Understanding Racism in Digital Capitalism. Racialisation and De-Racialisation in Platform Economies, Infrastructural Racism and Algorithmic Opacity
Stefania Animento
Humboldt University, Berlin, Germany, Stefania_animento@hotmail.it
Abstract: Recent literature has argued that digital technologies reinforce existent inequalities along race, class and gender. However, the relationship between digital capitalism and racism is yet to be explored in depth. How does digitalisation rework social relations and social cooperation to produce new forms of racialised hierarchisation and differentiation? The article is based on an international project on platform labour spanning seven European cities. It focuses on the sector of ride-hailing in Berlin and analyses the interactions between processes of platformisation and (de-)racialisation. Finally, it shows how racism becomes infrastructural when platforms organise its circulation.
Keywords: platform labour, racism, migration, digital capitalism, platform capitalism, racialisation, Uber
Acknowledgement: I warmly thank the Study Group “Migration, Labour, Digitalisation and Racism” at the BIM Institute of the Humboldt University of Berlin for creating an excellent environment and for the stimulating discussions at the intersection of research and politics.
In the past years, renewed interest has arisen in the relationship between racism and capitalism (Fraser 2022, Melamed 2016, Bhattacharyya 2018). Highly debated and even contested is the question of whether racism is necessary for capitalism to thrive (see Conroy 2022). Nevertheless, it is largely agreed that racism and capitalism as systems of domination have both a common history and a present marked by context-specific conjunctures and interactions. The expansion of capitalism rests on those labouring populations made disposable through disparate forms of primitive accumulation. To enter into relations of exploitation, they must be separated from modes of non-capitalist social reproduction. “Racism”, affirms Jodi Melamed, “enshrines the inequalities that capitalism requires”, it provides the ground for social separateness (Melamed 2016, 77). Capitalism thus flourishes when it can extract value from relationships of exploitation and expropriation put in place by the interaction between socially distinct, antagonistic, and unequal subjects. The renaissance of the concept of racial capitalism helps to understand this nexus, it both challenges the fallacies of Marxism regarding colonialism and imperialism and requires the analysis of racism to take newer developments of capitalism such as neoliberalism and exploitation of diversity into account (Melamed 2016). Similar to how materialist feminism has pushed to expand the analysis of capitalism beyond the focus on production and masculine labour, the analysis of racial capitalism pushes us to think of the modes of co-constitution of race and capital (Fuchs 2018).
The unity and homogeneity of capitalism, however, is now more contested than ever. The scripts of capitalism vary across geographies and temporalities and must be analysed accordingly, avoiding one-size-fits-all theories (Peck and Theodore 2007). Recent literature highlights distinct but crucial aspects and attributes of contemporary capitalism, which seem to coexist, depending on which angle of analysis we choose. It seems more correct to acknowledge that multiple vectors expand capitalist logics, modes of production and accumulation at the same time. Among these, the transformations of capitalism brought by digital technologies constitute an expanding field of studies. Zuboff mentions “surveillance capitalism” (Zuboff 2019), others refer to “digital capitalism” and “big data capitalism” (Schiller 1999, and more recently Fuchs 2019, 2021, Pfeiffer 2021) and Srnicek suggests the term “platform capitalism” (Srnicek 2017). In the past decade, the question of how race relations are reworked through the advent of digital capitalism has been more addressed but is still largely unexplored (McMillan Cottom 2020; Hamilton 2020; Benjamin 2023).
In the article, I investigate the ways racism as a system of unequal relations relates to the dynamics of the accumulation of capital sparked by digital platforms. While I will not be able to talk about all forms of digital capitalism, I will focus on platforms as one of the most important vectors of capital expansion in the present day.
Digital platforms are themselves not easy to define. Srnicek provided a useful classification of platforms according to their function and what they do, namely advertising, cloud, industrial, product and lean platforms (Srnicek 2017). According to Grabher and König (2020), platforms can be considered as infrastructures (Plantin and Punathambekar 2019), ecosystems, and match-making markets (Evans and Schmalensee 2016). Other authors have investigated the ideologies around the rise of the sharing economy, and the logic of organizing, distributing and governing resources which platforms propagate (Andersson Schwarz 2017).
In this article, I will relate to platforms as infrastructures able to organize the mobility and circulation of scripts, emotions, capital and labour. I am interested in showing how those platforms which organise labour and provide services for social reproduction relate to racism. Far from making universalising statements, I will illustrate empirically the specific contribution made by these platforms to the development and fixing of forms of racism, even when they might seem contradictory or trivial. I focus on the category of “lean platforms” (Srnicek 2017), including those platforms such as Airbnb, Deliveroo, Helpling and Uber. I will pay mostly attention to the latter, which I have researched more in-depth, and which stands out as particularly interesting to study the relationship of platforms to racism. In the next section, I will provide a methodological frame for the article. I will explain on which data it is based and my stance towards writing about racism as a white researcher. Following, I will proceed to highlight two main arguments. Firstly, I will show how platforms relate and interact with processes of racialisation and de-racialisation, considered as processes of race-making and unmaking. To do this, I will use the concepts of marketisation and racialisation. Secondly, I will focus on how platforms organise the circulation and blocking of racist scripts, making racism infrastructural. I suggest that such an enterprise can contribute to drawing a bigger picture of the restructuring of racism and capitalism in present times.
The article is mainly based on extensive research carried out between 2019 and 2022 at the international project “Platform Labour in Urban Spaces”. In the project, researchers from universities, labour cooperatives and third-sector organisations from eight European countries produced a qualitative and quantitative investigation on the impact of platforms on labour, urban economies, and regulation of the involved industries. The project focused on four platforms (Airbnb, Deliveroo, Helpling, and Uber) and seven European cities (Barcelona, Berlin, Bologna, Lisbon, London, Paris, and Tallinn). The present article discusses the findings of the qualitative part of the research, which entails over 220 interviews with platform workers, managers, and key stakeholders in the respective industries. In this research project, together with my colleagues, I coordinated the qualitative research in the other cities of the project and carried out part of the interviews with workers and stakeholders located in Berlin. Firstly, the article will refer mainly to interviews carried out in Berlin with platform workers, and in particular Uber drivers, and with two additional interviews with workers of the grocery delivery platform Gorillas. The Uber drivers were contacted via the app and were asked for an interview and interviewed during the ride. The interviews were carried out in German and Turkish. The interviews concerned three main topics, i.e., labour process, labour protection and skills, but also involved questions about socio-demographic factors and migration experiences. In the article, the analysis will focus on the implications of platform labour for processes of racialisation within and outside platforms. To deepen the understanding of the intricacies between platform labour and racial capitalism, in Spring/Sommer 2023, I carried out interviews and informal conversations with ten Uber drivers in Berlin.[1] These conversations centred on migration experiences, residency status, and experience of discrimination.
The article intends to shed light on the relationship of platforms and platform capitalism to racism. As a researcher, I use data collected and produced by and with interview partners. I am aware of the imbalance of power and social recognition entailed in researching poverty, exploitation, and inequalities of all sorts. As a white researcher, writing about racism makes it inescapable to risk reproducing othering processes and enforcing racialised social inequalities, hierarchies, and asymmetries. Further, researching racism from a white point of view means running the risk of neglecting crucial aspects of the social context under investigation (Duneier 2004).
I will try to give space to the words of the people interviewed, in particular an Uber driver who disclosed many crucial points about the relationship between racism and platform work.[2] I offer the data to the public for further analyses, critiques, and contestations. The research cannot be considered an ethnography, since too little time was dedicated to being with the interviewed subjects and participating in their everyday life (see Rosenblat 2018 for an ethnography of Uber drivers in the US). However, I will attempt to pay attention to the racialised and classed mismatch and to take the accounts of platform workers about their lives and their perspectives on the platform economy as seriously as they deserve.
Platform economies have spread across the world at a fast pace in the past ten to fifteen years. After the global financial crisis, huge masses of venture capital flowed into both established platform companies such as Uber and Airbnb, and newly rising unicorns, such as Gorillas and Getir. Platforms enter already existing economies and typically bring turbulences into their composition, regulation, and organisation. Through the injection of massive capital into these economies, they present themselves as newly arrived but lumbering economic actors. They lobby at local and supralocal institutions, present themselves to the public with bulky advertising across cities and the internet, and allure thousands of labouring bodies to work for them.
Behind the ideological façade of disruption and innovation, however, platforms adapt very pragmatically according to the social, economic, and institutional context in which they operate (Vallas and Schor 2020). For instance, recent developments show that in the West-European context, the original model of freelancing is not dominant. Rather, where sectors are more deeply regulated, platforms adopt hybrid approaches, including intermediary agencies and companies which provide them with the labour force, as it happens in many other sectors dominated by labour precarity (see Niebler et al. 2023a). In other cases, platforms let the workers formally decide whether they want to be employed or freelance, as in the case of Helpling in Berlin (Niebler et al. 2023b).
With the Polanyian term of marketisation, we can address the process of entering new markets sparked by platforms (Polanyi 1959, Grabher and König 2020). and connected to the mobilisation of the labour force and the rise of new consumption patterns. Rather than considering platformisation as a one-way and heterogeneous process, through which groups of labouring bodies are mobilised and practices of social cooperation and reproduction commodified, I adopt the suggestion to “push beyond the categories of market and non-market in the platform economy” developed by Grabher and König (2020). To do so, I accept that markets are formed at a crossroads of a host of different logics and rationalities”, among which the neoclassical economy is only one (Berndt et al. 2020). This perspective supports a “softer” understanding of platformisation as entailing multiple directions and a more complex relationship to processes of separation and racialisation of labour. Further, it helps to see labour agency beyond established, institutionalised and canonical forms of collective resistance. The research carried out and presented in this paper shows markets in the making and un-making and takes the perspective of markets as frontier regions (ibid.), in which negotiations and valuations are relatively fresh and particularly flexible.
The research on the four platforms Uber, Airbnb, Deliveroo and Helpling demonstrated that platform companies often relate in similar ways to processes of racialisation when they enter new markets. This article considers the variable of labour as a crucial one when exploring the market strategies of platforms and of course economic actors in general. Platforms need cheap, flexible, and interchangeable labour power, to stay flexible and adapt to the high dynamism of financial markets, upon which they depend. To do so, they change strategies at a very tight temporal pace. As elaborated elsewhere (Animento 2024), platformisation often follows two main phases. Firstly, platforms arrive in a new urban market accompanied by pervasive marketing campaigns which promote on the one side innovative, faster, and more efficient forms of consumption, on the other side labour opportunities based on autonomy, diversity and entrepreneurialism. As an example, in metropolises such as London and Paris, Uber and Deliveroo have actively advertised job opportunities in poor and peripheral districts, where a large part of the population has migrant backgrounds. In Berlin, Helpling has developed a marketing strategy targeting gender equality and diversity. Further, the initial phase is marked by relatively better working conditions, better pay and perks such as free gadgets and discounts. On the consumption side, perks and discounts also aim to attract as many clients as possible. This phase has the goal of producing a large pool of labour force available to the company to scale up quickly and become a leader in the sector. Our research shows that in newly arrived platforms workers present high diversity in terms of origins, migrant backgrounds, and social position. Here, the young white student and the racialised refugee might queue close to each other at a ghost kitchen waiting for their order before leaving, each one directed to his or her address of delivery.
While the first phase is marked by expansion and hyper-diversity, the second phase consists of shrinking and adjusting to the local market. Platforms usually adjust their institutional set-up for several reasons. One main reason is regulation: after the initial disruptive phase, regulative entities both at local and national levels might intervene to “curb” platform economies by applying existing laws or even producing ad hoc ones, as in the case of the so-called “ley de rider” in Spain. As research across countries shows, such attempts at regulation often fail their objectives, sometimes even worsening the labour conditions of workers. Platform companies excel in finding legal loopholes and semi-formal ways of operating in new markets. Other reasons for the readjusting of the business model are macroeconomic and financial shocks or, importantly, labour struggles and workers’ advocacy. This phase of adaptation, however, also entails changes in the labour composition. When platforms need to downsize or get settled, labour conditions usually worsen, often with unilateral and abrupt changes. The type of pay might switch from hourly to piece pay (as in the case of Deliveroo in the UK and Germany), bonus systems might get discarded, while many labour relations are completely terminated, either through logout (in the case of the pure freelancing model) or through layoffs or non-renovation of fixed-term contracts. The phase of shrinking usually comes with a loss of diversity in the labour force, particularly in terms of residency status, migrant background, and social position. More precisely, those workers who can afford to leave the platform will move to comparatively better jobs, while those who do not will have to stay. According to the type of platform and the urban and national contexts, the dynamics of exclusion from the platform might change. However, across cities and platforms, our research, as well as research from non-European contexts (Zhou 2022) shows that over time the labour force active in platform economies becomes more heavily racialized and poorer. Further, COVID-19 impacted the business of Uber in Berlin, as in most other cities. Many interviewees left Uber during the pandemic and started again once the measures of isolation were reduced. While the interviews carried out for the PLUS project took place between and right after the beginning of the pandemic in 2020, the interviews which I carried out took place in 2023, when many drivers who had stopped their work during the pandemic had come back to ride-hailing, once the business recovered.
The processes of marketisation which platforms trigger often follow a double movement. During the phase of “embeddedness”, in which regulatory instances intervene and tame the platforms, however, there is no improvement in the livelihoods of the labouring bodies which feed the platform economies. Rather, the alignment of platform models and regulations serves to move patterns of racialisation and ethnicity-making into the platform boundaries. The processes of racialisation which platforms interact with and even shape, thus, are more complex than the double movement could explain. They cannot be comprised within the boundaries of platform companies and cannot even be reduced to the present conjuncture. Rather, platforms rely on well-oiled mechanisms of hierarchisation of labouring bodies, leveraging the complex and cumulative matrix of migration regimes which have been applied for decades. Via their “more-than-double movement”, platforms draw and redraw boundaries around labouring populations, putting different social and political subjects in connection to one another, but also separating and hierarchising them. An example is the “caporalato” system which has developed in many platform economies in cities across the globe, as research (Peterlongo 2023). The selling out and renting of accounts enable illegalised migrants to work, by entering in relation with their “bosses”, who often share the same migrant background. This stratification and diversification well embody the “densely connected social separateness” which Melamed considers the “hallmark of racial capitalism” (Melamed 2016, 81). Hereby, the construction and fixation of subjects and identities via the attribution of characters and qualities which go back to older racial scripts is crucial.
Platform capitalism flourishes over processes of primitive accumulation stratified over time. Their need for a cheap and interchangeable labour force requires labouring populations that have few alternatives in the labour markets to choose from. Migration, while being a human trope and an act of autonomy (Mezzadra 2010), can also be viewed as the result of primitive accumulation, which separates people from the places they consider home with the hope of finding a livelihood elsewhere. Coming into conversation with Uber drivers, I asked them about their job, their families and how they liked it to be here. Many of them often ended up telling me similar stories and more specifically similar conclusions: if they could, they would go back to their home countries, which were mostly North African countries, such as Tunisia, and Middle Eastern countries, such as Syria or Afghanistan. Nostalgia for past, far away and imagined home countries is a common feature of migration and is part of the “double absence” experienced by migrants, as theorized by Sayad (1999). At the same time, the Uber drivers I talked to spoke of the feeling of being isolated, of living a life of work, with no kinship except for the closest nuclear family and a few more, with neither friends nor free time. The sense of being integrated into a system aimed at making profits must be seen along with their capacity to compare it to their home countries. This capacity of comparing different varieties of capitalism, and their different social and racialized position suggests how to understand the processes of capital accumulation exerted over their bodies. The social separateness that many platform workers experience started with their migration from their countries of origin, and often with the migration of their parents or grandparents.
This article contends that racialisation must be understood as relational, as it entails a never-ending process of re-articulating subjects in relation to others and to the fluctuations of capital accumulation which come along with technological transformations, changes in political entities and production of knowledge, and labour conflicts. I define racialisation as the process of allocating individuals and groups into a hierarchical and unequal system of race relations. Such a system, however, is never stable or settled, but rather prone to change and re-configuring. Racial formations, as theorised by Omi and Winant (1986), emerge at the intersection of politics, economy, and society. In the following, I will mention three examples to illustrate how platforms interact with processes of racialization. Finally, I will provocatively use the concept of de-racialisation to challenge static and one-way perspectives on the role played by platformisation in processes of racialisation.
The first example concerns the role of law and migration regimes in forging flows of mobility and immobility from and to platforms. When platforms start downsizing and worsening labour conditions, workers with secure residency can quit their jobs and move to a comparatively better one. For these groups, platform work can represent a stepping stone (Van Doorn 2020). This is the case of Margherita, whom I interviewed in 2021. At that time, she used to work in Berlin as a rider for the platform of grocery delivery Gorillas. Of Italian nationality, Margherita decided after one year to refuse to sign the renovation of the contract and instead receive unemployment benefits, while she could complete her German course and then apply for a job more fitting her educational background. On the contrary, migrants with fixed-term visas are often obliged to stay at their workplace, because they are not entitled to benefits. Via this process of “differential exclusion” (Mezzadra 2010), which goes hand in hand with the racialisation of certain groups and the re-racialisation of others, the platform labour force tends to become more homogenous in terms of status residency over time. Workers whose permit of residency is bound to a labour contract will have to stay put, with the platform able to squeeze conditions down to a lower level.
The second example refers to the formation of a labour force predicated upon inferiority, which has come to accompany processes of racialization concerning newly arrived migrants and refugees. The taxi drivers whom I interviewed in Berlin manifested a strong antagonism against Uber drivers, as in many other places elsewhere (Gebrial 2022; Hua and Ray 2020). Uber drivers were described by taxi drivers as not capable of doing their job, as illiterate and therefore hyper-exploitable, and dangerous. Across Facebook groups for taxi drivers, a recurrent narrative about the incapacity and disqualification of Uber drivers emerged, which oftentimes resulted in smear and hate speech. The taxi drivers interviewed were themselves migrants of the second or third generation, with roots in Turkey and the Balkans. They explained that their everyday work life is very similar to the one of Uber drivers; much digitalisation has taken place in the traditional taxi sector, too. In fact, the taxi and ride-hailing sectors are both largely dominated by racialised migrant labour, but the advent of platforms has created or cemented internal fractures between the deserving and the undeserving drivers. By adapting to the regulation of private vehicle transport (different from taxi transport), platforms enlarged the pool of labouring bodies, to encompass newly arrived migrants and refugees, who might have had otherwise difficulties in entering the highly regulated sector of taxis. Opposition to the Uber model by taxi drivers has easily become translated into hate against Uber drivers, who are considered less integrated and unworthy. However, digital inquiries across Facebook and WhatsApp Groups revealed that many taxi drivers also work parallelly for Uber and other ride-hailing apps. The stark division between taxi and Uber drivers, upon which such strong antagonism is predicated, is more than blurred. Again, the stratification of migration regimes emerged in the process of racialising Uber drivers as a tool for partition, social separateness, and conflict between migrant “generations”. In Berlin, as elsewhere, migrants of different generations share many experiences and systemic racism, but also experience important differences regarding the state treatment of their status as mobile, migrant, or post-migrant citizens. While the division is not at all clear, Uber drivers have come to represent the underprivileged and excluded workers. Even those taxi drivers, who used less harsh words when talking of Uber drivers, referred to them as too powerless to resist and face Uber for its merciless treatment of workers. Uber drivers become racialised even if they are not classified univocally, but are rather generically marked as inferior, unable, and still threatening because of their alleged compliance with the expansion of platform capitalism.
While platforms might redraw boundaries of the labour force considered inferior and disposable, however, in some cases, they might provide a bridge between different generations of migration and support the strengthening of kinship relations based on entrepreneurialism. This is the case of Mohammed, with whom I had a conversation during a taxi ride. He arrived in Germany from Tunisia at the age of 17 (he is now 30) to visit his uncle, and then stayed after finding a job. Then, he started a security business in East Germany together with his uncle and his family. Since the business performed poorly, the whole family moved to Berlin to open a private vehicle company, cooperating with Uber. Mohammed was positive about Uber and considered it a good company to start a business with, which in this case was successful and expanding. In his case, through the platform, he had been able to improve his livelihood by reinforcing his ties with his uncle’s family. However, when asked if he liked living in Germany, and if he was satisfied with his achievements and his life, he admitted that he considered his move to Germany a mistake, which had happened without him realising that it would be a decision for life.
Mohammed is now the co-owner of a subcontracting company which works for Uber. Other literature based on the PLUS project showed that workers in subcontracting companies have formal contracts, but often work for more hours than those written on their contracts and are exposed to multiple forms of exploitation (Niebler et al. 2023a; Animento 2024). On one side they are exploited in a more classical sense by the subcontractor, who is their formal employer, on the other side they are exposed to the algorithmic management developed by Uber. At the same time, subcontractors interviewed in the research admitted that only companies with a large fleet are successful when doing business with Uber. Two interviewees reported that they had started their own company, but after one or two years got back to simply driving because the business was not worthy and profitable. The separation between drivers and subcontractors among migrants, even when they belong to the first generation, shows a further albeit precarious partition brought by the platform.
Processes of platformisation of economies and industries are deeply intertwined with complex patterns of racialisation that are placed both inside and outside the boundaries of platforms, and both in present and past times. However, to tell the story that platforms intervene in race relations only to reinforce them would be superficial. Migrant workers with precarious status and a whole history of exclusion in the labour market often referred to platform work as the first job in which they did not feel considered as part of an ethnic or racialized group. Assan, a former refugee from Afghanistan, who had struggled since his childhood to achieve a decent livelihood in Germany, declared:
“I was left with no other job in this world. Uber was my one and only chance. Why Uber? Uber has an advantage for us foreigners, it doesn’t judge us. Really, that’s the only thing about Uber. Uber doesn’t care if you can speak German, if you’re good-looking, whether you are tall, or short, whether you close your eyes when you laugh or not. No one cares. So a very loyal company as far as drivers are concerned.”
[Uber_Ber_8]
Uber as most platforms, are not considered part of the “German” economy, of the economy of the majority in Germany. Rather, they were often referred to as “American companies”, the origin of which should be placed outside the national borders. I suggest that such detachment of platforms from the frame of the national economy is linked to the perceived de-racialization which Assan expresses. He goes even further, arguing:
“And you will never get into an Uber car where Peter Meier is driving, it’s either Mustafa, or Mohammed, or / Uber is actually an Arabic company [laughs], the driver is called Mohammed, Mustafa, Ahmed, Hussain … So only foreigners are driving. That’s the reason. Because you are not put in certain levels. You don’t have to kiss anybody’s ass, you don’t have you don’t have a superior, you are your own boss, you do what you want.”
[Uber_Ber_8]
Barot and Bird define de-racialisation as the “removal of the idea of race and its alleged consequences from a discourse, such that a group which is constructed and defined as a race may present itself in the public domain in non-racial terms” (Barot and Bird 2001, 614). They also admit that the concept is much less debated and used than racialisation (which is still part of ongoing debates, see Hochman 2019). Assan’s statements, which resonate with empirical data in recent literature (Purcell and Brook 2022; Anwar and Graham 2021), should be taken seriously. The subjective experience of platform workers might clash with accounts of the impact of digital technologies on racism, but it opens the question about how platform workers as classed and racialised subjects might develop spaces of autonomy.
Much literature has now been written on how digital capitalism interacts with race relations (Fuchs 2018; Noble 2018; Hamilton 2020; McMillan Cottom 2020; Benjamin 2023). While in “Algorithms of Oppression” Noble focuses on racialising bias in search engines, in “Race after Technology” Benjamin provides an extensive investigation of digital technologies which pretend to be neutral but instead reinforce racism. She adopts the term “New Jim Code” to define those “subtler and even alluring forms of coding [racial] inequity get a pass” (Benjamin 2023, 56) against the more visible forms of racism such as White supremacy discourses and hate speech against Blacks on social media. Digital technologies are often marketised as progressive and diversity-welcoming. Benjamin speaks about the “illusion of progress and neutrality” (54) which these technologies propagate. In our research, Uber drivers are very much aware that working for the app is not a good job, but they often argue that it is the best job that they could get, not secondarily because of the absence of a human boss in their everyday lives on the streets.
The liberating effect of not having to face a possibly racist employer, to escape some forms of everyday racism from their boss and superiors, is a crucial factor for workers. The feeling of emancipation from racialisation still holds even when workers recognize that their labour conditions are poor, that they are exploited and exposed to oppression and the unpredictable arbitrariness of algorithmic management. This applies to Assan, too, who later in the interview will remind one episode in which he was blocked by the app for no reason, to discover after many days, that it had been just a mistake of the Uber app itself. The ambivalent relationship of platform labour to migrant and racialised subjects resonates with Benjamin’s argument that digital capitalism values and devalues diversity and Blackness at the same time (Benjamin 2023, 63). As elaborated by Kornberger et al. (2017), platforms are evaluative, in that they undergo processes of valuation in which new economic subjects are forged. The liberating effect attached to valuation, however, comes with the price of exploitation and dependence on the platform.
Platform workers feed the algorithms of the owner companies with a bulk of personal data in real-time, over which they have little to no control. However, the extraction and processing of their data is hidden “behind the screen”, leaving them with the perception of autonomy. The opacity and “black boxing” which digital technologies pursue, enable workers to experience de-racialization when interacting with the app. Platforms hide social processes behind algorithmic calculation, which is presented as neutral from societal dynamics and failures. At the same time, opacity can also become a condition favouring collective agency, when it is used to make oneself unseen. While algorithms are more and more capable of seeing through people’s behaviours, ideas and bodies, there is always an outside that they cannot capture, an outside that can be productively turned into a tool for a better livelihood.
To complete the picture, I now pass to illuminate another crucial aspect of how platforms relate to racialization, beyond reinforcing or undermining it, that is by organizing the circulation and mobility of racist scripts. While racism can both constitute an input and an output of technologies processing data, I now want to focus on the definition of platforms as infrastructure, and the capacity of platforms to make racism circulate and connect subjects which were previously disconnected. Platforms affect the experience of everyday racism of workers in that they hide the racist boss behind algorithms. The figure of the boss, however, is only one among the subjects who come to judge and value platform workers. Here, I want to show how the patterns of labour relations produced by platforms structurally differentiate the economic subjects which they connect with each other. Workers, who are most often migrants, are exposed to multiple forms of oppression.
Lean platforms are usually part of the service economy, in which relations of labour are not reduced to the employer-employee one (Animento et al. forthcoming), but rather encompass the relationship between clients and workers. Each platform in each city has its specific structure of labour relations. For instance, Deliveroo links riders with restaurant workers and customers. Helpling connects cleaners to customers, with no further intermediation. In turn, due to increasing professionalization, Airbnb has come to connect customers to agencies which in turn employ workers to manage the apartments. Like Airbnb, Uber is a good example to show how different structures of labour relations have differentiated over time. It started with the model typical of the platform economy based on freelance entrepreneurs but is now often operated with a model based on subcontracting companies, drivers, and customers, at least in Europe (Niebler et al. 2023a).
Platforms can be defined as infrastructures (Plantin and Punathambekar 2019), which connect economic subjects, functioning as “match-makers” (Graber/König 2020; Evans and Schmalensee 2016). More specifically, they have been defined as “evaluative infrastructures” (Korneberger et al. 2017) since a key logic of their functioning and success is valuation. Korneberger et al. provide an analysis of platforms such as eBay, to show how the economic interactions they facilitate rest on a system of reciprocal rating and evaluations. Economic subjects who are active on platforms develop over time a sensitivity for navigating trust and willingness to risk. Crucially, valuation systems such as rating and ranking are not necessarily symmetric and equal and reinforce discrimination by race and gender (Schor and Attwood-Charles 2017). In our research, we identified several asymmetries concerning the role and function of rating, according to which subjects are addressed by it. At Helpling, for instance, the ratings of cleaners by customers are key to the cleaners’’ chances of getting more commissions. On the other side, cleaners can rate their customers, but such rating is not public. For this reason, Helpling cleaners in Berlin have organized to compile “blacklists” of “bad” clients, to protect themselves and avoid both scams and bad treatments at work (see Niebler et al. 2023b).
What do rating and ranking systems in platforms have to do with racism? How does the infrastructural quality of platforms affect race relations? Again, Assan’s words help to disentangle these questions:
“But I don’t know who’s sitting behind me, right? …and they are all wearing a mask [during Covid pandemic]. One driver could be in a bad mood, another one too slimy, or he will annoy you, he is too dreary, or too disrespectful. Everyone has a weakness. And in this short time, in this short moment that you have, you cannot really judge people, so they always get five stars from me, I don’t give a shit who they are. They can tell me a thousand times: you fucking foreigner. When they’ll get out of the car, I’ll still give them five stars. Because I know it won’t work on them, it won’t change anything. It’s just about a short while, he’s pissed off, I’ve caught him unfortunately with it, I have to go through with it.”
[Uber_Ber_8]
Everyday racism, according to Philomena Essed, concerns “injustices recurring so often, that they are almost taken for granted, nagging, annoying, debilitating, seemingly small, injustices one comes to expect” (Essed 2002, 203). Assan here seems to refer to this type of racist encounter, which he comes to expect. Even if he could rate customers who attack him with racist behaviour, he would not, because he has no hope that that rating neither will affect the customer’s ability to continue his racist behaviour, nor of course will change the racist matrix he lives in. Rating systems in apps almost or never support workers to defend themselves from this type of everyday offence, which is also very difficult to document.
On the other side, Assan reported that he had received a bad rate from a client whom he identified as a Mexican businessman. During the ride, the driver, who came to Germany as a refugee from Afghanistan, admitted that he had made a negative comment about Arabic people and about refugees. The client gave him a bad rate and even filed a complaint at the platform hotline. In another interview, a white German female driver admitted that she had made a racist comment about foreigners during a ride, and she had received a one-star rating, which had lowered her ranking in the app.
Racism can circulate through the infrastructure of platforms, but valuation systems affect this circulation by sanctioning racist behaviour and speech in unequal ways. Since platforms aim at maximizing profits and acquiring a large pool of consumers in the shortest period, they structurally tend to protect consumers and expose workers. Locating their operations at the nexus between structural and interpersonal racism, platforms make racism infrastructural, in that they organize its circulation. This implies that platforms have the power to moderate racism or to perform a public service, as Benjamin argues (2023). Since their goal is profit-making and not equality or justice, their moderation of racist behaviours and scripts ends up reinforcing unequal race relations.
A final but crucial point is yet to be made. As the examples mentioned in this chapter and the previous one show, racist scripts circulate across the platforms in such a way as to explode any pretence of coherence. Taxi drivers with migrant experiences in their lives or the lives of their families attribute to Uber drivers’ qualities which were used not so long time ago to discredit their ancestors. Assan, an Uber driver who came as a refugee from a Middle East country comes to be badly rated because of a racist comment about Arabic people. A white female Uber driver, on her turn, suggests that taxi drivers are not kind and respectful with customers, later stating that they are all foreigners. Racist scripts expand into a myriad of contradictory statements about ethnicities, races, classes, and groups. Here, I welcome Benjamin’s’ invitation to consider race itself as “a kind of technology, one designed to separate, stratify, and sanctify the many forms of injustice experiences by members of racialized groups, but one that people routinely reimagine and redeploy to their own ends” (Benjamin 2023, 84). In platforms, racism seems to be used by the subjects who are put in connection, to discredit others or evaluate themselves. It is used also to navigate interactions and transactions through the platform, to break down and explain structural antagonism and competition produced by the platform into everyday mundane pictures. Fluid encounters, such as those between taxi drivers and Uber drivers on the streets, trigger processes of boundary-making around the “illiterate”, “primitive” and “simple” Uber drivers. These processes are reinforced by the platform economy, either through rating systems or social media platforms, in which taxi drivers download their anger against surging competition and lowering income. At the same time, the platform can present itself as outside of race relations, while it is its role as mediator and infrastructure which allows the circulation of racism in the first place.
The previous chapters provided an investigation of the multiple and contradictory ways in which platforms interrelate with processes of racialisation. I began by deploying the Polanyian concept of “marketisation” to address the process through which platforms enter new markets. Our research on platform labour on four platforms in seven European cities shows that platformisation often entails a “more than double movement”. When they enter new urban markets, they try firstly to acquire as many workers and consumers as possible. Here, the degree of diversity among the labour force in terms of migrant backgrounds and residency status is very high. In a second moment, however, labour conditions at platforms worsen, and the most precarious and racialised workers might remain trapped in their jobs, while those who can find a comparatively better job leave the platform. Looking at this movement from the perspective of racial capitalism, I argued that racialisation can be both an input and an output of platformisation. Platformisation in its turn can both separate and connect groups of labouring bodies, reinforcing and blurring societal partitions of migrant labour at the same time, along with Melamed’s claim that racial capitalism “separates forms of humanity so that they may be connected in terms that feed capital” (Melamed 2016, 79). Along with a bulk of literature on the topic (Benjamin 2023; McMillan Cottom 2020; Hamilton 2020; Noble 2018), our research thus confirms that digital technologies reinforce racism. However, it also aligns with literature showing how the algorithm hides the relations of exploitation and racial domination by separating both physically and virtually the worker from their boss (Purcell and Brook 2022). The analysis suggests that platforms organize the mobility of racism along their infrastructures, in fact making racism “infrastructural”. Racist scripts circulate along the interactions and connections organised by platforms both online (through rating and ranking systems) and offline (through the fluid encounters generated by the operation of matchmaking carried out by platforms). The “infrastructuring” of racism seems to push further the disparateness of racist scripts across economies which are largely fed by migrants. Attributes of inferiority and superiority are increasingly detached from ethnic and national grouping in a world made increasingly by migrants.
The article leaves the question open about the kind of subjects which can arise in such a context, in which layers of migration regimes accumulate and produce singular intersections with relations of class and gender. How do digital technologies interact with these new re-configurations of racism? On one side, they make Blackness visible, in that they value migrant labour but also track, measure, and convert it into data, on the other, the White centricity of digital technologies cannot grasp or comprehend “generic Blackness” (Simone 2016; Benjamin 2023). What kind of practices and spaces of autonomy can arise among those who are “generalised” as inferior? While these practices are often informal, not narrated and not intelligible for algorithms, how can we – and should we? – investigate them? Further research might look into the multiple forms of simultaneous exposure to migration regimes, algorithmic management and labour exploitation, to understand which racial formations will emerge in the next future. To do this, the analysis suggests pushing the boundaries of analysis beyond the boundaries of platforms and to rather explore the connections between the insides and the outsides of platforms, to illuminate the making and unmaking of racism through platformisation.
Andersson Schwarz, Jonas. 2017. Platform Logic: An Interdisciplinary Approach to the Platform-Based Economy. Policy & Internet 9 (4): 374-394. https://doi.org/10.1002/poi3.159
Anwar, Mohammad 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. https://doi.org/10.1177/1024529420914473
Barot, Rohit and John Bird. 2001. Racialization: The Genealogy and Critique of a
Concept. Ethnic and Racial Studies 24 (4): 601-618. https://doi.org/10.1080/01419870120049806
Benjamin, Ruha. 2023. Race after Technology. New York: Routledge.
Berndt, Christian, Norma M. Rantisi and Jamie Peck. 2020. M/market frontiers. Environment and Planning A: Economy and Space 52 (1): 14-26. https://doi.org/10.1177/0308518X19891833
Bhattacharyya, Gaby. 2018. Rethinking Racial Capitalism: Questions of Reproduction and Survival. Lanham: Rowman & Littlefield.
Conroy, William. 2022. Race, Capitalism, and the Necessity/Contingency Debate. Theory, Culture & Society 41 (1):39-58. https://doi.org/10.1177/02632764221140780
Duneier, Mitchell. 2004. Three Rules I go by in my Ethnographic Research on Race and Racism. In Researching Race and Racism, edited by Martin Bulmer and John Solomos, 92-103. London: Routledge.
Essed, Philomena. 2002. Everyday Racism. In A Companion to Racial and Ethnic Studies, edited by David Theo Goldberg and John Solomos, 202-216. Hoboken: Blackwell.
Evans, David S. and Richard Schmalensee. 2016. Matchmakers: The New Economics of Multisided Platforms. Cambridge: Harvard Business Review Press.
Fraser, Nancy. 2022. Cannibal Capitalism: How our System is Devouring Democracy, Care, and the Planet and What We Can Do About It. London: Verso.
Fuchs, Christian. 2018. Capitalism, Patriarchy, Slavery, and Racism in the Age of Digital Capitalism and Digital Labour. Critical Sociology 44 (4-5): 677-702. https://doi.org/10.1177/0896920517691108
Fuchs, Christian. 2019. Karl Marx in the age of big data capitalism. In Digital Objects, Digital Subjects: Interdisciplinary Perspectives on Capitalism, Labour and Politics in the Age of Big Data, edited by David Chandler and Christian Fuchs, 53-71. London: University of Westminster Press.
Fuchs, Christian. 2021. Digital Capitalism: Media, Communication and Society: Volume Three. London: Routledge.
Gebrial, Delia. 2022. Racial Platform Capitalism: Empire, Migration and the Making of Uber in London. Environment and Planning A: Economy and Space, Online First. 1-25. https://doi.org/10.1177/0308518X221115439
Grabher, Gernot and Jonas König. 2020. Disruption, embedded. A Polanyian Framing of the Platform Economy. Sociologica 14 (1): 95-118. https://doi.org/10.6092/issn.1971-8853/10443
Hamilton, Amber M. 2020. A Genealogy of Critical Race and Digital Studies: Past, Present, and Future. Sociology of Race and Ethnicity 6 (3): 292-301. https://doi.org/10.1177/2332649220922577
Hochman, Adam. 2019. Racialization: A Defense of the Concept. Ethnic and
Racial Studies 42 (8): 1245-1262. https://doi.org/10.1080/01419870.2018.1527937
Hua, Julietta and Kasturi Ray. 2022. Beyond the Precariat: Race, Gender, and Labor in the Taxi and Uber Economy. Social Identities 24 (2): 271-289. https://doi.org/10.1080/13504630.2017.1321721
Kornberger, Martin, Dane Pflueger and Jan Mouritsen. 2017. Evaluative Infrastructures. Accounting for Platform Organization. Accounting, Organizations and Society 60: 79-95. http://dx.doi.org/10.1016/j.aos.2017.05.002
McMillan Cottom, Tressie. 2020. Where Platform Capitalism and Racial Capitalism meet: The Sociology of Race and Racism in the Digital Society. Sociology of Race and Ethnicity 6 (4): 441-449. https://doi.org/10.1177/2332649220949
Melamed, Jodi. 2015. Racial capitalism. Critical ethnic studies 1 (1): 76-85.
Mezzadra, Sandro. 2010. The Gaze of Autonomy: Capitalism, Migration and Social Struggles. In The contested politics of mobility, edited by Vicky Squire, 141-162. London: Routledge.
Niebler, Valentin, Giorgio Pirina, Michelangelo Secchi and Franco Tomassoni. 2023a. Towards ‘bogus employment?’ The contradictory outcomes of ride-hailing regulation in Berlin, Lisbon and Paris. Cambridge Journal of Regions, Economy and Society: 1-13. https://dx.doi.org/10.1093/cjres/rsad007
Niebler, Valentin, Moritz Altenried and Stefania Animento. 2023b. Der Fragmentierung begegnen. Zur Organisierung migrantischer Reinigungskräfte in der Plattformökonomie. WSI Mitteilungen 04/2023: 252-260.
Noble, Safiya Umoya. 2018. Algorithms of oppression. How Search Engines Reinforce Racism. New York: New York University Press.
Omi, Michael and Howard Winant. 1986. Racial Formation in the United States. New York: Routledge.
Peck, Jamie and Nik Theodore. 2007. Variegated capitalism. Progress in human geography. 31 (6): 731-772.
Peterlongo, Gianmarco. 2023. Imbrigliamento e rifeudalizzazione del lavoro nella gig economy. Una ricerca sul caporalato digitale tra Italia e Argentina. Cartografie sociali. Rivista semestrale di sociologia e scienze umane (15).
Pfeiffer, Sabine. 2021. Digitalisierung als Distributivkraft: Über das Neue am Digitalen Kapitalismus. Bielefeld: transcript.
Plantin, Jean-Christophe and Aswin Punathambekar. 2019. Digital Media Infrastructures: Pipes, Platforms, and Politics. Media, Culture & Society 41 (2) 163-174. https://doi.org/10.1177/0163443718818376
Polanyi, Karl. 1957. The Great Transformation. Boston: Beacon.
Purcell, Christina and Paul Brook. 2022. At least I’m my own boss! Explaining consent, coercion and resistance in platform work. Work, Employment and Society 36 (3): 391-406. https://doi.org/10.1177/0950017020952661
Sayad, Abdulmalek. 2016. La double absence. Des illusions de l‘émigré aux souffrances de l‘immigré. Média Diffusion.
Schiller, Dan. 1999. Digital capitalism: Networking the global market system. Chicago: MIT Press.
Schor, Juliet B. and William Attwood-Charles. 2017. The “Sharing” Economy: Labor, Inequality, and Social Connection on For-Profit Platforms. Sociology Compass 11 (8): 1-16. https://doi.org/10.1111/soc4.12493
Simone, AbdouMaliq. 2016. Urbanity and Generic Blackness. Theory, Culture & Society. 33 (7-8): 183-203. https://doi.org/10.1177/0263276416636203
Srnicek, Nick. 2017. Platform Capitalism. Cambridge: Polity.
Vallas, Steven and Juliet B. Schor. 2020. What Do Platforms Do? Understanding the Gig Economy. Annual review of Sociology 46: 273-294. https://doi.org/10.1146/annurev-soc-121919-054857
Van Doorn, Niels. 2020. Stepping stone or dead End? The ambiguities of platform-mediated domestic work under conditions of austerity. Comparative landscapes of austerity and the Gig economy: New York and Berlin. In Working in the Context of Austerity, 49-70. Bristol: Bristol University Press.
Zhou, Yang. 2022. Trapped in the Platform: Migration and Precarity in China’s Platform-Based Gig Economy. Environment and Planning A: Economy and Space, Online First. https://doi.org/10.1177/0308518X221119196
Zuboff, Shoshana. 2019. The Age of Surveillance Capitalism. The Fight for a Human Future at the New Frontier of Power. New York: Profile.
Stefania Animento
Stefania Animento is a post-doctoral researcher at the Humboldt University of Berlin. Her work is about migration, racism, labour, and digitalisation. She wrote her PhD thesis on the access to labour and housing of post-2008 European migrants in Berlin. Further, she worked on the project “Platform Labour in Urban Spaces” (https://project-plus.eu/), which investigated the impact of platformisation on labour across seven European cities.
[1] After a knee surgery that impaired my mobility for one and half months, I regularly used the app to go to my routine medical visits and physiotherapy. During this period, I came to know Uber as a company that provides the cheapest way to navigate the city for people whose mobility is strongly limited. I want to thank all the drivers who never missed an occasion to make me feel welcome in their car and always had kind words for me.
[2] The interview was carried out by my colleague Valentin Niebler. I want to kindly thank him for sharing the precious interview (of course in accordance with the informed consent obtained).