‌2

‌Enabling and resisting the platform economy from below: Platform immigrant workers in Ecuador

‌Henry Chávez and María Belén Albornoz

‘It’s like being in a toxic relationship: sometimes “she” treats you well; sometimes “she” punishes you and gives you nothing. You never know when, how or why’, Elmer, a middle-aged man from Venezuela, explains as he points to his phone with the delivery app [‘she’] for which he works.

He and three other colleagues have agreed to talk to us about their working conditions. They have been waiting fruitlessly for hours for an order to deliver, at the car park of a well-known fast-food restaurant in a high-class neighbourhood of Quito. It is almost noon, and the sun is beating down at 2,800 metres above sea level. All four are part of the half million immigrants who have arrived in Ecuador in the last six years, the largest migratory flow in the country’s history. Only Elmer has formal immigration status – for the others, this is the only job they could get without a work visa.

Over the last decade, online platforms for gig work have become a new vector for the technological globalisation process. These platforms have transformed work around the world by removing several barriers, facilitating workers’ access to the labour market, and providing a quick source of income. However, they have also brought harmful social and environmental externalities by undermining labour rights, reducing occupational health, increasing the use of disposable products, and creating new forms of exploitation, especially for vulnerable populations such as immigrants. This is the case in Ecuador, where the main delivery and ride-hailing platforms began operating around 2016 against a backdrop of economic recession resulting from falling oil prices and exports (the country’s main source of income) and a significant influx of migrants. Many of these immigrants have difficulty finding formal employment, not only because of adverse economic conditions but also because of the constraints caused by international inequalities, such as the limitations on the international circulation of people embodied in administrative barriers to obtaining work visas, a lack of local contacts, and discrimination. Thus, the platform economy has become an alternative for some of these workers, offering them flexibility and income opportunities but at the cost of poor working conditions and new forms of exploitation.

Elmer’s words somehow sum up a widespread sense of puzzlement among platform workers regarding the way in which the platforms allocate their orders, rates, compensations, and penalties. This idea of a ‘toxic relationship’ between the worker and the platform echoes the metaphor of the pharmakon (Stiegler 2007), an ancient Greek word used to designate at once the remedy, the poison, and the scapegoat of a disease. In the correct dose, this substance (or technology) could cure a person, but in excess, it becomes toxic and could kill them. Platforms seem to produce a similar effect. They help workers – many in critical situations (unemployed, without an income, immigrant) – to obtain an immediate and flexible source of income and somehow alleviate their condition. At the same time, these platforms put workers in a situation of dependence and vulnerability that forces them to accept poor working conditions and exposes them to new forms of disciplining and automated exploitation that are incomprehensible to the vast majority. The black box they hold in their hands thus appears to them as a capricious being that manages their time and rewards them as it pleases. They have no choice but to obey or cheat. In the words of José – another platform worker – ‘here you don’t have a boss; your boss is the platform’.

Platform companies rely heavily on algorithms to manage worker recruitment, task assignment, pricing, compensation, management, and evaluation. Although algorithms promise to improve efficiency and reduce costs, they also produce new forms of inequality, discrimination, exploitation, and injustice, as they can perpetuate bias and exclusion, oversimplify tasks and situations, or simply fail. A growing body of literature on platform economics and algorithmic management has already addressed some of these issues. This chapter attempts to build on these two approaches but focuses on a subaltern and vulnerable population: migrant platform workers in the Global South. In order to do that, we conducted about 115 semi-structured and 10 in-depth interviews, as well as ethnographic observations with platform workers in Ecuador between 2021 and 2023. Specifically, we sought to answer this question: How do workers learn to use and navigate these platforms, and how do they develop innovative tactics to circumvent their limitations or take advantage of their shortcomings?

The results of this research shed light on the specificities of the deployment of the platform economy and its algorithmic management systems in the Global South and how this process unveils some of the features of technological globalisation, as seen below. Specifically, they show how the interactions between workers and platforms through informal non-codified spaces and strategies allow them to resist and circumvent algorithmic management systems while enabling and expanding platforms’ functioning (Gago 2017).

This chapter is organised as follows: The next section briefly overviews the platform studies and algorithmic management literature. We then present the Ecuadorian context and our methodology. Then, we explain the role of the black markets for platform profiles and vehicles, as well as workers’ strategies of resistance to algorithmic management. This section is followed by a general discussion of subaltern human agency. Finally, we conclude by showing how vulnerable and subaltern populations, such as migrant workers, simultaneously enable and resist the globalisation of technological platforms from below.

Platform studies and algorithmic management from below

The concept of the ‘platform’ has emerged in the last two decades following the irruption and success of companies such as Google, Facebook, Amazon, Uber, Tinder, and many others (Courtois and Timmermans 2018; van Dijck 2013). While the term was initially associated with video games (Montfort and Bogost 2009), it later spread to other digital media and infrastructure (Plantin et al. 2018). Today, it encompasses social media platforms (Langlois and Elmer 2013), streaming services (Lobato 2019), and online marketplaces (Ipeirotis 2010). Several case studies have shed light on the transformations brought about by these new complex sociotechnical objects, particularly Facebook (Bucher 2012; Ellison et al. 2007; Tufekci 2015), YouTube (Abidin 2018; Baertl 2018; Burgess and Green 2009), and Amazon (Oestreicher-Singer and Sundararajan 2012). More recently, other studies grouped under the subfield of the ‘gig’ or ‘platform’ economy have focused on ride-hailing platforms (de Freitas et al. 2023; Rosenblat and Stark 2016), delivery platforms (Galiere 2020; Lord et al. 2023; Timko and van Melik 2021), and crowd-work platforms (Casilli and Bouquin 2020; Irani 2015; Le Ludec et al. 2020; Tubaro et al. 2020). Several scholars from various disciplines have highlighted the potential benefits of platform work (flexibility, autonomy for workers, and reduced transaction costs for consumers), but also the risks (employment precariousness, lack of benefits and social protection, harmful externalities and informality) (Albrieu 2021; Woodcock and Graham 2019). Others have focused on the experiences from below of subaltern, marginalised, and vulnerable workers, such as immigrants. The latter face the constraints imposed by international inequalities, language barriers, a lack of social networks, and discrimination (van Doorn and Vijay 2021) but also promote collective action and the organisation of workers (Albornoz and Chavez 2020; Woodcock and Cant 2022) or alternative value-creation strategies, such as platform cooperatives, geared towards creating worker-owned and democratic platforms that prioritise worker wellbeing and community building (Grohmann 2021).

Scholars in other branches have investigated the algorithmic management systems that power those platforms (Firmino et al. 2019; Rosenblat and Stark 2016; Vallejos Rivero 2021). These systems are used to match workers with tasks, set prices, and evaluate performance. So-called algorithmic work (Jarrahi et al. 2021; Schildt 2017), designed to enable collaboration between algorithms and humans, can be observed within the framework of ride-sharing services, delivery services, and the execution of micro-tasks. However, algorithms can also create new forms of control and surveillance (Wood et al. 2019). Algorithmic management is often opaque, partly due to companies’ secrecy regarding their systems but also because of the large amounts of data at stake, the complexity of the calculations, and the interactions between different algorithms involved in decision-making that render them very difficult for humans to fully understand (Faraj et al. 2018; Meijerink and Bondarouk 2023; Parent-Rocheleau and Parker 2022). Furthermore, workers may experience difficulty, confusion, and uncertainty in understanding whether they should comply with the algorithm’s instructions and the potential consequences of non-compliance (Zarsky 2016). They may also perceive the algorithm as unfair, inaccurate, or restrictive, yet they may hesitate to override it (Markus 2017). These conditions can lead to resistance to algorithmic work (Kellogg et al. 2020).

Building on this literature, we unpack the experiences of immigrant workers in the platform economy in Ecuador. By examining how algorithmic management affects their labour experiences and the strategies they develop, we seek to understand how this subaltern and marginalised population enables such platforms to adjust, adapt, and operate while simultaneously, from below, circumventing and resisting exploitation and the process of disciplining. This sheds light on how non-hegemonic actors in the Global South use tactics unforeseen by the platforms’ designers in the Global North to attempt to redefine the asymmetrical power relationship between platforms and workers, but, at the same time, they contribute to the expansion of the very same platform economy.

Recession, migration, and the platform economy in Ecuador

Ecuador is one of the smallest economies in South America and is structurally dependent on exports of primary goods, particularly oil. The lack of development of the non-extractive economy has resulted in high rates of poverty, inequality, unemployment, and informal employment. Since 2015, the country has faced an economic recession triggered by falling oil prices, rising public debt, and trade deficits (Burchardt et al. 2016). By the end of 2022, its GDP had barely returned to its 2015 level (Banco Central del Ecuador 2023). This recession has led to the growth of unemployment and the informal economy. In 2014, only 49% of the economically active population (about 8 million people) had an adequate job.1 The COVID-19 pandemic and generalised lockdowns exacerbated this situation. By June 2020, about 84% of the population was unemployed, underemployed, or working in informal activities. The situation has improved since, but not enough to return to pre-pandemic levels: by February 2023, about 67% of the population was still unemployed, underemployed, or working in informal activities (INEC 2023).

Over the past ten years, Ecuador has experienced a significant influx of migrants, with more than 630,000 people arriving there. The majority of these migrants originate from Venezuela (58%), followed by Colombia (22%), Peru (8%), and Cuba (7%). However, the outbreak of the COVID-19 pandemic in 2020 marked a turning point, with approximately 100,000 migrants leaving the country up to 2021 (INEC 2022).

In this challenging context, platforms like Cabify, Uber, UberEats, Glovo, and Rappi started operating in Ecuador, providing a lifeline for those who had lost their jobs, or immigrants seeking work, by offering a relatively simple and quick way to earn a regular income. Despite the lack of infrastructure, limited connectivity, and relatively weak demand for online services (DataReportal 2023), these companies have decided to expand their operations to countries such as Ecuador. The absence of regulation and the growing pool of workers accustomed to working in low-paying jobs without legal benefits provide an opening for these companies to earn greater profits.

No official records are available to estimate the number of individuals working on platform economy apps or the proportion of migrants participating in such activities. However, based on various reports, public declarations, and interviews, we estimate that platform workers in Ecuador account for about 1% of the underemployed population, that is, around 40,000 workers, about 35% of whom are immigrants (Albornoz et al. 2022; CITEC 2022; Maya et al. 2022; OIT 2022).

Between 2021 and 2023, we carried out ethnographic observations and 115 semi-structured interviews, including 10 in-depth interviews with workers from seven platforms: Rappi, Glovo, PedidosYa (formerly Glovo), Uber Eats, Cabify, Uber, and Didi. Workers were contacted through the platforms or at known worker meeting points using a snowball strategy. We interviewed platform workers who could share their real-life experiences of using algorithmic systems. In this way we gained an in-depth understanding of their practices and the contextual circumstances of their work.

The ages of the interviewees ranged from 19 to 71, but more than two-thirds were less than 38 years old. Eight out of ten were men, four out of ten were immigrants, and more than half had a technical or university diploma. Cabify, Rappi, and Uber were the most used platforms (two out of ten), followed by Didi, PedidosYa, Uber Eats, and Glovo (one out of ten). For the 10 in-depth interviews, we chose the most willing respondents to contribute to the research: six immigrants and four locals.

The interviews aimed to uncover workers’ latent understanding of algorithms as tools and managerial structures by focusing on their experiences with the platforms. The interview protocol included questions about the workers’ overall experience with the platforms; how they use them; their understanding of the underlying algorithms; their perception of the job assignment, pricing, and evaluation systems; how they look for information about the platforms’ functioning; and how they deal with algorithmic control.

In the following sections, we present the outcomes of the platforms’ expansion into the Global South and how intermediaries and informal markets enable the emergence of alternative value-creation strategies, novel forms of exploitation, and other harmful social and environmental externalities. Our analysis emphasises the workers’ capacity to interact with and resignify, adjust, and adapt to the platforms’ technological frameworks and demands, leading to new forms of resistance and technological appropriation from below.

Enabling the platform economy from below: Informal markets for platform profiles and vehicles

Until 2021, Ecuador had about half a million immigrants (INEC 2022). A survey of about 6,800 immigrants in Quito suggests that only 20% of them have formal immigration status (a work visa or residence permit) (Célleri 2020). This explains why only 55% claim to be employed and less than half have a formal contract. Consequently, these vulnerable workers are frequently subjected to exploitation and labour rights violations: 50% of them report working more than the legal 40-hour work week, only 12% receive more than the minimum legal wage of US$400 per month, and 44% experience discrimination based on their nationality.

In these conditions, many migrants turn to platforms as a relatively easy and quick way to earn money. According to CITEC (2022), about 37% of platform workers are immigrants. Platforms differ from traditional employment providers by operating in a non-codified informal space. From the period of their emergence, since their organisation and business models were entirely novel and most of their operations virtual, they were completely unregulated. For example, platforms could accept foreign documents to create worker accounts, allowing migrants to work under their profiles. But as they grew, local governments started pushing for regulation. Although several of these regulations in Ecuador are still under discussion, the platforms have already restricted access to new workers who lack documentation (work visa, identity card, or driver’s license). This strategy was to smooth out the new workers’ still informal and even illegal presence in the market. Indeed, during the first few years of operations, the platforms offered their drivers legal advice to recover the vehicles and cover the costs of the fines. However, few drivers used this service because it was easier to bribe the police and continue working than to lose several days of work without the car.

As a result, the subcontracting or renting of platform profiles on the black market became a common practice among undocumented immigrant workers. Most of the interviewees in this study stated that they had used or were currently using rented accounts due to a lack of suitable documentation. This alternative strategy has allowed immigrants to circumvent administrative and algorithmic restrictions to access the job opportunities offered by the platforms, and for the latter to continue expanding their markets and extracting profits from the grey areas created by the lack of local regulation and uncodified processes. In other words, by redefining the asymmetrical relationship with platforms (from which they were initially excluded by design), through an alternative strategy unforeseen by the platforms’ designers in the Global North, workers allow the platforms to have a larger workforce willing to accept their conditions and thus expand their operations. However, this strategy has also resulted in harmful externalities: a new system of exploitation that shifts risk from the employers to the employees, leading to unfair and unstable working conditions. Furthermore, many of these workers lack access to their vehicles, causing them to resort to a black market of ‘investors’, a pool of car or motorbike owners interested in renting out their vehicles or hiring drivers. Workers communicate directly with them to negotiate the terms and conditions of their agreements and subcontracting practices. This means that workers have to pay for both the vehicle and another person’s platform profile.

According to our interviews, two types of arrangements exist: a fixed weekly fee or 50% of the earnings. The fixed weekly fee for a ride-hailing app ranges from US$75 to US$100. Our respondents reported working an average of 11 hours per day and earning approximately US$200 per week, much more than the legal 8-hour working day, which shows the degradation of occupational health. After paying for the car rental, they were left with only 50% of their income, which amounts to around US$400 per month, the minimum legal wage in Ecuador. The delivery workers experienced a similar situation. Those renting profiles are charged between US$15 and US$40 per week. If a delivery worker rents a profile and a motorbike, they may pay up to US$50 per week. Interviewees reported earning an average of US$150 per week or US$400 per month after expenses. Unlike with a regular job, this does not include any form of health or risk insurance, social security benefits, sick leave, vacation, or other job-specific expenses (helmet, smartphone, internet use, etc.)

The employment status of platform workers is a contentious issue, as they face numerous challenges, including a lack of labour rights, unstable income, a lack of benefits, and precarious and exploitative working conditions. Platforms claim that workers are self-employed entrepreneurs who manage their own time and have no employer. Informal subcontracting practices thus make it even harder for workers to secure minimum working conditions and social benefits. Those who rent their profiles act as intermediaries between undocumented immigrant workers and the platforms, enabling the latter to operate and expand. However, this exposes these workers to high risks and harmful externalities and shifts the cost of operating in a non-codified informal grey area, and of the platforms’ expansion, onto their shoulders.

Resistance from below to algorithmic management: Dealing with a toxic relationship

Elmer’s intuition about the ‘toxicity’ of the relationship with the platforms suggests a pharmacological problem (Stiegler 2007): a critical situation in which their survival depends on a technology that may be toxic to them. Here, we draw on our respondents’ accounts and understanding of their situation to shed light on how subaltern actors resist, circumvent, and hack this technology to reduce or master its toxicity.

The term ‘La toxica’, which some of our interviewees (90% of them men) used to describe the app on their phones, reflects the embedded algorithms that govern these platforms and, consequently, the daily lives of platform workers. These algorithms, opaque and closed (Rosenblat and Stark 2016), serve as the black boxes managing the platforms. They assign orders and determine times, places, routes, prices, and payments. They also allocate rewards and penalties to ensure the supply of services and increase efficiency and quality.

During the high-demand period, they ask you to deliver at least two or three orders … but if you do just one, they lower your score. I mean, it is not your fault but that of the restaurant […] They also punish you based on the customers. […] A bad score […] reduces 3 points your general score. (Interview, May 2021)

The implementation of these global platforms at the local level is intricate and frequently results in conflicts, prompting the apps to regularly update, adjust, and adapt to local conditions. This creates an unpredictable relationship between the platforms and the workers, who must strive to decipher and navigate the constantly evolving regulations and demands to maximise their earnings and avoid sanctions.

December was crazy. One day, you were ‘diamond’; later on, you were ‘red’. They take you and give you points for no reason. […] The app has 508 updates so far. […] algorithm works very badly […] It sends you a triple order […] How can you be in three places at the same time? (Interview, February 2021)

The result is a kind of liquid Taylorism (Altenried 2020; Bauman and Lyon 2012) in which every worker has a planning office and a control system in their pocket but no workplace and no boss – at least not a human one.

That’s the problem. They don’t care. They don’t consider the time you lose when they make changes or updates without even warning you. Not an email, nothing. […] They just tell you, ‘keep your score’ […]. Everything is through messages. You have to upload a picture of the evidence of what is happening, and this is not an immediate procedure. There is no person to help you. (Interview, March 2021)

International inequality adds a layer of complexity to the analysis of algorithmic management presented above. Algorithmic scoring systems are designed to assign rewards or penalties to registered users who meet specific criteria, such as being legally eligible to work in a given country. Nevertheless, undocumented immigrants have found a way to enter the system through a black market of profiles and accounts, challenging the very foundations and assumptions of the scoring system. This Pavlovian algorithm aims to allocate, optimise, and evaluate tasks and payments to incentivise individual users and improve overall performance. However, rewards and penalties may not be distributed equally among profile owners and subcontracted workers.

I’ve worked with three different accounts. […] The owner of the first one told me, ‘My friend, this is an excellent account. It will drop you six orders per day […] That’s why I ask you for US$40 per week.’ […] I was fooled like a teenager in love. Then, another friend offered me a ‘green’ account […], but this time, I said, ‘Look, if I don’t get any orders during the day, I’ll give you back the account, and we keep being friends’ […] So, then, I asked a colleague, ‘My friend, how does it work this thing? Can you show me?’ […] He explained that you have to ‘upgrade’ the account to ‘diamond’ in order to get more orders. […] I tried for three more days, wasting fuel, going here and there, and I got nothing. So, I gave back this account […] Then, I took another one. He told me, ‘I will rent you the account and motorbike for US$50.’ It had three orders and 54 points. It is a very low number, as I know now, but at the time, I thought it was not so bad. I mean, it is far from zero, isn’t it? [they laugh] (Interview, February 2021).

The profile black market used by undocumented immigrants has resulted in the incentives of the algorithmic management system being split across and shared by multiple individuals with vastly different backgrounds and behaviours. The rewards are shared between profile owners and migrant workers, who end up being subject to a disciplinary algorithmic punishment regime but for a lower reward than regular users. Thus, while the penalties and physical effort are borne solely by the worker, the profile owners reap only the rewards. Ultimately, undocumented workers bear the brunt of the problems associated with this system, as their daily lives are controlled and influenced by the Pavlovian mechanism.

You must try to do things correctly, so the algorithm helps you improve. […] The application has some things you have to respect. You have to do this and that and do things correctly, such as deliver on time, be polite, and talk to the customer. (Interview, March 2021)

However, workers can also adjust and adapt to algorithms by altering and observing their data collection processes to better utilise some parts of the system. By grasping the inner workings of certain platform features, workers can enhance the performance of certain platform features or repurpose them altogether.

There are things I didn’t know. Because I was new, I started to understand by searching the internet. I found that the application likes you to talk to the customer, even if the customer doesn’t answer. […] The algorithm keeps [your score] stable, but if you start by being rushed, if you don’t write to the customer, the score starts to get lower. (Interview, March 2021)

By looking at the experience of undocumented immigrants, we can see that the effects of algorithmic management can go beyond the platforms themselves. In a kind of Black Mirrorian twist, the machine has broken free of its black box to exercise disciplinary measures not only on those who are officially registered in its database but also on those who should not be part of it. Here again, we can see how alternative strategies of non-hegemonic actors in the Global South attempting to change an asymmetrical power relationship end up contributing to the expansion of the platform economy beyond its limits in an unforeseen way. These insights show us some features and deviations of the technological globalisation process that can only be seen from below, which a Global North–based perspective would have missed.

Subaltern human agency vs. platforms’ technological power

Despite the opacity of algorithmic management technologies, it is important to highlight that they still rely heavily on non-automated human work, as several scholars have noted (for example, Casilli et al. 2019). Moreover, the dynamics linked to migration shed light on the offline human labour that underpins the platform economy. By examining these dynamics at the frontiers of the global system, we can better understand how subaltern populations appropriate, reinterpret and modify the new technological regime imposed on them by the platforms, revealing not only the importance of human labour in the digital age but also the different paths that technological globalisation may take as it moves into the Global South.

Although getting started on platforms may seem simple, they are intricate systems with numerous automated services. Workers invest a significant amount of time and energy in comprehending the platform’s ratings, deadlines, policies, and procedures (Mohlmann and Zalmanson 2017). To make sense of the complex workings of platforms, they might turn to social channels like WhatsApp chat groups to share knowledge and experiences with others. These chat groups serve as a resource for workers to seek advice on dealing with demanding customers, understanding companies’ policies, and staying up to date with platform changes. Some workers also learn directly from more experienced colleagues.

Sensemaking can benefit workers by providing them with a better understanding of algorithms and algorithmic management, which increases their confidence and sense of control. However, algorithms may also limit workers’ ability to perform their tasks. In such cases, platform workers may look for ways to avoid algorithmic processes or substitute them with outside tools. As they discover glitches in the algorithms, they may also learn to use these glitches to their advantage.

App-based platforms in peripheral contexts such as Ecuador rely on social practices and interactions beyond their algorithms. For example, the prevalence of immigrant workers on online platforms can be partially attributed to the convergence of the platforms’ need for cheap and flexible labour with the immigrants’ desire for accessible income, as they can work as much as they want. However, before this mutually beneficial relationship could be established, platforms had to overcome two significant obstacles. First, immigrant workers had to have access to the means of production: vehicles and phones. Second, platforms had to operate in a non-codified, informal, grey, unregulated, or even illegal space, which posed significant risks. In order to overcome these barriers, the platforms had to switch to an analogue approach and do extensive offline work and lobbying to ensure the viability of their algorithms. This was the case of the first ride-hailing platform that entered the Ecuadorian market. Like traditional analogue companies, this platform had to set up a physical office to meet potential candidates, provide them with training, and give promotional talks. It engaged in lobbying and reached informal agreements with workers, investors, and authorities, which were crucial to their start-up processes. These activities were eventually reduced after early adopters were secured, but the platform still supports a profile black market that operates outside the official system. This market creates a loophole in the algorithmic management system, resulting in lower earnings for immigrant workers at the end of the chain.

Far from being passive subjects, however, immigrants learn to deal with the platforms’ demands and innovate around them. Driven by their needs and hopes, they work over ten hours daily, follow the platform’s rules, try to comply with its algorithmic management, and even pay for its mistakes. But they also try to understand the platform, learn from it, and when possible, outsmart it. An interesting example of this was a sort of social hacking that a group of drivers carried out on one of the ride-hailing platforms operating in Quito. The route between the city and the airport was known to be one of the most profitable rides. However, the platform’s algorithm prevented individual drivers from doing only that route. This limitation was reinforced by the fact that drivers were only offered a ride if they were close to the pick-up point, so if they went to the airport, they would not be able to get another ride immediately from there unless they waited, which was not possible due to the traffic regulations in place at the airport and because the platform was officially illegal. This group of drivers, therefore, joined forces to create a kind of cooperative group that shared several smartphones with multiple profiles and parked at a gas station near the airport to get rides from there to the city. This social hack allowed them to get a large share of the rides between the city and the airport and to multiply their incomes beyond what the platform would foresee.

We started a sort of group, and we mounted a ‘platform’ down at the airport. We made US$2,000 per month. However, this was because we did something [the platform] could never imagine […] because the app lets you have another phone. [My second phone] was in a queue at the airport. So, when I arrived there, I already had another customer waiting. That was very profitable. (Interview, February 2021)

By harnessing technology and using it to their advantage, workers, especially immigrants, can resist the control imposed by algorithmic management and try to redefine an asymmetrical power relationship. This innovative approach is driven by the same force that compelled these people to leave their home country and seek a better life in a new place. However, the very same strategy and actors contribute to the expansion of the platform market, ensuring a constant flow of drivers willing to take customers to and from the airport at their own risk. It should not be forgotten that the use of platforms is still illegal in Ecuador and that police controls at airports are frequent.

This was, in fact, the cause that put an end to such a subaltern enterprise. At the height of the 2019 social revolt, the police arrested 19 platform drivers near the airport (mostly immigrants) and falsely accused them of planning an attack on the president. Later, it was revealed that they were only platform drivers parked ‘unusually’ near the airport (El Universo 2019; Vásconez 2020). This event put the platforms under the media spotlight and raised questions about how these still ‘illegal’ platforms were being used, especially around airports, where formal cab companies have an exclusive right to work. Shortly after, the platform updated its algorithm and removed the possibility of having the same profile on two different phones, closing the possibility of using the system invented by this group of drivers. Once again, this demonstrates how the strategies of subaltern actors to rebalance asymmetric relations with techno-transnational power also have an impact on the way technology is designed, adjusted, and globalised, even if it is against their interests and will. This particular feature highlights the importance of considering these subaltern populations’ role in the local adoption, appropriation, and resignification of global technologies.

Conclusions

The platform economy is one of the main vectors of the global dissemination of digital technologies and the attending datafication and algorithmic management process. It has removed several traditional barriers to employment, facilitated workers’ access to the labour market, and provided a quick source of income. However, these transformations have also undermined labour conditions and created new forms of exploitation and other harmful externalities, especially for vulnerable populations such as immigrants. Based on the case of immigrant platform workers in Ecuador, we have explored how subaltern actors deal with this process of technological globalisation from below.

Drawing on the growing literature on platforms, platform economics, and algorithmic management, and based on an ethnographic approach and more than 100 interviews with platform workers conducted between 2021 and 2023, we have shown how these workers interpret, interact, adapt, and resist the technological framework and demands of platforms and how the modern globalisation process driven by these platforms forces them to adjust, adapt, and negotiate their practices and technology in complex and unforeseen contexts. Platforms do so by taking advantage of the lack of regulation and the uncodified informal grey areas and practices in which they operate. These informal practices and areas that platforms use and help to perpetuate are the cradle of the detrimental externalities that the globalisation process brings to the Global South and one of the critical elements to understand the process of globalisation from below.

The main finding of this research is that in the context of international inequality, subaltern populations such as immigrant platform workers at once resist and enable the process of technological globalisation brought about by the platform economy. Indeed, workers are not merely passive recipients of algorithmic management and control systems imposed by those platforms. They develop strategies to learn, circumvent, adjust, and adapt to the algorithmic management exercised by the platform. By engaging in sensemaking activities, users acquire a working understanding of algorithms and their potential effects on their work. Through these sensemaking strategies, workers gain enough familiarity with the platforms’ functions to effectively work with and around them, even opening up the black box of algorithms used by digital platforms. However, by developing alternative strategies, such as informal black markets for platform profiles and vehicles, to circumvent legal, administrative, and algorithmic barriers, they enable platforms to keep working and expanding their operations from below. By contributing to the development of these non-codified (informal) grey areas, they have given new breath to the development of these platforms in the South. Despite the different conditions under which they operate in the Global North, for which they were initially conceived, it is very likely that marginalised actors in this context also engage in similar practices, but this will require further enquiry.

Finally, workers may also use their learning to circumvent the algorithms or manipulate them to their advantage. They can bypass algorithmic constraints by drawing on other resources and technologies or by using their knowledge of algorithms to achieve desired outcomes. However, platform technologies are not static and tend to adapt, adjust, and incorporate changes to regain power over workers’ practices. Consequently, a platform’s algorithms, which are essentially an automated representation of the platform organiser’s interests, are not deterministic rules. They iterate and are redesigned through feedback loops. In that sense, workers appropriate and expand the digital platform’s system of programmatic processes as part of an information infrastructure. Platform workers and users have agency and influence over the platform’s algorithms, its design, and redesign iterations. Algorithmic management is a sociotechnical process that results from continuous interactions between algorithms and humans. Workers’ encounters with algorithms shape not only their work practices but also algorithmic outcomes and decision-making, which fuel the globalisation of platform technologies.

Endnotes‌

References

Abidin, C., Internet Celebrity: Understanding Fame Online (Emerald Publishing, 2018).

Albornoz, M. B., and H. Chavez, ‘De La Gestión Algorítmica Del Trabajo a La Huelga 4.0’, Mundos Plurales – Revista Latinoamericana de Políticas y Acción Pública, 7.2 (2020): 43–54 https://doi.org/10.17141/mundosplurales.2.2020.4848.

Albornoz, M. B., H. Chavez, and others, Fairwork Ecuador Ratin‌gs 2022: Labour Standards in the Platform Economy (Fairwork, 2022) https://fair.work/en/fw/publications/labour-standards-in-the-platform-economy-ecuador-ratings-2022/.

Albrieu, R., ed., Cracking the Future of Work: Automation and Labour Platforms in the Global South (FOWIGS, CIPPEC, IDRC, 2021) https://fowigs.net/publication/cracking-the-future-of-work-automation-and-labour-platforms-in-the-global-south/.

Altenried, M., ‘The Platform as Factory: Crowdwork and the Hidden Labour Behind Artificial Intelligence’, Capital and Class, 44.2 (2020): 145–58 https://doi.org/10.1177/0309816819899410.

Baertl, M., ‘YouTube Channels, Uploads and Views: A Statistical Analysis of the Past 10 Years’, Convergence: The International Journal of Research into New Media Technologies, 24.1 (2018): 16–32 https://doi.org/10.1177/1354856517736979.

Bauman, Z., and D. Lyon, Liquid Surveillance: A Conversation (Polity Press, 2012).

Banco Central del Ecuador, Informe Estadístico Mensual (October 2023).

Bucher, T., ‘Want to Be on the Top? Algorithmic Power and the Threat of Invisibility on Facebook’, New Media & Society, 14.7 (2012): 1164–80 https://doi.org/10.1177/1461444812440159.

Burchardt, H.-J., R. Domínguez, C. Larrea, and S. Peters, Nada Dura Para Siempre: Neo-Extractivismo Tras El Boom de Las Materias Primas (UASB – ICDD, 2016).

Burgess, J., and J. Green, ‘The Entrepreneurial Vlogger: Participatory Culture Beyond the Professional/Amateur Divide’, in P. Snickars and P. Vonderau, eds., The YouTube Reader (National Library of Sweden, 2009), pp. 89–107.

Casilli, A., and S. Bouquin, ‘Il N’y a Pas D’automatisation Sans Micro-Travail Humain’, Les Mondes Du Travail, 24–25 (2020): 3–21.

Casilli, A., P. Tubaro, and others, Le Micro-Travail En France: Derrière L’automatisation, de Nouvelles Précarités Au Travail? [Rapport Final], Projet DiPLab ‘Digital Platform Labour’ (2019) http://diplab.eu/.

Célleri, D., Situación Laboral y Aporte Económico de Inmigrantes En El Centro/Sur de Quito-Ecuador(Rosa Luxemburg Stiftung, 2020) http://www.rosalux.org.ec/pdfs/SituacionLaboralYAporteEconomicoDeInmigrantes.pdf.

CITEC, Estudio de Impacto de Las Plataformas Digitales En Ecuador (Cámara de Innovación y Tecnología Ecuatoriana, 2022).

Courtois, C., and E. Timmermans, ‘Cracking the Tinder Code: An Experience Sampling Approach to the Dynamics and Impact of Platform Governing Algorithms’, Journal of Computer-Mediated Communication, 23.1 (2018): 1–16 https://doi.org/10.1093/jcmc/zmx001.

DataReportal, Digital 2023: Ecuador (2023) https://datareportal.com/reports/digital-2023-ecuador.

de Freitas, C., P. A. G. Pascoal, A. da S. Mello, and L. F. Silva, ‘Digital Platform: Uber’, Revista Tecnologia E Sociedade, 19.55 (2023): 176–88 https://doi.org/10.3895/rts.v19n55.14645.

El Universo, ‘Extranjeros Detenidos En Aeropuerto de Quito Tenían Agenda Presidencial, Según Ministra de Gobierno’, El Universo (10 October 2019) https://www.eluniverso.com/noticias/2019/10/10/nota/7554917/extranjeros-detenidos-aeropuerto-quito.

Ellison, N. B.,‌ C. Steinfield, and C. Lampe, ‘The Benefits of Facebook “Friends”: Social Capital and College Students’ Use of Online Social Network Sites’, Journal of Computer-Mediated Communication, 12 (2007): 1143–66 https://doi.org/10.1111/j.1083-6101.2007.00367.x.

Faraj, S., S. Pachidi, and K. Sayegh, ‘Working and Organizing in the Age of the Learning Algorithm’, Information and Organization, 28.1 (2018): 62–70 https://doi.org/10.1016/j.infoandorg.2018.02.005.

Firmino, R. J., B. de V. Cardoso, and R. Evangelista, ‘Hyperconnectivity and (Im)Mobility: Uber and Surveillance Capitalism by the Global South’, Surveillance & Society, 17.1–2 (2019): 205–12 https://doi.org/10.24908/ss.v17i1/2.12915.

Gago, V., Neoliberalism from Below: Popular Pragmatics and Baroque Economies (Duke University Press, 2017).

Galiere, S., ‘When Food-Delivery Platform Workers Consent to Algorithmic Management: A Foucauldian Perspective’, New Technology, Work and Employment, 35.3 (2020): 357–70 https://doi.org/10.1111/ntwe.12177.

Grohmann, R., ‘Rider Platforms? Building Worker-Owned Experiences in Spain, France, and Brazil’, South Atlantic Quarterly, 120.4 (2021): 839–52 https://doi.org/10.1215/00382876-9443392.

ILO, El Trabajo En Las Plataformas Digitales de Reparto y Transporte En Ecuador (International Labour Organization, 2022).

INEC (Instituto Nacional de Estadística y Censos), ‘Entradas y Salidas Internacionales’, Instituto Nacional de Estadística y Censos (31 May 2022) https://www.ecuadorencifras.gob.ec/entradas-y-salidas-internacionales-informacion-historica/.

INEC (Instituto Nacional de Estadística y Censos), ‘Estadísticas Laborales’, Instituto Nacional de Estadística y Censos (February 2023) https://www.ecuadorencifras.gob.ec/estadisticas-laborales-abril-2022/.

Ipeirotis, P. G., ‘Analyzing the Amazon Mechanical Turk Marketplace’, XRDS: Crossroads, The ACM Magazine for Students, 17.2 (2010): 16–21 https://doi.org/10.1145/1869086.1869094.

Irani, L., ‘Difference and Dependence among Digital Workers: The Case of Amazon Mechanical Turk’, South Atlantic Quarterly, 114.1 (2015): 225–34 https://doi.org/10.1215/00382876-2831665.

Jarrahi, M. H., G. Newlands, and others, ‘Algorithmic Management in a Work Context’, Big Data & Society, 8.2 (2021): 20539517211020332 https://doi.org/10.1177/20539517211020332.

Kellogg, K. C., M. A. Valentine, and A. Christin, ‘Algorithms at Work: The New Contested Terrain of Control’, Academy of Management Annals, 14.1 (2020): 366–410 https://doi.org/10.5465/annals.2018.0174.

Khan, A., F. T. N. Gabralla, and A. Musa, ‘Online Food Delivery Platforms and the Emerging Entrepreneurial Culture in Africa’, African Journal of Business Management, 14.8 (2020): 213–22 https://doi.org/10.5897/AJBM2020.9053.

Langlois, G., and G. Elmer. “The Research Politics of Social Media Platforms.” Culture Machine, 14 (2013): http://culturemachine.net/index.php/cm/article/view/505/531.

Le Ludec, C., E. Wahal, A. Casilli, and P. Tubaro. “Quel statut pour les petits doigts de l’intelligence artificielle ? Présent et perspectives du micro-travail en France.” Les Mondes Du Travail, 24–25 (2020): 99–112.

Leighton, T. G., ‘The Challenges of Algorithms in the Gig Economy’, Communications of the ACM, 62.11 (2019): 56–65 https://doi.org/10.1145/3357239.

Levi, M. L., ‘Los Modelos de Negocios y la Economía Colaborativa’, Revista Innovación y Ciencia, 21.3 (2018): 38–44.

Liem, M. C., J. Zhang, and Y. Tao, ‘Fairness and Transparency in Algorithmic Work Distribution: A Review’, AI and Ethics, 2.3 (2022): 155–72 https://doi.org/10.1007/s43681-022-00124-6.

Lobato, R. Netflix Nations: The Geography of Digital Distribution (NYU Press, 2019).

Lopez Galvez, D., ‘Digital Labor Platforms in Latin America: A Case Study of Delivery Workers in Quito’, Global Labor Journal, 11.1 (2020): 58–73.

Lord, C., O. Bates, A. Friday, F. McLeod, T. Cherrett, A. Martinez-Sykora, and A. Oakey. “The Sustainability of the Gig Economy Food Delivery System (Deliveroo, UberEATS and Just-Eat): Histories and Futures of Rebound, Lock-in and Path Dependency.” International Journal of Sustainable Transportation, 17.5 (2023): 490–502. https://doi.org/10.1080/15568318.2022.2066583.

Markus, M. L. “Datification, Organizational Strategy, and IS Research: What’s the Score?” The Journal of Strategic Information Systems, 26.3 (2017): 233–241. https://doi.org/10.1016/j.jsis.2017.08.003.

Martínez Fernández, G., Trabajo En Plataformas Digitales: Transformación, Regulación y Alternativas (Editorial Trotta, 2020).

Maya, N., T. Quevedo, D. Carrión, and P. Sánchez. Hacia una caracterización de las y los repartidores y de la economía de plataformas en Ecuador (Rosa Luxemburg Stiftung, 2022).

McClain, D. C., and K. A. Posner, ‘Worker Experiences on Food Delivery Platforms: Organizational Effects of Digital Intermediation’, Journal of Management Studies, 57.3 (2020): 125–38 https://doi.org/10.1111/joms.12542.

McKinsey, D., El Auge de Los Negocios de Plataforma En América Latina (McKinsey Global Institute, 2022).

Meijerink, J., and T. Bondarouk. “The Duality of Algorithmic Management: Toward a Research Agenda on HRM Algorithms, Autonomy and Value Creation.” Human Resource Management Review, 33.1 (2023): 100876. https://doi.org/10.1016/j.hrmr.2021.100876.

Mohlmann, M., and L. Zalmanson. “Hands on the Wheel: Navigating Algorithmic Management and Uber Drivers’ Autonomy.” ICIS 2017 Proceedings, December 10, 2017. https://aisel.aisnet.org/icis2017/DigitalPlatforms/Presentations/3.

Montfort, N., and I. Bogost. Racing the Beam: The Atari Video Computer System (MIT Press, 2009).

OECD, ‘The Future of Work: How Digital Platforms Are Shaping Labour Markets’, OECD Employment Outlook 2019, 2019 https://doi.org/10.1787/9ee00155-en.

Oestreicher-Singer, G., and A. Sundararajan. “Recommendation Networks and the Long Tail of Electronic Commerce.” MIS Quarterly, 36.1 (2012): 65–83. https://doi.org/10.2307/41410406.

OIT. El trabajo en las plataformas digitales de reparto y transporte en Ecuador (International Labour Organization, 2022).

Pang, H. S., ‘The Ethical Implications of Algorithmic Management on Gig Workers’, AI & Society, 36.1 (2021): 45–56 https://doi.org/10.1007/s00146-020-01035-y.

Parent-Rocheleau, X., and S. K. Parker. “Algorithms as Work Designers: How Algorithmic Management Influences the Design of Jobs.” Human Resource Management Review, 32 (2022): 1–17. https://doi.org/10.1016/j.hrmr.2021.100838.

Plantin, J.-C., C. Lagoze, P. N. Edwards, and C. Sandvig. “Infrastructure Studies Meet Platform Studies in the Age of Google and Facebook.” New Media & Society, 20.1 (2018): 293–310. https://doi.org/10.1177/1461444816661553.

Posenato, A., and D. Parker, ‘Digital Platforms as Drivers of Economic Innovation’, Journal of Economic Perspectives, 34.2 (2020): 28–45.

Rosenblat, L., Uberland: How Algorithms Are Rewriting the Rules of Work (University of California Press, 2018).

Rosenblat, A., and L. Stark. “Algorithmic Labor and Information Asymmetries: A Case Study of Uber’s Drivers.” International Journal of Communication, 10 (2016): 27. https://doi.org/10.2139/ssrn.2686227.

Schildt, H. “Big Data and Organizational Design – The Brave New World of Algorithmic Management and Computer Augmented Transparency.” Innovation, 19.1 (2017): 23–30. https://doi.org/10.1080/14479338.2016.1252043.

Scholz, R., and S. Schneider, ‘Workers’ Voice in the Digital Economy: Co-Determination and Representation’, Industrial Relations Journal, 49.4 (2018): 312–28 https://doi.org/10.1111/irj.12224.

Srnicek, A., Platform Capitalism (Polity Press, 2017).

Stein, C. ‘Algorithmic Cultures and the Futures of Work’, Global Media Journal, 19.36 (2019): 1–12.

Stiegler, B. “Questions de pharmacologie générale. Il n’y a pas de simple pharmakon.” Psychotropes, 13.3–4 (2007): 27–54. https://doi.org/10.3917/psyt.133.0027.

Sundararajan, J., The Sharing Economy: The End of Employment and the Rise of Crowd-Based Capitalism (MIT Press, 2016).

Timko, P., and R. van Melik. “Being a Deliveroo Rider: Practices of Platform Labor in Nijmegen and Berlin.” Journal of Contemporary Ethnography, 50.4 (2021): 497–523. https://doi.org/10.1177/0891241621994670.

Tubaro, P., A. Casilli, and M. Coville. “The Trainer, the Verifier, the Imitator: Three Ways in Which Human Platform Workers Support Artificial Intelligence.” Big Data & Society, 7.1 (2020): 2053951720919776. https://doi.org/10.1177/2053951720919776.

Tufekci, Z. “Algorithmic Harms Beyond Facebook and Google: Emergent Challenges of Computational Agency.” Colorado Technology Law Journal, 13.2 (2015): 203–218.

Vallejos Rivero, O. “Control Mechanisms of Uber Platform over Its Associate Drivers in the Metropolitan Region of Chile.” Cuhso-Cultura-Hombre-Sociedad, 31.1 (2021): 391–416. https://doi.org/10.7770/CUHSO-V31N1.ART2412.

van Dijck, J. The Culture of Connectivity: A Critical History of Social Media (Oxford University Press, 2013).

van Doorn, T. J., ‘Stealing Time: Labor, Capital, and the Temporalities of the Gig Economy’, New Media & Society, 19.2 (2017): 175–94 https://doi.org/10.1177/1461444816654346.

van Doorn, N., and D. Vijay. “Gig Work as Migrant Work: The Platformization of Migration Infrastructure.” Environment and Planning A: Economy and Space, 0308518X211065049 (2021): https://doi.org/10.1177/0308518X211065049.

Vásconez, L. “Juez Confirmó Inocencia de Ciudadanos Venezolanos Detenidos en el Paro.” El Comercio, January 21, 2020. https://www.elcomercio.com/actualidad/seguridad/juez-inocencia-venezolanos-detenidos-paro.html.

Webster, S., and L. Williams, ‘Social Protection in the Digital Age’, ILO Working Paper Series, 2020.

Woodcock, J., and C. Cant. “Platform Worker Organising at Deliveroo in the UK: From Wildcat Strikes to Building Power.” Journal of Labor and Society, 25.2 (2022): 220–236. https://doi.org/10.1163/24714607-BjA10050.

Woodcock, J., and M. Graham. The Gig Economy: A Critical Introduction (Polity Press, 2019).

Zarsky, T. “The Trouble with Algorithmic Decisions: An Analytic Road Map to Examine Efficiency and Fairness in Automated and Opaque Decision Making.” Science, Technology, & Human Values, 41.1 (2016): 118–132. https://doi.org/10.1177/0162243915605575.

Zysman, Y., and A. Kenney, ‘The Next Phase in the Digital Economy: Platforms, Apps, and the Growth of Contingent Work’, Journal of Industrial Relations, 58.1 (2016): 5–25 https://doi.org/10.1177/0022185615624011.