Britain) v The Secretary of the State for Work and Pensions and the Secretary of State for Business, Energy and Industrial Strategy,13 was a judicial review challenge to the restriction of certain EU-based health and safety rights (most notably, the provision of personal protective equipment and the ‘right to refuse’ unsafe work) to ‘employees’. Unlike the preceding two examples, this was a successful public law challenge that led to the extension of statutory protections to include limb (b) workers.14 It also represented a distinctive ‘public’ form of employment status litigation. Much employment status litigation in the gig economy is focused on the application of existing legal categories to new contractual forms of work. These conventional employment status cases are often litigated in order to achieve wider systemic effects beyond the parties themselves, for example by developing more inclusive legal tests through legal precedents. This judicial review was more radical in challenging the statutory restriction of fundamental health and safety rights to the narrower category of ‘employee’. It prompted important systemic changes as statutory coverage was extended to include limb (b) workers. According to Chamberlain J., the relevant EU Directives should be construed as broadly and inclusively as possible, particularly in view of the fundamental nature of the health and safety rights guaranteed in the EU Charter of Fundamental Rights. The exclusion of limb (b) workers from the scope of statutory coverage, many of them working in circumstances of significant vulnerability during the pandemic, was not compatible with that broad and inclusive approach. This provides a strong example of the direct political voice that can be achieved through judicial review, where this leads to specific changes to the general statutory framework. It is difficult to imagine that similar effects could have been achieved by the IWGB

40

Missing voice?

through ordinary political lobbying. Given the urgency of the situation, and the ongoing exposure of limb (b) workers to the risk of infection, it was also necessary to secure a rapid solution to the legislative exclusion.

5.

COLLECTIVE BARGAINING IN THE PLATFORM ECONOMY: LEGAL OPPORTUNITIES AND LEGAL OBSTACLES

Historically, the traditional model of collective voice was collective bargaining channelled through independent trade unions. As we have seen, there are two main legal mechanisms for collective bargaining in the UK system, voluntary and statutory (Bogg, 2009). To what extent does the existing legal model of collective bargaining provide effective support to collective voice in the platform economy? Apart from the exclusionary effects of employment status, there are three main problems. First, platform work involves the fragmentation of different employer functions amongst different intermediaries in complex corporate networks. This was a particular feature in the Uber litigation on worker status, for example. The use of subcontracting of work can also be a particular difficulty in contexts where there are high levels of self-employment (Cranford et al., 2005). The Schedule A1 model of single employer bargaining lacks the flexibility to adjust to the fragmentary nature of these contractual arrangements. The IWGB has sought to use litigation to extend the scope of Schedule A1 in situations of ‘contracting out’, where a university contracted out its security and support services. The argument that the university was the de facto employer was rejected by the CAC, and the judicial review challenge was unsuccessful.15 The English courts have emphasized that the specific design of collective bargaining machinery is a political matter on which the democratically elected legislature should be accorded a wide margin of appreciation by the courts. For this reason, the strategy of human rights litigation is very unlikely to be successful to secure a legal model of multi-entity bargaining. Human rights litigation will usually lead to incremental adjustments to the existing statutory model, reflecting the primacy of democratic arguments in restricting the judicial function. The selection of a single employer model by the legislature represents a reasonable legislative determination and the courts are very unlikely to disrupt it. Any changes to that scheme must occur through political change. New statutory models tailored to the particularities of platform work should experiment with models of sectoral and multi-employer bargaining, particularly important when workers increasingly work for multiple platform employers. Secondly, there are emerging difficulties in conflicts between established unions and newly emerging unions seeking to organize and represent workers

The legal framework of worker voice in the platform economy

41

in the platform economy (see also Wood et al., Chapter 5; and Martínez Lucio et al., Chapter 6). Organizations such as the IWGB have sometimes clashed with incumbent unions. These conflicts may be difficult to resolve through voluntary trade union channels because unions like the IWGB are not affiliated to the Trades Union Congress. In R (on the Application of the Independent Workers Union of Great Britain) v Secretary State for Business, Energy and Industrial Strategy, the Court of Appeal rejected the IWGB’s Article 11 challenge to the Schedule A1 procedure.16 This procedure requires a statutory application to be rejected where there is an existing voluntary recognition agreement with an independent union (Bogg and Dukes, 2017). In this case, the IWGB was seeking to displace an existing voluntary recognition between the employer and UNISON, on the basis that it was more representative of the sub-group of workers it had identified for the purposes of its own preferred bargaining unit. Schedule A1’s priority rule for voluntary recognition reflects a legislative choice to promote stable bargaining arrangements, to accord primacy to voluntary recognition over statutory recognition, and to avoid fragmented bargaining units. The Court of Appeal took the view that these were all political matters falling within the scope of reasonable legislative choice and determination. Any future statutory reform of union recognition should address the problem of representativeness of bargaining agents. In platform economy contexts, the alternative model of members-only collective bargaining would better enable the diverse identities and interests of workers to be represented more effectively. This would also avoid some of the difficulties of achieving majority support in larger bargaining units where platform workers are highly dispersed and work in atomized settings. On the other hand, the fragmentation involved in a members-only bargaining system might also militate against the unity and coordination needed to achieve strong collective organization. Finally, the restricted range of bargaining subjects in Schedule A1, limited to pay, hours and holidays, does not capture the emerging areas of importance for joint regulation in platform work. For example, De Stefano has argued for the centrality of collective bargaining in shaping and directing the interactions between technology and the governance of work, using the terminology of ‘negotiating the algorithm’ (De Stefano, 2018). The use and implementation of new technologies should be subject to joint workplace regulation, including the use of algorithms as disciplinary tools, privacy and monitoring of employees, and the management and ownership of personal data collected in the workplace. These subjects of bargaining are fundamental to the dignity of workers in the platform economy, who are likely to be particularly exposed to the impact of technology on the managerial prerogatives. There have also been very significant instances of collective action in response to sexual harassment and gender inequality in the workplace, for example as with the worldwide protests by employees of Google. Legal frameworks for collective bargaining

42

Missing voice?

must be flexible enough to anticipate those subjects which are of most concern to workers’ dignity and respect. While pay, hours and holidays will remain core concerns for all workers, bargaining duties in the platform economy must keep pace with new agendas as determined by the workers themselves. So far, we have considered the role of existing legal frameworks on collective bargaining, and the extent to which these may need to be modified so that they are sensitive to the specific context of work in the platform economy. We should finally consider the clash between competition law and collective agreements in situations where workers are categorized as self-employed. In this situation, the application of competition law prohibitions to collective bargaining in the platform economy is actively impeding trade union activity by treating it as an illegal anti-competitive activity. The seminal European law case on competition law and collective bargaining is Albany International BV v Stichting Bedrijfspensioenfonds Textielindustrie.17 It concerned a supplementary pension scheme negotiated in a sectoral collective agreement. At the request of the social partners, the Minister for Social Affairs and Employment decreed that the scheme required compulsory affiliation for all relevant workers in the textile industry. Albany sought an exemption from the compulsory scheme because it had entered into a supplementary pension scheme with an insurance company at a time when the sectoral scheme provided for less generous benefits. The exemption was refused. Albany challenged this on the basis that the social partners’ request to make the fund compulsory constituted an agreement between undertakings that was contrary to competition law provisions in Article 85 (1) of the Treaty. According to the Court, ‘agreements concluded in the context of collective negotiations between management and labour’18 directed at the improvement of ‘conditions of work and employment’19 were outside the scope of Article 85 (1). The agreement in Albany was in the form of a collective agreement concluded by organizations representing employers and workers, and it contributed directly to the improvement of the working condition of remuneration. It followed from this that the specific agreement did not fall within the scope of Article 85 (1) in view of its nature and purpose.20 The Albany case and its exclusion of the competition law provision were vital in creating a protected space for collective bargaining. Albany was confirmed and extended in FNV Kunsten Informatie en Media v Staat der Nederlanden (FNV Kunsten).21 FNV Kunsten was concerned with the scope of the Albany immunity, and whether it extended to collective agreements regulating the minimum fees of both employed and self-employed substitute musicians who were substituting for members of orchestras. The Court reaffirmed the limits of the Albany exclusion, and its restriction to collective agreements negotiated between management and labour on behalf of employees. The Court proposed a modest extension to Albany to encompass the ‘false

The legal framework of worker voice in the platform economy

43

self-employed’. A service provider would cease to be an ‘undertaking’ (and hence within the scope of competition law) where he ‘does not determine independently his own conduct on the market, but is entirely dependent on his principal, because he does not bear any of the financial or commercial risks arising out of the latter’s activity and operates as an auxiliary within the principal’s undertaking.’22 The court was required to scrutinize the economic substance of the work contract regardless of its legal classification in domestic law, to determine whether the service provider acts under the direction of his employer as regards, in particular, his freedom to choose the time, place and content of his work (see judgment in Allonby, EU:​C:​ 2004:​18, paragraph 72), does not share in the employer’s commercial risks (judgment in Agegate, C‑3/87, EU:​C:​1989:​650, paragraph 36), and, for the duration of that relationship, forms an integral part of that employer’s undertaking, so forming an economic unit with that undertaking (see judgment in Becu and Others, C‑22/98, EU:​C:​1999:​419, paragraph  26).23

The European Commission is currently engaged in consultation on reforming the law on collective bargaining for the self-employed, exploring the potential for excluding competition law for the solo self-employed. It is also consulting on the legal regulation of platform work, and the scope for developing EU-level labour standards in platform work. Legal changes to competition law will be critical in ensuring that the law does not obstruct collective voice in the platform economy, given the numbers of self-employed working in platform contexts. While some of those will count as ‘false self-employed’ within the definition provided in FNV Kunsten, many platform workers (particularly those working for multiple platform employers) may not (Countouris and De Stefano, 2019). They will nevertheless be working in circumstances of contractual inequality, and in need of the protections afforded by collective bargaining (Countouris and De Stefano, 2021).

6.

THE FUTURE OF VOICE AND THE LAW IN THE PLATFORM ECONOMY: TOWARDS STRUCTURAL EMANCIPATION

The platform economy is unlikely to diminish in significance and, indeed, is very likely to become more central to the organization of work. What does this mean for the future of voice and the law in the platform economy? One issue is ensuring that grassroots organizing and confrontational worker-led tactics develop into forms of voice that are enduring, institutionalized, and able to address the deep structural power imbalances in the platform economy. Platform employers deploy vast wealth and power, and their pursuit of network effects has enabled them to develop significant monopoly power

44

Missing voice?

in many labour markets. While workers and grassroots organizations have secured some remarkable successes through strikes and protest action, often in the shadow of legality, these victories may be Pyrrhic when the underlying structures of economic domination are left intact. We should also be clear-sighted about the risks that institutionalization of industrial conflict can often go hand in hand with strategies of pacification of labour unrest. Legal regulation is closely aligned with the maintenance of orderly dispute resolution and the deradicalization of economic conflict. Nevertheless, workers and trade unions may calculate that the trade-offs are worthwhile if they are better able to affect the structural determinants of precarious work in the platform economy. This is more likely to occur through sectoral or national level bargaining, and coordinated transnational organizing, than through more localized bargaining disputes. While spontaneous protest can be very effective in challenging discrete injustices, it may be less well-suited to addressing systemic or structural injustice. Another issue is the need for substantive scrutiny of voice arrangements implemented by platform employers. Following the recent Uber judgment in the UK, Uber entered into a voluntary recognition arrangement with a large TUC-affiliated union, the GMB. It did not recognize the IWGB and the ADCU. Controversially, this recognition agreement appears to exclude pay from the scope of negotiation.24 There have been similar issues in New York state, where Uber proposed sectoral bargaining arrangements which enabled Uber to retain significant unilateral control over the designation of its union bargaining partner.25 In the UK, there is a deep attachment to ‘industrial autonomy’ and a corresponding reluctance for courts and legislators to scrutinize the substantive quality of recognition and collective bargaining processes. Historically, this was regarded as an illegitimate form of state encroachment on the freedom of groups in civil society to formulate and pursue their own purposes (Bogg, 2009, chapter 1). Recent developments in the platform economy may support a more statist vision of legally regulated voice. Legal standards should be developed and applied to enable courts and labour boards to scrutinize the representativeness and substantive effectiveness of voice arrangements, and to intervene and displace those arrangements in circumstances where there are alternative bargaining representatives that are more representative. This should be guided by consideration of what workers themselves want, rather than the preferences of employers or more established union bureaucracies. Ultimately, it is vital to examine the structures of political participation and worker voice in the democratic polity. Platform employers have been actively engaged in shaping the law-making process, and this has created favourable regulatory conditions for the development of casualized employment models. In California, for example, platform employers secured an exemption for app-based workers from basic labour standards through a legislative enact-

The legal framework of worker voice in the platform economy

45

ment, ‘Prop 22’.26 If we are serious about genuine emancipation in labour markets, democratic agency and democratic exclusion must be central to debates about the legal future of voices at work. Whose voices are heard and whose are silenced in the legislative chambers and the corridors of political power? Ultimately, it is through the translation of popular sovereignty into democratic legislation that the platform economy will be subjected to democratic oversight and control. The legal future of worker voice will depend upon civic activism in the political sphere as much as it depends upon labour activism in the workplace.

NOTES 1. 2. 3.

[2021] UKSC 5. [2020] SCC 16. See the perceptive discussion of multi-apping in a recent Australian gig work case: Diego Franco v Deliveroo Australia Pty Ltd [2021] FWC 2818, [117]–[118]. 4. The inclusion of some self-employed workers who meet the statutory definition of ‘worker’ was recognized by the English Court of Appeal in Hospital Medical Group Ltd v Westwood [2012] EWCA Civ 1005. This case concerned the employment status of a self-employed doctor who provided his services to patients through a private clinic. Uber (n 1) [38]. 5. [2002] ICR 667. 6. 7. See https://​www​.ft​.com/​content/​b0743801​-4f64​-3b80​-844d​-cd5108f3b434. R (The IWGB) v Central Arbitration Committee [2018] EWHC 3342 (Admin). 8. See now the Court of Appeal judgment at [2021] EWCA Civ 952. The leading case is Sindicatul Pastoral Cel Bun v Romania (Pastoral Cel Bun) 9. (2014) 58 EHRR 10. 10. Gascoigne v Addison Lee [2018] ICR 1826. 11. [2020] EWCA Civ 1046. 12. [2020] EWHC 1554 (Admin). 13. [2020] EWHC 3050 (Admin). I should declare that I acted as an academic consultant on a pro bono basis in this case. 14. The Employment Rights Act 1996 (Protection from Discrimination in Health and Safety Cases) (Amendment) Order 2021. 15. R (on the application of the IWGB) v Secretary of State for Business, Energy and Industrial Strategy and Others [2021] EWCA Civ 260. 16. See note 15. 17. Case C-67/96 Albany International BV v Stichting Bedrijfspensioenfonds Textielindustrie [1999] ECR I-5751 (Albany). 18. [60]. 19. [59]. 20. [64]. 21. Case C-413/13 FNV Kunsten Informatie en Media v Staat der Nederlanden [2014] EUECJ C-413/13 (FNV Kunsten). 22. [33]. 23. [36].

46

Missing voice?

24. 25.

See https://​www​.adcu​.org​.uk/​news​-posts/​uber​-gmb​-union. See https://​onlabor​.org/​lawmakers​-should​-oppose​-new​-yorks​-uber​-bill​-workers​ -need​-real​-sectoral​-bargaining​-not​-company​-unionism/​. See https://​www​.theguardian​.com/​us​-news/​2021/​feb/​18/​uber​-lyft​-doordash​ -prop​-22​-drivers​-california.

26.

REFERENCES Bogg, Alan, 2009, The Democratic Aspects of Trade Union Recognition (Oxford: Hart). Bogg, Alan, 2020, ‘Employment Status in the Social Democratic Constitution: Law and Politics’, in Alan Bogg, Jacob Rowbottom and Alison Young (eds), The Constitution of Social Democracy (Oxford: Hart). Bogg, Alan and Ruth Dukes, 2017, ‘Article 11 ECHR and the Right to Collective Bargaining: Pharmacists’ Defence Association Union v Boots Management Services Ltd’, Industrial Law Journal 46(4): 543–65. Bogg, Alan and Cynthia Estlund, 2014, ‘Freedom of Association and the Right to Contest: Getting Back to Basics’, in Alan Bogg and Tonia Novitz (eds), Voices at Work: Continuity and Change in the Common Law World (Oxford: Oxford University Press). Countouris, Nicola and Valerio De Stefano, 2019, New Trade Union Strategies for New Forms of Employment (Brussels: ETUI). Countouris, Nicola and Valerio De Stefano, 2021, ‘The Labour Law Framework: Self-Employed and their Right to Bargain Collectively’, in Bernd Waas and Christina Hießl (eds), Collective Bargaining for Self-Employed Workers in Europe: Approaches to Reconcile Competition Law and Labour Rights (Alphen aan den Rijn: Kluwer). Cranford, Cynthia J., Judy Fudge, Eric Tucker and Leah F. Vosko, 2005, Self-Employed Workers Organize: Law, Policy, and Unions (Montreal & Kingston: McGill-Queen’s University Press). Davies, A. C. L., 2012, ‘Identifying “Exploitative Compromises”: The Role of Labour Law in Resolving Disputes between Workers’, Current Legal Problems 65(1): 269–94. De Stefano, Valerio, 2016, The Rise of the ‘Just In Time’ Workforce: On-Demand Work, Crowdwork and Labour Protection in the ‘Gig Economy’ (Geneva: ILO). De Stefano, Valerio, 2018, ‘Negotiating the Algorithm’: Automation, Artificial Intelligence and Labour Protection (Geneva: ILO). Dubal, Veena, 2017, ‘The Drive to Precarity: A Political History of Work, Regulation, and Labor Advocacy in San Francisco’s Taxi and Uber Economies’, Berkeley Journal of Employment and Labor Law. http://​doi​.org/​10​.15779/​Z387P8TD1D. Ford, Michael and Tonia Novitz, 2020, ‘There Is Power in a Union? Revisiting Trade Union Functions in 2019’, in Alan Bogg, Jacob Rowbottom and Alison Young (eds), The Constitution of Social Democracy (Oxford: Hart). Heeks, R. 2017, ‘Decent Work and the Digital Gig Economy: A Developing Country Perspective on Employment Impacts and Standards in Online Outsourcing, Crowdwork’, Paper No. 71 (Manchester: Centre for Development Informatics, Global Development Institute). Jacoby, Sanford, 1982, ‘The Duration of Indefinite Employment Contracts in the United States and England: A Comparative Analysis’, Comparative Labor Law. https://​www​.researchgate​.net/​publication/​303140735​_The​_Duration​_of​_Indefinite​

The legal framework of worker voice in the platform economy

47

_Employment​_Contracts​_in​_the​_United​_States​_and​_England​_An​_Historical​ _Analysis. McCann, Deirdre, 2021, ‘Mencap and Uber in the Supreme Court: Working Time Regulation in an Era of Casualisation’, Oxford Human Rights Hub Blog, April, https://​ohrh​.law​.ox​.ac​.uk/​mencap​-and​-uber​-in​-the​-supreme​-court​-working​-time​ -regulation​-in​-an​-era​-of​-casualisation/​. Moyer-Lee, Jason and Nicola Kountouris, 2021, Taken for a Ride: Litigating the Digital Platform Model (Washington, DC: International Lawyers Assisting Workers Network. Prassl, Jeremias, 2015, The Concept of the Employer (Oxford: Oxford University Press). Prassl, Jeremias, 2018, Humans As a Service: The Promise and Perils of Work in the Gig Economy (Oxford: Oxford University Press). Routh, Supriya, 2016, ‘Informal Workers’ Aggregation and Law’, Theoretical Inquiries in Law 17. http://​labourlawresearch​.net/​sites/​default/​files/​papers/​2016​_Workers​ _Aggregation​_​%26​_Law​.pdf. Standing, Guy, 2016, The Corruption of Capitalism: Why Rentiers Thrive and Work Does Not Pay (London: Biteback Publishing). Wood, Alex J., 2020, Despotism on Demand: How Power Operates in the Flexible Workplace (Ithaca, NY: Cornell University Press). Woodcock, Jamie and Mark Graham, 2020, The Gig Economy: A Critical Introduction (Cambridge: Polity).

3. Giving platform workers a say: regulating for voice in the gig economy Andrew Stewart and Jim Stanford 1. INTRODUCTION The capacity of digital platforms to mobilize labour from dispersed locations (in some cases anywhere in the world), and assign tasks performed for disparate (and sometimes equally dispersed) end-users, poses unique challenges to the ability of workers in these industries to have a say in their working conditions. It is common for a platform worker to be hired, directed, compensated, supervised, and even fired without meeting or even speaking to another human being in the enterprise for which they work. This fragmentation and individualization (Aloisi, De Stefano & Silberman, 2019; Heiland, 2020), driven by both technology and (in many cases) geography, constitutes a daunting barrier for these workers to express their opinions, priorities and demands. The very concept of worker “voice” takes on a surreal nature when there is no one to hear. Many platform workers are engaged in activities that are highly insecure, poorly paid, and often dangerous. While some platform models are used by specialized professionals and technicians with valuable skills and considerable bargaining power, most platform workers hold few credentials, have limited alternative employment or income options, are drawn from disempowered communities (such as migrant populations or international students), and are easily replaceable. They thus constitute an especially vulnerable and exploitable source of labour supply – making it all the more crucial they have access to channels and protections that counterbalance the concentrated economic power of the platforms for which they work. Making matters worse, the typical classification (or often misclassification) of platform workers as contractors or freelancers may deny them basic employment protections such as minimum wages, pensions and superannuation benefits, workers compensation, and paid leave for holidays and illness. In many (but not all) jurisdictions, this also 48

Giving platform workers a say

49

precludes traditional forms of collective organization and bargaining, which commonly underpin strong worker voice mechanisms in other settings. For all these reasons, there is a need to develop systems and channels through which platform workers can overcome their technologically and geographically mediated isolation, share common concerns, and advance demands for better treatment, whether from platform owners or regulators. Achieving this will be vital if this growing form of work is not to become an enclave of exploitation and despair. The importance of meaningful, collective voice for platform workers has been acknowledged in reports issued by or for transnational organisations such as the Organisation for Economic Co-operation and Development (OECD, 2019) and the European Union (EU) (Eurofound, 2018, p. 64). It has also been championed by the International Labour Organization (ILO, 2021, pp. 211–16, 249, 256), as well as government inquiries in some nations (see e.g. James, 2020, p. 200). However, achieving effective and genuine voice mechanisms in the gig economy is by no means a straightforward undertaking. Account has to be taken of the geographical organization of the work in question (Johnston, 2020), the likelihood of resistance by platform businesses (Prassl, 2018, pp. 14–15) and the variety of different labour models and practices they use, as well as “the associated variance in workers’ power resources” (Vandaele, 2018, p. 7). In some cases, platforms exercise a high degree of employer-like control and coordination; workers are assigned to specific tasks with predetermined revenue and compensation parameters, and have little autonomy or agency other than the ability to log on and off their apps. In other cases, the relationship between the platform and its workers is more nuanced. Some platform workers have more freedom to market, set fees, and choose customers. Genuinely self-employed suppliers may not want or need collective voice or representation; for them, regulation of the fairness of trading terms may be a higher priority (Eurofound, 2020, p. 11). But even in those cases, worker voice can empower relatively independent freelancers to negotiate collectively over such matters. Our focus in this chapter is on the use of regulation to facilitate either collective bargaining or other methods by which platform workers or their representatives can be given a meaningful say over work organization, compensation and working conditions. Although much of the chapter is concerned with instrumental state regulation at the national, regional or sectoral level, we also examine various types of co-regulation between platform businesses, worker representatives and/or the state that have emerged both nationally and, to a more limited extent, transnationally (Finck, 2018). The outcomes of these latter initiatives are typically codes of practice and other forms of “soft law”, rather than legally binding agreements.

50

Missing voice?

In Section 2 of the chapter, we assess the extent to which existing systems of labour regulation around the world can accommodate representation and collective bargaining for platform workers. As will be seen, this frequently depends on such workers being classified as employees, although some systems permit bargaining by independent contractors or freelancers. Collective bargaining for self-employed workers may also be constrained by competition laws, an issue we explore separately in Section 3. Section 4 considers examples of actual or proposed reforms to national laws to facilitate representation and collective action by platform workers. Some countries have moved forward quite forcefully in extending labour protections to cover platform workers; in other cases, legislative reforms have been aimed at ensuring their continued exclusion. Section 5 of the chapter is devoted to international and transnational norms. After examining the application of current international labour standards, and highlighting moves by the EU to fill what reports have identified as significant gaps in protection, we note some interesting examples of what can be achieved through co-regulation. Finally, our conclusion in Section 6 lists several of what we see as the most promising priorities for regulatory improvement to support the development of genuine and effective mechanisms of collective voice, representation, and negotiation for platform workers – and, by extension, the ultimate achievement of fairer and more sustainable outcomes in this growing segment of the labour market.

2.

REPRESENTATION AND BARGAINING WITHIN EXISTING SYSTEMS OF LABOUR REGULATION

In its most general sense, “voice” simply refers to the ability of workers to express their opinions and preferences regarding their jobs and workplaces; to resist unfair, unacceptable, or dangerous conditions and decisions; and to advocate for desired changes in the terms and conditions of their work (such as work organization, compensation, management, fairness and safety). Genuine voice cannot be understood solely as a form of expression or “venting” (Wilkinson et al., 2020, p. 3). To be meaningful and effective, voice must be embedded in a system in which expressed views are received, considered, and to some reasonable degree acted upon. In other words, employers and managers must have a responsibility to receive, respond and act on the information and opinions expressed. Hence genuine workers’ voice is inextricably linked to workers’ agency and power: that is, their ability to act to bring about the changes for which they have advocated (Stanford & Poon, 2021).

Giving platform workers a say

51

One obvious option for formalizing and solidifying meaningful channels of voice for platform workers would be the extension of traditional collective representation and bargaining mechanisms to platform workers (even those who are not classified as employees). For example, several efforts have been undertaken in Europe to create works councils in platform businesses. An early works council for platform workers was formed for Foodora couriers in Vienna, Austria in 2017 (Eurofound, 2018, p. 55; Aloisi, De Stefano & Silberman, 2019, pp. 4–5). Similar efforts have been made in Germany to establish platform works councils, such as one for Deliveroo couriers in Cologne (Haipeter et al., 2020, p. 18). A works council has been established in Italy for the food delivery company Laconsegna, whose workers are employees (Borelli, 2020, p. 386). Other examples have been noted in Finland, France, the Netherlands, Norway and Sweden (Mexi, 2019, p. 13). These initiatives build on the strong foundation of Europe’s existing use of works councils. However, as Johnston (2020) notes, the success of works councils in platform work depends on the legal and cultural context in which they aim to operate. She recounts the example of the Independent Drivers Guild (IDG, affiliated with the International Association of Machinists) which created a works council-like structure for Uber drivers in New York City (Maniscalco, 2016; Scheiber, 2017). This effort “proved unfruitful, if not divisive” (Johnston, 2020, p. 32). The arrangement was financially supported by Uber itself (through an agreement with New York City regulators), and adopted a non-threatening approach to dealing with Uber (ruling out collective bargaining or strikes). This raised concerns about the autonomy of the organization. The IDG campaigned for changes in working conditions and compensation (including helping win an option for customers to tip their drivers), mostly through community action and lobbying of the local government. But it does not constitute the sort of structured workplace dialogue process featured in European-style works councils. A later proposal to establish a sector-wide model of representation for rideshare workers in the state of New York, in which the IDG was involved, ultimately collapsed in the face of opposition from traditional unions, who viewed it as an effort to forestall more genuine organizing (Eidelson, 2021b). A similar arrangement was announced more recently in Canada. Uber’s Canadian subsidiary has reached agreement with UFCW (the United Food and Commercial Workers International Union, a US-based union with members in Canada) to provide representation to Uber workers in disputes with the company. Uber itself pays for the service, and there is no collective agreement, collective bargaining or industrial action involved in the relationship. It has been criticized by other unions as an effort by Uber to forestall more genuine representation for Uber’s Canadian drivers and couriers (Deschamps, 2022).

52

Missing voice?

Efforts to establish more traditional (and perhaps more powerful) collective bargaining for platform workers have always confronted the problem of worker classification. In many jurisdictions, contractors do not have access to conventional collective bargaining processes. In some countries, unions have challenged this prohibition by arguing that the workers are in fact employees, with mixed success.1 In other countries, in contrast, collective bargaining is permitted for contractors in certain circumstances. In several Canadian provinces, for example, workers who qualify as “dependent contractors” can form unions, negotiate collectively and go on strike. Under this criterion, Foodora couriers were found in 2020 to be able to join the Canadian Union of Postal Workers.2 In the wake of that decision, Foodora closed its Canadian operations and discharged its Canadian workforce (Mojtehedzedeh, 2020). Some other countries have also extended the operation of collective bargaining laws to self-employed service providers. Examples include economically dependent self-employed workers (Trabajador Autonomo Economicamente Dependiente or TRADE) in Spain (Cherry & Aloisi, 2019, pp. 321–2), employee-like persons (arbeitnehmeraehnliche Personen) in Germany (Cherry & Aloisi, 2019, pp. 322–3), and “hetero-organised collaborators” in Italy (Borelli, 2020, p. 372). Despite these possibilities, formal and binding collective agreements concluded specifically for platform workers have so far been rare. An early and notable example is a collective agreement concluded in 2018 between the Danish cleaning services platform Hilfr and the United Federation of Danish Workers (3F). Among other provisions, it allows freelance cleaners to transition to employee status (Kilhoffer et al., 2020, pp. 254–5; ILO, 2021, p. 214). But, as discussed further below, the agreement had to be subsequently altered to avoid a legal challenge by the competition regulator. One 2020 study could identify only eight collective agreements covering platform workers across Europe (through more were said to be pending). Five of those eight agreements involved food delivery or couriers (Kilhoffer et al., 2020, p. 119). The same study also reported several instances of broader sectoral agreements covering platform workers. Sometimes, as in Italy, these addressed at least some issues of particular concern to gig workers. But there were also examples of agreements applying in Germany without any apparent involvement from workers at the platforms concerned (Kilhoffer et al., 2020, p. 119). In Sweden, the app-based taxi service Bzzt is covered by an industry-wide collective agreement; its drivers are employees, which no doubt facilitated this agreement (Johnston & Land-Kazlauskas, 2018, p. 30). Similarly, a recent sectoral agreement for delivery riders concluded in Denmark by 3F with the employers’ organisation Dansk Erhverv presently covers only one platform, Just Eat (ETF, 2021). In Spain, an industry-wide agreement for the catering

Giving platform workers a say

53

and hotel sector was modified in 2019 to bring delivery platform riders within its scope (Rodriguez-Pinero Royo, 2020, p. 466). More significantly, in late 2021 the delivery platform Just Eat concluded a collective agreement with the CCOO and UGT trade union confederations. This has been described as representing “the first step in normalising platform work under the Spanish model of industrial relations and its labour market” (Rodríguez, 2022). It was concluded following the enactment of new labour legislation that deems delivery riders to be employees, as discussed below in Section 4. A more contentious instance of collective bargaining occurred in Italy, when Assodelivery (an association representing food delivery platforms) signed an agreement with the right-wing union UGL. The content of the agreement has attracted fierce criticism from other unions, and its validity has been challenged by the Ministry of Labour (Haipeter & Iudicone, 2020, p. 38; Tamma, 2020). In a similar vein, Rodriguez-Pinero Royo (2020, p. 468) cites a “professional interest agreement” signed by Deliveroo and the Spanish Association of Riders. Rather than being an instrument for improving working conditions, it denies riders employment status and is “mostly designed to avoid the application of collective agreements”. These examples again highlight the risk of platform firms favouring non-binding or relatively compliant forms of worker representation, in order to forestall more genuine and independent union organization. These strategies are also typically connected to advocacy efforts by platforms to avoid regulatory measures that would reclassify their workers as employees. Other relatively powerless representation initiatives steer clear of collective bargaining altogether, or exclude major platforms, such as an agreement for bike couriers in Switzerland that does not cover “main operators” such as Uber (Dunand, Magoga-Sabatier & Mahon, 2020). There has also been controversy about an agreement concluded in 2021 by Uber in the United Kingdom with a general trade union, GMB. The agreement has been hailed as “groundbreaking”, in so far as it commits Uber to recognize the union as a representative of its rideshare drivers and to bargain with it over matters such as health and safety (Butler, 2021). But the deal has been criticized by other, newer unions for not encompassing collective bargaining over pay, or for addressing aspects of Uber’s operations which are claimed not to comply with a landmark Supreme Court decision that found drivers are entitled to minimum wages and holiday pay (McGurdy, 2021).3 The difference of opinion reflects broader variations between the some of the tactics adopted by traditional and alternative unions in seeking to overcome legal constraints on the representation of platform workers (Bertolini & Dukes, 2021). In Australia, Airtasker (a general job-posting platform) entered into an agreement in 2017 with Unions NSW, a peak union body in the state of New South Wales which had previously criticized the platform’s operations (Unions NSW, 2016). The agreement recommended minimum standards

54

Missing voice?

for engagements arranged through the platform, including minimum rates of pay. Airtasker also committed to work with Unions NSW to improve safety, offer optional insurance to workers, and provide a dispute resolution process (Minter, 2017). However, the agreement was “met with skepticism due to its ‘entirely optional and possibly unenforceable’ nature” (Johnston & Land-Kazlauskas, 2018, p. 30). Even Unions NSW (2018, p. 10) concedes that the agreement “highlights the risks currently facing workers in the gig-economy and the failure of legislative tools to provide adequate protections”. Unions NSW continues to criticize the low wages and health and safety risks faced by Airtasker workers. In Switzerland, another non-binding code of conduct was created in 2019 by Mila, a repair and technology services platform, and Syndicom, a trade union for media and communication workers. It includes “eight principles aimed at creating trust and fairness in relations between the platform and the workers” concerning (among other matters) legality of projects, the right to collective representation and bargaining, clear definition of tasks and reasonable scheduling, constructive feedback and open communication, and data protection and privacy (Eurofound, 2021). The Canadian union Unifor has also organized a union of freelance media workers at a range of mostly digital outlets and platforms. Its activities include model contract terms, information sharing about unethical employer practices, training, and group insurance (Stanford & Poon, 2021, p. 81). Overall, it is perhaps unsurprising that “the most advanced examples of collective bargaining in the gig economy come from places like Sweden and Austria, which boast solid legal and regulatory frameworks, strong social partners, and a prevalence of industry-wide collective agreements” (Johnston & Land-Kazlauskas, 2018, p. 33). It is primarily in these countries, with their “strong union movements and robust and flexible bargaining systems”, that the “legal and practical obstacles to negotiating agreements for platform workers are being increasingly overcome” (ILO, 2021, p. 213).

3.

COMPETITION LAWS AS AN IMPEDIMENT TO COLLECTIVE ACTION

Collective bargaining over wages and working conditions is generally prohibited for self-employed workers (OECD, 2019, pp. 235–8). To the extent that platform workers are not employees (or at least are assumed not to be), then attempts to organize or set standards through collective action may be impeded by the operation of competition laws. This was illustrated in the case of an ordinance passed in 2015 by Seattle City Council, which granted collective bargaining rights to rideshare drivers even though they were not treated as employees. Its validity was challenged by the

Giving platform workers a say

55

US Chamber of Commerce, acting on behalf of Uber, which complained that it violated the Sherman Antitrust Act. The city ordinance was held not to fall within an exemption to that Act covering state action.4 The city then amended the ordinance to remove collective bargaining over wages (Steinbaum, 2019, pp. 59–60), although other provisions for collective representation in the ordinance were preserved. This case illustrates how platforms use antitrust law to prevent collective bargaining, even when they themselves engage in price coordination by unilaterally setting revenue parameters for their purportedly independent drivers (Paul, 2017). The situation is similar in Europe (Lianos, Countouris & De Stefano, 2019, pp. 298–302; Kilhoffer et al., 2020, pp. 246–50). EU competition laws are considered to apply only to “undertakings”, so it is not unlawful for collective agreements to fix the price of labour supplied by employees.5 But when unions or other bodies seek to negotiate on behalf of independent contractors or freelancers, they are constrained by competition law rules – unless they can claim to be representing “false self-employed” workers.6 In Denmark, the competition regulator intervened to require changes to the collective agreement mentioned earlier for Hilfr workers. The regulator insisted that the agreement could only set minimum prices for workers who were clearly designated as employees (Danish Competition and Consumer Authority, 2020). The decision has been criticized as a “flawed application of antitrust legislation” (Countouris & De Stefano, 2020), but it illustrates the barriers posed by competition law to collective representation and bargaining for platform workers. Schiek and Gideon (2018) have suggested ways in which EU law could be reinterpreted to permit collective bargaining on behalf of platform workers. Kim (2020) has suggested that the same could be achieved in the USA through revised guidance issued by the Federal Trade Commission, based on a rationale of promoting worker welfare. But Biasi (2018) takes a more pessimistic view, arguing that in both the USA and the EU legislative amendments would be needed to legalize collective bargaining for platform workers. As the OECD (2019, pp. 239–40) has pointed out, in several countries (including France, Italy and Spain) unions have negotiated working conditions on behalf of platform workers without hindrance from national antitrust authorities. But uncertainty over the long-term validity and viability of those efforts persists. Without legislative reform to clearly endorse collective bargaining for contractors, this implicit space could easily be closed by subsequent regulatory rulings or legal challenges. In other countries, such as Canada and Korea, certain types of “dependent contractors” are expressly allowed to engage in collective bargaining. Extensions to collective bargaining rights have likewise been made in Italy, Germany, the UK, Spain, Poland, Ireland and Japan for workers who fall into the “grey zone” between self-employment and conventional employees (ILO, 2021, pp. 212–13).

56

Missing voice?

An alternative approach is for exemptions from prohibitions on cartel conduct to be granted to self-employed workers in specified occupations or sectors, presumably reflecting the dependent or near-employment nature of contractual relationships in those circumstances. However, such targeted exemptions “are not always easy to define and apply; the list may need frequent updating, and the potential reversal of exemptions is a source of legal uncertainty for workers and businesses alike” (OECD, 2019, p. 240). To overcome such problems, a “class exemption” has been issued in Australia to permit small businesses to engage in collective bargaining with suppliers and processors (Australian Competition and Consumer Commission, 2020). In theory, this would allow self-employed platform workers to engage in collective bargaining with platforms, although it would still prohibit concerted action by workers and may not even permit information sharing (Hardy & McCrystal, 2020). Notwithstanding its limitations, however, it appears to accord with the international standards on freedom of association and collective bargaining mentioned below in Section 5 (McCrystal & Hardy, 2021). In June 2020, the European Commission announced a consultation process over new measures to “ensure that the EU competition rules do not stand in the way of collective bargaining for those who need it”. Specific reference was made to the need to improve the working conditions of platform workers, recognizing that “many individuals have no other choice than to accept a contract as self-employed” (European Commission, 2020). The European Parliament (2020, para. 27) expressed its support for “a targeted revision of EU competition law so as to allow collective price-setting for the precarious self-employed, in order to ensure a better balance in bargaining power and a fairer internal market”. The European Commission has since released draft guidelines concerning the application of competition laws to collective agreements regarding working conditions for “solo self-employed people”, in conjunction with other proposals for improving platform work discussed below in Section 5 (European Commission, 2021a, 2021b). The proposals indicate a willingness to acknowledge the unintended consequences of the application of competition law in preventing collective voice and action by platform workers – who, in most cases, have little bargaining power and are subject to the overwhelming direction and control of the platforms for which they work.

4.

REFORMING NATIONAL LAWS

There has been much research and public debate regarding the need to reform regulations governing platform work, but few countries have acted. In many countries, what has been described (in the case of the UK) as an “anemic” response by legislators (Mason, 2020, p. 340) has left courts to try (to the extent they are willing) to accommodate platform work within existing cate-

Giving platform workers a say

57

gories and forms of regulation. This is despite copious evidence provided by inquiries, reports and academic research on the need for improved regulatory intervention. Even in Europe, with its stronger tradition of labour rights, countries have generally preferred to “adapt platform work to the existing legislative framework rather than introduce new dedicated legislation” (Kilhoffer et al., 2020, p. 103). The platform industry itself has been an active participant in reinforcing this regulatory and legislative passivity. As the Californian example detailed below shows, major platforms have mobilized extensive financial resources, lobbying, and the political influence of their customer base to resist and overturn policy initiatives that would curtail their current business models. France was the first European country to pass a general law on platform work: the El Khomri Act, passed in August 2016.7 It amended the French Labour Code to create new rights for a particular type of self-employed worker (travailleur indépendant), defined as those using one or more digital platforms to provide personal services, where the platform determines the characteristics of those services and sets their price. The reform grants the rights to take collective action and engage in collective bargaining (Kilhoffer et al., 2020, pp. 105–6, 252–3). The Act was amended in 2019 to give platforms an option to extend certain rights to economically dependent platform workers who transport people or goods.8 Platforms do so mostly to reduce the chances of those workers asserting full employment status (Daugareilh, 2020, pp. 410–12). Italy has also been active in regulating platform work, although not with the same breadth of scope and effect. A national law passed in 2019 confirmed that platform workers may fall within a new category of “hetero-organized collaborators”, and thus have access to most (but not all) of the rights accorded to employees, including the rights to unionize and strike. It also created new rights and protections for self-employed delivery workers, entitling them (among other things) to hourly pay rates set by relevant national collective agreements (Borelli, 2020, pp. 378–9). After tripartite discussions with social partners, the Spanish government introduced legislation in 2021 to regulate platform work.9 The new “Riders’ Law” includes a presumption of employment for work involving the delivery or distribution of consumer goods or merchandise, carrying with it access to standard labour and social rights and protections. This was not unexpected, given the Supreme Court’s ruling in September 2020 that a rider for the food delivery platform Glovo was an employee.10 There is also a requirement to inform worker representatives about the design and operation of any algorithms used to determine employment conditions (Todoli-Signes, 2021). Other provisions in the new Spanish legislation put limits on the use of temporary employment contracts, create a new category of “fixed-discontinuous” con-

58

Missing voice?

tract for occupations with inherently unpredictable work patterns, and impose joint and several liability on contractors for social security and related obligations associated with work performed by contractors and sub-contractors (Faulkner, 2021). These provisions will further regulate the labour strategies of digital platforms, as well as other forms of non-standard employment. Spain’s leading business advocacy group, the Spanish Confederation of Business Organizations, actually endorsed many of these reforms – but the Association of On-Demand Service Platforms, which represents platforms like Deliveroo, Glovo and Uber Eats, opposed it (Aranguiz, 2021). An earlier, though less ambitious reform of this type was introduced in Portugal in 2018.11 It applies to passenger transport organized through platforms, and requires drivers to be engaged by an “operator” (who may be a person or a company) rather than contracting directly with the platform. The contract between the driver and operator is presumed to be one of employment, providing the drivers with access to labour and social protections, and there are also limits on working hours (Kilhoffer et al., 2020, p. 105). In the United States, by contrast, many of the most notable legislative initiatives, such as the “marketplace platform” laws adopted by several states (Duff, 2020), have sought to shield digital labour platforms from the normal operation of labour and social laws, thus confirming their continued use of contractor-based business models. For example, in November 2020 the electors of California were persuaded to modify a 2019 state law which effectively treated platforms such as Uber and Lyft as employers (Cherry, 2021). The law in question, commonly known as AB5 (the number of the original Bill),12 sought to enshrine an “ABC” test for determining employment status, focusing on whether workers genuinely had a business of their own. This test had been adopted by the California Supreme Court in a case involving delivery drivers.13 Although that precedent did not involve platform work, it was widely understood that both the purpose and effect of AB5 would be to categorize rideshare drivers and at least some other platform workers as employees under the California Labor Code. When Uber and Lyft did not comply with the new law, the state initiated litigation.14 But before the law could effectively be enforced, it was significantly amended by way of a referendum to exempt the rideshare platforms, following a campaign on which the platform businesses spent hundreds of millions of dollars (O’Donovan, 2020). The measure passed, Proposition 22,15 precludes “app-based rideshare and delivery drivers” from being treated as employees. It does not affect the status of other platform workers. It also specifies a number of purported benefits and protections for drivers, including a minimum earnings guarantee, platform-provided accident insurance, and protection against discrimination. In effect it creates “a new hybrid type of category” for the workers it covers (Cherry, 2021, p. 7). Following a legal challenge, a California court ruled that

Giving platform workers a say

59

several features of Proposition 22 were unconstitutional.16 The rideshare platforms are appealing (Roosevelt & Hussain, 2021). If they win before a higher court, the proposition will effectively preclude any future attempt to confer collective bargaining or other labour rights on rideshare and food-delivery platform workers in that state. The platform industry is now advocating aggressively for similar measures in other states, such as Massachusetts (Platoff, 2021). At a federal level, the Democrat-controlled House of Representatives has passed a Protect the Right to Organize Act (PRO Act), with outspoken support from President Biden. Among other provisions aimed at strengthening union representation across the economy, the Act would apply an ABC-style test to determine whether workers are eligible to organize unions. That would grant at least some platform workers (and other economically dependent workers currently classified as self-employed) collective bargaining opportunities. But the Bill may not have sufficient support to pass in the Senate, especially in the face of vocal and well-financed opposition from platform businesses (Goodkind, 2021; Mulvaney, 2021). A similar, though less far-reaching, proposal under consideration in the state of Connecticut would grant gig workers collective bargaining rights, without classifying them as employees (Eidelson, 2021a). As the ILO (2021, p. 213) has pointed out, even where platform workers can be brought within the scope of existing collective bargaining regimes, those laws may have to be adapted to meet their needs: For example, many countries require the identification of a “bargaining unit” or one or more “representative unions” in order for collective bargaining machinery to operate. These concepts are difficult to apply in the context of online web-based platforms, since the workforce is diffused across geographical and industry boundaries. The problem is especially acute in those systems where the law requires collective bargaining to occur predominantly at the level of an enterprise.

5.

INTERNATIONAL AND TRANSNATIONAL STANDARDS AND INITIATIVES

International labour standards are not consistent in their categorization of different types of worker (Creighton & McCrystal, 2016; De Stefano, 2021). But the core Conventions that underpin the ILO’s Declaration on Fundamental Principles and Rights at Work, which include the Freedom of Association and Protection of the Right to Organize Convention, 1948 (No. 87), and the Right to Organize and Collective Bargaining Convention, 1949 (No. 98), are broadly applicable to “workers” of any kind – not just employees. As such, they are “applicable to platform workers in the same way as to all other workers, irrespective of their employment status” (ILO, 2020, p. 142). Hence as a matter of international law, even if considered to be self-employed, platform workers

60

Missing voice?

should be entitled to unionize and engage in collective bargaining (see also Freedland & Kountouris, 2017; ILO, 2019a, p. 5). The status of platform workers for the purpose of EU laws is less clear (De Stefano & Aloisi, 2018; Kilhoffer et al., 2020, pp. 125–203). In theory, regulatory instruments such as Directive 2002/14/EC on informing and consulting employees should have high relevance to platform workers, to keep them and their representatives appraised of changes to their economic situation or the organization of their work. But its provisions have been described as “insufficiently, or not at all, adapted to the platform work’s digital business environment, making it relatively easy to circumvent” (Kilhoffer et al., 2020, p. 158). In February 2021, however, the European Commission commenced consultation with European social partners over measures to improve protections for platform workers. In December that year, after the conclusion of that process, the Commission released details of a proposed new Directive on improving working conditions in platform work. If and when adopted by the European Parliament and Council, member states of the EU would have two years to bring their national laws into conformity (European Commission, 2021a). Like the Spanish Riders’ Law discussed in Section 4, the proposed Directive would presume workers to be employees, if the digital platform they were using controlled certain elements of their work. The Commission has estimated that this would entitle some 5.5 million platform workers, currently believed to be misclassified as self-employed, to the labour and social rights associated with being a “worker” – including the right to engage in collective bargaining. But even platform workers still found to be self-employed would benefit from the other main element of the Directive, concerning transparency and oversight of algorithmic management. Among other things, platforms would be obliged to inform and consult workers and their representatives if they intend to introduce new forms of automated monitoring or decision-making, or to make substantial changes to their existing systems (European Commission, 2021c). What is missing from this initiative, however, is any requirement for platforms to actively engage in collective bargaining, or indeed to provide for collective representation (Fairwork, 2021). Besides international legal provisions, there is also potential for union federations to use global framework agreements (GFAs) with platforms operating multinationally (Novitz, 2020, pp. 285–6). These agreements are negotiated by global union federations (such as the ITF, UNI and IndustriALL) with major international firms or conglomerates, and set standards for minimum conditions, fair representation, and collective bargaining practices that are generally regarded as more meaningful and enforceable than voluntary codes of conduct (Wright et al., 2019). The application of GFAs to global platform businesses (like Uber) holds potential for mobilizing coordinated international pressure

Giving platform workers a say

61

on platforms to accept collective bargaining rights and respect principles of fair practice. It is also possible for national unions to engage internationally with platforms in less formal ways. An initiative of the German union IG Metall provides an intriguing example. With the support of unionists in Sweden and the UK, it developed a Crowdsourcing Code of Conduct. The Code applies to crowdwork platforms, which are typically staffed by workers based in many different countries (ILO, 2019b, pp. 105–6; Haipeter et al., 2020, p. 22), and was first negotiated with the software testing platform Testbirds. It sets out principles regarding (among other things) compliance with domestic law, fair payment, respectful interaction, clear tasks and reasonable timing, freedom and flexibility, constructive feedback, and data protection and privacy. Nine crowdwork platforms have now signed the Code, mostly from Germany but one from the UK (Digivante). The German Crowdsourcing Association is also an “official supporter” of the Code.17 The Code is supported by an Ombuds Office, established in 2017 to help resolve disputes between crowdworkers and signatory platforms, and administered by IG Metall. It comprises a board of five people: one worker, one representative each from a trade union, a platform and the Crowdsourcing Association, and a labour judge serving as a neutral chair (ILO, 2019b, pp. 105–6). It does not handle many cases: only 14 were submitted in 2019, only one of which required a decision, with the others being settled or withdrawn (Ombuds Office of the Code of Conduct for Paid Crowdsourcing, 2020). Nevertheless, it constitutes an interesting experiment in imparting meaningful oversight over platforms which participate in a voluntary code. These international initiatives hold some promise for facilitating international information sharing and cooperation among trade unions and others advocating for reforms to platform work practices. For the most part, however, the main arena for efforts to extend collective voice and representation for platform workers will continue to be the national level, where the main decisions regarding the applicability of collective bargaining law and other mechanisms to platform workers will continue to be made.

6.

CONCLUSION: PRIORITIES FOR REFORM

Efforts to improve fairness, compensation and safety for platform workers proceed on numerous fronts. Debates over the classification of platform workers continue in courts and parliaments around the world (De Stefano et al., 2021; Hießl, 2021; ILAW Network, 2021; ILO, 2021). A parallel stream of debate concerns how to provide platform workers (whether they are determined to be employees, or not) with meaningful and democratic channels for collective voice and representation. Various policy proposals have been

62

Missing voice?

advanced to grant more protection and bargaining power to platform workers, so they can address the numerous challenges most face in their work lives (including low pay, job insecurity, health and safety risks, and more). Reforms have been proposed in labour law, taxation policy, corporate and financial regulations, transportation and other sector-specific regulations. There has also been support for workers to form their own undertakings, such as cooperatives (Johnston & Land-Kazlauskas, 2018, p. 19; Sarina & Riley, 2018; Kilhoffer et al., 2020, p. 121; ILO, 2021, p. 88), to escape dependence on private platforms. In this chapter we have focused more narrowly on regulatory initiatives to support organized channels of collective voice and representation. Our review informs the following conclusions and recommendations. There is a broad but not perfect overlap between debates over the classification of platform workers, and their efforts to build channels of voice and representation. Many platform-mediated roles bear strong similarities to employment (due to the high degree of platform control over many aspects of the work, including pricing and compensation), and advocates will naturally continue their efforts to win recognition of employee status through legal challenges and legislative reform. However, the need for genuine and effective voice for platform workers can and should be separated from the ongoing dispute over classification. Many platform workers (such as those using crowdwork platforms) are clearly not employees in any conventional sense, yet they still have a shared interest in being able to communicate and negotiate with platforms over the terms of their engagement. Thus the effort to win regulatory and legislative support for platform worker voice cannot be equated with or subsumed within the effort to win employee status for some of them. In other work, we have criticized the idea of creating a new “intermediate” category of worker to reflect the supposedly unique circumstances of platform workers (Stewart & Stanford, 2017). In the first place, those circumstances are not as unique or novel as is often claimed: the engagement of workers under direction from an intermediary (including labour hire firms, gangmasters, putting-out merchants, and others) has a long and often ignoble history. The creation of a new, third category of worker would further muddy the regulatory waters, and open up new opportunities for regulatory arbitrage by employers, who may seek to reclassify existing employees into this new middle category. Major platforms would likely utilize this new category to forestall more traditional labour representation – alongside proposing “intermediate” packages of benefits and protections not equivalent to the entitlements received by conventional employees. This is a risky moment to consider a whole new class of worker. A more straightforward approach would clarify the applicability of existing standards and protections to workers whose reality belies the mirage of platform worker “independence”, while also recognizing that certain rights (such as to safety and non-discrimination at work, privacy, or access to timely

Giving platform workers a say

63

and affordable dispute resolution) should be afforded even to genuinely self-employed workers (Stewart & McCrystal, 2019). One protection that could be directly extended to platform workers, without requiring a solution to the broader classification problem, is the right to organize and bargain collectively. Indeed, the more powerful and reliable mechanisms for collective voice and representation tend to be those associated with unionization and collective bargaining. Therefore, a potentially effective avenue for creating more opportunity for voice for platform workers is to simply ensure that they have access to conventional channels for unionization and bargaining – as international labour standards plainly require, and regardless of whether those workers are ever classified as employees. As we have noted, in several countries (including Canada, Italy, Germany and Spain) certain types of contractor already have access to conventional collective bargaining mechanisms. Those avenues should be utilized to the fullest – in addition to, or instead of, ongoing battles over classification or proposals to develop entirely new structures of representation. In other countries, industrial relations laws should be reformed to clarify that the basic tools of collective representation and bargaining are accessible to platform workers. Pragmatic tests (akin to the ABC test in California) could be applied to determine whether platform workers have genuine independence and control over their “businesses”. In many cases they clearly do not, and those workers should be granted explicit rights to organize under conventional law. In some cases this will require parallel measures (perhaps including legislative reform) to relax competition laws so that workers who meet pragmatic tests for dependence are not prevented from developing and exercising collective voice and power. Clearly, the purpose of antitrust policy is to protect relatively powerless and fragmented agents in the economy from the concerted power of large, centralized forces – not the other way around. The use of competition law to forestall organizing by platform workers is a grotesque misuse of its original purpose. Another direct and obvious step to ensure that efforts to organize collective voice for platform workers are genuine and accountable, is to ensure that any such initiatives (whether achieved through conventional collective bargaining channels, or via alternative means such as non-union councils or voluntary codes of conduct) must be independently ratified by the workers who are covered by them. The history of trade unionism is peppered with examples of employers who sought to foster captive, non-threatening forms of worker representation as a way of preventing the establishment of more genuine unions (Dunn, 1926; Nelson, 1982; Debicky, 2001; Taras & Kaufman, 2006; Donaghey et al., 2012). Platform businesses have now adopted a similar strategy, invoking overstated claims about how the “innovative” or “technology-intensive” nature of the work renders traditional forms of worker

64

Missing voice?

representation inappropriate – an analogue to their claims about conventional labour protections, such as minimum wages. The platforms may even find willing partners within the union movement to participate in such ventures, as with some of the initiatives mentioned earlier. To prevent the emergence of “sweetheart” arrangements, and to ensure that voice mechanisms are ultimately grounded in the democratic approval of the workers they cover, voice structures should be subject to endorsement from relevant populations (in the same way that conventional bargaining units and collective agreements must be ratified, in most jurisdictions, by affected employees). Our proposals largely rely on the incremental extension of existing structures and practices in labour law. We do not think, in general, that the playbook of labour regulation – from the perspective of either regulators, or unionists and advocates – must be fundamentally rewritten to deal with the challenges arising from the gig economy. Most of the problems that raise most concern in platform work (oversupply of workers, low compensation, lack of autonomy, health and safety risks) are ultimately similar to those that have been challenged by unionists and labour advocates for centuries. Even the gig economy’s core labour model is not fundamentally new – even if some of the technologies used for mobilizing, supervising, and compensating this highly precarious labour are (Finkin, 2016; Stanford, 2017). What is missing is clarity and determination that tried-and-true principles of labour regulation and collective voice are, indeed, still legitimate and relevant in the platform economy. In the case of collective voice for platform workers, which will be critical for improving working conditions in ways that reflect the unique characteristics of each platform, there is no need to reinvent the wheel. We have only to make sure that wheel is still attached, and functional, to this modern digital vehicle.18

NOTES 1.

For a successful claim of this type in the Netherlands, see Deliveroo Netherlands BV v Federatie Nederlandse Vakbeweging, Gerechtshof Amsterdam, 200.261.051/01, 16 February 2021. Deliveroo is reported to be appealing this decision to the Supreme Court. In the meantime, a similar claim against Uber in the same country has also ruled that drivers are employees, and eligible for coverage under Netherlands’ sector collective agreement for taxi drivers (ILAW Network, 2021, pp. 60–61; Toh, 2021). Compare the unsuccessful claim brought against Deliveroo in the UK: R (on the application of Independent Workers’ Union of Great Britain) v Central Arbitration Committee [2021] EWCA Civ 952. In Argentina, the Platform Staff Association (APP) took action to require the platform Rappi to comply with freedom of association laws and reinstate three union delegates it had dismissed. The complaint was dismissed by an appeal tribunal on the basis that it was impossible to determine whether the workers were employed (Hartmann-Cortés, 2021).

Giving platform workers a say

65

2.

Canadian Union of Postal Workers v Foodora Inc, OLRB Case No 1346-19-R, 25 February 2020. 3. Uber BV v Aslam [2021] ICR 657. 4. US Chamber of Commerce v City of Seattle, 890 F 3d 769 (9th Circ 2018). 5. Albany, C-67/96, EU:​C:​1999:​430. 6. FNV Kunsten Informatie en Media, C-413/13, EU:​C:​2014:​2411. 7. Law no 2016-1088 of 8 August 2016 on labour, modernization of social dialogue and securing of professional careers. See now French Labour Code, Arts 7341-1, L. 7342-1. 8. Law no 2019-1428 of 24 December 2019 on mobility orientation. 9. Royal Decree Law 9/2021 of 11 May 2021. 10. Tribunal Supremo, Sala de lo Social, STS 2924/2020, 25 September 2020. 11. Law no 45/2018 of 10 August 2018, Legal regime of individual transport activity and paid passengers in unregistered vehicles from electronic platforms. 12. AB 5, 2019 Leg Reg Sess (Cal 2019). 13. See Dynamex Operations West Inc v Superior Court of Los Angeles County, 4 Cal 5th 903 (2018). 14. See People v Uber Techs, Inc, 56 Cal App 5th 266 (Cal Ct App. 2020). 15. See https://​vig​.cdn​.sos​.ca​.gov/​2020/​general/​pdf/​topl​-prop22​.pdf [accessed 27 January 2022]. 16. Castellanos v State of California, Case Number S266551, Alameda County Superior Court, 20 August 2021. 17. See http://​crowdsourcing​-code​.com/​[accessed 27 January 2022]. 18. This chapter draws on research funded by the Australian Research Council for a Discovery Project on Working the Gig Economy: The Organisation of Digital Platform Work (DP180101191), as well as on research undertaken by Andrew Stewart, Robyn Mayes, Penny Williams and Simon Guthrie for the International Labour Organization (ILO).

REFERENCES Aloisi, A., De Stefano, V. & Silberman, M. (2019), “A Manifesto to Reform the Gig Economy”, Regulating for Globalization, 1 May. Available: http://​regula​tingforglo​ balization​.com/​2019/​05/​01/​a​-manifesto​-to​-reform​-the​-gig​-economy/​?doing​_wp​ _cron​=​1588474862​.11​75210475921630859375 [accessed 27 January 2022]. Aranguiz, A. (2021), “Spain’s Platform Workers Win Algorithm Transparency”, Social Europe, 18 March. Available: https://​www​.socialeurope​.eu/​spains​-platform​ -workers​-win​-algorithm​-transparency [accessed 27 January 2022]. Australian Competition and Consumer Commission (2020), Competition and Consumer (Class Exemption—Collective Bargaining) Determination 2020, 19 October. Available: https://​www​.accc​.gov​.au/​system/​files/​public​-registers/​documents/​ Legislative​%20instrument​%20​%E2​%80​%93​%2019​%20October​%202020​.pdf [accessed 27 January 2022]. Bertolini, A. & Dukes, R. (2021), “Trade Unions and Platform Workers in the UK: Worker Representation in the Shadow of the Law”, Industrial Law Journal, Vol. 50, No. 4, pp. 662–88. Biasi, M. (2018), “‘We Will All Laugh at Gilded Butterflies’: The Shadow of Antitrust Law on the Collective Negotiation of Fair Fees for Self-Employed Workers”, European Labour Law Journal, Vol. 9, No. 4, pp. 354–73.

66

Missing voice?

Borelli, S. (2020), “Fitting the Panoply in a Binary Perspective: The Italian Platform Workers in the European Context”, Comparative Labor Law & Policy Journal, Vol. 41, No. 2, pp. 365–93. Butler, S. (2021), “Uber Agrees Union Recognition Deal with GMB”, The Guardian, 27 May. Cherry, M. A. (2021), “Proposition 22: A Vote on Gig Worker Status in California”, Comparative Labor Law and Policy Journal, Dispatch No. 31 – United States. Cherry, M. & Aloisi, A. (2019), “‘Dependent Contractors’ in the Gig Economy: A Comparative Approach”, American University Law Review, Vol. 66, No. 3, pp. 635–89. Countouris, N. & De Stefano, V. (2020), “Collective-bargaining Rights for Platform Workers”, Social Europe, 6 October. Available: https://​ www​ .socialeurope​ .eu/​ collective​-bargaining​-rights​-for​-platform​-workers [accessed 27 January 2022]. Creighton, B. & McCrystal, S. (2016), “Who is a Worker in International Law?”, Comparative Labor Law and Policy Journal, Vol. 37, No. 3, pp. 691–725. Danish Competition and Consumer Authority (2020), Commitment Decision on the Use of a Minimum Hourly Fee, 24 August. Available: https://​www​.en​.kfst​.dk/​nyheder/​ kfst/​english/​decisions/​20200826​-commitment​-decision​-on​-the​-use​-of​-a​-minimum​ -hourly​-fee​-hilfr/​[accessed 27 January 2022]. Daugareilh, I. (2020), “The Legal Status of Platform Workers in France”, Comparative Labor Law & Policy Journal, Vol. 41, No. 2, pp. 405–521. De Stefano, V. (2021), “Not as Simple As It Seems: The ILO and the Personal Scope of International Labour Standards”, International Labour Review, Vol. 160, No. 3, pp. 387–406. De Stefano, V. & Aloisi, A. (2018), European Legal Framework for Digital Labour Platforms (Luxembourg: European Commission). De Stefano, V., Durri, I., Stylogiannis, C. & Wouters, M. (2021), Platform Work and the Employment Relationship, ILO Working Paper 27 (Geneva: International Labour Office). Debicky, C. (2001), “Review of Nonunion Employee Representation: History, Contemporary Practice, and Policy”, Industrial Relations, Vol. 65, No. 4, pp. 816–19. Deschamps, T. (2022), “Uber Canada Signs Deal with Union Offering Workers Dispute Representation”, Toronto Star, 27 January. Donaghey, J., Cullinane, N., Dundon, T. & Dobbins T. (2012), “Non-union Employee Representation, Union Avoidance and the Managerial Agenda”, Economic and Industrial Democracy, Vol. 33, No. 2, pp. 163–83. Duff, M. C. (2020), “All the World’s a Platform?: Some Remarks on ‘Marketplace Platform’ Employment Laws”. Available: https://​papers​.ssrn​.com/​sol3/​papers​.cfm​ ?abstract​_id​=​3520723 [accessed 27 January 2022]. Dunand, J.-P., Magoga-Sabatier, S. & Mahon, P. (2020), “Status of Platform Workers in the Swiss Legal System”, Comparative Labor Law & Policy Journal, Vol. 41, No. 2, pp. 343–63. Dunn, R., ed. (1926), American Company Unions (Washington, DC: Trade Union Educational League). Eidelson, J. (2021a), “Lawmakers Look to Spruce Up Gig Work Rather Than Replace It”, Bloomberg, 18 March. Eidelson, J. (2021b), “N.Y. Lawmaker Drops Bid for Uber Driver Union Deal This Year”, Bloomberg, 8 June.

Giving platform workers a say

67

ETF [European Transport Workers’ Federation] (2021), “3F Secures Ground-Breaking National Sectoral Agreement for Delivery Riders”, 4 February. Available: https://​ www​.etf​-europe​.org/​3f​-secures​-ground​-breaking​-national​-sectoral​-agreement​-for​ -delivery​-riders/​[accessed 27 January 2022]. Eurofound (2018), Employment and Working Conditions of Selected Types of Platform Work (Luxembourg: Eurofound). Eurofound (2020), Back to the Future: Policy Pointers from Platform Work Scenarios, New Forms of Employment series (Luxembourg: Publications Office of the European Union). Eurofound (2021), Syndicom and Mila Code of Conduct, Platform Economy Repository, European Foundation for the Improvement of Living and Working Conditions). Available: https://​www​.eurofound​.europa​.eu/​data/​platform​-economy/​initiatives/​ syndicom​-and​-mila​-code​-of​-conduct [accessed 27 January 2022]. European Commission (2020), “Competition: The European Commission Launches a Process to Address the Issue of Collective Bargaining for the Self-employed”, press release, 30 June. Available: https://​ec​.europa​.eu/​commission/​presscorner/​ detail/​en/​IP​_20​_1237 [accessed 27 January 2022]. European Commission (2021a), “Commission Proposals to Improve the Working Conditions of People Working Through Digital Labour Platforms”, press release, 9 December. Available: https://​ec​.europa​.eu/​social/​main​.jsp​?langId​=​en​&​catId​=​89​&​ newsId​=​10120​&​furtherNews​=​yes [accessed 27 January 2022]. European Commission (2021b), Approval of the Content of a Draft Communication from the Commission: Guidelines on the Application of EU Competition Law to Collective Agreements Regarding the Working Conditions of Solo Self-Employed Persons, Annex, C(2021) 8838 final. European Commission (2021c), Proposal for a Directive of the European Parliament and of the Council on Improving Working Conditions in Platform Work, COM(2021) 762 final. European Parliament (2020), Report on a Strong Social Europe for Just Transitions, A9-0233/2020, 24 November. Fairwork (2021), “Fairwork Response to the European Commission’s Proposal for a Directive on Platform Work”, 9 December. Available: https://​fair​.work/​en/​fw/​ blog/​fairwork​-response​-to​-the​-european​-commissions​-proposal​-for​-a​-directive​-on​ -platform​-work/​[accessed 27 January 2022). Faulkner, C. (2021), “New Laws: How Spain Plans to Empower its Precarious Workers”, The Local, 27 December. Available: https://​www​.thelocal​.es/​20211227/​ [accessed 27 new​-laws​-how​-spain​-plans​-to​-empower​-its​-precarious​-workers/​ January 2022]. Finck, M. (2018), “Digital Co-regulation: Designing a Supranational Legal Framework for the Platform Economy”, European Law Review, Vol. 43, No. 1, pp. 47–68. Finkin, M. (2016), “Beclouded Work in Historical Perspective”, Comparative Labor Law and Policy Journal, Vol. 37, No. 3, pp. 603–18. Freedland, M. & Kountouris, N. (2017), “Some Reflections on the ‘Personal Scope’ of Collective Labour Law”, Industrial Law Journal, Vol. 46, No. 1, pp. 52–71. Goodkind, N. (2021), “Here’s How the PRO Act Would Impact Freelance and Gig Workers”, Fortune, 1 April. Haipeter, T. & Iudicone, F. (2020), “New Social Initiatives on Cloud- and Gigwork – Germany and Italy Compared”, Hungarian Labour Law E-Journal, Vol. 2020, No. 2, pp. 17–40.

68

Missing voice?

Haipeter, D., Owczarek, D., Faioli, M. & Iudicone, F. (2020), Don’t Gig Up! Final Report (Brussels: European Commission). Hardy, T. & McCrystal, S. (2020), “Bargaining in a Vacuum? An Examination of the Proposed Class Exemption for Collective Bargaining for Small Businesses”, Sydney Law Review, Vol. 42, No. 3, pp. 311–42. Hartmann-Cortés, K. (2021), “How Did a Food-Delivery Platform’s Judgement Transform Freedom of Association into a Second-Class Right?”, Comparative Labor Law and Policy Journal, Dispatch No. 37 – Argentina. Heiland, H. (2020), Workers’ Voice in Platform Labour: An Overview, Study No. 21, WSI (Institute of Economic and Social Research) (Düsseldorf: Hans-Böckler-Stiftung). Hießl, C. (2021), “The Classification of Platform Workers in Case Law: Cross-European Comparative Analysis”, Comparative Labour Law & Policy Journal, Vol. 42, No. 2, pp. 465–517. ILAW [International Lawyers Assisting Workers] Network (2021), Taken for a Ride: Litigating the Digital Platform Model, Issue Brief. Available: https://​www​ .ilawnetwork​.com/​wp​-content/​uploads/​2021/​03/​Issue​-Brief​-TAKEN​-FOR​-A​-RIDE​ -English​.pdf [accessed 27 January 2022]. ILO (2019a), Policy Responses to New Forms of Work: International Governance of Digital Labour Platforms (Paper prepared for the 2nd Meeting of the G20 Employment Working Group under Japan’s Presidency 2019, 22–4 April, Tokyo). Available: https://​www​.ilo​.org/​global/​about​-the​-ilo/​how​-the​-ilo​-works/​multilateral​ -system/​g20/​reports/​WCMS​_713378/​lang​-​-en/​index​.htm [accessed 27 January 2022]. ILO (2019b), Interactions Between Workers’ Organizations and Workers in the Informal Economy: A Compendium of Practice (Geneva: International Labour Office). ILO (2020), Promoting Employment and Decent Work in a Changing Landscape, Report of the Committee of Experts on the Application of Conventions and Recommendations (articles 19, 22 and 35 of the Constitution) Report III (Part B) (Geneva: International Labour Office). ILO (2021), World Employment and Social Outlook 2021: The Role of Digital Labour Platforms in Transforming the World of Work (Geneva: International Labour Office). James, N. (2020), Report of the Inquiry into the Victorian On-Demand Workforce (Melbourne: Victorian Government). Johnston, H. (2020), “Labour Geographies of the Platform Economy: Understanding Collective Organizing Strategies in the Context of Digitally Mediated Work”, International Labour Review, Vol. 159, No. 1, pp. 25–45. Johnston, H. & Land-Kazlauskas, C. (2018), Organizing On-Demand: Representation, Voice, and Collective Bargaining in the Gig Economy, Conditions of Work and Employment Series No. 94 (Geneva: International Labour Office). Kilhoffer, Z. et al. (2020), Study to Gather Evidence on the Working Conditions of Platform Workers, VT/2018/032, Final Report (Luxembourg: Publications Office of the European Union). Kim, E. K. (2020), “Labor’s Antitrust Problem: A Case for Worker Welfare”, Yale Law Journal, Vol. 130, No. 2, pp. 428–76. Lianos, I., Countouris, N. & De Stefano, V. (2019), “Re-thinking the Competition Law/ Labour Law Interaction: Promoting a Fairer Labour Market”, European Labour Law Journal, Vol. 10, No. 3, pp. 291–333.

Giving platform workers a say

69

Maniscalco, J. (2016), “Union Agreement with Uber Jeered as a ‘Surrender’”, Labor Press, 16 May. Available: https://​www​.laborpress​.org/​union​-agreement​-with​-uber​ -jeered​-as​-a​-surrender/​ [accessed 27 January 2022]. Mason, L. (2020), “Locating Unity in the Fragmented Platform Economy: Labor Law and the Platform Economy in the United Kingdom”, Comparative Labor Law & Policy Journal, Vol. 41, No. 2, pp. 329–42. McCrystal, S. & Hardy, T. (2021), “Filling the Void? A Critical Analysis of Competition Regulation of Collective Bargaining Amongst Non-Employees”, International Journal of Comparative Labour Law and Industrial Relations, Vol. 37, No. 4, pp. 355–84. McGurdy, C. (2021), “Uber and GMB’s Deal: A New Era for Trade Unionism?”, Social Review, 1 August. Available: https://​www​.thesocialreview​.co​.uk/​2021/​08/​ 01/​uber​-and​-gmbs​-deal​-a​-new​-era​-for​-trade​-unionism/​ [accessed 27 January 2022]. Mexi, M. (2019), Social Dialogue and the Governance of the Digital Platform Economy: Understanding Challenges, Shaping Opportunities, background paper for discussion at the ILO-AICESIS-CES Romania International Conference (Bucharest, 10–11 October 2019). Available: https://​www​.ilo​.org/​global/​about​-the​-ilo/​how​-the​ -ilo​-works/​departments​-and​-offices/​governance/​dialogue/​WCMS​_723431/​lang​-​-en/​ index​.htm [accessed 27 January 2022]. Minter, K. (2017), “Negotiating Labour Standards in the Gig Economy: Airtasker and Unions New South Wales”, Economic and Labour Relations Review, Vol. 28, No. 3, pp. 438–54. Mojtehedzedeh, S. (2020), “Foodora Couriers Win $3.46 million Settlement Following Food-Delivery Giant’s Abrupt Canadian Closure”, Toronto Star, 25 August. Mulvaney, E. (2021), “Uber Will Push to Shape Direction of Biden Gig Worker Regulation”, Bloomberg Law, 12 March. Nelson, D. (1982), “The Company Union Movement, 1900–1937”, Business History Review, Vol. 8 (Autumn), pp. 335–57. Novitz, T. (2020), “The Potential for International Regulation of Gig Economy Issues”, King’s Law Journal, Vol. 31, No. 2, pp. 275–86. O’Donovan, C. (2020), “Uber and Lyft Spent Hundreds of Millions to Win their Fight Over Workers’ Rights. It Worked”, BuzzFeed News, 21 November. Available: https://​www​.buzzfeednews​.com/​article/​carolineodonovan/​uber​-lyft​-proposition​-22​ -workers​-rights [accessed 27 January 2022]. OECD (2019), Negotiating Our Way Up: Collective Bargaining in a Changing World of Work (Paris: OECD Publishing). Ombuds Office of the Code of Conduct for Paid Crowdsourcing (2020), Report of the Activities of the Ombuds Office of the Code of Conduct for Paid Crowdsourcing, 2019. Available: https://​ombudsstelle​.crowdwork​-igmetall​.de/​pdf/​Ombuds​_Office​ _for​_paid​_Crowdsourcing​_Report​_2019​_EN​.pdf [accessed 27 January 2022]. Paul, S. (2017), “Uber as For-Profit Hiring Hall: A Price-Fixing Paradox and Its Implications”, Berkeley Journal of Employment and Labor Law, Vol. 38, No. 2, pp. 233–64. Platoff, E. (2021), “Gig Worker Petitions Move One Step Closer to 2022 Ballot”, Boston Globe, 27 December. Prassl, J. (2018), Collective Voice in the Platform Economy: Challenges, Opportunities, Solutions (Brussels: European Trade Union Confederation). Rodríguez, L. (2022), “First Collective Agreement for Platform Workers in Spain”, Social Europe, 13 January. Available: https://​socialeurope​.eu/​first​-agreement​-for​ -platform​-workers​-in​-spain [accessed 27 January 2022].

70

Missing voice?

Rodriguez-Pinero Royo, M. (2020), “Platforms and Platform Work in Spanish Industrial Relations”, Comparative Labor Law & Policy Journal, Vol. 41, No. 2, pp. 443–71. Roosevelt, M. & Hussain, S. (2021), “Prop. 22 is Ruled Unconstitutional, a Blow to California Gig Economy Law”, Los Angeles Times, 20 August. Sarina, T. & Riley, J. (2018), “Re-crafting the Enterprise for the Gig-Economy”, New Zealand Journal of Employment Relations, Vol. 43, No. 2, pp. 27–35. Scheiber, N. (2017), “Uber Has a Union of Sorts, but Faces Doubts on Its Autonomy”, New York Times, 12 May. Schiek, D. & Gideon, A. (2018), “Outsmarting the Gig-Economy Through Collective Bargaining – EU Competition Law as a Barrier to Smart Cities?”, International Review of Law, Computers & Technology, Vol. 32, Nos. 2–3, pp. 275–94. Stanford, J. (2017), “The Resurgence of Gig Work: Historical and Theoretical Perspectives”, Economic and Labour Relations Review, Vol. 28, No. 3, pp. 382–401. Stanford, J. & Poon, D. (2021), Speaking Up, Being Heard, Making Change: The Theory and Practice of Worker Voice in Canada Today (Vancouver: Centre for Future Work). Steinbaum, M. (2019), “Antitrust, The Gig Economy, and Labor Market Power”, Law and Contemporary Problems, Vol. 82, No. 3, pp. 45–64. Stewart, A. & McCrystal, S. (2019), “Labour Regulation and the Great Divide: Does the Gig Economy Require a New Category of Worker?”, Australian Journal of Labour Law, Vol. 32, No. 1, pp. 4–22. Stewart, A. & Stanford, J. (2017), “Regulating Work in the Gig Economy: What Are the Options?”, Economic and Labour Relations Review, Vol. 28, No. 3, pp. 420–37. Tamma, P. (2020), “Italy’s Labor Ministry Slams Delivery Company Deal with Far-Right Union”, Politico, 17 September. Available: https://​ www​ .politico​ .eu/​ article/​italys​-labor​-ministry​-slams​-delivery​-company​-deal​-with​-far​-right​-union/​ [accessed 27 January 2022]. Taras, D. & Kaufman, B. (2006), “Non-union Employee Representation in North America: Diversity, Controversy and Uncertain Future”, Industrial Relations Journal, Vol. 37, No. 5, pp. 513–42. Todoli-Signes, A. (2021), “Spanish Riders Law and the Right to be Informed about the Algorithm”, European Labour Law Journal, Vol. 12, No. 3, pp. 399–402. Toh, J. (2021), “Another Win for Workers: Uber Drivers are Employees”, Social Europe, 22 September. Available: https://​socialeurope​.eu/​another​-win​-for​-workers​ -uber​-drivers​-are​-employees [accessed 27 January 2022]. Unions NSW (2016), Innovation or Exploitation: Busting the Airtasker Myth (Sydney: Unions NSW). Unions NSW (2018), “Submission – Victorian Inquiry into On-demand Workers”. Available: https://​engage​.vic​.gov​.au/​download/​document/​7374 [accessed 27 January 2022]. Vandaele, K. (2018), Will Trade Unions Survive in the Platform Economy? Emerging Patterns of Platform Workers’ Collective Voice and Representation in Europe (Brussels: European Trade Union Institute). Wilkinson, A., Donaghey, J., Dundon, T. & Freeman, R. B. (eds) (2020), Handbook of Research on Employee Voice (2nd ed., Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing). Wright, C. et al. (2019), “Towards a New Web of Rules: An International Review of Institutional Experimentation to Strengthen Employment Protections”, Employee Relations, Vol. 41, No. 2, pp. 313–30.

4. How do workers gain voice on digital work platforms? Hotspots and blind spots in research on platform worker voice Laura Thäter, Thomas Gegenhuber, Elke Schüßler and Markus Ellmer INTRODUCTION Having emerged about one decade ago, digital work platforms have become an established phenomenon in the contemporary work landscape. In the most general terms, digital work platforms mediate between demand (clients consuming a service) and supply (workers providing the service) of labour (Howcroft & Bergvall-Kåreborn, 2019; Kenney & Zysman, 2016; Kuhn & Maleki, 2017), using internet and mobile technology to coordinate dispersed actors and decentralize work processes. At the same time, platforms centralize information and discipline workers through algorithms, rankings and reputation systems (e.g. Attwood-Charles, 2019; Kornberger et al., 2017). The work on these platforms can either take place completely online, commonly called cloudwork or crowdwork1 (Kaganer et al., 2013; Shafiei Gol et al., 2019) or offline, commonly called gig work (Maffie, 2020). For example, crowdworkers from around the world perform micro-tasks (e.g. training AI) on platforms such as Amazon Mechanical Turk or Clickworker. But crowdwork also takes place in the professional services sector, where a high skill level is required. This includes consulting, IT services, or graphic design (e.g. UpWork). Gig work takes place on platforms for food delivery,2 mystery shopping or ride-hailing.3 Whether online or offline, an increasing number of scholars from different disciplines, ranging from information systems to legal studies as well as sociology and management studies, are concerned about the precarity of these new work arrangements, with platform workers falling outside of existing classifications for work regulation and protection (e.g. De Stefano, 2016; Dubal, 2017) and being subjected to exploitative, controlling 71

72

Missing voice?

algorithmic management techniques (e.g. Griesbach et al., 2019; Kellogg et al., 2020; Rahman, 2021). As Wilkinson et al. (2021) argue, social, economic and technological developments regularly disrupt existing patterns and norms for how work is approached, organized, and designed. One critical aspect when it comes to platform-based work is the provision of worker voice, that is, “the ways and means through which employees attempt to have a say, formally and/or informally, collectively and/or individually, potentially to influence organizational affairs relating to issues that affect their work, their interests, and the interests of managers and owners” (Wilkinson et al., 2021). While acts of speaking up pursue the goal of bringing about change rather than accepting the current states of affairs or exiting the work arrangement (Farrell, 1983; Hirschman, 1970), the ways in which workers can actually participate in co-shaping the conditions of their work arrangements are highly affected by social, economic and technological context conditions. Digital work platforms use technological developments – particularly advances in internet technology, cloud and mobile services (e.g. Srnicek, 2017) – to reorganize how work is distributed and governed, advancing contingent work structured around ad hoc tasks rather than jobs and careers (Davis, 2016a, 2016b; Kirchner & Schüßler, 2020; Lehdonvirta et al., 2019; Vallas, 2019; Vallas & Schor, 2020). This implies individualization and a lack of formal representation, but also increasing contestation by workers that, depending on the specific work context, are nonetheless able to mobilize for a collective platform worker voice (Wood & Lehdonvirta, 2021). In this chapter, we aim to provide an overview over the “hotspots” and “blind spots” regarding voice in digital work platform contexts. We understand hotspots to mean the research areas on which much literature is currently focusing, not least because of pressing empirical developments, whereas blind spots are topic areas where existing literature is meagre and to which future research needs to pay more attention. We see a growing body of literature investigating the multiple ways in which platform workers may gain voice and what barriers workers face in these contexts (e.g. Gegenhuber et al., 2020; Maffie, 2020; Vandaele, 2018; Wood et al., 2018), with voice initiatives driven by either workers, unions, regulators or, sometimes, by platforms themselves. However, the current state of the literature remains quite fragmented and lacks a systematic, bigger picture. Against this background, the main goal of our chapter is to give an overview of how platform workers, platforms, unions and regulators attempt to build and extend the means through which platform workers can influence platform-related affairs. We are thereby building on a framework that differentiates the continuums of “formal” and “informal” voice as well as on “individual” and “collective” voice. Following Mowbray et al. (2015), we define formal forms of voice as codified, pre-arranged, and

How do workers gain voice on digital work platforms?

73

regular/concrete structures that foster consistent implementation (see also Klaas et al., 2012; Marchington & Suter, 2013). Informal forms of voice refer to ideas or concerns expressed directly and outside a structured process through ad hoc or non-programmed interactions between management and their staff. Based on the emerging picture of different forms and shapes of voice in platform work, their related key issues, as well as missing or under-researched forms of voice, we derive four policy implications: First, workers need to be able to network and organize. Second, traditional unions need to be able to access platform workers directly. Third, unions must develop sustainable digital unionizing strategies. Fourth, policymakers need to consider voice in platform regulations. Our chapter contributes to the growing body of industrial relations literature pointing out the relevance of voice in digital work contexts (Wilkinson et al., 2018), to the sociology of work and platform literature concerned with labour standards in the digital economy (e.g. Ravenelle, 2019; Robinson, 2017; Rosenblat & Stark, 2016), and to the growing debates in management literature on how to counter the “dark sides” of the platform and sharing economy (Maurer et al., 2020; Trittin-Ulbrich et al., 2021)

THE NEED FOR VOICE ON DIGITAL WORK PLATFORMS AND ITS HURDLES Within public debates as well as scholarly literature, we find increasing criticism regarding the labour conditions on digital work platforms: platforms fuel precarious work conditions, pay dumping “wages” far below the legal minimum, have opaque work processes disadvantaging and ignoring workers’ needs while favouring clients, have a negative effect on gender equality and ignore instances of mobbing and sexual violence (Barzilay & Ben-David, 2018; Cook et al., 2018; Wood et al., 2019a, 2019b). These issues raise concerns regarding the opportunity for platform workers to express their voice (Healy et al., 2017; Wilkinson et al., 2021). Indeed, there are many structural obstacles to platform workers’ voice. First, the platform workers’ social structure impedes organizing efforts. Particularly in crowdwork, the workers are dispersed, anonymous, and often cannot interact with each other. The many-to-many relationships platforms establish, combined with minimal opportunities for expression on the platforms’ infrastructures, imply a large number of parallel and mutually invisible employment relations (Felstiner, 2011; Hoose et al., 2019; Howcroft & Bergvall-Kåreborn, 2019; Irani & Silberman, 2013; Johnston & Land-Kazlauskas, 2018). Similarly, large digital work platforms seek to mobilize large numbers of crowdworkers. The resulting oversupply of labour combined with the crowdworkers’ global dispersion fuels the platforms’ power, as it strengthens their bargaining position by enabling mechanisms

74

Missing voice?

of global labour arbitrage (Aytes, 2012; Bauer & Gegenhuber, 2015; Ellmer, 2015). Although these mechanisms do not apply to gig work in the same way because of the workers’ local embeddedness and the resulting lower labour supply, research suggests that gig workers also lack voice opportunities and thus suffer under the power imbalance (Vandaele, 2018). Second, platforms are in a powerful position because of their roles as market organizers (Kirchner & Schüßler, 2019). Platform managers set the rules of the game and control the platform interface and corresponding algorithms for managing the work process. This encompasses elements such as “career systems” (e.g. who sees and can access what kind of jobs), deploying evaluation mechanisms and defining opportunity structures (e.g. clients evaluate crowdworkers but not vice versa), payment systems (e.g. payment only if the client accepts the task), and whether or how the platforms use community management (Bucher et al., 2021; Dickel & Thiem, 2018; Hoose et al., 2019; Vandaele, 2021). Workers could, of course, exit the platform in protest at the prevailing precarious working conditions. However, as platform reputation is not transferrable from one platform to another, workers – at least those depending on platform income – face a high exit barrier and are, as a result, locked in to the platform. Consequently, the relationship between a platform and its workers is characterized by information asymmetries and power disparities, typically to the disadvantage of the workers. Third, voice needs to be institutionally embedded in employment relations or collective worker organization. Such institutions are typically missing in platform work, because platforms seek to avoid legally classifying their workers as “workers” (Aloisi, 2019; De Stefano, 2016). Additionally, platform work is a global phenomenon that reaches across national boundaries, so even if changes to national regulations are made, they are unlikely to have a widespread impact, at least when it comes to globally dispersed crowdwork.

HOW DO WORKERS GAIN VOICE ON DIGITAL WORK PLATFORMS? Given that platform worker voice is a relatively new phenomenon, but one which has received much interest in the 2020s, we should start by reviewing extant literature to provide an overview of how platform workers gain voice and structure these insights before identifying the hotspots and blind spots in the existing research (see the Appendix for an overview of the articles we examined, as well as the search and selection procedure). We differentiate four avenues towards platform worker voice: worker-driven, platform-driven, union-driven and regulation-driven, depending on who speaks up or enables voice. The transition between these categories is fluid, however. This becomes particularly apparent when looking at initiatives over time; for instance, many

How do workers gain voice on digital work platforms?

75

initiatives start as worker-driven organizing efforts, but unions get involved later on throughout the process (e.g. Niebler & Kern, 2020). Worker-driven Voice Worker-driven voice comprises all worker-initiated efforts to influence platform-related affairs. Here, the clear focus of the current literature is on how workers attempt to organize autonomously. The form of these autonomous organizing efforts varies greatly. As portrayed in different case studies in the food delivery industry (see e.g. Animento et al., 2017; Tassinari & Maccarrone, 2017), worker-driven organizing can take the form of spontaneous direct-action initiatives (e.g. strikes). However, we also found evidence for engagement in more democratic forms of organization, such as grassroots unions (see e.g. Niebler & Kern, 2020) or worker associations (see e.g. Heiland & Brinkmann, 2020). The empirical studies by Aloisi (2015), Anwar & Graham (2020), Lehdonvirta (2016), Niebler (2020) and Wood et al. (2018) show that workers engaging in crowdwork rely heavily on digital communication tools such as social media groups, forums, or instant messenger applications to build a shared identity, find a collective voice and structure their organizing efforts. However, Wood et al.’s (2018) study among crowdworkers suggests that the type of digital communication tool used depends on the type of task (routine or creative tasks) the workers engage in. They argue that workers performing creative tasks use social media groups (e.g. on Facebook), whereas workers fulfilling routine tasks use distinct, platform-related forums (e.g. TurkerNation on Reddit). Following the authors’ line of argumentation, the explanation for this pattern lies in the different nature of the tasks the workers do, as creative workers build relationships with other creatives (e.g. getting feedback, outsourcing minor tasks to other creatives) and seek to form direct ties to clients. This requires a certain amount of trust, which can be built better via social media platforms because users do not interact anonymously. Sutherland et al.’s (2020) findings support the importance of platform-related forums for workers performing routine tasks. Moreover, Lehdonvirta (2016) found that different socio-economic backgrounds make it difficult for crowdworkers to build a shared identity or find a collective voice. A shared identity – traditionally a class consciousness (Fantasia, 1995), which has already largely disappeared during globalization and marketization (Hyman, 1999) – is important for workers’ collective organization. Workers have different needs and expectations due to their global diffusion (e.g. workers’ needs in developed countries like the USA differ greatly from those in emerging countries like the Philippines). Furthermore, Lehdonvirta (2016) did not find evidence that the building of a shared identity

76

Missing voice?

via digital communication tools in the context of crowdwork resulted in effective collective action. In contrast, Aloisi’s (2015) findings from a comparative case study on crowd and gig workers suggest that organization via digital communication tools is often the first step to a more sustainable form of collective organization (e.g. worker associations or grassroots unions). Niebler and Kern (2020) investigated the formation of a social movement among globally dispersed YouTube creators. The authors found that the use of social media platforms can enhance worker power and, by extension, collective action that goes beyond the usually fragmented and spontaneous forms. While echoing Aloisi’s (2015) findings, Niebler (2020) nonetheless argues that the YouTuber case might be an exception rather than a rule, because a critically large number of creators have been affected by the platform’s governance changes at the same time. This might have been a driver for the autonomous self-organization. Aloisi’s (2015) and Maffie’s (2020) studies support the importance of digital communication tools also in the context of gig work. Various studies (e.g. Hoose et al., 2019; Tassinari & Maccarrone, 2017) have found that spatial proximity is a driver for successful collective action in the platform economy because face-to-face contact between workers makes organization significantly easier. Thus, engaging in collective action is generally more widespread among gig workers. The study by Animento et al. (2017) investigating direct action in the Italian food delivery industry explains that engaging in on-site strikes not only fulfils the aim of improving working conditions, but also of strengthening the network among the workers. Approaches to institutionalizing worker voice structures in gig work are still in their infancy. Successful examples are a rare exception (but see Chapter 9 by Cianferoni et al.). Heiland and Brinkmann (2020) conducted a study in the German food delivery industry in which they found that labour platforms intentionally exploit their position of power to reduce transparency and increase uncertainty for the workers. They thereby actively hinder the establishment of formal participation structures like works councils. On the other hand, Wood and Lehdonvirta (2019) revealed that it is exactly that structural antagonism between workers and the platform that fuels collective action (see also Wood & Lehdonvirta, 2021). Moreover, the studies by Barratt et al. (2020) and Karanović et al. (2020) show that the broader institutional context in which workers are situated influences their autonomous organizing. Barratt et al. (2020) conducted their study in the Australian food delivery industry. The Australian labour market is characterized by high rates of unemployed and underemployment, as well as a high number of migrant workers. These circumstances influence how the workers perceive their role in the production process and hinder their collective agency. As a result, platform workers adopt a rather entrepreneurial mindset. This

How do workers gain voice on digital work platforms?

77

mindset, in turn, leads to workers aiming at improving working conditions on an individual level and in alignment with the general business model of the labour platform rather than in opposition to it. Research by Karanović et al. (2020) further shows that the regulatory environment affects worker agency. Their comparative case study in the ride-hailing industry suggests that, when platform regulation is rather strict, platform workers tend to supplement the platform’s solution (e.g. by providing other workers with additional information). If the regulatory environment for the platforms is more lenient, workers react antagonistically and oppose the platform. The authors attribute this finding to the mutual dependence of platform workers and labour platforms in directly regulated environments. To sum up, digital communication tools play a vital role in the formation of collective identity and voice, both in crowd and gig work. However, the majority of studies suggests that acting collectively based on a shared identity is somewhat more likely in gig work due to the workers’ spatial proximity.4 Even though there are some exceptional cases, analysis of the empirical studies further demonstrates that sustaining the momentum after spontaneous collective action is difficult in general, regardless of the type of platform work. Nonetheless, since we identified more cases of sustainable worker organization in gig work, we conclude that the spatial proximity of the workers is an essential factor. Furthermore, at least in gig work, there seem to be context factors, like the regulatory environment and the nature of the national labour market, that influence worker agency as well as the aim of collective action. Platform-driven Voice Platform-driven voice encompasses all instances where voice structures that enable workers to influence platform-related affairs are established by for-profit platforms themselves, as well as by academics providing voice platforms and by alternative platform cooperatives. The various governance forms and the corresponding type of motivation impact the voice arrangements’ design. As already stated, platforms generally have low interest in empowering their workers because the supply of labour exceeds demand. Therefore, it is no surprise that platform-driven voice and participation initiatives are a rare exception. However, if we move away from large, US-based platforms to medium or smaller-sized ones based in other countries, the conditions appear to change. Gegenhuber et al.’s (2020) findings from a comparative case study of six medium-sized platforms in Germany show that these platforms did provide voice mechanisms in many different forms (e.g. contact forms, forums, rating systems, votes); however, mostly on a functional level. This means that the degree to which workers can influence the platform’s operations is largely

78

Missing voice?

limited to improving and optimizing organizational performance and client satisfaction rather than empowering them to exert control. One of the two cases in the literature reviewed of more far-reaching platform-initiated improvements to working conditions is the collective agreement between the Danish platform Hilfr, which mediates housekeeping tasks, and the union 3F. According to Ilsøe and Jesnes (2020), the motivation for the platform to initiate this agreement was rooted in its business strategy, as it perceived fair working conditions as a way to differentiate itself from other platforms in the market. Thus, the platform was aiming for a competitive advantage. The second case is the German Code of Conduct for crowdworkers analysed by Gegenhuber et al. (2021). A handful of German platforms mediating (mostly) crowdwork, together with the German platforms’ industry association, initiated a Code of Conduct for fair platform work that encompasses guidelines on governance, payment systems, community management and feedback systems. The platform providers in this sample wish to distinguish themselves from the negative example of American work platforms, thus following a similar motivation as the Danish example. Additionally, there are a few cases in the platform economy where the platform’s governance deliberately embraces worker participation and voice. Examples here are platform cooperatives and platforms established as part of academic social activism initiatives. Academic activism initiatives range from building a platform add-on to empower the workers, to setting up a new platform as an alternative to the existing ones. In all these cases, the governance of the platform is designed to give workers a more active role right from the beginning. An example of an academic social activism initiative is Turkopticon, which is a browser add-on that enables MTurk workers to rate “requesters” (i.e. the parties putting work on MTurk) on the basis relevant criteria (fairness, communicativeness, promptness) and making their payment structures transparent by adding a rating system to the MTurk interface (Irani & Silberman, 2013). Another example of academic social activism was the crowd-governed marketplace Daemo (Salehi et al., 2015; Whiting et al., 2017). Within the Daemo initiative, a so-called constitution was ratified by all stakeholders that created a formal process for crowdworkers to propose changes and participate actively in the governing processes of the platform. Meanwhile, however, the platform has gone offline. Platform cooperatives give workers a key role in platform governance from the start. Just as in traditional cooperatives,5 workers (or all involved stakeholders) own the platform and collectively decide on the platform’s governance, its further development, and strategies (Scholz, 2016; Thäter & Gegenhuber, 2020). This set-up mitigates the shareholder-driven conflict between short-term profits and workers’ interests. However, research on platform cooperatives is mostly anecdotal. It is an open question to what extent a cooperative automatically results in “good”

How do workers gain voice on digital work platforms?

79

working conditions since even a cooperative is unlikely to escape all rules of the game within an industry. Together, platform-driven worker participation and voice are rare in the platform economy, especially in the for-profit sector and among larger platforms. Our analysis shows that platform entrepreneurs who are willing to provide workers with voice and participation structures often do so out of strategic motives. As particularly smaller to medium-sized platforms struggle with worker turnover rates, the platforms’ decision to provide workers with a voice and participation structures on a functional level is motivated by a desire to positively differentiate themselves from their competitors and create an incentive for the workers to keep working on the platform. Therefore, voice on a functional level is the rule, whereas worker voice on a more fundamental level (e.g. the ability to control or influence strategic decisions) is the exception. In the cases where the platforms’ governance is set up differently from the beginning (as a cooperative or as part of academic efforts), studies suggest that there are opportunities for more fundamental worker voice and participation resulting from the initiators’ worker-oriented value and belief system. However, in both cases of academic engagement (Daemo and Turkopticon), sustaining the operation of the platform over time is difficult because academia is not designed for engaging in social activism over the long term. The Turkopticon initiators tried to solve this problem by involving the open-source community, whereas the Daemo initiative ceased to exist.6 Union-driven Voice Union-driven voice includes all organizing initiatives where an already established labour union of the “old economy” gets involved in influencing platform-related affairs.7 This involvement ranges from initiating organizing efforts to supporting bottom-up initiatives (e.g. collaboration with grassroots unions) or forging new partnerships (e.g. negotiating governance directly with platform managers). The degree and duration of union involvement vary from one-time engagements for action at the micro-level (e.g. supporting a strike) to long-term involvement in the form of collective bargaining at the meso and macro level (Fernàndez & Barreiro, 2020; Ilsøe & Jesnes, 2020; Vandaele, 2020). Union engagement in the context of platform work is generally rather low, as the phenomenon is still quite novel. Vandaele (2020) argues that the unions engaging in platform worker organizations do so to strengthen their power resources and to portray unions as modern, relevant organizations to the public (see also Gegenhuber et al., 2021). Several studies (e.g. Cant & Woodcock, 2020; Vandaele, 2020) found that the novel form of work organization in digital labour markets generally renders traditional union strategies unsuitable. Therefore, unions are faced with the necessity of developing crea-

80

Missing voice?

tive new strategies to support platform workers in their fight for better working conditions. In a comparative case study of union strategies in the British food delivery sector, Cant and Woodcock (2020) uncovered that the evolution of successful strategies usually adheres to the following pattern: first, unions identify a workplace-specific grievance on a small (mostly local) scale. Having mapped the existing solidarities as well as informal workgroup structures, unions move to support collective action on that small-scale grievance. In the next step, unions can then escalate collective action to larger-scale grievances. The findings from Vandaele’s (2020) comparative study in the Belgian and the Dutch food delivery sector are in line with those of Cant and Woodcock’s study. Vandaele (2020) uncovered that union support for worker organization in the platform economy is strongly driven by individual unionists on a local level. This might also be an explanation for the fact that most cases of union-supported organizing that we identified (e.g. Fernàndez & Barreiro, 2020; Ilsøe & Jesnes, 2020; Johnston & Pernicka, 2021) are situated in the context of gig work. Moreover, the majority of cases we reviewed are cases where union engagement was short-term only (e.g. by supporting workers in the organization of (spontaneous) direct action initiatives) (e.g. Fernàndez & Barreiro, 2020; Vandaele, 2020). However, Fernandez and Barreiro’s (2020) study of Spanish food delivery riders showed that even the short-term union support was crucial to sustaining the workers’ autonomous organization efforts by helping to foster collective voice and enhance learning. Johnston and Pernicka (2021) found in their comparative case study situated in the European and the US ride-hailing industry that union strategies also vary depending on their societal position within countries. In places where unions traditionally have a rather strong position of power, as in the EU, unions tend to rely on existing advocacy mechanisms (like political lobbying for a minimum wage). Therefore, it comes as no surprise that the first collective agreement between a labour platform and a union was concluded in Norway (Ilsøe & Jesnes, 2020). The rare examples of union-supported organizing in crowdwork deviate from the previously elaborated bottom-up pattern uncovered in gig work. To give an example, the case of the FairTube campaign (Niebler & Kern, 2020) illustrated that the support of the German trade union IG Metall for YouTubers had strong backing within the union. The study also shows that the chosen union strategy differs from those uncovered in gig work: here the union leveraged its discursive power by generating international media attention for the social movement. The German Code of Conduct case (Gegenhuber et al., 2021) reveals similar patterns. Having been approached by several German platform providers and the German platform industry association, IG Metall initiated a new format to encourage a social dialogue between capital

How do workers gain voice on digital work platforms?

81

and labour. While the platforms had already self-committed to the Code of Conduct, which was mostly about procedural concerns, the union pushed for stronger accountability, leading to a revision of the code and the establishment of an “ombuds office” responsible for hearing and ruling on crowdworker complaints across several platforms. In this case, union engagement led to the establishment and formalization of voice and participation structures. Since national and international policymakers consider the German “social partnership” model, including enforcement and accountability mechanisms, in their policy papers, the German model might serve as a template for further platform worker voice initiatives. Gegenhuber et al. (2021) contrast the German initiative with other approaches such as the World Economic Forum’s Charter of Principles for Good Platform Work, which is signed by major platforms such as Uber and Deliveroo. The World Economic Forum approach clearly lacks accountability mechanisms and can therefore be seen as more of a corporate social responsibility (CSR) initiative that is decoupled from the actual work practices on the platforms. Regulation-driven Voice Regulation-driven voice comprises all regulatory efforts on the part of public actors on different levels (local, state or national) to strengthen workers’ rights and systemically embed avenues for voice to influence platform-related affairs. Even though there is a severe time lag between the rapidly growing platform labour markets and the political processes leading to regulations (Hinings et al., 2018; Kirchner & Schüßler, 2019), public regulators, on a state or local level (often pushed by worker protests), have reacted to the platform work phenomenon in various ways. They can therefore use a variety of more accommodating and more restrictive approaches and follow different strategies to pave this new legal territory (Frenken et al., 2020; Gorwa, 2019). First, regulators can apply existing laws to platforms (e.g. by classifying them as temp agencies or even employers because of their direct influence over the work process). Several lawsuits in the EU have been fought over this issue, with varying outcomes (De Stefano & Aloisi, 2018; Urzì Brancati et al., 2019). For instance, the Independent Workers Union of Great Britain fought for platform workers in the UK to be guaranteed a minimum wage and to receive holiday pay. However, they were not successful in guaranteeing a right to collective bargaining. In California, a 2019 law makes it harder for platforms to classify rideshare drivers as self-employed. However, large platforms like Uber and Lyft successfully mobilized their users and workers to fight against this law (Hawkins, 2020). This also indicates that workers’ interests are not homogeneous (see Vallas & Schor, 2020). The European Union has acknowledged that, if certain criteria defined by the European Court of Justice are met,

82

Missing voice?

existing laws apply to platform workers (DGB, 2021). For example, a German court ruled in favour of a worker’s claim against the deactivation of his account on the ground that the worker cannot, among other things, freely decide when and where to work, and is thus practically an employee. Therefore, the worker is covered by worker protection (Fuhlrott, 2020). Second, new laws can be shaped that create a third category between dependent and independent workers, which allows them to maintain flexibility but provides for some basic social and economic security as well as reduces the workers’ legal insecurity (Harris & Krueger, 2015). Although a legal category for dependent self-employed workers already exists in many European countries, the category does not adequately fit the situation of platform workers (Ellmer et al., 2019). For this reason, different governments, such as the UK government, are investigating the possibility of creating a new category for “dependent contractors” or “independent workers” (Taylor et al., 2017). Furthermore, regulators could provide incentives, such as tax allowances, to platforms to strengthen more democratic governance structures (Martin et al., 2017). Apart from that, laws can regulate participation structures, such as the right to stage a walk-out, engage in strikes and to organize collectively. However, to the best of our knowledge, regulation has paid scant attention to this issue to date.

THE HOTSPOTS AND BLIND SPOTS IN PLATFORM WORKER VOICE RESEARCH Our literature review identifies four key sources of voice initiatives (i.e. workers, platforms, unions and regulators). In the following, we take a look at the different sources of initiatives through our analytical framework to subsequently identify hotspots and blind spots of extant research. Worker-driven initiatives have a strong focus on more informal and collective forms of voice. Through organizing in online forums or through local protests, these initiatives aim at improving the situation for many platform workers at once. In crowdwork contexts, these initiatives face difficulties of forming a collective identity via digital communication tools with decreasing spatial proximity and a high heterogeneity of needs and interests among platform workers. Initiatives driven by workers are rather spontaneous and local, and the effectiveness of voice resulting from these initiatives is quite mixed. Platform-driven voice mostly results in more formal, individual forms of voice. In the for-profit context, platforms provide features through which platform workers can speak up (e.g. contact forms, forums, rating systems, votes); academic initiatives either run platforms (such as Daemo) or provide add-ons to extend existing platforms (e.g. Turkopticon on MTurk). Some rare examples of platform cooperatives in the non-profit context give workers more formal

How do workers gain voice on digital work platforms?

83

and collective forms of voice. A key issue here is that for-profit platforms facilitate voice mainly based on instrumental motives and thus limit speaking up to rather functional issues (e.g. for improving workflows). In summary, platform-related initiatives mostly depend on the values of those who run the platform. Union-driven voice covers a broad range of formal as well as informal initiatives. Yet, we see a clear focus on collective voice mechanisms. Unions either support more informal, bottom-up initiatives (e.g. collaboration with workers or grassroots unions) or forge new partnerships (e.g. negotiating governance directly with platform managers), resulting in more formal forms of voice. In worker-driven initiatives, research uncovers a main procedural pattern: first, support of small grievances, then summing it up to a more collective movement to put pressure onto platforms. Most union-supported initiatives remain short-term; their success often depends on the societal position of unions in a country. Finally, regulation-driven voice focuses on formal and collective forms of voice. These initiatives attempt to apply existing laws to platforms (e.g. minimum wage) or to shape new laws (e.g. platform worker category in law). So far, however, they have had no success in guaranteeing freedom of association and collective bargaining. Table 4.1 sums up the review and gives a detailed overview. Looking at the reviewed literature from the lens of our analytical framework reveals hotspots, where research is concentrated, as well as blind spots, where further research is needed to better inform future policies related to platform worker voice (Figure 4.1). Overall, we can identify a clear hotspot in research focusing on formal types of voice aiming at speaking up in a collective fashion, covered by regulation, union- and worker-driven voice. Initiatives related to regulation-driven voice are clearly situated in the “formal-collective” quadrant: regulators attempt to provide formal schemes and laws across many platforms that make it possible for platform workers to improve their working conditions and organize. To some extent, union-led initiatives can also be found in this quadrant (e.g. when negotiating local policies with platforms). In some cases, regulation-driven voice and union-driven voice overlap, as one strategy of unions is to influence working conditions via regulation (see Vandaele, 2018). Yet, most of the union-led initiatives can be found in the “collective-informal” quadrant, as most unions only support the spontaneous and autonomous organization of platform workers for a specific cause and for a limited period of time. Worker-driven voice initiatives such as strikes can also be found in this quadrant. Most platform-driven initiatives fall into the “individual-formal” quadrant. Platforms provide mostly formal channels for individuals to speak up. In the for-profit context, this makes sense, as platforms are interested in voice contributing to business success (Barry &

84

Table 4.1

Missing voice?

The hotspots and blind spots in platform worker voice research

  Focus

formal

informal

individual

collective

 

+++

+

+++

– Online forums (Facebook, Reddit, etc.) Examples

– Local protests (e.g. in cities) – Grass roots unions

Worker-driven voice

– Difficulties of forming a collective identity via digital communication Key findings and issues

tools with decreasing spatial proximity – Heterogeneity of needs and interests with decreasing spatial proximity – Entrepreneurial identity of platform workers – Initiatives are rather spontaneous and local – Effectiveness of voice resulting from these initiatives is quite mixed

Focus

+++

 

+++

+

– For-profit: features in platform interfaces (e.g. contact forms, forums, Examples

rating systems, votes) – Academic: platforms (Daemo), add-ons (Turkopticon) – Non-profit: platform cooperatives

Platform-driven voice

– For-profit platforms allow voice mainly based on entrepreneurial Key findings and issues

motives (positive image on digital labour market, business strategy) and thus limit voice to rather functional issues – Platform-related initiatives are mostly driven by the values of those who run the platform

Focus

++

++

+

+++

– Unions supporting bottom-up initiatives (e.g. collaboration with Examples Union-driven

with platform managers)

voice Key findings and issues Focus Regulation-driven voice

grassroots unions) – Unions forging new partnerships (e.g. negotiating governance directly

Examples Key findings and issues

– Main pattern as process: first, support of small grievances, then consolidating it into a more collective movement – Most initiatives remain short-term – Success depends on the societal position of unions in a country +++

 

+

+++

– Applying existing laws to platforms (e.g. minimum wage) – Shaping new laws (e.g. platform worker category in law) – No success in guaranteeing freedom of association and collective bargaining to date – Outcomes for heterogeneous platform workers still unclear

Wilkinson, 2016; Dundon et al., 2004). In their study on voice on German crowdwork platforms, Gegenhuber et al. (2020) draw a parallel to the metaphors of “microphones” and “megaphones”: most platforms provide workers

How do workers gain voice on digital work platforms?

Figure 4.1

85

Hotspots and blind spots in platform worker voice

with “microphones” (i.e. individual channels that can easily be controlled and muted), rather than “megaphones” (i.e. channels thorough which workers could collectively co-determine platform settings). In the rare examples of cooperative platforms, things are different: these platforms usually provide formal ways through which platform workers can collectively influence platform management. Platforms differ in the extent to which they have an interest in integrating crowdworkers’ voice, often based on founders’ values, which is consistent with findings in more “classical” organizations (Franca & Pahor, 2014; Helfen & Schuessler, 2009). The “individual-informal” quadrant in Figure 4.1 remains empty and thus is clearly a blind spot. Here, research documenting voice at the individual or team level in spontaneous forms could be interesting. In voice research, such forms of voice mostly happen in teams, with members speaking up to their team manager (Mowbray et al., 2018). In current research on platform voice, however, these “organizational entities” in between workers and platforms have not been considered yet, as worker collective organization or worker–platform relationships have been the almost exclusive focus to date. However, as found by Gegenhuber et al. (2020), crowdwork platforms install support crowds to collect input from individual crowdworkers and give it to the platform. Anecdotal evidence suggests that some platforms in the food delivery sector do have some form of middle management (e.g. in the form of

86

Missing voice?

“rider captains”). Besides delivering food, these rider captains have additional managerial responsibilities, such as planning shifts by coordinating riders. Moreover, gig work platforms often run local centres where food couriers meet and come together, suggesting the existence of some local middle management and team-building. Given that informal voice mechanisms rely on the discretionary powers with which direct managers handle voice (Harlos, 2001), we think such informal, individual voice mechanisms should receive further attention. Do platform workers speak up in such contexts, and if so, how? How do middle managers encourage or inhibit that voice? How do they cope with suggestions or grievance from platform workers? How effective are such voice mechanisms? Another open question is what forms of voice middle managers themselves have (Mowbray, 2018). Evidence of such informal dynamics could inform policy design, as it could bring insights on the “right mix” of voice channels to address the heterogeneous workforces on platforms and give them effective voice (Bryson, 2004; McCloskey & McDonnell, 2018; Vallas & Schor, 2020).

FOUR POLICY IMPLICATIONS TO FOSTER PLATFORM WORKERS’ VOICE AND EMBED IT AT A SYSTEMIC LEVEL Based on our literature review, we derive four policy recommendations which aim at improving the structural conditions for strengthening platform worker voice. Many recommendations address the jurisdiction of policymakers, but we also consider unions to play a vital role in turning these recommendations into action. The recommendations are generic because we are unable to account for the differences of national contexts (e.g. labour laws, cultural and historical contingencies shaping the role and power of labour unions), despite these clearly shaping the ways in which platforms insert themselves into labour markets and industries (Davis, 2016a, 2016b; Davis & Sinha, 2021; Mair & Reischauer, 2017; Rahman & Thelen, 2019; Uzunca et al., 2018). Creating Spaces for Workers to Network and Exchange In the “offline” world, organizing for voice occurs in the workplace. Digitalization has spurred a further delocalization, delegation and intermediation of work and industrial organization, a dynamic that started with outsourcing and offshoring as early as the 1990s (Kirchner & Beyer, 2016; Kirchner & Schüßler, 2020). As platform workers, especially crowdworkers, do not meet at the workplace, digital spaces need to be created so that workers have the opportunity to network and exchange ideas and opinions. Extant research revealed that, depending on the platform context, crowdworkers either already

How do workers gain voice on digital work platforms?

87

use various tools provided by the platform (e.g. direct messaging or platform forums), or initiate their own channels if not provided by the platform (social media, direct messaging on the platform, smartphone messaging tools). Location-based platform workers that convene regularly at certain points seem to rely on smartphone messaging tools (e.g. Telegram, WhatsApp), whereas crowdworkers or workers for location-based platform services that meet infrequently (e.g. Uber, mystery shopping) use Facebook groups or create their own forums. It should be made easier for workers to create and use such digital spaces, not least by platforms themselves. This would allow everyone working on the platform to easily access them, which is not necessarily the case for digital spaces created on third-party websites or applications. Even though previous research is unclear about the extent to which digital spaces foster long-term worker organizing efforts, our literature review provides clear evidence that those digital spaces are used to exchange best practices (e.g. how to increase income given certain algorithmic rules) and thereby are an important tool to improve working conditions on an individual level. These digital spaces clearly serve as a venting and community space, providing emotional support, which can be extended to serve as a mobilizing tool as a basis for collective action initiatives. Policymakers could enforce regulations requiring platforms to offer public (e.g. platform-wide forums) and private (e.g. direct messaging tools) means of communication among workers. Granting Labour Unions Access to Platform Workers Regardless of the type of work, platform workers feel the need to organize in order to collectively fight for the improvement of their working conditions. However, our analysis reveals that autonomous organizing efforts mostly lead to incremental improvements within the existing digital labour market system rather than to systemic improvements. To achieve substantive changes such as higher wages and stronger process rights rooted in increased transparency or participation opportunities, union involvement is needed. Because of their resources, unions have the ability to sustain and foster workers’ organizing efforts. Our literature review clearly shows that union-supported organizing has a significant and, in comparison to worker or platform-driven initiatives, a more substantial impact that has the potential to achieve changes across several platforms, or even across industries. However, many platforms are “unsurprisingly hostile to any efforts at organizing genuinely independent worker representation” (Prassl, 2018, p. 65) and actively restrict their workers’ access to unions. The platform Deliveroo, which refuses to recognize the Independent Workers’ Union of Great Britain (IWGB) in the UK, is only one of many examples. What further hinders the unions’ access to workers is the classification of workers as “independent

88

Missing voice?

contractors”. This impedes their right to collective bargaining and corporate co-determination structures. Regulators therefore need to adapt labour laws to enable unions’ access to workers. France, for example, has already done so by introducing the Act of August 2016 that explicitly recognizes platform workers’ right to unionize and take industrial action. In Germany, the German Trade Union Federation (Deutscher Gewerkschaftsbund (DGB)) supports this approach and explicitly proposes digital access rights for interest groups and trade unions in its position paper on platform work, published this spring. The DGB demands that “digital communication channels must be made accessible to trade unions and statutory interest groups, such as works councils, by means of documented rights and must be able to take place via these channels” (DGB, 2021, p. 5).8 Unions Need to Develop Sustainable Digital Unionizing Strategies Becoming active in the platform work context is challenging for unions, and the development of suitable and effective union strategies for organizing platform workers is still in its infancy. The invisibility of crowdworkers, coupled with the unions’ mostly national scope of action, complicates union-driven organizing in the context of digital platform work. Union strategies traditionally rest on organizing in large (industrial) firms and on union members providing a steady source of income to finance the operations (Hyman, 2007). In certain national contexts, industrial relations are further based on a system giving power to unions through social dialogue and collective bargaining institutions. This stands in stark contrast to autonomous bottom-up crowdworker initiatives with a social movement character (Vandaele, 2018). Therefore, we see three challenges for labour unions: first, coping with the different organizing logic (with different practices, values, norms and language); second, finding a sustainable way to finance union operations in platform economy contexts; and third, how to organize for change and incorporate new logics (Yu, 2013). However, our analysis has shown that small-scale testing seems to be a common approach for unions to learn and to develop these new strategies. Additionally, rare examples of large-scale union involvement in crowdworker organizing exist, which could serve as blueprints for future initiatives. Unions can also learn from each other. Therefore, unions on both national and international levels should engage and foster an exchange of know-how and best practices with regard to how to efficiently organize in the context of platform work. This exchange should not be limited to specific initiatives, but should also encompass more general topics such as unions’ internal challenges, contradictions and complexities in organizing, and representing platform workers. Additionally, as platform work is a phenomenon spanning national boundaries, national unions should investigate possibilities of trans-

How do workers gain voice on digital work platforms?

89

national cooperation, as already observed in other sectors (e.g. regarding transnational wage bargaining across Europe in the metalworking sector) (Pernicka & Glassner, 2014) or attempts to set collective standards for building safety and living wages in the garment sector with the help of global union federations (Ashwin et al., 2020). There are some exemplary initiatives, such as the Frankfurt Declaration on Platform-Based Work that show that transnational cooperation between unions is possible. These initiatives could be a starting point for further initiatives that deepen cooperation between national unions and make that cooperation sustainable over the long-term. Clarifying Worker Classification and Ensuring Fair Democratic Debate An investigation into how the regulatory framework could be adapted to strengthen workers’ and unions’ position is vital. In addition to issues such as regulating workers’ ability to connect with each other and providing access for unions to platform, we would like to discuss two additional issues related to the classification of platform workers. First, a key issue is determining under what conditions platform workers fall into the category of a worker and when they are classified as independent contractors, or whether a third category, such as “dependent contractors” or “independent workers” (Taylor et al., 2017) is necessary. Regulators, policymakers and courts are determining this issue around the world for gig workers and crowdworkers alike, sometimes with conflicting outcomes. Being classified as a worker goes hand in hand with certain formal rights to organize for voice (which may vary according to the national regulatory context). Even though some crowdworkers may be classified as independent contractors, regulators need to allow these independent contractors to engage in collective action. This might require adapting cartel laws or solutions based on the aforementioned third category. However, a major force in ensuring that gig workers remain contractors is the platforms themselves. Uber and Lyft spent over $200 million for the Proposition 22 campaign, a public ballot initiative that resulted in gig workers keeping their status as independent contractors. According to Hawkins (2020), the platforms used their apps to bombard both their users and riders with pop-up messages asking them to support this initiative, which was seemingly critical to their gaining sufficient support for their proposition. While the platforms seem to have operated within the confines of the law, it raises a question as to how this massive power imbalance between labour and the platforms represents a fair democratic competition of ideas. Regulators and policymakers need to pay close attention to this issue and consider regulatory action.

90

Missing voice?

CONCLUSION In this chapter, we have investigated how workers gain voice in the platform economy. Our approach to synthesizing the current state of the literature was a search for the sources of worker voice, that is to say, who speaks up or enables voice. We identified four sources: worker-driven, platform-driven, union-driven and regulator-driven. Summarizing the findings of this review, we first provided guidance for future research. We pinpointed formal, collective channels as the hotspots at the centre of current research on platform worker voice. While research shows that some platforms provide formal channels through which individual workers can speak up, a clear blind spot in current research can certainly be found in informal voice channels that require more attention in future research. Where do informal voice channels exist on platforms? And how effective are they? In the second step, we leveraged our literature review to arrive at recommendations for how to improve worker voice through policy, namely, by creating spaces for workers to interact with each other, granting unions the right to access platform workers, a call for unions to develop sustainable digital unionizing strategies, and the need to clarify worker classification and ensure fair, democratic debate on these issues. We think these suggestions have great potential. Consider the suggestion that workers should be able to interact with each other. As we write this chapter, a bottom-up worker initiative in Berlin is engaged in a strike and blockade of a distribution centre of the food delivery service Gorillas. This initiative has its own Twitter channel, where it also informs the public about the protest. One of the workers stated that he wrote a letter to upper management criticizing inadequate rider safety and equipment. He then shared this letter in the public chat of the Gorillas riders’ app, which workers usually use to manage shifts (e.g. swapping shifts). This letter received positive feedback from workers, and the Gorillas management team responded by removing the chat function for a week (Twitter, 2021). The platform only allowed for voice that served functional purposes, enabling communication among workers to reduce the platform’s coordination costs, but turning it off once workers began using it as a megaphone.

NOTES 1. 2.

We will henceforth use the term “crowdwork” to refer to digital platform work. Food delivery platforms, strictly speaking, connect supply (e.g. restaurants) with the demand (i.e. consumers). In this case, there is no intermediation process. The platforms seek to outsource the delivery service to crowdworkers while remaining in full control of the work process.

How do workers gain voice on digital work platforms?

3. 4.

5.

6.

7.

8.

91

Other platform workers comprise the architects and technologists of the platforms themselves and social media content producers and influencers (Vallas & Schor, 2020), but these categories are, to date, still heavily understudied. Mystery shopping is an exception in this context. Though the workers operate in the same location, they cannot identify each other. This is different in, for example, the food delivery industry, where workers wear a “uniform” by which they can recognize each other and convene at meeting points to wait for assignments. Cooperatives are defined as “an autonomous association of persons united voluntarily to meet their common economic, social, and cultural needs and aspirations through a jointly-owned enterprise” (ICA, 2018). To give an example, cooperatives are a form of organization that can often be found in the agricultural sector. The announcement about closing Daemo resulted in criticism – some perceived it as sudden and called for more transparency about why the project failed (to enable learning for future initiatives). See https://​ twitter​ .com/​ ntnsndr/​ status/​ 1105148290868248577 (with the root tweet deleted, probably due to the assertive voiced criticism). The term “old economy” describes traditional industry sectors that produce tangible goods. To give an example, both the automotive and the chemical industries are part of the old economy, and these sectors are typically still unionized, albeit in varying degrees across countries. The quote is translated from German.

REFERENCES Aloisi, A. (2015). Commoditized workers: the rising of on-demand work, a case study research on a set of online platforms and apps. SSRN Electronic Journal. https://​doi​ .org/​10​.2139/​ssrn​.2637485. Aloisi, A. (2019). Negotiating the digital transformation of work: non-standard workers’ voice, collective rights and mobilisation practices in the platform economy. https://​papers​.ssrn​.com/​sol3/​papers​.cfm​?abstract​_id​=​3404990. Animento, S., Di Cesare, G. & Sica, C. (2017). Total eclipse of work?: Neue protestformen in der gig economy am beispiel des foodora streiks in Turin. PROKLA. Zeitschrift Für Kritische Sozialwissenschaft, 47(187). https://​doi​.org/​10​.32387/​ prokla​.v47i187​.145. Anwar, M. A. & Graham, M. (2020). Hidden transcripts of the gig economy: labour agency and the new art of resistance among African gig workers. Environment and Planning A, 52(7). https://​doi​.org/​10​.1177/​0308518X19894584. Ashwin, S., Oka, C., Schuessler, E., Alexander, R. & Lohmeyer, N. (2020). Spillover effects across transnational industrial relations agreements: the potential and limits of collective action in global supply chains. ILR Review, 73(4). https://​doi​.org/​10​ .1177/​0019793919896570. Attwood-Charles, W. (2019). Dimensions of platform labor control and the experience of gig couriers. Boston College. Aytes, A. (2012). Return of the crowds: Mechanical Turk and neoliberal states of exception. In Digital Labor: The Internet as Playground and Factory. https://​doi​ .org/​10​.4324/​9780203145791.

92

Missing voice?

Barratt, T., Goods, C. & Veen, A. (2020). ‘I’m my own boss …’: active intermediation and ‘entrepreneurial’ worker agency in the Australian gig-economy. Environment and Planning A, 52(8). https://​doi​.org/​10​.1177/​0308518X20914346. Barry, M. & Wilkinson, A. (2016). Pro-social or pro-management? A critique of the conception of employee voice as a pro-social behaviour within organizational behaviour. British Journal of Industrial Relations, 54(2). https://​doi​.org/​10​.1111/​ bjir​.12114. Barzilay, A. R. & Ben-David, A. (2018). Platform inequality: gender in the gig-economy. SSRN Electronic Journal. https://​doi​.org/​10​.2139/​ssrn​.2995906. Bauer, R. M. & Gegenhuber, T. (2015). Crowdsourcing: global search and the twisted roles of consumers and producers. Organization, 22(5), 661–81. Bryson, A. (2004). Managerial responsiveness to union and nonunion worker voice in Britain. Industrial Relations, 43(1), 213–41. https://​doi​.org/​10​.1111/​j​.0019​-8676​ .2004​.00324​.x. Bucher, E. L., Schou, P. K. & Waldkirch, M. (2021). Pacifying the algorithm – anticipatory compliance in the face of algorithmic management in the gig economy. Organization, 28(1). https://​doi​.org/​10​.1177/​1350508420961531. Cant, C. & Woodcock, J. (2020). Fast food shutdown: from disorganisation to action in the service sector. Capital & Class, 44(4), 513–21. https://​doi​.org/​10​.1177/​ 0309816820906357. Cook, C., Diamond, R., Hall, J., List, J. & Oyer, P. (2018). The gender earnings gap in the gig economy: evidence from over a million rideshare drivers. National Bureau of Economic Research. https://​doi​.org/​10​.3386/​w24732. Davis, G. F. (2016a). What might replace the modern corporation? Uberization and the web page enterprise. Seattle University Law Review, 39, 501–15. Davis, G. F. (2016b). Can an economy survive without corporations? Technology and robust organizational alternatives. Academy of Management Perspectives, 30(2). https://​doi​.org/​10​.5465/​amp​.2015​.0067. Davis, Gerald F. & Sinha, A. (2021). Varieties of Uberization: how technology and institutions change the organization(s) of late capitalism. Organization Theory, 2(1). https://​doi​.org/​10​.1177/​2631787721995198. De Stefano, V. (2016). The rise of the “just-in-time workforce”: on demand work, crowdwork, and labor protection in the “gig economy”. Comparative Labor Law and Policy Journal, 37(3), 461–71. De Stefano, V. & Aloisi, A. (2018). European Legal Framework for Digital Labour Platforms. Luxembourg: European Commission. DGB (2021). DGB – Position zur Plattformarbeit. Deutscher Gewerkschaftsbund. https://​www​.dgb​.de/​downloadcenter/​+​+​co+​+​231aa4e0​-a1ea​-11eb​-a42d​ -001a4a160123. Dickel, S. & Thiem, C. (2018). Zur Organisation von Arbeit 4.0: Crowdsourcing als Sozialtechnologie. In J. P. Redlich, Tobias Moritz & Manuel Wulsfberg (eds), Interdisziplinäre Perspektiven zur Zukunft der Wertschöpfung (pp. 247–54). Berlin: Springer Fachmedien. Dubal, V. B. (2017). The drive to precarity: a political history of work, regulation, & labor advocacy in San Francisco’s taxi & uber economies. Berkeley Journal of Employment and Labor Law, 38(1). https://​doi​.org/​10​.15779/​Z38KW57H5S. Dundon, T., Wilkinson, A., Marchington, M. & Ackers, P. (2004). The meanings and purpose of employee voice. International Journal of Human Resource Management, 15(6). https://​doi​.org/​10​.1080/​0​9585190410​0016773359.

How do workers gain voice on digital work platforms?

93

Ellmer, M. (2015). The digital division of labor: socially constructed design patterns of Amazon Mechanical Turk and the governing of human computation labor. Momentum Quarterly, 4(3), 174–86. Ellmer, M., Herr, B., Klaus, D. & Gegenhuber, T. (2019). Platform workers centre stage! Taking stock of current debates and approaches for improving the conditions of platform work in Europe. June. www​.boeckler​.de​%22Platfor​mworkersce​ ntrestage!​%22von Fantasia, R. (1995). From class consciousness to culture, action, and social organization. Annual Review of Sociology, 21(1). https://​doi​.org/​10​.1146/​annurev​.so​.21​ .080195​.001413. Farrell, D. (1983). Exit, voice, loyalty, and neglect as responses to job dissatisfaction: a multidimensional scaling study. Academy of Management Journal, 26(4). https://​ doi​.org/​10​.2307/​255909. Felstiner, A. (2011). Working the crowd: employment and labor law in the crowdsourcing industry. Berkeley Journal of Employment & Labor Law, 32(1). https://​www​ .jstor​.org/​stable/​24052509. Fernàndez, A. & Barreiro, M. S. (2020). The algorithm is not my boss anymore: technological appropriation and (new) media strategies in Riders x Derechos and Mensakas. Revista Contracampo, 39(1). https://​doi​.org/​10​.22409/​contracampo​.v39i1​.38404. Franca, V. & Pahor, M. (2014). Influence of management attitudes on the implementation of employee participation. Economic and Industrial Democracy, 35(1). https://​ doi​.org/​10​.1177/​0143831X12464070. Frenken, K., van Waes, A., Pelzer, P., Smink, M. & van Est, R. (2020). Safeguarding public interests in the platform economy. Policy and Internet, 12(3). https://​doi​.org/​ 10​.1002/​poi3​.217. Fuhlrott, M. (2020). Ein Crowdworker war Arbeitnehmer. Legal Tribune Online. https://​www​.lto​.de/​recht/​hintergruende/​h/​bag​-urteil​-9azr​-10220​-crowdworker​-war​ -arbeitnehmer​-auftrag​-vermittlung​-online​-arbeitsvertrag/​. Gegenhuber, T., Ellmer, M. & Schüßler, E. (2020). Microphones, not megaphones: functional crowdworker voice regimes on digital work platforms. Human Relations, 74(9). http://​doi​.org/​10​.1177/​0018726720915761. Gegenhuber, T., Schüßler, E., Reischauer, G. & Thäter, L. (2021). Building collective institutional infrastructures for decent platform work: the development of a crowdwork agreement in Germany. In Organizing for Societal Grand Challenges (Research in the Sociology of Organizations, Vol. 79). https://​doi​.org/​10​.1108/​ S0733​-558X20220000079004. Gorwa, R. (2019). What is platform governance? Information Communication and Society, 22(6). https://​doi​.org/​10​.1080/​1369118X​.2019​.1573914. Griesbach, K., Reich, A., Elliott-Negri, L. & Milkman, R. (2019). Algorithmic control in platform food delivery work. Socius: Sociological Research for a Dynamic World, 5. https://​doi​.org/​10​.1177/​2378023119870041. Harlos, K. P. (2001). When organizational voice systems fail: more on the deaf-ear syndrome and frustration effects. Journal of Applied Behavioral Science, 37(3). https://​ doi​.org/​10​.1177/​0021886301373005. Harris, S. D. & Krueger, A. B. (2015). A proposal for modernizing labor laws for twenty-first-century work: the “independent worker.” Hamilton Project Discussion Papers, 10. https://​www​.hamiltonproject​.org/​assets/​files/​modernizing​_labor​_laws​ _for​_twenty​_first​_century​_work​_krueger​_harris​.pdf.

94

Missing voice?

Hawkins, A. J. (2020). Uber and Lyft had an edge in the Prop 22 fight: their apps. The Verge. https://​www​.theverge​.com/​2020/​11/​4/​21549760/​uber​-lyft​-prop​-22​-win​-vote​ -app​-message​-notifications. Healy, J., Nicholson, D. & Pekarek, A. (2017). Should we take the gig economy seriously? Labour & Industry: A Journal of the Social and Economic Relations of Work, 27(3), 232–48. https://​doi​.org/​10​.1080/​10301763​.2017​.1377048. Heiland, H. & Brinkmann, U. (2020). Liefern am Limit. Wie die Plattformökonomie die Arbeitsbeziehungen verändert. Industrielle Beziehungen. Zeitschrift Für Arbeit, Organisation Und Management, 27(2–2020). https://​doi​.org/​10​.3224/​indbez​.v27i2​ .02. Helfen, M. & Schuessler, E. S. (2009). Uncovering divergence: management attitudes towards HRM practices and works council presence in German SMEs. Economic and Industrial Democracy, 30(2). https://​doi​.org/​10​.1177/​0143831X09102420. Hinings, B., Gegenhuber, T. & Greenwood, R. (2018). Digital innovation and transformation: an institutional perspective. Information and Organization, 28(1), 52–61. https://​doi​.org/​10​.1016/​j​.infoandorg​.2018​.02​.004. Hirschman, A. O. (1970). Exit, Voice and Loyalty: Responses to Decline in Firms, Organizations, and States. Cambridge, MA: Harvard University Press. Hoose, F., Haipeter, T. & Ittermann, P. (2019). Digitalisierung der Arbeit und Interessenvertretungen. Arbeit, 28(4), 423–44. https://​doi​.org/​10​.1515/​arbeit​-2019​ -0025. Howcroft, D. & Bergvall-Kåreborn, B. (2019). A typology of crowdwork platforms. Work, Employment and Society, 33(1), 21–38. Hyman, R. (1999). Imagined solidarities: can trade unions resist globalisation? In P. Leisink (ed.), Globalization and Labour Relations. Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing. Hyman, R. (2007). How can trade unions act strategically? Transfer: European Review of Labour and Research, 13(2). https://​doi​.org/​10​.1177/​102425890701300204. ICA (2018). Cooperative identity, values & principles. International Cooperative Alliance. https://​www​.ica​.coop/​en/​cooperatives/​cooperative​-identity. Ilsøe, A. & Jesnes, K. (2020). Collective agreements for platform workers: two cases from the Nordic countries. In K. Jesnes & S. Oppegaard (eds), Platform Work in the Nordic Models: Issues, Cases and Responses (pp. 53–67). Copenhagen: Nordic Council of Ministers. Irani, L. C. & Silberman, M. S. (2013). Turkopticon: interrupting worker invisibility in Amazon Mechanical Turk. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 611–20. Johnston, H. & Land-Kazlauskas, C. (2018). Organizing on-demand: representation, voice, and collective bargaining in the gig economy. Conditions of Work and Employment Series, No. 94. Geneva: International Labour Office. Johnston, H., & Pernicka, S. (2021). Struggles over the power and meaning of digital labour platforms: a comparison of Los Angeles, New York City, Berlin and Vienna taxi markets. In J. Drahokoupil & K. Vandaele (eds), A Modern Guide to Labor and the Platform Economy (pp. 308–22). Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing. Kaganer, E., Carmel, E., Hirschheim, R. & Olsen, T. (2013). Managing the human cloud. MIT Sloan Management Review, 54(2). https://​sloanreview​.mit​.edu/​article/​ managing​-the​-human​-cloud/​.

How do workers gain voice on digital work platforms?

95

Karanović, J., Berends, H. & Engel, Y. (2020). Regulated dependence: platform workers’ responses to new forms of organizing. Journal of Management Studies, 58(4). https://​doi​.org/​10​.1111/​joms​.12577. Kellogg, K. C., Valentine, M. A. & Christin, A. (2020). Algorithms at work: the new contested terrain of control. Academy of Management Annals, 14(1). https://​doi​.org/​ 10​.5465/​annals​.2018​.0174. Kenney, M. & Zysman, J. (2016). The rise of the platform economy. Issues in Science and Technology, 32(3), 61–9. Kirchner, S. & Beyer, J. (2016). Die Plattformlogik als digitale Marktordnung. Zeitschrift Für Soziologie, 45(5). https://​doi​.org/​10​.1515/​zfsoz​-2015​-1019. Kirchner, S. & Schüßler, E. (2019). The organization of digital marketplaces: unmasking the role of internet platforms in the sharing economy. Organization Outside Organizations, 131–54. https://​doi​.org/​10​.1017/​9781108604994​.006. Kirchner, S. & Schüßler, E. (2020). Regulating the sharing economy: a field perspective. In I. Maurer, J. Mair & A. Oberg (eds), Theorizing the Sharing Economy: Variety and Trajectories of New Forms of Organizing (pp. 215–36). Bingley: Emerald Publishing. https://​doi​.org/​10​.1108/​s0733​-558x20200000066010. Klaas, B. S., Olson-Buchanan, J. B. & Ward, A. K. (2012). The determinants of alternative forms of workplace voice: an integrative perspective. Journal of Management, 38(1). https://​doi​.org/​10​.1177/​0149206311423823. Kornberger, M., Pflueger, D. & Mouritsen, J. (2017). Evaluative infrastructures: accounting for platform organization. Accounting, Organizations and Society, 60, 79–95. https://​doi​.org/​10​.1016/​j​.aos​.2017​.05​.002. Kuhn, K. M. & Maleki, A. (2017). Micro-entrepreneurs, dependent contractors, and instaserfs: understanding online labor platform workforces. Academy of Management Perspectives, 31(3), 183–200. https://​doi​.org/​10​.5465/​amp​.2015​.0111. Lehdonvirta, V. (2016). Algorithms that divide and unite: delocalisation, identity and collective action in ‘microwork’. In Space, Place and Global Digital Work. https://​ doi​.org/​10​.1057/​978​-1​-137​-48087​-3​_4. Lehdonvirta, V., Kässi, O., Hjorth, I., Barnard, H. & Graham, M. (2019). The global platform economy: a new offshoring institution enabling emerging-economy microproviders. Journal of Management, 45(2), 567–99. https://​doi​.org/​10​.1177/​ 0149206318786781. Maffie, M. D. (2020). The role of digital communities in organizing gig workers. Industrial Relations: A Journal of Economy and Society, 59(1), 123–49. Mair, J. & Reischauer, G. (2017). Capturing the dynamics of the sharing economy: institutional research on the plural forms and practices of sharing economy organizations. Technological Forecasting and Social Change, 125. https://​doi​.org/​10​.1016/​j​ .techfore​.2017​.05​.023. Marchington, M. & Suter, J. (2013). Where informality really matters: patterns of employee involvement and participation (EIP) in a non-union firm. Industrial Relations, 52(SUPPL.1). https://​doi​.org/​10​.1111/​irel​.12004. Martin, C. J., Upham, P. & Klapper, R. (2017). Democratising platform governance in the sharing economy: an analytical framework and initial empirical insights. Journal of Cleaner Production, 166, 1395–406. https://​doi​.org/​10​.1016/​j​.jclepro​.2017​.08​ .123. Maurer, I., Mair, J. & Oberg, A. (2020). Variety and trajectories of new forms of organizing in the sharing economy: a research agenda. In Theorizing the Sharing Economy: Varieties and Trajectories of New Forms of Organizing. https://​doi​.org/​ 10​.1108/​s0733​-558x20200000066002.

96

Missing voice?

McCloskey, C. & McDonnell, A. (2018). Channels of employee voice: complementary or competing for space? Industrial Relations Journal, 49(2). https://​doi​.org/​10​.1111/​ irj​.12209. Mowbray, P. K. (2018). Giving a voice to managers: forging the desire line through the creation of informal employee voice channels and productive resistance. International Journal of Human Resource Management, 29(5). https://​doi​.org/​10​ .1080/​09585192​.2017​.1396552. Mowbray, P. K., Wilkinson, A. J. & Tse, H. H. M. (2015). An integrative review of employee voice: identifying a common conceptualization and research agenda. International Journal of Management Reviews, 17(3), 382–400. https://​doi​.org/​10​ .1111/​ijmr​.12045. Mowbray, P. K., Wilkinson, A. J. & Tse, H. H. M. (2018). High performance strategy: influence on HR department and line managers’ design of employee voice. Academy of Management Proceedings, 2018(1). https://​doi​.org/​10​.5465/​ambpp​ .2018​.11034abstract. Niebler, V. (2020). ‘YouTubers unite’: collective action by YouTube content creators. Transfer: European Review of Labour and Research, 26(2). https://​doi​.org/​10​.1177/​ 1024258920920810. Niebler, V. & Kern, A. (2020). Organizing Youtube: a novel case of platform worker organizing. Friedrich-Ebert-Foundation. https://​www​.academia​.edu/​44269481/​ Organising​_YouTube​_A​_novel​_case​_of​_platform​_worker​_organising. Pernicka, S. & Glassner, V. (2014). Transnational trade union strategies towards European wage policy: a neo-institutional framework. European Journal of Industrial Relations, 20(4). https://​doi​.org/​10​.1177/​0959680113518232. Prassl, J. (2018). Humans As a Service: The Promise and Perils of Work in the Gig Economy. Oxford: Oxford University Press. Rahman, H. A. (2021). The invisible cage: workers’ reactivity to opaque algorithmic evaluations. Administrative Science Quarterly. https://​doi​.org/​10​.1177/​ 00018392211010118. Rahman, K. S. & Thelen, K. (2019). The rise of the platform business model and the transformation of twenty-first-century capitalism. Politics and Society, 47(2). https://​doi​.org/​10​.1177/​0032329219838932. Ravenelle, A. J. (2019). Hustle and Gig: Struggling and Surviving in the Sharing Economy. Berkeley: University of California Press. Robinson, H. C. (2017). Making a Digital Working Class: Uber Drivers in Boston, 2016–2017. Cambridge, MA: MIT Press. Rosenblat, A. & Stark, L. (2016). Algorithmic labor and information asymmetries: a case study of Uber’s drivers. International Journal of Communication, 10. https://​ doi​.org/​10​.2139/​ssrn​.2686227. Salehi, N., Irani, L. C., Bernstein, M. S., Alkhatib, A., Ogbe, E., Milland, K. & Clickhappier. (2015). We are dynamo: overcoming stalling and friction in collective action for crowd workers. Conference on Human Factors in Computing Systems – Proceedings, 2015 – April. https://​doi​.org/​10​.1145/​2702123​.2702508. Scholz, T. (2016). Platform cooperativism: challenging the corporate sharing economy. Rosa Luxemburg Stiftung, New York Office. https://​ rosalux​ .nyc/​ wp​ -content/​ uploads/​2020/​11/​RLS​-NYC​_platformcoop​.pdf. Shafiei Gol, E., Stein, M. K. & Avital, M. (2019). Crowdwork platform governance toward organizational value creation. Journal of Strategic Information Systems, 28(2). https://​doi​.org/​10​.1016/​j​.jsis​.2019​.01​.001.

How do workers gain voice on digital work platforms?

97

Srnicek, N. (2017). Platform Capitalism. Cambridge, UK and Malden, MA: Polity Press. Sutherland, W., Jarrahi, M. H., Dunn, M. & Nelson, S. B. (2020). Work precarity and gig literacies in online freelancing. Work, Employment and Society, 34(3). https://​doi​ .org/​10​.1177/​0950017019886511. Tassinari, A. & Maccarrone, V. (2017). The mobilisation of gig economy couriers in Italy. Transfer: European Review of Labour and Research, 23(3). https://​doi​.org/​10​ .1177/​1024258917713846. Taylor, M., Marsh, G., Nicol, D. & Broadbent, P. (2017). Good work: the Taylor review of modern working practices. Department for Business, Energy and Industrial Strategy. https://​assets​.publishing​.service​.gov​.uk/​government/​uploads/​system/​ uploads/​attachment​_data/​file/​627671/​good​-work​-taylor​-review​-modern​-working​ -practices​-rg​.pdf. Thäter, L. & Gegenhuber, T. (2020). Plattformgenossenschaften: mehr Mitbestimmung durch die digitale Renaissance einer alten Idee? https://​doi​.org/​10​.1007/​978​-3​-658​ -32276​-2​_13. Trittin-Ulbrich, H., Scherer, A. G., Munro, I. & Whelan, G. (2021). Exploring the dark and unexpected sides of digitalization: toward a critical agenda. Organization, 28(1). https://​doi​.org/​10​.1177/​1350508420968184. Twitter (2021). Another juicy story. https://​twitter​.com/​hueso​_gorillas/​status/​ 1403722317419667459​?s​=​20. Urzì Brancati, M. C., Pesole, A. & Fernandez Macias, E. (2019). Digital labour platforms in Europe: numbers, profiles, and employment status of platform workers. Joint Research Centre (Seville Site), No. JRC117. Uzunca, B., Rigtering, J. P. C. & Ozcan, P. (2018). Sharing and shaping: a cross-country comparison of how sharing economy firms shape their institutional environment to gain legitimacy. Academy of Management Discoveries, 4(3). https://​doi​.org/​10​ .5465/​amd​.2016​.0153. Vallas, S. P. (2019). Platform capitalism: what’s at stake for workers? New Labor Forum, 28(1). https://​doi​.org/​10​.1177/​1095796018817059. Vallas, S. P. & Schor, J. B. (2020). What do platforms do? Understanding the gig economy. Annual Review of Sociology, 46. https://​doi​.org/​10​.1146/​annurev​-soc​ -121919​-054857. Vandaele, K. (2018). Will trade unions survive in the platform economy? Emerging patterns of platform workerss collective voice and representation in Europe. SSRN Electronic Journal. https://​doi​.org/​10​.2139/​ssrn​.3198546. Vandaele, K. (2020). From street protest to improvisational unionism. Trade Unions in Transformation 4.0. http://​library​.fes​.de/​pdf​-files/​iez/​16608​.pdf. Vandaele, K. (2021). Collective resistance and organizational creativity amongst Europe’s platform workers: a new power in the labour movement? In J. Haidar & M. Keune (eds), Work and Labour Relations in Global Platform Capitalism (pp. 206–35). Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing. Whiting, M. E., Gamage, D., Goyal, S., Gilbee, A., Majeti, D., Richmond-Fuller, A., Salih, M. A., Sarma, T. S., Mathur, V., Pandey, M., Gaikwas, S., Vaish, R. & Bernstein, M. (2017). Designing a constitution for a self-governing crowdsourcing marketplace. Collective Intelligence Conference. https://​wtf​.tw/​ref/​whiting​.pdf. Wilkinson, A., Gollan, P. J., Kalfa, S. & Xu, Y. (2018). Voices unheard: employee voice in the new century. International Journal of Human Resource Management, 29(5). https://​doi​.org/​10​.1080/​09585192​.2018​.1427347.

98

Missing voice?

Wilkinson, A., Knolee, M., Mowbray, P. & Dundon, T. (2021). New trajectories in worker voice: integrating and applying contemporary challenges in the organization of work. British Journal of Management, 32(3): 693–707. Wood, A. & Lehdonvirta, V. (2019). Platform labour and structured antagonism: understanding the origins of protest in the gig economy. SSRN Electronic Journal. https://​ doi​.org/​10​.2139/​ssrn​.3357804. Wood, A. J. & Lehdonvirta, V. (2021). Antagonism beyond employment: how the ‘subordinated agency’ of labour platforms generates conflict in the remote gig economy. Socio-Economic Review, 19(4). https://​doi​.org/​10​.1093/​ser/​mwab016. Wood, A. J., Lehdonvirta, V. & Graham, M. (2018). Workers of the Internet unite? Online freelancer organisation among remote gig economy workers in six Asian and African countries. New Technology, Work and Employment, 33(2), 95–112. Wood, A. J., Graham, M., Lehdonvirta, V. & Hjorth, I. (2019a). Good gig, bad gig: autonomy and algorithmic control in the global gig economy. Work, Employment and Society, 33(1). https://​doi​.org/​10​.1177/​0950017018785616. Wood, A. J., Graham, M., Lehdonvirta, V. & Hjorth, I. (2019b). Networked but commodified: the (dis)embeddedness of digital labour in the gig economy. Sociology, 53(5). https://​doi​.org/​10​.1177/​0038038519828906. Yu, K. H. (2013). Institutionalization in the context of institutional pluralism: politics as a generative process. Organization Studies, 34(1). https://​doi​.org/​10​.1177/​ 0170840612464755.

APPENDIX The search and selection procedure for the literature review was conducted as follows. First, we identified empirical key works by searching for keywords such as “digital work/labour”, “voice”, “platform economy”, and “worker organizing”. We then employed a snowball system to arrive at a literature sample of 21 studies. After an initial analysis, we realized that our sample did not cover all phenomenon-related areas of interest. Therefore, we conducted another round of literature research, this time including industry reports as well as book chapters. As a result, we added another three studies to our sample, which in its final version comprises 24 studies in total.

gig work

Platform-driven and union-driven voice in gig and crowdwork

workers’ power resources (structural, institutional, symbolic) differ. That affects the strategies which unions choose to support platform workers. In Europe, structural and associational worker power is rather low, while institutional and symbolic power is medium to high, because unions

The Code of Conduct as an example of a new private regulation model that draws on elements from the traditional German social partnership model to improve the working conditions for platform workers. Even though the

of the ridesharing industry

in Europe and the USA

(Vienna, Berlin, New

York City, Los Angeles)

Qualitative case study

of the German Code of

Conduct initiative

Gegenhuber,

Schüßler, Reischauer,

Thäter

2021

an institutional infrastructure.

mechanisms which, in turn led to the emergence of the Code of Conduct as

that formalized the existing structures and implemented accountability

platforms initiated the Code of Conduct, it was the union’s involvement

comparably high.

level). Additionally, chances are better in places where symbolic power is

power at a scale similar to the regulation they seek (e.g. at the municipal

struggles’ outcomes are favourable if workers demonstrate associational

low to medium. In places, where associational power is medium/high, the

In the USA, where unions are rather weak, all four power resources are

tend to rely on existing tools (e.g. minimum wage).

traditionally have a strong institutional position. In these countries, unions

Union-driven voice in

Depending on the geographic location and thus the institutional context, the

Comparative case study

Category

Johnston and Pernicka

Key constructs and findings

2021

Methods and empirical

context

Author(s)

Literature review data

publication

Year of

Table 4A.1

How do workers gain voice on digital work platforms? 99

in gig work

role in the production process. If the workers perceive themselves as entrepreneurs, they also develop an entrepreneurial mindset. This, in turn, leads to low levels of agency that aim at improving individual working

two major platforms in the

Australian food delivery

industry

Cant and Woodcock

Cant and Mogno

2020

2020

gig work

the following four-step development: first, accurately identifying a winnable workplace-specific grievance; second, mapping the workforce to identify existing embryonic solidarities and informal work group structures;

of union organizing

strategies in the British

food delivery industry

Worker-driven voice in gig work

The case of the TFC has shown that gig worker organization across national borders strengthens the workers’ power resources. It does so by fostering their discursive power and by enhancing a broader working-class alliance. The transnationality of the workers’ organization strengthens their collective identity and voice. This also makes it more relevant on the political level.

Single quantitative case

study of the Transnational

Federation of Couriers

(TFC) in the European

food delivery industry

direct action).

novel representation strategies (e.g. prioritizing informal bargaining via

setting, unions need to change their traditional modus operandi and find

over larger-scale grievances. As traditional union strategies fall short in this

escalating from this initial collective action towards larger-scale action

third, moving to collective action on that winnable grievance; and fourth,

Union-driven voice in

Union organizing efforts were most successful when they followed roughly

Comparative case study

is discouraged.

workers, coupled with the platforms’ organizational fixes, collective agency

unemployment and underemployment rates and a high number of migrant

opposing it. Due to the nature of the Australian labour market with high

conditions and aligning with the platform’s business model, rather than

Worker-driven voice

The form of worker agency is influenced by the perception of the workers’

Qualitative case study of

Category

Barratt, Goods, Veen

Key constructs and findings

2020

Methods and empirical

context

Author(s)

publication

Year of

100 Missing voice?

gig work

protests), generate awareness for their coop, and connect with workers from

delivery industry

2020

2020

union-driven voice in

RidersxDerechos generate awareness for their direct action (strikes,

study on the Spanish food

in gig and crowdwork

firm, corporate) and degrees (= the extent of worker influence they provide (information–control)). Costs, control and a crowd’s social structure, and

of six medium-sized,

German platforms

Schüßler

in gig work

councils) in the German food delivery industry are still in their infancy. Even though there are examples where riders managed to establish a works council, this is the exception, not the rule. The platforms intentionally

of two platforms in the

German food delivery

industry

Brinkmann

existing fluid industrial relations can be institutionalized.

point of conflict. The outcome of this conflict determines whether the

industrial relations among workers. Communication logics are a central

decreasing the transparency and thereby fostering uncertainty about the

to establish formal participation structures. Platform providers do this by

exploit their position of power over the workers to hinder their attempts

Worker-driven voice

Approaches to institutionalizing participation structures (e.g. works

Comparative case study

Heiland and

performance.

way and within the confines of improving and optimizing organizational

levels and degrees. Platforms do embrace voice, but largely in a functional

platforms’ varying interest in fair work standards all lead to the variance in

Platform-driven voice

Platforms provide voice mechanisms in various levels (= task, department,

Comparative case study

Gegenhuber, Ellmer,

traditional struggle is still useful in the new digital economy.

made the project successful (learning, finding a voice). This shows that

different industries and countries. The support of a traditional union alone

Worker-driven and

A media communication strategy helped the grassroots union

Qualitative ethnographic

Solina-Barreiro

Category

Fernàndez and

Key constructs and findings

2020

Methods and empirical

context

Author(s)

publication

Year of

How do workers gain voice on digital work platforms? 101

platform-driven voice in gig work

Worker-driven voice in gig work

agreements in the platform economy can be implemented and how they could be designed. Yet, they differ at the level of worker involvement in the process. While the collective agreement in Denmark was initiated by the platform itself, it was the Norwegian workers who approached the union. Therefore, the degree of unionization among the platform workers differs strongly in the two cases: in Denmark it is close to zero; in Norway, it is The regulatory environment influences the way Uber drivers react to Uber’s way of organizing work. In cities with direct regulation (= comparably strict regulations for platforms and drivers), structures tend to supplement Uber’s solution (e.g. by providing information) rather than oppose it. In

case study of two

collective agreements

between labour platforms

and workers in Norway

(Foodora) and Denmark

(Hilfr)

Comparative case study of

Uber drivers in different

regulatory environments

(direct or indirect)

Karanović, Berends,

Engel

2020

power imbalance between platform and worker.

creates mutual dependence, whereas indirect regulation accentuates the

against the backdrop of resource dependence theory, direct regulation

not be entirely content with the solutions provided by Uber. Considered

environments, drivers prioritize working with Uber, even though they might

for platforms and drivers), the exact opposite happens. In directly regulated

cities with indirect regulatory structures (= more discretion and fewer rules

considerably higher (as workers initiated the cooperation with the union).

Union-driven and

The two agreements are both a unique examples of how collective

Comparative qualitative

Category

Ilsøe and Jesnes

Key constructs and findings

2020

Methods and empirical

context

Author(s)

publication

Year of

102 Missing voice?

in gig work

(including collective norms, rituals, and communal activities). The more a worker is connected online, the more (s)he is interested in collective

on the ride-hailing

industry

Niebler

Niebler and Kern

2020

2020

union-driven voice in crowdwork

channel shutdowns since YouTube’s advertisement crisis, a critically large number of creator groups were affected at the same time. A shared

YouTubers’ Union and

the FairTube campaign

union-driven voice in crowdwork

context of crowdwork. While collective action in this setting is usually spontaneous and fragmented, the YTU is an actor with strategic capabilities. Furthermore, the case of the YTU shows that platform workers that do

study on the YouTubers’

Union (YTU) and the

FairTube campaign

YouTubers’ voice by generating media attention.

IG Metall fostered the increase in associational power and amplified the

and societal power. The cooperation with the traditional German trade union

exclusively digital work can use social media to build up their associational

Worker-driven and

The YTU goes beyond previous forms of collective action in the

Qualitative single case

advantage, as it is YouTube’s main asset to attract advertising investors.

a crucial power resource. The workers can use the high brand value to their

the group’s collective actions, its ability to draw public attention has been

self-organization. The collective discontent of creators was mobilized. In all

(occupational) identity as YouTubers was successfully invoked to build

Worker-driven and

While not all YouTubers have been affected by content strikes and

Single case study on the

perception becomes.

associations. The more frequent the interaction, the more positive the

online community affects the worker’s perception of unions and labour

labour organization. Additionally, the frequency of interaction in an

Worker-driven voice

Platform workers use online communities to build a shared identity

Ethnographic case study

Category

Maffie

Key constructs and findings

2020

Methods and empirical

context

Author(s)

publication

Year of

How do workers gain voice on digital work platforms? 103

gig work

Worker-driven voice in crowdwork

engaged in direct action, which was supported by traditional unions. In the beginning, union support was driven by strong bottom-up dynamics – initiated by individuals within the unions. It also led to unions leaving their

The structural antagonism that exists between workers and the platform is the fuel that powers collective worker action. The findings highlight that collective worker action remains rooted in experiences generated by

case study on the Dutch

and Belgian food delivery

industry

Ethnographic fieldwork

in North America, the

Philippines and the UK

Wood and

Lehdonvirta

2019

economy offers a potential new source of members.

can represent the interests of workers vis-à-vis platforms, and that the gig

the organization of labour. It suggests that there is a need for unions that

organizations, and thereby strengthen the unions’ power resources.

is a way to enhance the public perception of unions as modern and relevant

organizing strategies in the platform context in general. At the same time, it

a union perspective, the engagement for riders is a learning journey for

to organize the riders, which is described as improvisational unionism. From

traditional paths of collective bargaining and finding new and creative ways

Union-driven voice in

Both in Belgium and the Netherlands, riders organized collectively and

Qualitative comparative

Category

Vandaele

Key constructs and findings

2020

Methods and empirical

context

Author(s)

publication

Year of

104 Missing voice?

self-organization efforts. Unlike micro-workers, macro-workers organize themselves primarily via social media groups – not distinct forums. There are several reasons for that: (1) Macro-workers outsource tasks to other workers – trust is important. Social media groups foster trust better than

online freelancers (macro

crowdwork) in Southeast

Asia and sub-Saharan

Africa

2017

2017

2017

in crowdwork

among workers. These communities are the main tool for structuring

quantitative survey among

in gig work

in the gig economy is hard to organize – but possible. Being in the same location is an advantage. Despite that, sustaining momentum is hard – union

Foodora riders in Turin,

Italy

an alternative to the governance models of the existing platforms. A constitution ratified by all stakeholders formalizes the governance process and allows workers to participate in the platform’s strategic decisions as

a self-governing

crowdsourcing

marketplace (Daemo)

Gilbee, Majeti,

Richmond-Fuller,

Salih, Sarma, Mathur,

Vaish, Bernsetin

well as propose operational changes.

in crowdwork

platform Daemo to demonstrate that open-source governance is

in the context of

Dilrukshi, Shirish,

Pandey, Gaikwas,

Platform-driven voice

As part of a social activism initiative, the authors set up the crowdwork

Action research

Whiting, Gamage,

support would help make the efforts more sustainable.

Worker-driven voice

Due to lack of representation and collective bargaining, collective action

Single case study on

organizing labour via a platform also requires new forms of protest.

Italy

Maccarrone

in gig work

network and to fight for better working conditions. The new form of

Foodora riders in Turin,

Sica

Tassinari and

Worker-driven voice

Direct action (protests and strikes) has a dual function: to strengthen their

Single case study on

Animento, Di Cesare,

nature of the tasks).

(3) Macro-workers collaborate, whereas micro-workers do not (due to the

direct relationships with their customers outside the working platform.

the (mostly anonymous) forums. (2) Macro-workers want to develop

Worker-driven voice

Online communities foster the support and exchange of information

Qualitative study and

Graham

Category

Wood, Lehdonvirta,

Key constructs and findings

2018

Methods and empirical

context

Author(s)

publication

Year of

How do workers gain voice on digital work platforms? 105

backgrounds differ immensely. Empirical findings show that the use of online communities (forums, chats, apps), be they third-party or hosted by the platform, open up a space to interact and help build a shared identity. However, the actual collective actions that stemmed from these places were

(MTurk, Mobileworks,

Cloudfactory);

ethnographic research and

interview data

Scholz

Aloisi

2016

2015

shareholder-driven platforms. The alternative ownership model of platform

empirical examples

in crowdwork and gig work

and traditional means of unionization. Soft tools for organization are used to pool bargaining power and are often the first step towards more sustainable organizing efforts (e.g. worker associations/grass-root unions).

Uber, Lyft, TaskRabbit,

Freelancer, Handy,

InnoCentive)

Upwork, Wonolo,

Worker-driven voice

Different soft tools for organization (online communities, forums, add-ons)

nine platforms (MTurk,

in gig and crowdwork

Case study research on

marketplaces or labour platforms.

forms: from producer- or city-owned platforms to worker-owned online

mechanisms are similar, platform cooperativism can take various

all decision-making processes. Although the platform coops’ governance

coops formalizes the workers’ opportunity to engage and participate in

Worker-owned platform cooperatives (coops) as an alternative to

Conceptual piece with

Platform-driven voice

in crowdwork

a shared identity is much harder using labour platforms, as socio-economic

of three platforms

modest in scale, and it was difficult to find evidence of effectiveness.

Worker-driven voice

Successful collective action is contingent on a shared identity. Building

Comparative case study

Category

Lehdonvirta

Key constructs and findings

2016

Methods and empirical

context

Author(s)

publication

Year of

106 Missing voice?

in crowdwork

academic social activism approach to enhancing the exchange among Amazon Mechanical Turk workers, increasing transparency by giving workers the opportunity to evaluate the clients, thus increasing worker voice

research (setting up

Turkopticon)

by making workers visible in the work process.

Platform-driven voice

Turkopticon (browser extension and web-based app) as the result of an

not be replicated.

requesters)

on MTurk and action

(academic requesters with university-specific ethical guidelines) and it may

Are Dynamo for academic

Ethnographic research

creative mechanisms to enforce labour standards. But the setting is special

research (setting up We

Ogbe, Clickhappier

Irani and Silberman

in crowdwork

action for a highly dispersed workforce and to identify sites of leverage and

on MTurk and action

2013

Platform-driven voice

The approach demonstrates the capacity of forums to facilitate collective

Ethnographic research

Bernsetin, Alkhatib,

Category

Saheli, Irani,

Key constructs and findings

2015

Methods and empirical

context

Author(s)

publication

Year of

How do workers gain voice on digital work platforms? 107

PART III

Meso forces

5. How anger sparks voice: exploring individual and collective voice in the remote gig economy Alex J. Wood, Nick Martindale and Vili Lehdonvirta INTRODUCTION The gig economy can be divided into local gig work, which includes transport and delivery services, and remote gig work, which consists of digital tasks that can be delivered over the Internet, such as data entry, graphic design and content writing. Remote gig work can be further broken down into microwork (or digital piecework; Lehdonvirta, 2018) and ‘macrowork’ consisting of larger projects (Wood et al., 2019a). Deliveroo and Foodora delivery workers undertook dramatic examples of collective voice in 2016 which generated a great deal of public interest (Cant, 2019; Tassinari and Maccarrone, 2020). However, long before these events, remote gig workers, using platforms such as Amazon Mechanical Turk, Upwork and Fiverr, had been engaging in other forms of voice, which attracted less attention but were likewise anomalous in that the participants were seemingly self-employed workers using platforms to connect with clients (Irani and Silberman, 2013; Johnston and Land-Kazlauskas, 2018; Lehdonvirta, 2016; Panteli et al., 2020; Salehi et al., 2015; Wood and Lehdonvirta, 2021; Wood et al., 2018). Employment Relations and Human Resource Management research has tend to investigate voice in terms of the existence of collective and individual mechanism within firms by which ‘employees have “a say” in organisational decision-making’ (Wilkinson et al., 2018, p. 713) with particular focus on unions (e.g. Freeman and Medoff, 1984). However, when researching the gig economy there is a need to take a broader perspective which is not limited to employees and organisational mechanisms. Taking inspiration from Hirschman’s (1970) original ‘Exit, Voice and Loyalty’ framework, we instead adopt a broader action-orientated approach. Hirschman (1970, p. 30) defines voice as ‘any attempt at all to change, rather than to escape from, an objectionable state of 109

110

Missing voice?

affairs … through individual or collective petition … actions and protests, including those which are meant to mobilise public opinion’. In terms of collective voice an index of media reports finds that there have been 330 incidences of platform worker collective voice globally since January 2015 (Joyce et al., 2020). Examples of collective voice in the local gig economy include strikes, demonstrations and legal action aimed at influencing a variety of targets such as online platforms, policymakers, customers (Cant, 2019; Joyce et al., 2020; Tassinari and Maccarrone, 2020). Wood and Lehdonvirta (2021) highlight an example of social media being used by remote gig workers as a means to collectively protest against a platform which altered the fees they charged workers. However, other researchers indicate that in most cases social media facilitates individual voice among remote gig workers towards both platforms and clients (Anwar and Graham, 2020; Gegenhuber et al., 2020; Gerber, 2020; Gerber and Krzywdzinkski, 2019). In this chapter we investigate the circumstances under which workers in the remote gig economy engage in and support three different voice activities: individual voice towards clients, support for collective organisation, and state regulation. Investigating voice in the remote gig economy may be particularly revealing of wider voice dynamics in the gig economy, as this sector most closely resembles the ideal typical form of such work, that is, the use of platforms (or multi-sided markets) by self-employed workers to match with clients, with both parties being free to choose who they contract with, the rate charged to do so, and whether other workers can be brought into a project (Lehdonvirta et al., 2019). In contrast, the local gig economy perhaps more closely resembles the conventional employment relationship, as these workers often experience less discretion over their fees, methods of work and clients (Cant, 2019; Tassinari and Maccarrone, 2020). In fact, in some cases local gig platforms have actually employed their workforce (Ivanova et al., 2018). In investigating the drivers of different voice activities in the gig economy our research situates platform work within industrial relations and highlights that this field should not be limited to employees as has previously been suggested.

A HISTORY OF VOICE AT WORK Understanding worker collective voice has been central to the academic field of industrial relations (Kelly, 1998). Industrial relations as a field developed in both the UK and USA in reaction to explosive episodes of collective voice known as the ‘Labour Question’, ‘Labour Problem’ or ‘Social Question’: the numerous labour strikes and demonstrations of the mid-19th and early 20th centuries that often descended into violence (Kaufman, 2004). For much of the labour movement’s history, workers influenced their employers not through formal collective bargaining, but through ‘sabotage and direct action’

How anger sparks voice

111

(Hobsbawm, 1952, p. 58). This could usually be counted on to improve workers’ conditions, creating minimum standards which employers understood needed to be maintained to avoid negative sanctions such as sabotage. The mainstream establishment of industrial relations as both a field of practice and of academic study was largely a consequence of two violent episodes of collective voice: the 1914 Colorado miners’ strike and the Standard Oil of New Jersey strikes of 1915–16. Rockefeller Jr., the world’s richest person and the owner of the companies at the centre of this violence, responded to the public outcry by becoming ‘a life-long proponent of IR [Industrial Relations] and, in particular, collective employee voice’ (Kaufman, 2008, p. 328). Rockefeller not only instituted reforms in his own companies but was a major funder of the new academic field of IR (Kaufman, 2004). The ‘golden age’ of this new academic field began in the USA during the 1930s. Of particular importance was the passing of the 1935 Wagner Act. This had the effect of shifting the focus of IR as a field from spontaneous strikes, demonstrations and riots towards other forms of collective voice, specifically formal collective bargaining between employers and recognised trade unions (Kaufman, 2004). The central text of this golden age was Dunlop’s (1958) Industrial Relations System. Dunlop’s analysis centred on explaining the ‘body of rules created to govern the actors at the workplace and work community’ (Dunlop, 1958, p. 7). The three main actors producing and governing this emergent web or rules were the state, managers, and workers and their representatives. Given that IR emerged as a direct consequence of the perceived threat posed by worker protest and was then legitimised as an area of academic inquiry through the institutionalisation of this collective voice via formal collective bargaining, it is unsurprising that trade unions would be the focus of this field (Dundon et al., 2004; Wilkinson et al., 2020). However, this ‘golden age’ of IR was not to last and the web of rules that Dunlop had identified began to shrink (Wright et al., 2019). Trade unions, key actors in his system, entered decline. Industrial relations scholars in both the UK and the USA sought to maintain the field’s relevance by recentring it on the wider conceptual terrain of the ‘employment relationship’ (Edwards, 1995; Kaufman, 2008, 2010; Sisson, 2008). As Edwards (1995, p. 9) explains, ‘a focus on collective bargaining no longer captured the key issues of the regulation of the employment relationship’. Instead of collective action, the starting point of inquiry became the employment relationship. As a side effect of this refocusing on the employment relationship, IR came to be understood as excluding any ‘domestic labour and also the self-employed and professionals who work on their own account: the contractual relations between a self-employed plumber and his customers are not “industrial relations”’ (Edwards, 2003, p. 2). At a similar time some writers in the human resource tradition began to broaden research

112

Missing voice?

on worker voice away from trade unions and instead highlighted the variety of channels through which it could operate (Dundon et al., 2004; McCabe and Lewin, 1992; Millward et al., 2000; Wilkinson et al., 2020). Most importantly, this literature stressed the growth and potential of voice via direct employee involvement mechanisms, such as ‘upward problem solving groups, quality circles, suggestion schemes, attitude surveys, [and] self-managed teams’ (Dundon et al., 2004, p. 1152). Voice in the Gig Economy Voice in platform work has the potential to be more complex than posited by traditional industrial relations. As outlined above, IR traditionally focused on the employment relationship and the relations between employees, employers and the state (Dunlop, 1958; Edwards, 2003). However, platforms act as two-sided markets matching customers who request goods and services with self-employed workers who provide them. Extant research mirrors this conceptual complexity, highlighting the existence of a diversity of voice behaviour across the many industries and platforms that make up the gig economy. Voice activities include collective participation in strikes, demonstrations and legal actions (Joyce, 2020), as well as the individual use of platform communication and consultation channels (Gegenhuber et al., 2020). Voice activities have been documented in which the targets of workers’ demands have variously been: customers (Irani and Silberman, 2013; Lehdonvirta, 2016; Wood et al., 2018); platforms (Cant, 2019; Lehdonvirta, 2016; Tassinari and Maccarrone, 2020; Panteli et al., 2020; Salehi et al., 2015; Wood and Lehdonvirta, 2021); and the state (Aslam and Woodcock, 2020; Johnston and Land-Kazlauskas, 2018). Workers have participated in strikes, demonstrations and legal actions; sought to influence price norms; and utilised and contributed to client-rating software. However, across the existing scholarship, there appears to be little evidence of any industry-based pattern in the forms or targets of voice. This is perhaps surprising as most literature on employee voice assumes that employers will be the target of workers’ attempts to have a say over their working lives – a situation that reiterates that voice in the gig economy is more complex than in conventional employment settings. An analysis of media reports undertaken by Joyce et al. (2020) finds substantial variation in the patterning of the protest actions undertaken by platform workers. This raises the question of which factors might then explain voice dynamics in the gig economy. Tassinari and Maccarrone (2020: 5) argue that collective voice in the gig economy is underpinned by the degree to which the work naturally generates ‘collective feelings of reciprocity and responsibility’. The term solidarity relates to ‘a communal sense of obligation to support collective action’ (Heckscher and McCarthy, 2014, p. 629) and can be understood in terms of

How anger sparks voice

113

Putnam’s (2000) discussion of social capital in which solidarity develops via bonding and bridging within and between groups with situational commonalities (Jarley, 2005; Morgan and Pulignano, 2019; Saundry et al. 2012). Wood (2015, 2020a, 2020b) has demonstrated how inter-worker communication via digital technologies can lead to the development of a sense of ‘networked solidarity’ amongst spatially and temporally fragmented workers. Likewise, in Tassinari and Maccarrone’s (2020) study of Foodora and Deliveroo workers, solidarity grew between gig workers as they provided each other with support while gathered at order waiting points and via digital networks. The importance of communication for solidarity has been additionally supported by studies that find gig workers engage in digital communication in search of help and support for dealing with customers (Anwar and Graham, 2020; Aslam and Woodcock, 2019; Maffie, 2020; Wood et al., 2018; Wood et al., 2019b). Such embryonic solidarity has been found to form the basis of protest against platforms and support for unions (Aslam and Woodcock, 2020; Cant, 2019; Cant and Woodcock, 2020; Maffie, 2020; Newlands et al., 2018; Tassinari and Maccarrone, 2020). Extant research on collective voice in the local gig economy also highlights the role of antagonisms and anger over issues such as low pay, changes to the payment method, uncertainty over earnings and working hours, absence of sick pay, health and safety, non-transparent performance management, shift allocation, and deactivation (Cant, 2019; Tassinari and Maccarrone 2020). Similarly in the remote gig economy, Wood and Lehdonvirta (2021) detail how antagonism manifests as anger towards platform fees, low pay rates, and lack of voice over changes to how the platform operated. Wood and Lehdonvirta go on to document how this antagonism led to collective protest and to workers supporting collective organisation and unions as a means for better representing their interests and enhancing their influence over platform decision-making. However, Wood and Lehdonvirta (2021) maintain that such collective voice is only likely when workers are unable to easily exit their platforms due to dependency resulting from network effects, data lock-in and the role of platforms as gatekeepers. Likewise, earlier research found that gig workers dependent on platform work were more likely to support collaborative efforts to raise their wages (Wood et al., 2018). Joyce (2020) argues that platform dependence fuels collective voice by creating an unstable ‘cash nexus’ between workers and their platforms. Indeed, dependent workers experience greater levels of dissatisfaction, less autonomy, lower hourly wages, and more precarity than workers who use platforms to supplement their income from other sources (Schor et al., 2020). Moreover, Tassinari and Maccarrone (2020) argue that workers who are less attached to gig work might differ in their interests and demands regarding issues, such as renumeration, contracts

114

Missing voice?

and flexibility. Meanwhile Barratt et al. (2021) find that weakly attached food delivery platform workers are less likely to utilise voice. In summary, existing gig economy research has highlighted three important aspects of gig economy social relations that lead both to the emergence of and support for collective voice: (1) anger generated by poor working conditions; (2) experiences of dependence on particular platforms; (3) emergence of solidarity between workers on the same platform. However, researchers have yet to explain why these factors lead to different forms of voice at different times, including less well-studied instances of individual voice towards customers or claims made on the state. Methodologically, existing research has largely relied on interviews with workers involved in singular instances of protest or cycles of contention. This makes it difficult to draw broader conclusions about the determinants of voice in other settings, why some workers fail to protest when others do, or why the targets and forms of voice have been so variable. There has been very little quantitative research into the conditions under which workers support or engage in voice; only Maffie (2020) and Newlands et al. (2018) use survey data to study aspects of the phenomenon. Appropriate survey data can contribute towards our understanding of voice in the gig economy by reaching a broader range of workers and by assessing the simultaneous influence of individual and contextual factors, which influence both support for protest and individual acts of resistance. In the rest of this chapter, we seek to address the limitations of previous research on gig economy protest by using novel survey data to investigate the role of antagonisms, dependence and solidarity in explaining gig worker support for different forms of voice.

DATA AND METHODS Data for the analyses comes from a recent survey of freelancers (N = 430) working on two prominent online labour platforms for remote digital services. The services include, for example, data entry, graphic design, digital marketing, translation, transcription and programming. The survey was commissioned by the European Centre for the Development of Vocational Training of the European Union to investigate online freelancers working in six European countries: Finland (N = 76), Germany (63), Italy (70), Romania (78), Spain (69), and the UK (74). The survey included questions on workers’ experiences of and attitudes towards online platform work. The survey targeted workers aged 18 and over, with experience on the target platform, and located in one of the six target countries. Workers belonging to this population were identified using platform search functions and invited to complete the survey until equal quotas for sex and country were approximately filled. Workers were picked randomly from each search results page as opposed to systematically following

How anger sparks voice

115

the results, to mitigate algorithmic bias and to introduce a degree of randomization to the sample. A response rate of 20 per cent for successfully completed questionnaires was achieved. Survey respondents were paid USD 9.50, EUR 8.50 or GBP 7–7.50 depending on their preferred means (in-platform payment or Amazon gift card), exchange rates and platform fees. Despite best efforts, the sampling procedure does impose limitations on the generalisability of the findings. In particular, there is likely self-selection into the sample by preference for answering surveys, which could be due to differences in opportunity costs. However, our respondents are similar in age and educational characteristics to the respondents of a large-scale survey conducted for the European Union’s COLLEEM project (Pesole et al., 2018). Respondents are young and highly educated: more than three quarters are under 40, and 68 per cent have completed at least an undergraduate degree. This suggests that at least in demographic characteristics, the sample is, despite conventional wisdom, reflective of the broader European gig worker population. To investigate the extent to which respondents engage in or support different forms of voice, we constructed three composite dependent variables. The variables measure the extent to which respondents (1) have taken action against their clients (‘Client Action’), (2) support collective organisation (‘Collective Organisation’), and (3) support greater application of state regulation on platforms (‘State Regulation’). Each of these composite dependent variables was constructed by summing relevant five-category Likert-scale items from the survey. Client Action comprises three items measuring the extent to which workers’ treatment by clients led them to (i) threaten to end a contract with a client, (ii) end a contract, and (iii) refuse to continue working for a client (  0.72). While ending a contract or refusing to continue to work for a client might appear to be examples of exit rather than voice, the fact that workers are not exiting the platform but attempting to change their experience on it by sanctioning clients means that conceptually this behaviour fits with Hirschman’s (1970) definition of voice. Collective Organisation likewise comprises three items: (i) wanting to improve wages collaboratively, (ii) interest in joining, and (iii) interest in forming a freelancer union (  0.83). State Regulation comprises two items: whether (i) national minimum wages should be paid by platforms and whether (ii) states should do more to regulate platforms (  0.38). While support for state regulation is not a conventional channel for worker voice, supporting the provision of such public goods does fit with mainstream conceptions of collective action (Olson, 1971) and relates to the view of the state as one of the main actors in industrial relations (Dunlop, 1958). Greater detail on the distributions of the underlying factors in these composite dependent variables (and the independent variables described

116

Missing voice?

below) can be found in the first stage of the analysis in the next section of this chapter. To investigate the determinants of support for different forms of voice, we also constructed three composite independent variables. These variables measure aspects of the social relations of gig work, which previous research has argued are associated with support for protest. The variables measure the extent of workers’ (1) anger at their working conditions (‘Anger’), their (2) dependence on the platform for work (‘Dependence’), and (3) how frequently they communicated with other workers (‘Communication’). The Anger variable measures how frustrated workers feel about working on the platform by summing nine relevant five-category Likert-scale items that capture attitudes towards pay from clients (too low, lower than deserved, late), fees charged by the platform (too high), and experiences of working for clients and platforms (treated unfairly, not supported, exploitative, too much competition from abroad, platforms have too much power) (  0.74). These items reflect the antagonism built into the labour process of platform work. The Dependence variable measures the extent to which workers are unable to act on their grievances or exert power over their clients or platforms. The variable comprises nine relevant items: unable to replace clients, unable to criticise clients, unable to switch platform without loss of income, unable to redress unfair treatment, feeling replaceable, feeling afraid to ask for better conditions, feeling helpless, having to work too many hours, fearing platform fee increases, and fearing the effect of unfair feedback on income (  0.72). 1 The Communication variable sums three relevant seven-category items which measure the frequency of communicating with other workers via online forums, social media (including text messaging and email), and face to face (  0.66).

FINDINGS Anger, Dependence and Communication Among Platform Workers First, we examine the distribution of responses for the independent variables. Our findings suggest that, although our respondents generally feel fairly treated, there is a high level of discontent related to the material rewards of platform work. Figure 5.1, which plots the distributions of underlying items in Anger, reveals that among our survey respondents at least 75 per cent of workers believe that the fees platforms charge are too high and that the pay which workers receive is too low. At least 50 per cent believe that their pay is less than they deserve, and that foreign competition should be limited.

How anger sparks voice

Figure 5.1

117

‘Anger’ independent variable underlying items

Sizable minorities believe that platforms benefit shareholders at the expense of workers (42%) and that platforms have too much power (24%). Figure 5.2, which plots the distributions of the underlying items in Dependence, reveals that many workers experience dependence on the platform on which they work. For instance, more than half could not switch the platform on which they work without substantial loss of income. More than 75 per cent feared that their platform would increase the fees they are charging. A significant minority of workers indicated that they had a weak bargaining position relative to clients, suggesting that they are likely to also be in a position of dependence with regards to their clients. For instance, more than 25 per cent of workers indicated that they are afraid to demand better work terms and hourly rates, that they feel helpless when unfairly treated by clients, that they are unable to easily replace clients, and that they feel replaceable. Additionally, 62 per cent of workers feared the effect of unfair feedback on their future income. Finally, with regard to the Communication variable, survey respondents had little face-to-face communication with other workers, with 85 per cent of workers never or rarely communicating in this way. However, as Figure 5.3 demonstrates, many workers were making use of digital technologies to communicate with each other on a regular basis, with 45 per cent communicating through either social media or online forums at least once a week. In com-

118

Figure 5.2

Missing voice?

‘Dependence’ independent variable underlying items

parison, Southeast Asian and sub-Saharan African platform workers working through similar platforms, surveyed by Wood et al. (2018), were slightly more likely to be in weekly contact with each other (+13 percentage points (pp)). As for the items comprising our dependent variables, we find that 55 per cent of workers have engaged in some form of voice towards individual clients: ending or threatening to end their contract or refusing to continue to work with them. Support for government regulation and minimum wages were also widespread, with 39 per cent supporting the former and 55 per cent the latter. Many of the workers we surveyed were also supportive of collective organisation: 59 per cent responded that they were interested in collaboratively improving wages and 50 per cent agreed or strongly agreed that, if there were a trade union or workers’ association for freelancers, they would join. That support for unions among the surveyed platform workers here was similar to that found among standard workers in several high-income countries (e.g. Charlwood, 2002; D’Art and Turner, 2008) is surprising given the individualist attitudes that autonomous, self-employed workers have been argued to hold (Langsæther and Evans, 2020). The union support figure in our survey is slightly higher (+12pp) than what Vandaele et al. (2019) find among Belgian Deliveroo workers. Again this is surprising, as the geographic isolation and spatial barriers to communication entailed by this remote online gig work might be expected to make the emer-

How anger sparks voice

Figure 5.3

119

‘Communication’ independent variable underlying items

gence of collective solidaristic attitudes more difficult than in food delivery. However, the percentage of surveyed workers who support collaboratively improving wages (59%) is slightly lower than the 71 per cent figure obtained in a survey of similar African and Asian workers by Wood et al. (2018). As one would expect, far fewer workers were interested in organising a union themselves, with 9 per cent agreeing or strongly agreeing. This is again lower than the 25 per cent figure for African and Asian workers (Wood et al., 2018). Lower support for collective organisation among European platform workers is surprising and will be returned to in the discussion. Using Anger, Dependence and Communication to Explain Protest To assess the influence of our three independent variables on workers’ propensity to support and engage in different forms of protest, we estimate three OLS regressions, specifying the same model (1) for each of the three dependent variables: Individual Action, Collective Organisation and State Regulation. Models include controls, Xk, for workers’ age, sex, education level (whether possessing an undergraduate degree),2 country of residence, and platform. We

120

Missing voice?

standardise the constructed dependent and independent variables to facilitate the interpretation of results.3 Y    1 Anger   2 Dependence  3Communication 



n

 X k 4 k k

  (1)

Results in Table 5.1 reveal that platform workers’ feelings of anger are significantly associated with all three dependent variables: individual voice against clients (Individual Action), support for collective organisation (Collective Organisation), and support for the greater application of state regulations to platforms (State Regulation). The directions of the associations are consistent with the hypothesis that angrier workers are more likely to engage in or support voice. This means, for example, that a worker answering ‘Strongly Disagree’ instead of ‘Disagree’ on five of Anger’s underlying items is associated with answering ‘Strongly Disagree’ instead of ‘Disagree’ on one of the items in State Regulation. Communication with other platform workers is significantly associated with both Individual Action and Collective Organisation but not with State Regulation. Workers who are more frequently in contact with other workers on their platforms are more likely to engage in individual voice towards a client and to support Collective Organisation. Substantively, answering three categories more extreme on Communication (e.g. going from ‘Never’ to ‘Several Times a Week’) is associated with answering between a half and a whole category more extreme on both Individual Action and Collective Organisation. Workers who are more dependent are more likely to support Collective Organisation but not Individual Action or State Regulation. The association is not particularly strong, however. Substantively, answering five categories more extreme on Dependence items is associated with answering half a category more extreme on Collective Organisation. Results are not sensitive to which of the two platforms in the sample workers use. There is also little significant variation by country of residence; however, workers in the UK are less likely to support Collective Organisation than those in the reference country, Finland. Workers in Romania are more likely to have taken individual action against clients. Age is only significantly associated with support for Collective Organisation, where support declines with age. Gender is only significantly associated with support for State Regulation, where men are less likely to support regulation than women. Education levels are not associated with any of the dependent variables. The R2 values of the models are low, suggesting that a significant amount of variation in the dependent variables is either idiosyncratic or explained by unobserved variables. This is not unusual for research which seeks to explain complicated social phenomena such as attitudes towards protest.

121

How anger sparks voice

Table 5.1

Results of multivariate analyses Dependent variable State Regulation

Client Action

Collective Organisation

Anger Dependence

(1)

(2)

(3)

0.371***

0.209***

0.178***

(0.052)

(0.054)

(0.054)

0.086

0.053

0.132*

(0.051)

(0.054)

(0.054)

Communication

−0.002

0.228***

0.211***

(0.044)

(0.046)

(0.046)

Age

−0.0004

0.008

−0.012*

(0.005)

(0.005)

(0.005)

Male (Ref.: Female) Graduate (Ref.: Non-Graduate) Platform 2 (Ref.: Platform 1)

−0.204

*

−0.144

−0.015

(0.087)

(0.092)

(0.091)

0.111

0.094

−0.023

(0.093)

(0.098)

(0.097)

−0.145

−0.034

−0.061

(0.116)

(0.122)

(0.121)

Country (Ref.: Finland) Germany Italy Romania Spain United Kingdom Observations R

−0.061

0.297

−0.208

(0.154)

(0.162)

(0.160)

0.296

−0.036

−0.119

(0.156)

(0.164)

(0.163)

0.043

0.513***

−0.130

(0.146)

(0.154)

(0.153)

0.191

0.071

0.163

(0.154)

(0.162)

(0.161)

0.049

0.229

−0.320*

(0.152)

(0.160)

(0.159)

430

430

430

0.236

0.156

0.169

Adjusted R2

0.212

0.129

0.143

Residual Std. Error (df = 416)

0.887

0.933

0.926

9.897***

5.903***

6.511***

2

F Statistic (df = 13; 416)

Note: *p