When discussing politics and economics on a global scale, it is important to center the debate not only around actors such as states and international organizations, but also corporations that have come to accumulate an immense amount of political, economic, social and technological capital and power. One such corporation that continues to make headlines is Amazon. While Amazon has experienced exponential growth over the course of the last decade, it is particularly during the pandemic that it skyrocketed. As governments announced lockdowns and stores were closed, there was a further shift to the digital sphere. The growing mountain of orders necessitated the employment of more warehouse and delivery workers. In the process, the excitement around Amazon stocks and investments grew larger. Although the future trajectory of such a corporation remains uncertain, given the recent announcement of layoffs, it is crucial to take companies like Amazon seriously due to their expanding ecosystem and influential power dynamics.
I focus on Amazon, a corporation with multiple platforms, in Work and Alienation in the Platform Economy: Amazon and the Power of Organization to illuminate the realities of the individuals who drive this company. These workers are instrumental to Amazon’s success, making it imperative to include them in any discussion on Amazon’s transnational power.
While Amazon has become synonymous with online shopping, considering its e-commerce platform, its ecosystem is in fact much larger and also includes acquisitions such as Whole Foods. Amazon also created other platforms like Amazon Web Services (AWS), its cloud service platform, and Amazon Mechanical Turk (MTurk), its digital labor platform. While these two platforms were initially created for Amazon’s internal use, they have since been made available to the public. AWS essentially mediates the infrastructure of the Internet via the cloud, while MTurk operates the market for digitally outsourced labor. The ways in which these platforms organize their workers, and in turn, how these workers organize themselves can strongly differ. This is worth examining as it has implications for our understanding of political-economic, technological and social power of platforms, as well as for devising policies and regulations to support the workers behind them.
In my work, I contrast two types of Amazon workers: its warehouse workers and its MTurk workers. These workers are organized in fundamentally different ways. The Amazon warehouse workers are organized in more traditional ways within a physical space (here the warehouse) and receive a traditional wage (a rather steady hourly wage). MTurk, on the other hand, is less conventional, bringing the market for outsourced labor to the online world. It can be imagined as a market for microtasks termed ‘Human Intelligence Tasks’ (HITs). These can be surveys or tasks revolving around data such as labeling (like identifying an instrument within a line up of images). Such human labor is crucial for the development and training of machine learning algorithms for AI. MTurk workers can be located anywhere in the world, and labor remotely from behind their screens. Unlike the warehouse workers, they are paid by task – much like an Uber driver is paid by gig based on their drives and is part of the gig economy.
Both of these Amazon platforms are global in scope, albeit in distinct ways. Amazon opens warehouses across the world and has been growing in existing regions while expanding into new ones in South America, the Middle East, and North Africa. Yet, Amazon organizes the working day and division of labor in relatively similar ways. Warehouse workers’ productivity is monitored using a Units Per Hour (UPH) rate. Workers follow a clear division of labor and tasks, including prepping items, stowing them on the shelves, picking them off shelves, and packing them. This organization of work is coupled with a digital panopticon, as workers face both the gaze of their supervisors, as well as the technology which monitors UPH rates. Having to labor as efficiently and productively as possible, workers express that they feel alienated and like cogs in the machine or robots. This sentiment is reinforced by the slogan “work hard. have fun. make history.” prominently displayed on warehouse walls, which portrays a collective work culture, overshadowing the individual dimension of the UPH on whose basis workers are then evaluated. These conditions have the potential to fragment the workforce and create a sense of estrangement from both the work itself and fellow laborers. The global dimensions of the corporation result in turn in varying productivity rates, degrees of surveillance, wages, working hours, and (precarious) employment conditions depending on the local and national context.
These global dimensions are part and parcel of the MTurk platform where workers and those posting tasks are located in different time zones and spaces – though predominantly in the US and India. In this sense, workers are tied to their local contexts while simultaneously participating in the global labor market through remote work behind their screens. Workers are automatically anonymized with an alphanumerical ID on MTurk, but requesters too can choose to anonymize themselves. Workers, paid as independent contractors, are further alienated by not necessarily being able to discern who they are working for and for what their data is being used from the HITs. In contrast to the Amazon warehouse workers, who encounter each other at the workplace, MTurk workers do not interact with other workers through the interface. As labor is exclusively mediated digitally through the platform, social surveillance and management are replaced by algorithms and algorithmic management. The global context factors into this in different ways: from requesters filtering workers based on their location – as requesters can wish to have mainly MTurk workers from the US for specific tasks, or India for others.
Additionally, the value of the gig wages differs, as these are posted in US$ and depending where you are in the world, this amounts to a different wage based on different exchange rates. Another point would be the crucial role of different time zones: if requesters are predominantly located in the Global North, and workers generally in the Global South, then it means workers have to structure their working day around the locations of the requesters. This reflects more generally a specific international division of labor which carries implications for the chain of data production and usage. This is embedded in the power relations and dynamics between the Global North and the Global South, where the former essentially extracts data and labor from the latter.
These working realities of different platforms and the geographical dimensions thereof also have implications for the ways in which workers collectively organize to improve their working conditions. The local and national context, along with the industrial relations and rights connected to them, open up certain forms of resistance while closing off others. Given the more traditional organization of warehouse workers, they pursue unionization and industrial action where legally permissible. The landscape varies significantly across different countries: Amazon warehouse workers in Germany have been striking regularly since 2013 in pursuit of a collective bargaining agreement, while those in the US and the UK have recently been fighting for unionization. However, both Amazon’s anti-union efforts, and the conditions of their employment (i.e. depending on whether they have a fixed or permanent contract) may still undermine their labor organization. Given the transnational nature of Amazon, workers underline the importance of not only organizing on a local and national level but also transnationally and internationally.
In addition to local and national efforts, we have also seen transnational networks like the Amazon Workers International and UNI Amazon Global Union Alliance which aim to build solidarity between workers and unions, foster communication and coordinate campaigns especially on Black Friday. One campaign that has been growing over the course of the last years has been Make Amazon Pay, which brings together more than 80 organizations from unions to civil society organizations – including those two transnational networks. Targeting Amazon from an array of different angles, it works “towards labour, tax, climate, data and racial justice.” Global days of action are crucial in order to frame and further push for these goals, which manifested for instance in the last Black Friday of 2022, where three dozen countries across the globe showed solidarity and action in various forms. What such action looks like differs according to local and national contexts from solidarity actions to walkouts and strikes.
The global nature of the MTurk platform, which is essentially an online gig global labor market, makes it difficult for workers to organize more traditionally. Unionization is complicated by workers not being classified as workers in the first place and workers being distributed across the globe. Going on strike is also extremely difficult to imagine and carry out. If a worker somewhere in the world chooses to refuse to work tasks as an act of resistance and strike, workers elsewhere will continue to carry out those tasks – often without even being aware of the striking worker. Accordingly, the ways in which MTurk workers can organize fundamentally differ from those of warehouse workers. Given that the interface does not bring workers together and that these workers are always employed precariously, they organize in more alternative ways. They may exchange tips, ask questions and provide answers to one another as to how to best navigate MTurk and find higher-paying tasks. We can perceive these as online communities, such as those found on platforms like Reddit, including the subreddit Turker Nation or Turkopticon.
Turkopticon, initially created by researchers Lilly Irani and Six Silberman, has been centered on workers’ demands and concerns, and has now become a “worker-led non-profit organization!” While MTurk allows requesters to rate workers by approving or rejecting their tasks, this de facto determines workers’ “approval rating.” This approval rating is instrumental, as it gives workers access to future tasks that may filter for a certain percentage. Accordingly, on Turkopticon workers evaluate requesters according to different criteria like communicativity, fairness, and promptness – thereby signaling to one another who to look out for or avoid because they are likely to reject tasks. Turkopticon flips thereby the power relation on its head, as workers become the ones rating requesters. As MTurk’s labor market is global in nature, this form of solidarity takes on a transnational and global dimension through which workers instrumentalize the potentials of the Internet.
Considering how transnational these platforms are and how international their workers, it is important to recognize platforms as manifestations of economic, political, technological, and social power. Given the time we live in, we are likely to be users of these platforms in one way or the other, and must therefore acknowledge those laboring behind them. As workers fight their battles in improving working conditions which are worth examining, their efforts must be accompanied by regulation from above. This is integral to creating a fairer working world and more democratic society, especially if we take the concentration of power within the hands of these transnational corporations into consideration. Regulation can be imposed both in relation to the capital, i.e. increasing taxation and limiting subsidies, but also in relation to workers. The latter can be achieved by eliminating subcontracting and agency work, reclassifying workers, labor inspections where possible generally establishing fairer working conditions and safer environments for organizing. Therefore, when we engage in discussions concerning the influential actors shaping our present world and society on an international scale, it is crucial to broaden the discourse to encompass transnational corporations and the platform economy.