Are platform workers willing to unionize? Exploring survey evidence from 14 European countries
Abstract
Embedded in the particular model of work organization of digital labour platforms, platform workers face several hurdles discouraging them from becoming trade union members. These relate to algorithmic management, regulatory arbitrage regarding the employment arrangements and the promotion of an entrepreneurial orientation among platform workers. Nevertheless, based on data from a representative survey in 14 European countries, union density in the platform economy stands at 13.4 per cent. This should be interpreted as a kind of “platform unionism” that exists by coincidence, however, as union membership is most likely rooted in the labour market status of platform workers in the conventional economy. Compared to the general population, platform workers have stronger pro-union attitudes and are more receptive to union membership. Probably partly reflecting difficulties in the ability to unionize, there is still a gap, though, between attitudes and willingness to unionize: whereas about two-thirds of platform workers hold positive attitudes towards unions, only over a quarter state that they would like to join a union. Apart from those positive pro-union attitudes, the propensity to unionize also seems to be determined by engagement in offline networks that promote a social norm of union membership and online participation in digital work-related communities. While these findings could inform union recruitment and organizing strategies, it is needless to say that the heterogeneity of the platform workforce, strongly influenced by the different ways in which workers participate in the platform economy, requires at the same time tailor-made strategies.
Introduction
There is no lack of evidence on social protest among platform workers. Several studies have analysed how a critical mass of them are mobilizing against digital labour platforms,1 although individual forms of resistance are probably at least equally important. Often of a spontaneous and ephemeral nature, the collective resistance of platform workers has spread across the globe since around 2015.2 Local acts of ‘disrupting the disrupter’ in urban settings have been quantitatively documented and mapped globally by the ‘Leeds Index of Platform Labour Protest’.3 This index reveals that protest is dominant in location-based platform work such as transportation, especially in app-based food delivery, and in Europe, at least before the Covid-19 pandemic. While this demonstrates the ‘willingness to act’ of platform workers, it is an open question whether they also have the ‘willingness to pay’;4 that is, to become formal trade union members and pay union dues.
Associational power through collective organization has been evaluated as crucial since the workplace and marketplace bargaining power of platform workers is considered almost futile,5 albeit with differences between types of platform work.6 Exercising freedom of association would allow platform workers to represent their interests and needs better and to gain bargaining strength for improving their wages and employment terms and conditions, including on health and safety issues, with trade unions able to bring more stability to the platform labour market and act as a countervailing power vis-à-vis the digital labour platforms. Various obstacles to building workers’ solidarity and developing associational power in the platform economy have been identified, however. The particular organizational work model of most digital labour platforms hampers unionization. This model is based on: (1) algorithmic management; (2) regulatory arbitrage regarding the employment arrangement; and (3) an asset-light model framed and promoted as ‘micro-entrepreneurship’.
First, although surveillance via algorithmic management is never complete, its disciplinary control over task allocation and performance tends to suppress worker voice.7 Algorithmic management also implies that “face-to-face dialogue with an authoritative figure”,8 needed for establishing a relationship between management and platform workers and their representatives, is simply absent, while proper dispute resolution mechanisms are lacking. Platforms act as “shadow employers”,9 as direct supervision is unwarranted due to individual online ranking and reputation systems which often reflect or even reinforce forms of discrimination and impel self-disciplining behaviour among platform workers. Moreover, participation in industrial action might lead to an outright de-activation of a worker by the platform, effectively discouraging any further action.10
Second, workers are in general not employed by digital labour platforms.11 Using a contractor relationship instead grants platforms more scope for adjusting labour supply and minimizing labour costs, with associated risks being shifted to the shoulders of platform workers. In cases where the (mis)classification of their employment status as “independent contractors” has been undone via litigation actions or rulemaking, employment terms and conditions have rarely been improved.12 The limited impact of successful court cases is either due to a deficiency of state enforcement or because rulings usually apply only to the plaintiff and not to all workers on the same platform or the whole platform economy in a country.13 Moreover, digital labour platforms have responded to unfavourable rulings by unilaterally changing the terms and conditions of the contracts, or their phrasing, to evade the applicability of a ruling. The contractor status of platform workers poses a major challenge to collective bargaining on wages and working conditions as competition rules and laws are seen as precluding the setting of minimum prices in such labour markets.14 Equally, platform workers classified as independent contractors might hinder unionization as they are simply not allowed to become a trade union member due to internal union rules or legislation.15
Third, the combination of algorithmic management, induced gamification, the reliance on client rating systems and independent contractor status can foster an “entrepreneurial orientation”16 among platform workers and stimulate between-worker competition. Research has hinted at the possible internalization by workers of an “entrepreneurial framing”17 constructed by the platforms. This, in turn, can hamper workers’ collective identity, solidarity and their openness to unionization. Such a claim has been made for food delivery couriers in on-location platform work,18 for instance, as well as remote, online platform workers.19 Also, the subjective understandings of workers about their platform work might ‘fit’ well with their individual life circumstances (like in app-based food delivery20 ) so that they see fewer reasons to mobilize or to unionize for better wages and employment terms and conditions. Competition between workers is further exacerbated by the algorithmic recruitment policies of digital labour platforms, with low entry barriers ensuring a continuous influx of new workers, resulting in high labour market turnover and a structural oversupply of labour relative to available tasks.21
As such, platform work is characterized by short tenure and job instability:22 it tends for most platform workers to be a side-job or to be perceived as a temporary, transitory stage in a working life, although there are exceptions.23 The general volatility of platform work might lower the willingness to invest time and energy in improving employment terms and conditions via collective organization in trade unions. Equally, from a cost-benefit perspective, it could be questioned if platform workers are willing to pay union dues given their generally (very) low earnings,24 while the small minority of platform workers who gain a decent income from their platform work might feel less need to unionize.
Based on a quantitative analysis of individual-level survey data from 14 European countries, this working paper explores and analyses to what extent platform workers are willing to become union members. Three main arguments are formulated and discussed in the paper. First, studying the unionization levels of platform workers and their propensity to unionize cannot be detached from their labour market status in the conventional, offline economy. Given that platform work tends to be juggled with other forms of employment,25 the specific insertion of workers into the platform economy should be considered jointly with status in the conventional economy. Second, the drivers for unionization among platform workers are fairly similar to those for non-platform workers in the conventional economy. Pro-union attitudes, offline embeddedness in union-friendly social networks and online engagement in work-related digital communities all help to explain the likelihood of unionization in the platform economy. Moreover, workers performing platform tasks are more likely to join a trade union compared to similar groups in the general population. Third, in terms of practical implications, unions should develop diverse recruitment and organizing strategies accounting for the heterogeneity of the workforce within the platform economy. While the drivers for unionization are generally similar across different types of platform work, workers differ in their patterns of participation in the platform economy and in their levels of economic dependence on and attachment to it, which are at least partly linked to the type of platform work. This calls for tailored union recruitment and organizing approaches.
Inspired by the literature on trade union revitalization, this working paper aims to contribute to the understanding of the determinants of unionization in four ways. First, it focuses on a new group of workers, platform workers, who have been much less studied in terms of their possible drivers for unionization. Second, instead of focusing on a single type of platform work, it takes a comprehensive approach by comparing workers in various types of platform work. Platform workers are by no means a homogeneous group; they include both precarious workers in bogus contractual arrangements and the genuine solo self-employed, with the latter tending to exercise considerable labour market bargaining power, at least in the conventional economy.26 Third, while previous research on platform work has largely centred on western European countries, in terms of its geographical scale, this working paper also includes countries from central and eastern Europe. Finally, the likelihood of unionizing in the platform economy is not studied in isolation, but in comparison to the conventional economy.
Workers’ various patterns of participation in the platform economy and union strategies
The collective agency of platform workers seems almost futile given the many structural constraints and systematic obstacles to unionization in the platform economy. While such a pessimistic account simply ignores the realities of their mobilization and resistance, it also tends to conceive of the “online” platform economy as a separate ecosystem, disentangled from the “offline” conventional economy.27 Such a division may be fruitful for analytical reasons, but the engagement of workers in platform work cannot be considered entirely separately from the conventional economy. Calling for a nuanced approach for studying their unionization, this section builds the argument that the specific insertion of workers into the platform economy should be taken into consideration in understanding their willingness to become trade union members. Those insertions are contingent on the status of platform workers in the conventional labour market. The peculiarities of platform work should therefore be de-emphasized and, in a similar vein, its halo of novelty should be dimmed. The organizational work arrangements of platform work are reminiscent of older practices,28 while the work is more than often a type of precarious or vulnerable employment fitting into long-term tendencies towards the relative erosion of standard employment relationships, workplace fissuring29 and the deregulation, flexibilization and informalization of labour markets in the conventional economy, at least in the Global North.30
Platform work has both regulatory and economic links to labour markets and welfare state regimes in the conventional economy. In terms of regulation, for instance, there are generally few dedicated or specific rules regulating the employment status of workers in the platform economy so far. Economically, too, motives for engaging in platform work are linked to specific dynamics within local labour markets, with relatively poor offline working conditions and fewer employment opportunities locally encouraging workers to combine these with work on digital labour platforms.31 Labour market status in the conventional economy can thus influence, though not determine, the pattern of engagement in platform work in terms of, for instance, tenure, economic attachment to it, as measured by the number of working hours or income, and the type of tasks performed in platform work, the latter of which is associated with different skill levels.32 Equally, relatively low levels of social welfare benefits or workfare policies permitting or promoting their combination with additional income sources might in a similar way encourage beneficiaries to perform platform work. The operation of digital labour platforms could therefore be considered “parasitic” since they (opportunistically) hinge on fractures or openings within traditional welfare state regimes as they shift to workfare regimes.33 The platform economy is thus able to attract workers beyond those being formally employed in the conventional economy, especially due to its low entry barriers and hitherto unregulated nature. This particular influx beyond the traditional labour market contributes to the heterogeneity and segmentation of the platform workforce.
The different ways in which workers participate in the platform economy require that trade unions integrate this into their recruitment and organizing strategies. Established workplaces are usually the key locus of such strategies, often facilitated by union recognition agreements with management or statutory union access at company level.34 Platform work obviously challenges this workplace-oriented approach of unions. The physical co-presence of platform workers and formal common workplaces are generally absent in the platform economy. This holds true for both on-location and remotely performed platform tasks, although functional equivalents can be found in on-location platform work such as parking areas in private ride-hailing35 and restaurant zones in app-based food delivery or other socializing spaces.36 Nevertheless, what kind of strategies could unions embark on to recruit and organize a spatially dispersed and isolated workforce in the platform economy?
The argument is put forward here that such strategies could hinge on three relevant drivers for unionization. Those drivers are the individual pro-union attitudes of platform workers, their degree of embeddedness in union-friendly offline social networks and their online activity in digital work-related communities – see Figure 1. The latter two drivers point to the importance of community unionism and digital unionism, respectively: two types of networked unionism complementing and overcoming a solely workplace-oriented approach.37
Figure 1 – Studying unionization in the platform economy38
First, assuming that job characteristics affect employment relations and attitudes towards trade unions,39 the various patterns of insertion of workers into the platform economy based on status in the traditional labour market might be indicative of their attitudes towards unions, with pro-union attitudes considered an important driver for unionization.40
Second, the relative novelty of digital labour platforms, with their particular organizational work model, might differentiate platform workers from the rest of the labour market, potentially fostering perceptions of common interests and grievances, and has opened space for informal, bottom-up initiatives among them.41 Offline network-driven processes and structures of solidarity tend to emerge “outside the traditional union framework”,42 particularly in on-location platform work such as ride-hailing43 and food delivery.44 Such offline social networks and bottom-up collectives are considered breeding grounds for mobilization and collective action. They are communities of support, often structured along ethnic lines. They could also foster shared identities and build solidarity among platform workers, and this is something that trade unions can use to connect and engage with platform workers, in particular in location-based platform work. At the same time, however, there is no guarantee of success as such networks and collectives might equally want to be independent of (mainstream) unions or even hostile to them.45
Third, online networks arise when platform workers are engaging in work-related digital, online communities for overcoming their spatial dispersion. Such communities operate in the first place as a space for sharing experiences or complaints, either in online platform work46 or in on-location platform work where online interactions could complement offline encounters. The evidence is hitherto limited whether online communities go beyond individual types of resistance and individual approaches to coping with the uncertainties produced by the platforms. Digital communication channels and tools might not always be autonomous of the digital labour platforms, thus suffering from the risk of being monitored by the platform or used to enhance its business performance rather than serve workers’ needs and interests. As such, online professional platform workers might tend to align themselves functionally to “improving and optimising [the] organizational performance”47 of the platforms. Online communities might nevertheless stimulate a sense of shared identities and the unified needs and interests which are prerequisites for collectivism.48 On the condition that online communities are independent of digital labour platforms and go beyond accommodating to the existing work practices of the platforms, they have the potential of becoming a transformative entry point for unionization.49
Data and methods
To analyse the attitudes of platform workers towards trade unions and their willingness to unionize, this paper makes use of large-N individual-level data from the Internet and Platform Work Survey (IPWS) carried out by the European Trade Union Institute (ETUI).50 This is a representative cross-national, multi-lingual survey among a simple random sample of the working age population between 18-65 years old, with the fieldwork harmonized and coordinated by Ipsos. Interviews are conducted via computer-assisted telephone interviewing implying answer options being kept simple and the questionnaire reasonably compact.
The analysis uses data pooled from two survey waves, carried out in Spring and Autumn 2021, and including in total 36,124 respondents from fourteen European countries of which seven are in central and eastern Europe: Austria, Bulgaria, Czechia, Estonia, France, Germany, Greece, Hungary, Italy, Ireland, Poland, Romania, Slovakia and Spain. The country selection offers a good geographical representation of Member States of the European Union (EU), spanning all regions apart from the Nordic countries – the latter being marked by high union density. Moreover, the 14 countries included in the survey in fact cover 84 per cent of the working age population in the 27 EU Member States. This, together with the large representative random sample of respondents, allows the generalizing of conclusions to the workforce at EU level with a reasonable degree of confidence. This working paper adopts a more holistic approach, however, instead of highlighting country differences or similarities. Post-stratification weights have been applied in both the descriptive analyses and regression models.
The survey contains a battery of questions on the prevalence and intensity of different types of digitally-mediated work.51 Platform workers are those respondents who self-report having performed any of the following digitally-mediated services in the past 12 months, with overall six types of platform work discerned on this basis: (1) remote platform workers performing relatively simple micro-tasks; (2) remote professional freelancers performing more complex tasks such as graphic design and multimedia, sales and marketing support, software and web development, writing and translation, although differentiating between requester- or worker-initiated transactions is not possible; (3) location-based private ride-hailing; (4) location-based app-oriented food delivery; (5) other types of location-based services carried out in the private sphere, usually in clients’ households, such as care, tutoring or handyperson work; and (6) workers who are combining more than one of the above activity types, the latter being labelled as “multiple platform work”.52 To validate these self-reports of digitally-mediated work, respondents are also asked to provide the name of the website or app they use. Only when this name matches an actual online labour platform, algorithmically matching workers with clients, collecting client ratings and handling payments, are respondents classified as platform workers.
The main focus of the analysis in this working paper is on non-unionized platform workers,53 defined as respondents who have performed work through a digital labour platform in the last twelve months and who are not a member of a trade union. Willingness to unionize is measured by the question “Would you be willing to become a trade union member in the near future?”, with “yes” and “no” as possible responses.54 The final sample for the analysis includes 1,961 platform workers of whom 1,698 are non-unionized; 28.3 per cent of the latter declared that they would like to join a union in the near future. To contextualize the findings about platform workers and to reveal to what extent they might be distinct from other workers, throughout this working paper they are compared to the general population; thus, working age adults who have not worked on digital labour platforms in the past 12 months.
The individual-level data from the IPWS provide rich information on the respondents, allowing for comparisons between similar workers in the online, platform economy and the offline, traditional economy. Apart from several sociodemographic characteristics of the respondents, their attachment to and tenure in the platform economy and their labour market status are considered. Regarding the latter, it is assumed that self-reported status mostly refers to status in the conventional economy, although this might be different for platform workers with a strong attachment to the platform economy. Overall, five possible combinations of platform work with “offline” labour market statuses are distinguished: the active, either (1) employees (i.e. wage- and salary-earners either on temporary or open-ended contracts), (2) own-account workers or the solo self-employed and (3) the unemployed; and the inactive, either (4) students or (5) persons economically inactive because of care responsibilities or other reasons like pension. The formal contractual status of platform workers vis-à-vis the digital labour platform is not known, however. This implies that country differences in the regulatory or legal framework on the employment status of platform workers cannot be analysed. Put differently, only the probability of platform workers becoming unionized is researched; their formal ability to become a union member is thus not part of the analysis due to a lack of relevant data.
Accounting for the heterogeneity of platform work
The workforce in the conventional economy is not uniform in its interests and needs. The workforce in the platform economy has equally been found to be diverse and non-homogenous.55 Accounting for the heterogeneity of platform work is therefore crucial for trade unions. Heterogeneity might refer, for instance, to the economic risks of the work or the effective job autonomy over which tasks are undertaken and how they are performed. All of this is likely to influence union strategies and tactics for extending their offer to platform workers as well as informing their priorities in selecting target groups and understanding their (diverse) interests and needs.
Two major distinctions for considering the heterogeneity of platform work are made in the empirical analysis below: the first is based on the degree of economic dependency or attachment to platform work;56 and the second is connected to the type of tasks performed in platform work. In addition, differences between workers with various lengths of tenure in platform work are considered. The overview below provides an overall picture of the heterogeneity and segmentation of the workforce in the platform economy, with a focus on the links and commonalities with the conventional, offline economy.
Table 1: Comparison between the general population and platform workforce57
|
General population (n=31,953) |
Platform workforce (n=1,961) |
|
|
Employees |
61.6% |
56.0% |
|
Solo self-employed |
7.8% |
15.4% |
|
Unemployed |
10.0% |
11.2% |
|
Students |
5.6% |
10.8% |
|
Inactive |
14.8% |
6.6% |
Table 1 provides an overview of the specific insertion of workers in the platform economy based on their labour market status in the conventional economy. A comparison of the labour market status of platform workers with that of the general population – that is, those who have not worked on a platform in the last twelve months – reveals that the two groups are statistically significantly different. The overall share of the economically active (i.e. employees, solo self-employed and the unemployed) does not differ too much between the two groups and is around 80 per cent. However, a much higher proportion of platform workers are solo self-employed, with over fifteen per cent defining themselves as such, although the contractual status – genuine or bogus self-employed, “independent contractor” or otherwise – in the platform economy is unknown. In addition, more than half of the respondents engaged in the platform economy are employees – a difference of five percentage points from the general population. Just over ten per cent of platform workers are either unemployed or students, while almost seven per cent are economically inactive in the traditional labour market. The platform workforce thus also differs from the non-platform workforce in terms of the higher proportion of students and the lower share of the economically inactive.58 These two groups traditionally do not belong to the workforce in the conventional economy, whereas students and the inactive can be part of the platform workforce.
Following previous studies,59 a distinction is made between two degrees of economic attachment to platform work, based on the number of working hours and the share of personal income derived from platform work. A higher level of attachment is defined as spending 20 hours or more per week on platform work or earning more than 50 per cent of personal income through platforms.60 This group is called “main platform workers”, and a little over one in four (27.2%) platform workers fall into this category. Put differently, main platform workers account for 1.6 per cent of the general working age population, the latter including both the workforce (employees, the solo self-employed and unemployed) as well as the non-workforce (students and the economically inactive). The remaining platform workers are considered to be weakly attached to platform work as they work fewer than 20 hours per week or earn less than half their income from platform work.
Figure 2: Platform workers’ attachment to the platform economy by labour market status61
Note: Attachment is statistically significantly different across the labour market status (p<0.01).
Figure 2 shows differences in the degree of economic attachment to the platform economy depending on labour market status in the conventional economy – a variation that is statistically significant. The solo self-employed, followed by the unemployed, are most likely to be highly attached to platform work, with 42.5 per cent and 31.6 per cent respectively being main platform workers. It can only be speculated why this is the case. One possible reason might be that some solo self-employed consider working via digital labour platforms as an expansion of their client base. Alternatively, they presumably engage in platform work more intensely to compensate for their relatively more precarious position and insufficient incomes, or indeed to fill in income gaps during periods of job search and drops in demand in the traditional labour market. Conversely, the shares of main platform workers are the lowest among those economically inactive in the conventional economy (21.7%), those in dependent employment (23.4%) and students (24.9%).
Apart from degree of attachment to the platform economy, a second feature that differentiates platform workers is the type of platform work they do, defined by type of tasks and format of task delivery, with a major distinction being made between “geographically tethered” location-based platform workers and online, remote platform workers performing “cloud work”.62 Perhaps contrary to conventional wisdom, shaped by first-hand observation and public debate, most of the platform work in Europe actually does not take place in public settings, with food delivery and ride-hailing being the main examples.63 Conversely, the largest group of platform workers in the survey sample are micro-taskers (38.4%), followed by remote professional platform workers (15.0%) and those performing on-location platform work in private settings (14.8%). Delivery work (11.7%) and ride-hailing (7.3%) are relatively less common. Some 12.7% of platform workers perform more than one type of platform work. In terms of the split between location-based and remote platform work, more workers perform remote work (59.4%) than any type of location-based platform work (40.6%).
Figure 3: Workers’ main labour market status by types of platform work64
Note: Ordered by share of employees. Labour market status is statistically significantly different across platform type (p<0.01).
Figure 3 illustrates the differences between platform workers doing different types of platform work in terms of their main labour market status in the conventional economy. Overall, there are statistically significant differences in status between the types, although there is no clear division between on-location and remote forms of task delivery. While most platform workers are dependent employees in the conventional economy, their share is the highest among those doing more than one type of platform work and among food delivery couriers, while it is by far the lowest among ride-hailing drivers. There is also considerable variation in the prevalence of the solo self-employed, with the highest share in private ride-hailing followed by multi-platform work and remote professional platform work. Students are especially common in micro-task remote platform work, among on-location platform workers in the private sphere and remote professional work, while they are relatively less active in ride-hailing and multi-platform work. The unemployed are similarly likely to engage in each of these types of platform work, with on-location platform work in the private sphere being the most attractive for them. Finally, those economically inactive in the conventional economy are somewhat more likely to engage in ride-hailing, remote micro-tasking and other types of on-location platform work.
Figure 4 Degrees of attachment by types of platform work65
Note: Platform attachment is statistically significantly different across platform type (p<0.01).
Previous studies have pointed out that platform work is usually only supplementary income, yet at the same time that the algorithms allocating work tend to favour workers with a higher degree of attachment to platform work, thus those who declare greater availability and have a greater number of tasks completed.66 Figure 4 shows that attachment to platform work varies considerably depending on the type of tasks performed. Unsurprisingly, those who work on multiple platforms and perform several types of tasks have the highest economic attachment to platform work, with nearly half of them (45.8%) classified as main platform workers. Conversely, the attachment of micro-taskers is by far the weakest, with 85.3% of workers considering it a marginal commitment which should come as no surprise given the extremely low rates of pay in this type of work. For all other types of platform work, the level of attachment is fairly similar, with around one-third of respondents being main platform workers.
Existing unionization levels in the platform economy
Depending on status in the traditional labour market, workers are likely to bring a certain degree of unionization with them into the platform economy. Levels will be relatively higher for employees, albeit with considerable variation across occupations and economic sectors, than for groups who are temporarily or completely outside the labour market.67 This section therefore maps and describes existing levels of unionization among platform workers and compares these levels with similar groups in the conventional economy.
Two caveats are warranted here. First, it should be remembered that platform workers are defined here as those who have done tasks through a digital labour platform at least once in the past twelve months, thus also including occasional platform workers. Second, to avoid comparing apples with oranges, the calculation of union density includes non-workers in the conventional economy. This means that density in the conventional economy is underestimated, as the non-working population (students and the economically inactive) is usually not included in such calculations. In other words, union density rates in the conventional economy are actually higher than what is presented here. As many platform workers are not part of the workforce in the traditional labour market, union density is thus captured as the share of union members in the total working age population. This allows a more meaningful comparison of the density rates of platform workers and those who do not engage in platform work.
Figure 5: Unionization rates among platform workers and the general population, by individual characteristics (%)68
Note: Differences in unionization between platform and non-platform workers are statistically significant (p<0.05) for migrants and those with low and high educational attainment.
In general, the proportion of platform workers who are union members is remarkably similar to the general population in the sample and amounts to 13.4 per cent.69 However, a comparison of groups with similar sociodemographic characteristics reveals some noticeable differences – see Figure 5. While women are slightly more likely to be union members among platform workers than men (14.2% and 12.6%, respectively), the reverse is observed in the general population (12.9% and 13.9%). There is little difference among younger workers but, among those older than 54, unionization is visibly higher among platform workers (20.2%) than in the rest of the population (16.0%). Union membership among people with a migrant background is much more common for those who work on digital labour platforms (16.2%) compared to those who do not (10.7%). Platform workers with a higher level of education are less likely to be union members (13.1%) than those with a similar level of education who do not work on digital labour platforms (16.6%); while the opposite is found for those with lower educational levels, with 16.9 per cent of platform workers and only 9.4 per cent of others being union members.
Figure 6: Unionization rates among platform workers and the general population, by labour market status, working time and sector (%)70
Note: Differences in unionization between platform and non-platform workers are statistically significant (p<0.05) for the solo self-employed; the inactive; full-time workers; and in the construction sector.
In terms of labour market status, trade union membership is relatively high among platform workers who are not employees in their main paid job – see Figure 6. Thus, among those who earn from platform work but occupy “marginal” or “outsider” positions in the traditional labour market, such as the solo self-employed, the unemployed, students or the economically inactive, unionization rates surpass by a considerable margin – between 1.3 and 6.6 percentage points – those of the rest of the population. The reverse is observed among employees, with membership rates of 16.0 per cent among platform workers and 17.7 per cent among the rest. A parallel pattern is found among those with full-time jobs in the conventional economy, with platform workers having lower unionization rates (13.6%) than offline workers (16.8%). As could be expected, the sectors with the highest shares of unionized workers in the sample are manufacturing and public services (i.e. education, healthcare and public administration); platform workers with jobs in these sectors are more likely to be unionized than non-platform workers. In other industries, apart from the residual category of “other services”, platform workers are less likely to be unionized than offline workers.
Figure 7: Unionization rates among platform workers by heterogeneity of platform work (%)71
Note: Ordered within types of heterogeneity. Differences by platform type are statistically significantly different (p<0.05); other differences are not.
In devising trade union strategies for the platform economy, it is key to recognize and understand not only the differences between platform workers and those working in the conventional economy, although there is a large overlap between the two, but also the heterogeneity among platform workers in their views on and experiences with trade unions.
Figure 7 reveals this heterogeneity in terms of union membership. There is virtually no variation in membership by level of platform economic attachment, with main platform workers and those with weak attachment to platform work standing at the same unionisation level. Unionization rates are lower among platform workers with the shortest tenure, of less than one year, however. Furthermore, while the overall unionization rate among platform workers stands at 13.4 per cent, as shown earlier, it is by far the lowest in food delivery, with 7.5 per cent of couriers being union members, followed by remote professional work and ride-hailing, where union density hovers around 10 per cent. Low unionization rates among food delivery couriers might be surprising in view of the high visibility of bottom-up organizing and collective initiatives,72 but this is at least partly due to union membership being confounded in the conventional and platform economies. Furthermore, micro-taskers and workers in on-location private work show middle-range unionization rates, while those engaging in multiple types of platform work report by far the highest rate of just over 20 per cent.
As in the conventional economy, all of this suggests that there is still considerable scope for unions to set up initiatives to organize the platform workforce, given that 86.6% overall are not unionized in the 14 European countries considered. The remainder of the analysis will therefore focus on this group, with the aim of improving understanding of their openness to becoming union members in the future and the possible drivers for such a decision.
Drivers for unionization among platform workers
This section addresses one of the key questions posed in this paper, namely which factors can be considered as drivers for unionization among platform workers. Three factors are identified as potential drivers and their association with the propensity to become a trade union member is explored: (1) individual pro-union attitudes; (2) embeddedness in offline union-friendly social networks; and (3) online engagement in digital work-related communities – all of which are drivers that are not necessarily tied to physical workplaces at company level.
Attitudes towards unions are measured by the question “Do you think trade unions help to improve working conditions?” Responses of “no” are considered an expression of negative attitudes while the responses “yes, somewhat” and “yes, considerably” are expressions of positive ones. With regard to the second driver, embeddedness in union-friendly offline social networks is derived from the question “Are many of your colleagues or friends in a trade union or similar organization?”, with “yes” and “no” as answer options. Finally, participation in digital, online work-related communities is measured by the question “Do you discuss online with other workers any issues related to your working conditions?”, with “yes” and “no” as response options. This question is only asked to respondents who are employed, either as an employee or solo self-employed. The aim is to see whether platform workers are affected differently by these factors in their propensity to unionize or whether the effects are similar for all non-union members.
The vast majority (69.2%) of platform workers have pro-union attitudes, while less than one-third (30.8%) express negative attitudes towards trade unions. Nearly one in three platform workers (30.6%) say they are embedded in union-friendly networks. This finding shows that, in general, a critical mass of non-unionized platform workers are actually embedded in social networks that are considered union-friendly. Finally, participation in work-related digital communities is even more common among platform workers than being embedded in union-friendly social networks. Such online activity is reported by almost 40 per cent of non-unionized platform workers and is somewhat more common than among non-platform workers.
Table 2: Drivers for unionization among non-platform and platform workers73
|
|
Pro-union attitudes |
Union-friendly offline networks |
Online communities |
|||
|
|
Non-platform |
Platform |
Non-platform |
Platform |
Non-platform |
Platform |
|
Overall |
65.0% |
69.2% |
29.5% |
30.6% |
37.5% |
38.4% |
|
Employees |
65.7% |
68.4% |
30.1% |
30.9% |
37.7% |
38.0% |
|
Solo self-employed |
52.1% |
57.8% |
26.3% |
25.3% |
35.8% |
39.6% |
|
Unemployed |
56.9% |
72.4% |
31.1% |
36.7% |
n.a. |
n.a. |
|
Students |
82.2% |
88.7% |
18.2% |
29.6% |
n.a. |
n.a. |
|
Inactive |
67.1% |
62.7% |
32.8% |
32.7% |
n.a. |
n.a. |
Table 2 compares non-unionized non-platform workers with platform workers to determine what accounts for their willingness to become trade union members. Overall, compared to non-union members in the general population, platform workers are generally more likely to hold pro-union attitudes, they are slightly more likely to be embedded in union-friendly social networks and they are slightly more likely to engage in online communities discussing work-related issues. As platform workers differ to some extent from non-platform workers in their labour market status and sociodemographic profiles, however, it is useful to consider whether such relationships hold when comparing people with similar characteristics.
In terms of pro-union attitudes, platform workers show more positive attitudes towards unions than non-platform workers with the same labour market status. This difference is greatest among the unemployed. Concerning offline embeddedness in pro-union social networks, platform workers who are unemployed or students are more likely to report being embedded in such networks than non-platform workers. There is no such difference among employees, the economically inactive or solo self-employed in platform work. However, the latter are much more likely to engage in discussions about working conditions in online communities than non-platform workers. There is no real difference among employees.
Figure 8: Attitudes towards trade unions among non-union members, by heterogeneity of platform work (%)74
Note: Ordered by negative attitudes within types of heterogeneity. Differences in attitudes by type of tasks, attachment and tenure are not statistically significantly different.
Turning to heterogeneity within the group of platform workers, Figure 8 shows that food delivery platform workers are the most likely to hold negative views about unions (38.2%), while micro-taskers are the most likely to be positive about them. Overall, however, attitudinal differences between types of platform work are small. Furthermore, weaker attachment to the platform economy is associated with being slightly less likely to hold negative attitudes towards unions, as is shorter tenure in platform work.
Figure 9 indicates that there is some variation between types of platform work in terms of engagement in union-friendly networks. For instance, platform workers carrying out on-location tasks in the private sphere and food delivery couriers are the most likely to be embedded in such social networks (34.1% and 31.8%, respectively), as are workers with longer tenure in platform work. There is little difference, however, in relation to levels of economic attachment to platform work. There are some small differences between types of platform work in terms of participation in work-related digital communities, but no clear patterns emerge. For instance, ride-hailing drivers and micro-taskers are the least likely to engage in online discussions (34.8% and 35.4%, respectively), while remote professional platform workers (42%) and on-location platform workers in private settings (41.6%) are the most likely. Perhaps surprisingly, main platform workers are less often active in online communities (37%) as are those with the shortest tenure in platform work (34.7%).
Figure 9: Engagement in union-friendly offline social networks and participation in online work-related communities among non-union members, by heterogeneity of platform work (%)75
Note: Differences in engagement and participation by type of tasks, attachment and tenure are not statistically significantly different.
Willingness to become union members among platform workers
From the perspective of trade unions and their recruitment and organizing drives, it is crucial to know the extent to which workers who are still outside the labour movement are likely or willing to join its ranks. This section therefore focuses on the propensity of platform workers to unionize, taken to be a dependent variable in the following analysis. Irrespective of their (legal) ability to do so, all platform workers in our sample were asked “Would you be willing to become a trade union member in the near future?”, with the response options “yes” and “no”. Given the interconnectedness of the conventional and platform economies, it is likely that the openness of most platform workers to unionization is not solely rooted in their experience with the platform economy. Moreover, they share many of the characteristics of the general population, although platform workers have also been found to show some particularities in terms of age, educational attainment or vulnerability in the labour market.76 Therefore, in order better to understand their willingness to unionize, the focus should not only be on platform workers themselves, across different types of platform work, but also on comparing their propensity to unionize with similar groups in the conventional economy. If significant differences emerge when comparing the two groups, it is reasonable to attribute these to experiences in the platform economy rather than the conventional economy.
Figure 10: Comparing willingness to unionize between platform workers and the general population, by individual characteristics (%)77
Note: Differences in willingness to unionize between platform and non-platform workers are statistically significant (p<0.05) overall; for men; for non-migrants; for migrants; and for the middle educated.
Figure 10 reveals that over a quarter (28.0%) of non-unionized platform workers actually express a willingness to join a trade union, which is slightly higher than in the general population (23.8%). This discrepancy, whereby platform workers declare a willingness to join a union more often compared to non-platform workers, is apparent for each one of the individual characteristics considered here. It is especially noticeable among men and those in the youngest age category, as well as among platform workers with a migrant background and with the lowest levels of educational attainment. These last groups might feel in a more precarious and vulnerable position in the platform economy, without many of the safety nets applying in the regulated labour market which, in turn, might fuel their openness to join a union.
Figure 11: Comparing willingness to unionize between platform workers and the general population, by labour market status, working time and sector (%)78
Note: Differences in willingness to unionize between platform and non-platform workers are statistically significant (p<0.05) for the manufacturing sector.
There are no significant differences between non-platform and platform workers in their willingness to unionize that would be related to their labour market status – see Figure 11. Overall, students and the unemployed are more often open to the possibility of joining a trade union, particularly so for platform workers who declare being unemployed. Finally, there are some differences between economic sectors, with platform workers whose main job is in either manufacturing, agriculture or financial services more often willing to unionize than non-platform workers. In contrast, platform workers employed in professional, scientific and technical services in the conventional economy, as well as in administrative and support services, are less likely to become union members.
Figure 12: Willingness to unionize among platform workers by heterogeneity of platform work (%)79
Note: Ordered within types of heterogeneity. Differences in willingness to unionize by platform type are statistically significantly different (p<0.05); other differences are not.
Figure 12 illustrates the variation in propensity to unionize across the heterogeneity of platform workers. Experience in platform work seems to be related to this propensity, with those who have worked on platforms for longer and those who are more economically attached to platform work being less inclined to become union members. There may be a variety of reasons for such a link, but it is not possible to provide definitive answers with the data available. Presumably, however, greater involvement in the platform economy weakens workers’ attachment to the conventional economy, which is currently a main breeding ground for unionization. But it is also conceivable that those with more extensive experience in platform work may have witnessed, or even been exposed to, a certain hostility towards unionization on the part of platform companies which may have increased their reluctance to seek union membership. Furthermore, propensity to unionize is visibly higher among ride-hailing drivers and multi-type platform workers, with just over 35 per cent expressing a willingness to join a trade union. By contrast, only 19.6 per cent of remote professional workers expressed the same intention, with other categories of platform workers falling in between. There thus appears to be some link between willingness to join a union and labour market bargaining power, with higher skilled workers not only being less unionized (see Figure 7), but also presumably having less need to be represented by a union.80 Alternatively, or additionally, it may indicate that these workers are unable to join a union, either because of internal union rules or because it is not legally possible.
Figure 13: Differences in willingness to unionize between platform workers and the general population, by heterogeneity of platform work81
Note: Positive values show higher willingness among platform workers compared to non-platform workers. Effects are estimated from logistic regression controlling for gender, age, migrant background, education, place of residence, trade union membership in the past and labour market status, including union presence in the workplace for those employed, and including fixed effects. Four separate models are estimated, divided by horizontal lines in the figure.
Finally, Figure 13 maps the willingness to join a trade union among workers in the platform economy compared to the general population, accounting for differences that may be linked to sociodemographic profile. The latter includes individual characteristics such as gender, age, educational attainment, type of place of residence and migrant background, as well as labour market status, and information on past union membership and the presence of a union in the workplace. In general, the results confirm that platform workers are more willing to unionize than the rest of the population – a statistically significant difference of nearly three percentage points – once other differences between these groups are accounted for. Main platform workers do not differ from non-platform workers in their declared willingness to unionize, while those weakly attached to platform work are significantly more likely to join a union although the differences in tenure in platform work turn out not to be statistically significant. When considering different types of platform work, professional remote workers stand out as the only category that is significantly less inclined to unionize, with nearly six percentage points fewer than the general population expressing an intention to join a union in the near future. Those engaged in several different types of platform work are much more likely to unionize than non-platform workers (by over 11 percentage points), with other categories of platform workers not significantly different from the rest of the population.
The relationship between unionization drivers and willingness to unionize
This section ties together the issues discussed in the previous two and examines the impact of three drivers for unionization on willingness to join a union in the future. As a reminder, the drivers include: (1) pro-union attitudes; (2) union exposure in terms of being embedded in union-friendly social networks; and (3) engagement in digital, online communities. Figure 14 shows the overall relationship between each of these three drivers and the share of respondents likely to unionize in the near future, accounting for individual differences between workers in terms of their sociodemographic profile and labour market status.
In the general working age population, the overall effect of each of these three drivers on willingness to unionize is positive and statistically significant. For instance, people who express pro-union attitudes are nearly 12 percentage points more likely to unionize. Those embedded in a union-friendly social network are nearly 10 percentage points more likely, while those engaged in online communities are nearly eight percentage points more likely to become a union member.
Figure 14: The effects of drivers for unionization on willingness to join a union82
Note: Effects are estimated from logistic regressions controlling for gender, age, migrant background, education, place of residence, trade union membership in the past and labour market status, including union presence in the workplace for those employed, and country fixed effects. Each block separated by horizontal lines indicates a separate model.
The next step is to test whether these three drivers for unionization are indeed different for platform workers. The results are shown below the dotted lines in Figure 14. They reveal that the effect of pro-union attitudes is much stronger among platform workers and is statistically significantly different from the effect among the rest of the population. Thus, platform workers who express pro-union attitudes are also nearly 18 percentage points more likely to want to join a union, compared to just over 11 percentage points among the rest. The other two drivers for unionization – union-friendly network and online communities – have the same effects among platform workers as among the general population.
Figure 15: The effects of drivers for unionization on willingness to unionize, by heterogeneity of platform work83
Note: Effects are estimated from logistic regressions for platform workers controlling for gender, age, migrant background, education, place of residence, trade union membership in the past and labour market status, including union presence at the workplace for those employed, and country fixed effects. Each block separated by horizontal lines indicates a separate model. The driver of unionization is interacted with platform heterogeneity variables.
Finally, whether or not the drivers of unionization affect all platform workers in a similar way is examined, with the results shown in Figure 15. Overall, pro-union attitudes are positively associated with willingness to unionize for all types of platform workers. Having union-friendly social networks is also positively associated with the likelihood of unionizing for platform workers overall. While the differences are not statistically significant, the effect seems weaker for platform workers engaged in ride-hailing or food delivery work and stronger for micro-taskers, on-location private work and multiple types of platform work. Engagement in online communities for discussing work-related issues is positively associated with a willingness to unionize but seems to affect some groups more positively than others. It is particularly important for platform workers in ride-hailing, those carrying out remote professional work and those engaged in multiple types of platform work. It does not seem to be important for on-location workers in the private sphere or food delivery couriers. Online communities also appear to be important only for platform workers who have been engaged in platform work for at least one year, possibly due to their longer exposure to platform organizational practices.84
Discussion, limitations and future directions
To the best of our knowledge, this working paper is the first to examine both existing levels of unionization among platform workers and their propensity to unionize in the future from a comprehensive cross-national, quantitative perspective. The analysis covers six types of platform work in 14 European countries. In contrast to previous studies,84 attitudes towards collective organization via trade union membership in the platform economy are explicitly compared with the conventional economy. The individual-level data allow accounting for the range of individual and sociodemographic characteristics of platform workers, as well as the heterogeneity of platform work. Where feasible and relevant, workplace and industrial characteristics are also considered. Despite these strengths offered by a unique, representative individual-level dataset, the analysis carried out also has a number of limitations due to which it can therefore only be considered exploratory.
To begin with, many of the measures used are based on dichotomous variables and single items in the survey. For instance, only the embeddedness in and engagement with on- and offline networks per se are measured, not their frequency and the dynamics of those contacts. Follow-up research could include measures with more complex response options and derived from multiple items to allow testing of the reliability and validity of the findings. The measurement context may also have had an impact on the results as the survey was conducted during the Covid-19 pandemic; it is well known, for instance, that food delivery couriers suffered from poor working conditions in this context. Obviously, only attitudes and intentions to join a trade union were surveyed; no claims could be made about actual behaviour.
Furthermore, with the data at hand, it is not possible to determine to what extent willingness to unionize is actually rooted in work experiences in the platform economy as opposed to the conventional economy or other types of life experience. This issue has been addressed, at least in part, by comparing platform workers with similar groups in the conventional economy which revealed, among other things, that platform workers hold stronger pro-union attitudes and have a relatively higher propensity to unionize. Moreover, assuming that trade unions are active in the platform economy, and that platform workers are aware of their right to unionize, it is not known which union they would like to join. The survey did not differentiate between grassroots unions and mainstream unions, although their conceptions of membership are likely to differ, with grassroots union members tending to be more active than in mainstream unions.85 Generalization of the findings should also be undertaken with caution as they apply predominantly to formal unionism in the European context whereas informal types of worker representation are more prevalent in the Global South.86
Possible differences in union activity in the platform economy between countries has not been considered here, such as outreach to platform workers, union efforts to mobilize and recruit platform workers or legal action to reclassify their employment status. In this sense, it might be useful for future analysis to include information on the number and type of trade unions active in the platform economy and their activities to date. At individual level, the survey did not ask about instrumental reasons for unionization, such as perceptions of income from platform work and working conditions, even though dissatisfaction with these seems to drive support for collective organization in the platform economy.87 Similarly, although offline embeddedness in union-friendly social networks and online participation in work-related communities may indicate platform workers’ own potential involvement in social protest, this has not been measured directly. The organizational status of these work-related communities is also unknown: they could simply refer to online participation or to self-organized online collectives. Neither have other determinants of unionization, such as the political attitudes of platform workers, been included in the analysis although such a link seems of less importance in the platform economy.88 Other contextual (macro) variables have also not been accounted for, such as the changing regulatory environment in which digital labour platforms operate, as the costs and benefits of unionization may vary between different jurisdictions.89 Equally, dispute resolution mechanisms within platform companies may influence the decision to unionize.
Finally, one dimension of the heterogeneity of platform work that has received much less attention in this study are workers with a migrant background, including circular or undocumented migrants, who are particularly active in the platform economy because of the low entry barriers with minimal formal requirements and limited alternative labour market opportunities in the conventional economy.90 Some of them may have been less familiar with trade unions in their home country, which may translate into a lesser willingness to unionize in their current country of residence, although union experience is also a function of individual biographies. Even if they are positive towards collectivism, they have been found to be simply “too pressed for time and caught up in their daily hustle to free up space in their heads and agendas for collective organisation”.91 This may be a general observation: while the overall majority of platform workers are generally positive about unions, far fewer of them express a willingness to unionize in the future. Thus, similar to findings in the conventional economy,92 there is a considerable gap between pro-union attitudes, unionization levels and actual willingness to unionize which might be a further research avenue to be explored.
Conclusion
Although (still) limited in relative terms, with 5.6 per cent of the working age population engaged in platform work in Europe,93 the platform workforce is quite substantial in absolute terms and is growing rapidly. This study contributes to the understanding of unionization in the platform economy, both in terms of membership rates among different groups of platform workers and the potential for unionization among those who are not yet members. The findings are based on a unique cross-national and comparative survey, dedicated to the analysis of various forms of digitally-mediated work and trade union-related behaviour, which allows for an explicit link between the platform economy and the conventional economy.
Four main conclusions can be drawn from the analysis about the possibilities to foster associational power via trade unions in the platform economy. These conclusions relate to trade union membership per se; the attitudes of platform workers towards unions; network-like drivers for unionization; and the heterogeneity of the platform economy – all of which have implications for possible union recruitment and organizing strategies in the platform economy.
First, the platform economy is not, by default, completely free of trade union membership. A small share of platform workers are already unionized: the survey results indicate that union density in the platform economy in Europe stands at a (low) level of around 13.4 per cent. This percentage should not be all that surprising. In all likelihood, it is less the result of any dedicated or specific union activity in the platform economy: it is simply related to the labour market status of platform workers rooted in the conventional economy. Already unionized platform workers largely follow unionization patterns in the traditional labour market that are specific to their personal sociodemographic characteristics. Exceptions are workers aged over 54, those with lower educational attainment and migrants, all of whom report relatively higher unionization rates in the platform economy. Workers who carry out platform work are particularly active in relatively well-unionized sectors of the traditional economy, such as manufacturing and public services. In terms of labour market status, union density is relatively higher among platform workers who are unemployed, the solo self-employed and the economically inactive in the conventional economy. All in all, however, “platform unionism” merely exists by coincidence; that is, through the specific inclusion of platform workers in relation to their labour market status in the conventional economy. This small share of platform workers who are already unionized could nevertheless be an entry point for union organizing strategies.
Second, the generally pro-union attitudes of platform workers and their intentions to unionize stand at odds with conventional wisdom about the platform economy. Induced by Californian ideology,94 this wisdom is often based on the newspeak of the digital labour platforms themselves. It tends to portray this economy as a union-free zone with highly independent, flexible and autonomous platform workers showing little interest in unionization or even being hostile towards trade unions. Yet, in fact, over two-thirds – 69.2 per cent – of platform workers express a positive attitude towards unions, a percentage that is even higher than in the general population. Equally, more than one in four platform workers say they would be willing to join a union – again a slightly higher proportion than in the general population. Openness to future union membership is particularly high among younger platform workers, those with a migrant background and those with lower levels of education.
Third, the drivers for unionization are quite similar in the conventional and platform economies, with pro-union attitudes alone being a relatively more important driver for platform workers than for the general population, but it is unknown to what extent those drivers are rooted in the conventional economy or the platform economy. Nevertheless, trade unions tapping into offline pro-union social networks and online work-related communities seems to be a fruitful strategy for unions in the platform economy as well. Online communities might thus be a breeding ground for future unionization, acting as a stepping stone to union membership95, and such online communities are thus likely to pose less of a risk of simply replacing unions.96 The importance of pre-existing off- and online networks calls respectively for networked organizing tactics and strategies, and digital unionism. Tapping into pre-existing networks is of course more feasible in the on-location platform economy due to the spatial proximity of workers, engaged in physical, time- and place-dependent platform jobs that depend on local customers. It remains an empirical question, however, to what extent such on- and offline networks are marked by stability in terms of their organizers, given the generally high turnover in the platform economy.
There is nevertheless a gap between the level of pro-union attitudes and willingness to unionize. A much higher proportion of platform workers hold positive attitudes towards unions than are willing to become members. Possible explanations for this gap could be related to the different insertion of workers into the platform economy and the degree of openness of trade unions to groups beyond their traditional core members, like those on workfare and the solo self-employed. Union membership may also be less open to the unemployed, especially as countries in the survey sample do not have unemployment systems in which unions are involved in their administration. The unemployed who also work on platforms are nevertheless more likely to unionize compared to those who do not work online. Furthermore, while students and the economically inactive are likely to be sympathetic to unions, they may have far fewer reasons actually to join them as they are outside the traditional workforce. Also, while some unions in certain occupations have a tradition of organizing the solo self-employed, and others have been catching up by opening up union membership and introducing dedicated structures for them, there is still work to be done.97
Fourth, looking more closely at the platform economy and the heterogeneity of its workforce, differences in existing union density are small when it comes to the degree of attachment to platform work. In general, those who are more economically attached to platform work are less inclined to join a union, but they represent a minority of all platform workers. The majority are only weakly economically attached to their platform job, primarily relying on other sources of income, especially those engaged in micro-tasking, and they are relativity more likely to be pro-union. Moreover, students engaging in platform work are demonstrating a greater willingness to unionize compared to students outside the platform workforce. As this is a young population that is relatively unfamiliar with trade unions, this is thus an interesting target group for outreach activities in the platform economy, especially as a first personal impression of unions seems to be important for future union membership.98
There are some differences in unionization rates depending on type of platform work, ranging from 7.5 per cent in food delivery to the one in five platform workers engaged in multi-type platform work. The latter thus seems an attractive group for brownfield organizing strategies, and they account for about 13 per cent of the platform workforce. Moreover, not only are multiple platform workers the most unionized of all platform workers, but they are also highly dependent on their platform jobs and they are more receptive to union membership than workers in other types of platform work. Together with food delivery couriers, who are the most strongly embedded in union-friendly social networks, and ride-hailing drivers, multiple platform workers are also comparatively more inclined to unionize in the future. Yet, perhaps even more than other types, multiple platform workers may be difficult to integrate into existing union structures of representation; of all platform workers, they may be the most in need of a dedicated union representation structure as their work spans multiple sectors. Needless to say, some, but not all, grievances and issues of platform workers are likely to vary between different types of platform work.
Trade unions appear to be mainly active in the delivery and private ride-hailing sectors. Unions with a tradition of organizing a similarly precarious, transient or dispersed workforce, such as in cleaning or transport, are seen as more successful in connecting with platform workers and mobilizing them.99 Reflecting again the interconnection between the conventional and platform economies, it is in particular unions with previous experience in network-driven organizing that are engaging with those platform workers. Although delivery work and private ride-hailing are the most visible types of platform work, and therefore emblematic of the platform economy, most platform work – about 80 per cent – does not take place in public settings, however.
Platform workers in the private sphere are relatively strongly embedded and engaged in union-friendly offline social networks and online digital communities. While micro-taskers are the largest group, on-location platform work in private settings and professional remote workers are almost of equal size. Performing the most complex and highly skilled tasks in the platform economy, more than half of professional remote workers has employee status, but a large share is made up of the solo self-employed. Remote professional workers differ from other types of platform work in that they hold relatively less pro-union attitudes and are less likely to unionize compared to the general population – those results might also reflect the inability of those workers to unionize related to their labour market status. Nevertheless, trade unions are not without inspiration, strategies and tactics for recruiting and organizing them. Tapping into online, digital communities of remote professional workers seems most feasible here, especially as they are among the most active compared to other types of platform work. Yet, attention to offline social networks should not be dismissed. Remote platform work is not only still place-bound and likely city-based100 but possibly also marked by regional specialization,101 all of which might open opportunities for offline (transnational) union organizing tactics and strategies.
While some mainstream trade unions have opened up membership to the solo self-employed,102 their membership base is unlikely to be solid among remote professional workers. This is probably also true of grassroots unions. Yet the latter’s engagement with workers in location-based platform work and bottom-up activism add to their legitimacy to speak on behalf of and in solidarity with these workers.103 In contrast, the claim of mainstream unions to represent workers in the platform economy are largely based on their expert role and embeddedness in social dialogue institutions.104 These claims might not always reflect the interests and needs of platform workers, however, which might strengthen their perceptions of being less represented by mainstream unions.105 The legitimacy of their representative claims in the platform economy could therefore be brittle if the membership of platform workers does not grow sufficiently.
References
Aleksynska, Mariya, Andrey Shevchuk and Denis Strehkov. 2021. “The geographic and linguistic variety of online labour markets: the cases of Russia and Ukraine”. In A modern guide to labour and the platform economy, by Jan Drahokoupil and Kurt Vandaele (eds). Cheltenham: Edward Elgar.
Aloisi, Antonio. 2022. “Platform work in Europe: lesson learned, legal developments and challenges ahead”. European Labour Law Journal Special Issue, 13 (1): 1–26.
Anwar, Mohammed Amir and Mark Graham. 2021. “Between a rock and a hard place: Freedom, flexibility, precarity and vulnerability in the gig economy in Africa”. Competition & Change 25 (2): 237–258.
Atzeni, Maurizio. 2020. “Worker organisation in precarious times: Abandoning trade union fetishism, rediscovering class”. Global Labour Journal 11 (3): 311–314.
Barratt, Tom, Caleb Goods and Alex Veen. 2020. “‘I’m my own boss . . .’: Active intermediation and ‘entrepreneurial’ worker agency in the Australian gig-economy”. Environment and Planning A: Economy and Space 52 (8): 1643–1661.
Berg, Janine, Marianne Furrer, Ellie Harmon, Uma Rani and M. Six Silberman. 2018. Digital labour platforms and the future of work: Towards decent work in the online world. Geneva: International Labour Office.
Bessa, Iouliya, Simon Joyce, Denis Neumann, Mark Stuart, Vera Trappmann and Charles Umney. 2022. A global analysis of worker protest in digital labour platforms. Geneva: International Labour Office.
Bogg, Alan. 2022. “The legal framework of worker voice in the platform economy”. In Missing Voice? Worker voice and social dialogue in the platform economy, by Adrian Wilkinson, Tony Dundon, Paula Mowbray and Sarah Brooks (eds). Cheltenham: Edward Elgar.
Cant, Callum. 2019. Riding for Deliveroo. Resistance in the new economy. Cambridge: Polity Press.
Cini, Lorenzo. 2022. “Resisting algorithmic control: Understanding the rise and variety of platform worker mobilisations”. New Technology, Work and Employment 38 (1): 125–144. https://doi.org/10.1111/ntwe.12257
Cini, Lorenzo, Vincenzo Maccarrone and Arianna Tassinari. 2022. “With or without U(nions)? Understanding the diversity of gig workers’ organizing practices in Italy and the UK”. European Journal of Industrial Relations 28 (3): 1–22.
Clark, Paul F. 2009. Building more effective unions. 2nd ed., Cornell University Press.
Countouris, Nicola and Valerio De Stefano. 2021. “The labour law framework: Self-employed and their right to bargain collectively”. Bulletin of Comparative Labour Relations, accessed 18 January at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3763214.
della Porta, Donatella, Riccardo Emilio Chesta and Lorenzo Cini. 2022. Labour conflicts in the digital age. A comparative perspective. Bristol: Bristol University Press.
Drahokoupil Jan and Kurt Vandaele. 2021. “Introduction: Janus meets Proteus in the platform economy”. In A modern guide to labour and the platform economy, by Jan Drahokoupil and Kurt Vandaele (eds). Cheltenham: Edward Elgar.
Farinella, Domenica and Davide Arcidiacono. 2023. “Beyond formality: the informalisation and tertiarization of labour in the gig economy”. In Routledge Handbook of the Gig Economy, by Immanuel Ness (ed.). New York: Routledge.
Ferrari, Fabian and Mark Graham. 2021. “Fissures in algorithmic power: platforms, code, and contestation”. Cultural Studies 35 (4-5): 814–832.
Galière, Sophia. 2020. “When food delivery platform workers consent to algorithmic management: a Foucauldian perspective”. New Technology, Work and Employment 35 (3): 357–370.
Gandini, Alessandro. 2019. “Labour process theory and the gig economy”. Human Relations 72 (6): 1039–1056.
Garben, Sacha. 2021. “The regulation of platform work in the European Union: mapping the challenges”. In A modern guide to labour and the platform economy, by Jan Drahokoupil and Kurt Vandaele (eds). Cheltenham: Edward Elgar.
Gegenhuber, Thomas, Markus Ellmer and Ellie Schüßler. 2021. “Microphones, not megaphones: functional crowdworker voice regimes on digital work platforms”. Human Relations 74 (9): 1473–1503.
Gerber, Christine. 2021. “Community building on crowdwork platforms: Autonomy and control of online workers?”. Competition & Change 25 (2): 190–211.
Gerber, Christine and Martin Krzywdzinkski. 2019. “Brave new digital work? New forms of performance control in crowdwork”. In Work and labor in the digital age. Research in the sociology of work, by Steven Vallas and Anne Kovalainen (eds). Vol. 33. Bingley: Emerald.
Geyer, Leonard, Kurt Vandaele and Nicolas Prinz (2023). “Riding together? Why app-mediated food delivery couriers join trade unions in Austria”, Economic and Industrial Democracy.
Godard, John. 2008. “Union Formation”. In The Sage Handbook of Industrial Relations, by Paul Blyton, Nicolas Bacon, Jack Fiorito and Edmund Heery (eds). London: Sage.
Goods, Caleb, Alex Veen and Tom Barratt. 2019. “‘Is your gig any good?’ Analysing job quality in the Australian platform-based food delivery sector”. Journal of Industrial Relations 61 (4): 502–527.
Hau, Mark F. and Owen G. Savage. 2022. “Building coalitions on Facebook: ‘social media unionism’ among Danish bike couriers”. New Technology, Work and Employment https://doi.org/10.1111/ntwe.12261
Hadziabdic, Sinisa and Lorenzo Frangi. 2022. “Rationalizing the irrational: Making sense of (in)consistency among union members and non-members”. European Journal of Industrial Relations 28 (2): 147–174.
Heiland, Heiner. 2021. “Controlling space, controlling labour? Contested space in food delivery gig work”. New Technology, Work and Employment 36 (1): 1–16.
Herr, Benjamin, Philip Schörpf and Jörg Flecker. 2021. “How place and space matter to union organizing in the platform economy”. In A modern guide to labour and the platform economy, by Jan Drahokoupil and Kurt Vandaele (eds). Cheltenham: Edward Elgar.
Howcroft, Debra and Birgitta Bergvall-Kåreborn. 2019. “A typology of crowdwork platforms”. Work, Employment and Society 33 (1): 21–38.
ILO. 2021. Work employment and social outlook. The role of digital labour platforms in transforming the world of work. Geneva: International Labour Office.
Jansen, Giedo and Alex Lehr. 2022. “On the outside looking in? A micro-level analysis of insiders’ and outsiders’ trade union membership”. Economic and Industrial Democracy 43 (1): 221–251.
Johnson, Hannah. 2020. “Labour geographies of the platform economy: Understanding collective organizing strategies in the context of digitally mediated work”. International Labour Review 159 (1): 25–45.
Johnston, Hannah and Chris Land-Kazlauskas. 2019. Organizing On-Demand: Representation, Voice, and Collective Bargaining in the Gig Economy. Conditions of Work and Employment Series 94. Geneva: International Labour Office.
Johnson, Hannah and Susanne Pernicka. 2021. “Struggles over the power and meaning of digital labour platforms: A comparison of the Vienna, Berlin, New York and Los Angeles taxi markets”. In A modern guide to labour and the platform economy, by Jan Drahokoupil and Kurt Vandaele (eds). Cheltenham: Edward Elgar.
Joyce, Simon and Mark Stuart. 2021. “Trade union response to platform work: an evolving tension between mainstream and grassroots approaches”. In A modern guide to labour and the platform economy, by Jan Drahokoupil and Kurt Vandaele (eds). Cheltenham: Edward Elgar.
Joyce, Simon, Mark Stuart and Chris Forde. 2022. “Theorising labour unrest and trade unionism in the platform economy”, New Technology, Work and Employment 38 (1): 21-40. https://doi.org/10.1111/ntwe.12252
Kalum Schou, Peter and Eliane Bucher. 2022. “Divided we fall: The breakdown of gig worker solidarity in online communities”. New Technology, Work and Employment https://doi.org/10.1111/ntwe.12260
Lehdonvirta, Vili. 2016. “Algorithms that divide and unite: delocalisation, identity and collective action in microwork”. In Space, place and global digital work, by Jörg Flecker (ed.). Basingstoke: Palgrave Macmillan.
Maffie, Michael D. 2020. “The role of digital communities in organizing gig workers”. Industrial Relations 59 (1): 123–149.
Maffie, Michael D. and Mark D. Gough. 2023. “Bargaining against the machine: a theory of bargaining power in the gig economy”. In Advances in Industrial and Labor Relations Vol. 27, by David Lewin D and Paul J. Gollan (eds). Bingley: Emerald.
Newland, Gemma, Christophe Lutz and Christian Fieseler. 2018. “Collective action and provider classification in the sharing economy. New Technology, Work and Employment 33 (3): 250–267.
Nissen, Bruce and Paul Jarley. 2005. “Unions as social capital: Renewal through a return to the logic of mutual aid?”. Labor Studies Journal 29 (4): 1–26.
Offe, Claus and Helmut Wiesenthal. 1980. “Two Logics of Collective Action: Theoretical Notes on Social Class and Organizational Form”. Political Power and Social Theory 1, 67–115.
Panteli, Niki, Andriana Rapti and Dora Scholarios. 2020. “‘If he just knew who we were’: Microworkers’ emerging bonds of attachment in a fragmented employment relationship”. Work, Employment and Society 34 (3): 476–494.
Pesole, Annarosa, Cesira Urzí Brancati, Enrique Fernández-Macías, Federico Biagi and Ignacio González Vázquez. 2018. Platform Workers in Europe. EUR29275 EN. Luxembourg: Publications Office of the European Union.
Piasna, Agnieszka. 2022. “Precariousness in the Platform Economy”. In Faces of Precarity: Critical Perspectives on Work, Subjectivities and Struggles, by Joseph Choonara, Annalisa Murgia and Renato M. Carmo (eds). Bristol: Bristol University Press.
Piasna, Agnieska and Jan Drahokoupil. 2021. “Flexibility unbound. Understanding the heterogeneity of preferences among food delivery platform workers”. Socio-Economic Review 19 (4): 1397–1419.
Piasna, Agnieszka, Wouter Zwysen and Jan Drahokoupil. 2022. The Platform Economy in Europe: Results from the Second ETUI Internet and Platform Work Survey (IPWS). Working Paper 2022.05, Brussels: ETUI.
Prassl, Jeremias. 2018. Humans as a service. The promise and perils of work in the gig economy. Oxford: Oxford University Press.
Schor, Juliet B. 2020. After the gig: How the sharing economy got hijacked and how to win it back. Oakland: University of California Press.
Schor, Juliet B., William Attwood-Charles, Mehmet Cansoy, Isak Ladegaard and Robert Wengronowitz. 2020. “Dependence and precarity in the platform economy”. Theory and Society 49: 833–861, https://doi.org/10.1007/s11186-020-09408-y
Shanahan, Genevieve and Mark Smith. 2021. “Fair’s fair: psychological contracts and power in platform work”. International Journal of Human Resource Management 32 (19): 4078–4109.
Simms, Melanie and Duncan Adam. 2021. “Precarious solidarities. Unions, young workers and representative claims”. In Forms of collective engagement in youth transitions. A global perspective, by Valentina Cuzzocrea, Ben Gook and Bjørn Schiermer (eds). Leiden: Brill.
Srnicek, Nick. 2017. Platform capitalism. Cambridge: Polity Press.
Stanford, Jim. 2017. “The resurgence of gig work: historical and theoretical perspectives”. Economic and Labour Relations Review 28 (3): 382-401.
Tassinari, Arianna and Vincento Maccarrone. 2020. “Riders on the storm: workplace solidarity among gig economy couriers in Italy and the UK”. Work, Employment and Society 34 (1): 35–54.
Thäter, Laura, Thomas Gegenhuber, Elke Schüßler and Markus Ellmer. 2022. “How do workers gain voice on digital work platforms? Hotspots and blind spots in research on platform worker voice”. In Missing Voice? Worker voice and social dialogue in the platform economy, by Adrian Wilkinson, Tony Dundon, Paula Mowbray and Sarah Brooks (eds). Cheltenham: Edward Elgar.
Vallas, Steven and Juliet B. Schor. 2020. “What do platforms do? Understanding the gig economy”. Annual Review of Sociology 46 (1): 1–22.
Vandaele, Kurt. 2018a. Will trade unions survive in the platform economy? Emerging patterns of platform workers’ collective representation and voice. Working Paper 2018.05. Brussels: ETUI.
Vandaele, Kurt. 2018b. “How can trade unions connect with young workers?”. In Youth labor in transition: Inequalities, Mobility, Policies in Europe, by Jacqueline O’Reilly, Janine Leschke, Renate Ortlieb, Martin Seeleib-Kaiser and Paola Villa (eds). New York: Oxford University Press.
Vandaele, Kurt. 2021 “Collective resistance and organizational creativity amongst Europe’s platform workers: a new power in the labour movement?”. In: Work and labour relations in global platform capitalism, by Julieta Haidar and Martin Keune (eds). Cheltenham/Geneva: Edward Elgar/ILO.
Vandaele, Kurt. 2023. “Vulnerable food delivery platforms under pressure: protesting couriers seeking ‘algorithmic justice’ and alternatives”. In Routledge Handbook of the Gig Economy, by Ness I. (ed.). New York: Routledge.
Vandaele, Kurt, Jan Drahokoupil and Agnieszka Piasna. 2019. “Algorithm breakers” are not a different “species”: attitudes towards trade unions of Deliveroo riders in Belgium. Working Paper 2019.6. Brussels: ETUI.
Vandaele, Kurt and Agnieszka Piasna. 2023. “Sowing the seeds of unionisation? Exploring remote work and work-based online communities in Europe during the Covid-19 pandemic”. In The future of remote work, by Agnieszka Piasna, Silvia Rainone, Nicola Countouris and Valerio De Stefano (eds). Brussels: ETUI.
van Doorn Niels, Fabian Ferrari and Mark Graham. 2022. “Migration and migrant labour in the gig economy: an intervention”. Work, Employment and Society 37 (4): 1099-1111. https://doi.org/10.1177/09500170221096581
Veen, Alex, Tom Barratt and Caleb Goods. 2020. “Platform capital’s ‘app-etite’ for control: a labour process analysis of food delivery work in Australia”. Work, Employment and Society 34 (3): 388–406.
Visser, Jelle. 2002. “Why fewer workers join unions in Europe: a social custom explanation of membership trends”. British Journal of Industrial Relations 40 (3): 403–430.
Visser, Jelle. 2019. Trade unions in the balance. Geneva: International Labour Office.
Walker, Michael. 2021. “Peer-to-peer online voice as emergent collective action”. Journal of Industrial Relations 63 (5): 777–797.
Wilkinson, Adrian, Tony Dundon, Paula Mowbray and Sarah Brooks. 2022. “Missing voices? Integrating worker voice and social dialogue in the platform economy”. In Missing Voice? Worker voice and social dialogue in the platform economy, by Adrian Wilkinson, Tony Dundon, Paula Mowbray and Sarah Brooks (eds). Cheltenham: Edward Elgar.
Wood, Alex J. 2020. “Beyond mobilisation at McDonald’s: Toward networked organising”. Capital & Class 44 (4): 493–502.
Wood, Alex J. and Vili Lehdonvirta. 2021. “Antagonism beyond employment: how the ‘subordinated agency’ of labour platforms generates conflict in the remote gig economy”. Socio-Economic Review 19 (4): 1369–1396.
Wood, Alex J., Vili Lehdonvirta and Mark Graham. 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.
Woodcock, Jamie and Mark Graham. 2020. The gig economy: A critical introduction. Cambridge: Polity Press.
Zwysen, Wouter and Agnieszka Piasna. 2023. Juggling online gigs with offline jobs. How local labour markets are driving the growth in internet and platform work. Working Paper 2023.02. Brussels: ETUI.
Acknowledgements
The authors are very grateful to Maria Helena André for her insightful comments on the draft version of this paper and Calvin Allen for his excellent language-editing.
Earlier versions of this ILO Working Paper have been presented at the international workshops ‘Work in the post-Covid era: The end of platform labor as we know it?’, in Amsterdam in November 2022, and ‘Governance by Numbers and its Discontents’, in Dublin in June 2023.
