With respect to platform-based gig work, one key challenge shared by labor organizers and researchers is recruitment. How do you study and/or organize workers who mostly operate in isolation, who are not always visible, and who frequently lack a strong attachment to a collectively crafted professional identity, let alone a social class? These are certainly not the easiest people to find and to mobilize. Sometimes, however, gig workers are one step ahead and have taken it upon themselves to organize and let their voice be heard. Over the past few years, Uber drivers in different cities across the globe have protested against unfair working conditions and rate cuts, while food delivery workers in the UK, Italy, Germany, and the Netherlands have more recently attracted significant public exposure for taking delivery platforms like Deliveroo and Foodora to task for their malpractices . By leveraging existing communication infrastructure, such as a company-initiated Telegram chat group (in the case of Deliveroo in the Netherlands), in combination with the generative possibilities of meeting in public spaces during work time, these couriers have forged grassroots and improvised modes of resistance that did not only capture the media’s attention but also spurred unions and other labor organizations to take a serious interest in a category of workers they previously considered to be by and large ‘ unorganizable ’.
No longer isolated, invisible, or disempowered, food deliver workers were quickly embraced by smaller, rank and file unions in the aforementioned countries, which positioned them as the new vanguard in a rekindled class struggle. Moreover, for larger unions in countries like the Netherlands, supporting the actions of self-organized food delivery workers did not only promise to refresh their membership base and campaign strategies, it (thereby) also enhanced their public profile and validity. Meanwhile, to the media these young men (because there are relatively few female and/or middle-aged couriers doing this kind of work) wearing their green or pink branded outfits looked photogenic and relatable, or at least familiar enough to justify a widely published focus on their plight. Here it certainly helped that there was a clearly identifiable conflict connecting the protests across national boundaries: these protests arose after different food delivery platforms announced that they would change their payment structure from an hourly wage to a per-delivery (piece rate) remuneration system, which in some countries coincided with a contractual overhaul that reclassified couriers from employee to independent contractor status. In other words, these were concrete and often poorly communicated business decisions with far-reaching repercussions for a young workforce that was quickly becoming a staple of urban landscapes all over Europe, which made it a palatable issue for the public – including politicians – to take notice and express indignation.
So when, in September 2017, I was looking for gig workers who would be willing to participate in the Reshaping Work conference that I co-organized, it was relatively easy to recruit a number of Deliveroo couriers who had recently collectivized into the Riders Union (backed by the largest trade union in the Netherlands – the FNV). It was, however, much more difficult to find cleaners working through Helpling, a platform for domestic cleaning services that operates in nine countries worldwide. In contrast to the newfound militancy of the predominantly young and male delivery workers who were claiming their rights in both on- and offline public spaces, Helpling cleaners were conspicuously absent – from such spaces but also from the public debate about on-demand gig work (on this last absence, see this piece by Julia Ticona and Alexandra Mateescu). This presented me with a problem and a question. The problem was practical and pertained to recruitment: where to find Helpling cleaners and how to get them motivated to participate in our conference? The question was both practical and conceptual in nature, and it is one that I’d like to reflect on here, in this first research note since I started the Platform Labor project in February: what are the conditions of possibility of worker organizing in platform-based gig economies?
We need a more granular approach to the study of gig work
Let me start with a statement: there is no such thing as “the gig economy”. We need a more differentiated and granular approach to studying platform-based gig work that focuses on particular econom ies (plural), markets, and/or industries which platforms have sought to “disrupt”, while taking into account a number of factors that co-determine how/to what extent gig workers in each industry will be able and willing to organize. For example, one significant difference between the situation of Dutch Deliveroo couriers and Helpling cleaners is that, at the time, the latter were not forced to deal with a sudden managerial decision that directly affected their livelihood. Helpling cleaners had always been independent contractors and continued to be paid an hourly wage. There was thus no controversy, or labor conflict, around which an engaged and antagonistic public of stakeholders could organize. Beyond this “incidental” difference, there are a number of structural factors that – I believe – need to be taken into account by researchers, labor organizers, and policymakers alike, as they seek to grasp and improve the conditions/dynamics of gig work in its multiple forms.
The nature of the work itself The particular type or kind of gig work that is being performed matters a great deal. This may perhaps be obvious but unfortunately it isn’t always registered sufficiently. Whereas food delivery work is done outdoors, in public spaces, and is thus highly visible, domestic cleaning takes place in private homes, away from the public eye. The barrier to public protest will thus be lower for couriers, who are “at home” in the streets (where they are already visible to each other and the public at large), compared to cleaners who operate in a more isolated workplace. Moreover, while food delivery workers usually have very little engagement with customers (whether restaurants or consumers), domestic cleaners often develop a more extensive relationship with their clients, whose homes form intimate workplaces where the labor process is monitored and evaluated by the client more than by an algorithm. In other words, food delivery work produces a much more fungible , or interchangeable, service compared to domestic cleaning. This is likely to affect a gig worker’s proclivity to protest their conditions and organize for change: if you decide to strike, who will be negatively affected by this action? As a cleaner, you may be hesitant to let down a client with whom you have built up a professional and personal relationship and whom you know depends on you.
A worker’s relationship to/investment in the work Gig work is often presented as a good way to earn some supplemental income, but in various industries there are many people who have turned – or are trying to turn – ride-hailing, (food) delivery, cleaning and other kinds of platform-based contingent work into a full time occupation. The measure of time and capital invested in gig work, and thus the level of dependency on this work as a source of income, will greatly affect the extent to which a gig worker is prepared to fight for better labor conditions. Here in New York City, where I am conducting interviews with workers who provide app-based food delivery and cleaning services, couriers who do food delivery full time have expressed relatively more interest in labor organizing than those who have a more casual relationship to this work. It should be added here that many of the more “militant” full time couriers – who are currently attempting to organize worker resistance in the city’s very crowded and fragmented food delivery industry where most couriers alternate between multiple apps – have previously worked as bike messengers, which imbued them with a strong professional identity and a deep sense of bike worker-centric solidarity. In contrast, the full time cleaners whom I’ve been talking to – who offer their services through Handy and/or TaskRabbit – are generally much less interested in labor organizing. Instead of understanding themselves as belonging to a class of workers, they are much more likely to adopt an entrepreneurial perspective in which they embrace their position as independent contractors and sometimes aspire to start their own cleaning company. Handy, for them, is primarily a lead generator serving them clients whose business may potentially be taken off the platform. This illustrates how professional self-identification shapes one’s relation to the work at hand, which in turn affects one’s inclination to organize for better labor conditions. Finally, it also matters whether platform-mediated gig work is seen as a temporary income opportunity that allows them to get out of an economic tight spot on their way to other jobs/sources of income, or is embraced as a longer term engagement that demands particular lifestyle and mindset. (This itself brings up the issue of “gig work temporality”: many people I interviewed lived from week to week, if not day to day, whereas labor organizing demands a different, more protracted relationship to time that includes long-term goals which may be overshadowed by the more immediate exigencies of making ends meet. One concrete suggestion would be to set as many concrete, short-term goals as possible when trying to get gig workers on board within a campaign.)
The social situation and legal status of the worker When assessing the extent to which gig workers are able/willing to organize or protest their conditions, it is also important to expand one’s purview beyond the immediate sphere of production to include a worker’s social/domestic situation as well as their legal status. For instance, when a worker has a high level of private care-giving responsibility, or relies extensively on care-giving facilities, s/he will have limited time to devote to labor organizing activities (and may even have little opportunity to sit down and answer some questions about her/his work). Labor organizers should take this situation into account and seek ways to assist and support workers in their daily lives, like community-based organizations have been doing all along. Moreover, and more pertinently, undocumented workers will rightfully be much more reluctant to join protest or organize against their employer, for fear of retaliation. In New York City, food delivery has traditionally been a highly informal and unregulated sector of the city’s economy where the work is done by (mostly Chinese and Latin American) immigrant men, many of them undocumented and laboring under hyperprecarious conditions . As food delivery platforms entered this space, undocumented workers have recently also started to experiment with this new form of gig work, even though they are technically not allowed on the platform without a social security number, a driver’s license/State-issued ID, and a bank account. Although most undocumented couriers continue to work directly for restaurants, the distinction between traditionally organized and platform-orchestrated food delivery is becoming more blurred, generating a need to think creatively about how undocumented workers can be safely and responsibly included in struggles for better platform-based work. Worker centers and community-based organizations such as Make the Road will play a vital role moving forward.
Platform/app design Another factor that co-determines the experience and conditions of app-mediated gig work, thereby also affecting possibilities and inclinations with respect to labor organizing, is user interface design. Most directly, this pertains to the availability of features that facilitate communication between workers, such as the Telegram chat groups that Deliveroo initially used in the Netherlands to manage its courier fleet in a decentralized manner (i.e. small groups of couriers managed by team leaders, called “lead riders”). As mentioned above, this communication channel was quickly exploited for more radical purposes when the news about the changing payment structure reached the couriers. Yet platform-facilitated communication between gig workers is very rare, if not non-existent by now. More indirectly, then, the nature and quality of a platform’s app features – such as push notifications about incoming gigs (how much information about the gig is given before it is accepted and how much time does a worker have to accept it?) and live support options (can the worker call a helpdesk or does s/he have to write an email?) – will influence the level of frustration workers experience as they are trying to maximize their income opportunities while minimizing stress-inducing situations. This will in turn make them more or less likely to develop an antagonistic relationship to the platform and become susceptible to labor organizing initiatives or other, more casual forms of pushback against their conditions. It may, for instance, foment the desire to seek solidaristic connections to other workers via social media and online forums . Some couriers and cleaners have told me that they felt a strong sense of relief and even empowerment when they learned that there were other workers out there who shared their frustrations with an app and were dealing with similar problems which could be discussed in Facebook groups or subreddits dedicated to particular platforms or types of gig work. It transformed work-related issues and perceived injustices from individual concerns into collective problems that required crowdsourced, makeshift solutions in the absence of platform accountability.
- Management of the labor process Each platform company governs its workforce in a particular – app-based, data-intensive – way, iteratively developing its own take on what has been termed “algorithmic management” . Decisions made during this process depend to a large extent on the previously discussed nature of the work that has to be managed and are materialized as various control and evaluation protocols aimed at optimizing the labor process with respect to quality and productivity. It is important to note that while the data generated in algorithmic management processes may be harvested and analyzed in a centralized manner, control and evaluation protocols are distributed practices in which management tasks are outsourced to customers who (are expected to) monitor and judge gig workers to varying degrees – again depending on the nature of the service provided. When I ask Handy cleaners who, in their view, has the most control over their work, the most common answer is that this control is unevenly split between Handy, the client, and themselves. It is the client who is requested to evaluate the job, not just through a rating system but also by completing a survey/check list and writing a review (only 5-star reviews are publicly visible on a cleaner’s profile). Nevertheless, it is Handy which penalizes cleaners with fees when they arrive late or cancel a job within 48 hours before the start of a job. Even though food delivery workers do not incur such fees, they are much more likely to point to the platform as the primary controlling agent. While most food delivery platforms use customer ratings, this evaluative score makes up a relatively minor part of a larger set of performance metrics and incentives mobilized to manage courier fleets in real time. (It also matters here whether the platform expects couriers to sign up for so-called “blocks” [i.e. work shifts] or if couriers can log on and work whenever they want to, giving them a stronger sense of control over their schedule.) When studying gig work, we should thus not only look at the extent to which the work is controlled and how this control is achieved technically, but also examine who is expected to participate in this process . (For instance, who controls the labor process when a courier decides to chase an incentive such as a bonus or a guarantee?) Beyond generating a richer understanding of app-based workforce management practices, such an approach also enables the development of more customized labor organizing strategies that are responsive to the (mal)practices of specific platforms.
Besides these first five immediately pertinent factors, researchers, labor organizers and policymakers should take into account three other structural factors that I now briefly address at the end of this research note, but which will certainly be elaborated on in future notes and publications: the platform business model ; the composition and regulation of an industry ; and the socio-historical position of that industry and its workforce . First, the way a platform is engineered – including the design of the apps that interface with workers and clients – and how workers are managed through this technical infrastructure is to a large extent shaped by the particular business model a company chooses to adopt. Decisions regarding asset acquisition or outsourcing, level of dependency on venture capital, revenue model, governing structure, workforce classification, brand strategy, payment structure, and provision of particular benefits and/or compensations all either directly or indirectly affect the remuneration and quality of the gig work performed through the platform and thus the necessity as well as the likelihood of labor organizing initiatives. I have already mentioned how a sudden change in payment structure has instigated a wave of courier protests that continues to haunt European food delivery platforms. Likewise, we should not underestimate the power of branding: as I suggested earlier, it was not just the public visibility of the Deliveroo and Foodora couriers that managed to attract the media’s attention, it was also the due to the fact that they were widely recognizable as a branded workforce – in green or pink outfits, carrying green or pink containers. This evoked a sense of unity, as couriers were also clearly visible to each other as co-workers laboring for the same company – indeed until recently they were employees of these companies. I would thus not be surprised when, under the independent contractor regime, the rules and expectations regarding workforce branding will loosen up considerably at Deliveroo and Foodora, with couriers being allowed to leave their green/pink jackets at home (if they even get one these days). This would more closely resemble the situation in New York City, where couriers have never been employed by the food delivery platforms they work for and nobody uses branded outfits or gear – save for the occasional bag with “Caviar”, “Grubhub”, “Postmates”, or “Doordash” printed on it. Here, these companies would rather not be visibly connected to their workforce, or so it seems.
Second, the particular business model a platform company decides to adopt – and the likelihood that it will pivot along the way – will be significantly influenced by the existing composition and regulation of the industry it is seeking to transform and dominate. Both food delivery and domestic cleaning are service industries that have traditionally been highly informal, fragmented, and un(der)regulated. In the case of food delivery, it could hardly be called an industry at all before companies like Grubhub and Takeaway.com enabled consumers to conveniently place online orders with their local restaurants, in many cases expanding the markets of these businesses – which managed their own delivery fleet. When a new wave of companies like Deliveroo and Doordash started offering delivery services on top of online (now app-based) order processing, they made food delivery available to restaurants which had previously been dine-in only, again significantly transforming and expanding this relatively new industry. As I noted above, NYC’s food delivery industry is rather crowded compared to most European cities, with six companies offering relatively high sign-up rewards and low base rates per delivery, which has resulted in a market flooded with couriers who – especially when they work full time – alternate between multiple apps to make ends meet. This situation makes it less effective for labor organizers to target one particular company in their fight for better protections and compensations (as has been happening in Europe), instead requiring a sectorial or industry-level approach which demands that these companies share a collective responsibility to their workforce. One important issue moving forward, also for policymakers and regulators, will be ensuring a greater measure of data transparency that, for instance, can help with problems such as pro-rating minimum wage and compensation payments among companies.
Finally, the socio-historical position of the industry and especially its workforce should also be taken into account, as it will affect the level of public interest in and support for labor struggles. One important reason for the success of many food delivery and cleaning platforms is precisely that they entered – and thereby selectively formalized – traditionally informal and poorly regulated industries that have always relied on the precarious labor of marginalized minority – increasingly immigrant – men and women who service households by delivering their food or cleaning their homes. In other words, long before the market entry of these platforms, food delivery and domestic cleaning were already devalued – because deeply classed, gendered, and racialized – types of “menial” labor that were excluded from many protections and rights bestowed on other, more respectable (i.e. white working/middle class) kinds of work. One of the main questions I aim to answer during the Platform Labor project is to what extent (and how) platform companies operating in these low-wage service industries perpetuate, intensify, and/or counter these existing inequalities – while perhaps also creating new forms of inequality. I believe that platforms like Deliveroo, UberEATS, and Handy could potentially – when looking at their technical capabilities – play a positive and empowering role in the lives of their workers, many of whom have never experienced the privilege of secure employment with a firm that pays a living wage and provides the benefits and protections many of us (rightfully) feel entitled to. Yet I also think it is safe to say that these companies are currently far from living up to this potential. There is a lot of work to be done by all stakeholders involved.
With this in mind, I want to close this inaugural research note by emphasizing that all the factors discussed here should be understood as interconnected and layered. Together they co-determine the labor process and income opportunities of particular forms of gig work, thereby shaping the conditions of possibility for worker organizing in industries where platform-driven companies are continually changing labor standards, norms, and expectations.