Parasitic platform urbanism in Dhaka and Melbourne

Abstract

This paper explores the relationship between platform-mediated labour on Uber, Pathao, UberEats and DoorDash, and the public infrastructures of two cities – Dhaka and Melbourne. Due to geographical location and economic difference, the provision of public amenities varies greatly between Dhaka and Melbourne. However, our research shows that there are some similarities in how labour platforms make use of public amenities in each city. In this paper, we argue that digital labour platforms rely on cities’ existing public infrastructures to run their businesses and that this causes various challenges for both gig workers and cities. Our paper contributes to infrastructure studies, platform studies, and labour studies by demonstrating that digital labour platforms operate like parasites. Just as parasites benefit at the expense of the other, or the ‘host’, digital labour platforms rely on cities’ existing public infrastructure without investing to that infrastructure. Following this argument, we conclude that there is a need for platforms to provide more amenities whether they recognise platform workers as ‘employees’ or not.

Introduction

Digital labour platforms offer a convenient mode of connection between people looking for work and those looking for workers to perform tasks on their behalf. These platforms are marketplaces, where jobs are performed on an as-needed basis, with no obligation for either party to remain connected after the completion of a given task. Platform work is particularly appealing to precarious workers, for whom other forms of employment are scarce or unattainable (Altenried, 2019; Metawala et al., 2021; Lata et al., 2023). While identity markers such as gender, race, language, visa status, and social standing can inhibit people from securing other forms of work, these markers do not factor into one's capacity to sign up to work on a platform (Attoh et al., 2024; Orth, 2024). However, these markers influence how platforms allocate tasks to platform workers. Vallas and Schor (2020) note that platform workers experience algorithmic discrimination due to their gender, race, and ethnicity. For example, workers of colour on TaskRabbit are given lower algorithmic priority which reduces their earnings and employability (Hannák et al., 2017). With low barriers to entry, digital labour platforms provide an accessible means by which workers can earn an income. However, in each instance where labour platforms operate, the income afforded is low relative to other forms of work, and the work comes with few, if any, employment rights (Maury et al., 2024; van Doorn and Vijay, 2024). In addition, platform workers often personally bear the risk involved in the work. When a delivery rider's scooter is stolen, their capacity to earn is directly impacted, with the delivery platforms for whom they work taking no responsibility. By bearing the risks of platform work individually, and by adhering to sophisticated algorithmic management techniques, many platform workers, particularly in the Global South, have become indebted to the platforms for which they work (Dib, 2022; Ray and Sam, 2023; Zhou, 2024). For example, in Mexico, Didi and UberEats offer in-app loan service to riders. However, the interest rate is very high. If riders take loan using this in-app service, they might have to work longer hours to pay the loan (Dib, 2022).

Labour relations within the platform economy reconfigure the space and resources of the city. However, there is dissensus among scholars regarding the extent to which platform industries organise work differently to other, historical industries. Finkin (2016) notes, for example, the similarities between the atomisation of work within the platform economy, and within the ‘putting out’ system which was common up until the twentieth century. The putting out system saw merchants have materials delivered to multiple home-based subcontractors, who would complete their respective tasks before passing the goods onto the next worker or returning the goods to a central location. Yet, while task-based work done outside of the workplace is not new (Adams-Prassl, 2018), digital labour platforms mediate this work in new ways, thereby producing new spatial configurations. Instead of work being done at home as in the putting out system, in location-based platform work such as rideshare and food delivery, the workplace becomes dematerialised and mobile. The city itself becomes the workplace, with the associated resources and infrastructures spread across urban space. While scholars have noted the platform economy's reliance on digital communications infrastructure, such as cellular towers (Wiig and Masucci, 2020: 75), the infrastructure which supports platform labour has received less critical consideration. Within this context, this paper addresses the following research questions: how do the gig workers in Bangladesh and Australia navigate the gig economy sector using existing public infrastructure? And how is the platform economy transforming urban space? Following these questions, in this paper, we clarify the novel relationship which digital labour platforms, namely Uber, Pathao (a local ridesharing app), UberEats and DoorDash, have established between urban workers and urban infrastructure, and how this relationship is defined by a lack of responsibility.

The next section of this paper presents the conceptual framework. Following this, we discuss the method we employed for conducting our study. Next we present our findings, conclusion and policy recommendations.

Conceptual framework: infrastructure and platform urbanism

Infrastructure is often thought of in terms of big material projects such as railroads, buildings, dams, and so on. There are disciplinary differences in understanding and using the concept of ‘infrastructure’. The most common definition of infrastructure is that infrastructures are the ‘basic physical and organisational structures and facilities (e.g. buildings, roads, power supplies) needed for the operation of a society or enterprise’ (Lexico, 2020). Infrastructures also ‘involve ‘assets and facilities’ such as airports, bridges, canals, dams, ports, rail, roads, sewage, telecommunications, water supply and wastewater’ (Kanoi et al., 2022: 2). This understanding of infrastructure is often used in planning and geography whereas the infrastructural studies that focus on digital environments mainly focus on two trends. The first one focuses on historical perspective on large technical systems (Plantin et al., 2018). This perspective is known as ‘sociotechnical systems’ that analyse systems such as electric power grids, telephone networks and air traffic control (Hughes, 1983; Bijker et al., 1987; Plantin et al., 2018; Plantin and Punathambekar, 2019). This perspective highlights that infrastructures are ‘centrally designed and controlled, typically in the invention and development phases of new technologies’ (Plantin et al., 2018: 295). However, users and other developers can modify or extend infrastructures to make them compatible with new technologies and new enterprises; ‘consider the many similar, but incompatible devices and standards developed during the early days of railroads, electric power, or digital computers’ (Plantin et al., 2018: 295).

The second group of scholars such as Star and Bowker expanded the discussion on phenomenology and sociology of infrastructures (Star and Ruhleder, 1996; Graham and Marvin, 2001; Edwards et al., 2009; Ribes and Finholt, 2009; Plantin et al., 2018). They noted, for example, ‘interacting dependencies on infrastructures, which create social chaos during episodes of breakdown, such as urban blackouts or major Internet outages’ (Plantin et al., 2018: 296). This dependency arises from the ‘qualities of ubiquity, reliability, and especially durability because major infrastructures such as railroads, telephone networks, Linnaean taxonomy, and the Internet’ (Plantin et al., 2018: 296) provide services for many years. This discussion indicates that although media scholars have mostly focused on the discussion of digital infrastructure, sociologists are more interested in exploring the relations between human and infrastructure or the ways some groups (such as people with disability – deaf, blind or wheelchair-bound individuals) might find it difficult to access certain infrastructures (Plantin et al., 2018). Following this discussion, we build our argument in this paper by employing the terminology infrastructure to refer to urban infrastructure. Geographers mostly have contributed to the literature on urban infrastructure (Amin, 2014; Graham and McFarlane, 2014; Barnes, 2017).

Urban infrastructure refers to the complex physical systems and services that support the functioning of cities such as transport infrastructure, business infrastructure, water, energy, sanitation and communication networks (Amin, 2014; Graham and McFarlane, 2014; Barnes, 2017; Stokes and De Coss-Corzo, 2023). Historically, the development of urban infrastructure is tied to the growth and expansion of cities. Industrialisation and technological advances particularly played an important role in shaping and reshaping our cities. The most recent phase in urban development is the emergence of smart cities. Smart cities are emerging as a solution for improving urban infrastructure through the integration of digital technologies. The use of Internet of Things, data analytics and smart grids enables real-time monitoring and optimisation of urban systems. Over the last decade, there has been a plethora of scholarship attending to the rapid rise of digital networks, data and digital platforms in cities, especially how these digital phenomena are producing and reproducing urban built environments and our everyday lives (Kitchin, 2015; Datta, 2019; Fields et al., 2020; Leszczynski, 2020; Van Doorn and Badger, 2020). To date, most research has criticised the techno-solutionism of smart city developments (Kitchin, 2014, 2016; Leszczynski, 2016; de Almeida Fagundes & Matias, 2018; Datta, 2019; Cugurullo et al., 2023). More recently, digital platforms such as Uber, Didi, Deliveroo and DoorDash have successfully changed the critical narratives of techno-solutionism by presenting platforms as fixtures of urban landscapes in different parts of the world (Leszczynski, 2020; Lata et al., 2023). These platforms have ‘rapidly change(d) how people experience cities and how cities work’ (Rosenblat, 2018: 38). Urban studies scholars refer to this condition as ‘platform urbanism’ – a new trend in urbanism that has changed the narratives of smart city formations through combining not only data and space but also labour, mobilities, consumption, governance, civic citizenship, urban infrastructure and a range of urban services (Leszczynski, 2020).

The term ‘platform urbanism’ emerges from Srnicek (2017)'s concept of ‘platform capitalism’. The popularity of the concept ‘platform urbanism’ demonstrates a recognition that platform capitalism significantly changes urban space (Davidson and Infranca, 2016; Barns, 2019; Sadowski, 2020; Caprotti et al., 2022; Cirolia et al., 2023). The concept further shapes current debates within political economy discourse by pointing out the importance of data and data capture as new forms of capital. As Leszczynski (2020: 193) notes:

Platform urbanism discursively signals and provides a theoretical framework for researching the unprecedented scale, scope, agency, and urban ambitions of platform economy actors and their effects, which are held to be unique to platform entities and distinct from antecedent digital-urban configurations, namely those of the smart city and its corollary smart urbanism in several key ways.

However, as any conceptual framework develops, the perspectives on what platform urbanism is and how to study it are not homogeneous in the literature. Some scholars recognise platform urbanism as a process which reconfigures relations between urban actors (Barns, 2020). Other scholars focus on understanding platform urbanism as a framework for studying the ways in which platforms act as extractive agents in the city (Srnicek, 2017; Sadowski, 2020; Veen et al., 2020; Lata et al., 2023). For instance, while writing about cities and digital platforms, Sadowski (2020) argues that platforms are centralised in cities for similar reasons that capital is. This is because platforms benefit from the population density and spatial proximity of users and workers in cities and can seize the ‘opportunities for mediating social relations and extracting economic value in large, diverse markets’ (Sadowski, 2020: 450). In addition to extracting economic value, platforms use cities public infrastructure to operate their business. They rely on urban infrastructure like parasites (Serres, 2013) which extract benefits from their hosts but do not give back. We call this trend ‘parasitic platform urbanism’. Although platforms heavily rely on existing urban infrastructure, little is known about their reliance on existing urban infrastructure and how they are changing the urban landscape through their operations. Within this context, in this paper, we focus on existing public infrastructure that several platform companies rely on to conduct their business and the impact this infrastructural use has on platform workers in Dhaka and Melbourne.

Method

We present data from three research projects in this paper. The first project was a case study focusing on ridesharing drivers’ everyday life in Dhaka. For the Dhaka case study, the first author conducted qualitative interviews with 27 Uber and Pathao drivers including two leaders of App-based Workers’ Federation of Bangladesh, one ex Uber official and one focus group discussion (FGD) with 10 members of the Dhaka ride sharing Drivers’ Union (DRDU). The first author recruited 25 Uber and Pathao drivers through trips taken on the Uber and Pathao platforms and two drivers were recruited using DRDU Facebook group page. The first author used convenience sampling approach to recruit participants due to the nature of gig work. The interviews and FGD were conducted between June 2023 and April 2024 and combined with participant observations in July 2023 where the author observed drivers during several trips taken with Uber and Pathao drivers. The semi-structured interviews lasted between 60 and90 min and the FGD lasted nearly 3 hours. All drivers were given a gift card (valued at 1000 Bangladeshi taka) as an appreciation of their time and participation in the study. All participants were male. The ages of drivers ranged from 19 to 60. Most participants were from lower socio-economic background. However, some participants were from lower middle-class background. The interview checklist includes information on participants’ entry into the gig economy, motivations for joining ridesharing platforms, employment history, length of current employment, understanding of how ridesharing apps work, benefits and challenges of working in the gig economy, the role of social networks in accessing to and navigating the gig economy, information on their involvement with DRDU or other similar organisations, migration history, age, income and expenditure, education and strategies utilised to navigate the gig economy. The data were analysed using thematic approach. This project received ethical clearance from the first author's university's Ethics Committee.

The second project was conducted by the first and second authors from October 2023 to February 2024. In this project, we focus on the food delivery riders’ everyday life in Melbourne. We collected data from 17 food delivery riders and six key informants including four government officials: a representative from the Victorian Government's Industrial Relations department, three representatives from Melbourne municipality – two from the Strategic Transport team and one from the City Data team, one Transport Workers’ Union staff, one DoorDash staff and one bike shop owner.

The third project is part of the second author's PhD project. As part of this project, the second author conducted a survey with 35 food delivery riders in Melbourne to explore how far they live from their preferred work area. The delivery riders surveyed worked for platforms including UberEats, DoorDash, Menulog and HungryPanda. The survey also collected information from riders about their preferences regarding the use of amenities in the city, such as which bathrooms they find most accessible. The second author also gathered first-hand experience while conducting 10 e-bike deliveries with DoorDash, and 22 with Uber Eats, between March and July 2023. Conducting these deliveries provided the second author with direct experience of the challenges faced regarding lack of access to resources in the city. Working as a delivery rider during this time also meant developing friendships with fellow riders, who conveyed personal insights to the second author regarding the experience of delivery work.

Data from our second and third projects demonstrate that while the vast majority of our participants worked on UberEats, most had accounts with multiple platforms, including DoorDash, Menulog and HungryPanda. The ages of delivery riders interviewed and surveyed ranged from 19 to 35 years. Of the riders interviewed, four were Australian citizens, and four were female. The remaining 30 were male migrants, predominantly hailing from South America and Southeast Asia. Both projects received ethical clearance from the first and second authors’ university's Ethics Committee. Thematic approach was used to analyse data. Pseudonyms are used for all participants to maintain their confidentiality.

Findings

Infrastructural barriers

When delivery riders in Melbourne were asked about the kind of amenities they would like greater access to, many focused on toilets. ‘Toilets are most important and there should be more of them around. They have toilets in parks, but they’re all locked’. This statement from Chitapanya, a 25-year-old Thai food delivery rider working in Melbourne, suggests that public amenities are insufficient for the needs of those working in the street at night. Similarly, Uber drivers in Dhaka pointed out how the lack of public toilets negatively impacted their health. For example, Riad said:

I do not drink enough water when I drive. As there are not many public toilets and public parking available in Dhaka city, it's hard to find a toilet. Also, we can’t go to toilet if we have a customer. But it has impacted my health now. I have diabetes now.

Other drivers also echoed similar concerns pointing out that ‘many of us do not drink enough water as we are not able to find and use a public toilet if we are on a ride request’. As there are not many available public toilets in Dhaka, Uber and Pathao drivers often use toilets located in several petrol stations.

As a convenient and affordable mode of transport, e-bikes are the most common mode of delivery in central Melbourne. One key challenge of delivery work is therefore keeping both a smartphone and an e-bike battery charged. In the words of Harrison, an Australian-born rider in his 30s, to do delivery work ‘you don’t have to be tech literate, but you have to be good at managing battery power’. The alternative, as experienced by multiple respondents, is running out of battery, which often results in having to walk a heavy e-bike for kilometres before finding a suitable place to recharge.

Delivery riders spoke of microwave access for food and shelter from rain as other amenities that would improve their working conditions. Riders also reported wanting somewhere to purchase work-related equipment. ‘There have been plenty of occasions where I've had to go home due to a faulty this or that, [so it] would be great if I could go to a space to repair or buy something’. This comment by Harrison was echoed in the following comment by Drevan:

Expanding on a stationary toolkit, maybe having a vending machine or even a manned store that sells common things like patch kits, small snacks, batteries, power banks and disposable batteries for lights, for example. If possible, clothing even, like spray jackets, neck scarfs, gloves, and delivery bags.

Drevan also highlighted the value of having somewhere to exchange information.

‘It would be nice to have a space to discuss where is busy, and any incidents that have occurred, with other delivery drivers’. The desire for a space in which to share advice and socialise was common among respondents, illustrating the potential of any dedicated rider amenity to also act as a social space. Where possible, Drevan chooses to wait ‘wherever there is a busy, well-lit area with bike racks’, highlighting the impact that urban design has on riders’ perceptions of safety. Further analysis of how platform workers use and understand urban space is critical to ensuring that platform work changes cities to not only be more convenient, but also more equitable.

Parking

Parking is a major issue for Uber and Pathao drivers in Dhaka. Without dedicated parking spaces, these drivers have difficulty parking their cars (Figure 1), especially in several affluent neighbourhoods such as Gulshan, Banani and Baridhara. Uber and Pathao drivers receive more trip requests from these areas. However, traffic police are very vigilant in these areas and police often issue drivers fines if they find them parking illegally on the road.

Figure 1.

Uber drivers parked their cars in front of the Bir Shreshtha Shaheed Shipahi Mustofa Kamal stadium.

A number of participants mentioned police harassment due to a lack of designated parking spot in different parts of Dhaka. The idea of paid public parking is almost non-existent in Dhaka. The little paid public parking that exists here can mostly be found in large shopping malls. However, Uber and Pathao drivers do not use those parking areas because they are expensive. The President of DRDU mentioned this issue:

DRDU President: Our Prime Minister says Bangladesh is a digital country, but there is no trace of digitalisation in the transport sector. Traffic police trouble us when we stop on the road, even if you talk to them, they don't listen. We can't convey how oppressed and tortured we are from all sides. The police constantly harass us. When we pick up passengers or drop them off, they charge us for ‘illegal parking’ and take 5000 taka from us, or they take 1000 taka as a bribe.

Interviewer: I understand that parking is a big problem in Dhaka city. Have you ever discussed this with Uber or Bangladesh Road Transport Authority (BRTA)?

DRDU President: We mentioned this issue every time we organised a movement against Uber, including movements we organised for the ridesharing workers in Dhaka, Chittagong, and Sylhet. Every time we organised a movement, we made arrangements for parking for ridesharing drivers. We have plans to meet with our advisors and the Mayor to discuss our next steps and see if they can do anything for us. There are some places where it's written that we can stop here, meaning bus stands. Even there, they can charge us if we pick up or drop off passengers.

In Melbourne, delivery riders park their scooters and e-bikes on footpaths, much to the frustration of local residents and retailers. Irrespective of where delivery riders park their e-bikes, the risk of bike theft is ever present. Due to the risk of fire from lithium-ion batteries, many landlords and apartment building managers forbid residents from bringing e-bikes into their own homes. When left inside apartment common areas such as hallways or garages, reports of e-bike theft are common. This threat leads many riders to lock their bikes up at busy train stations in the city centre. Yet even here, thieves cut through heavy-duty bike locks using bolt cutters or angle grinders.

Generally, e-bikes are not insured, and despite many common e-bikes costing thousands of dollars, police do not treat bike theft with the same gravity as car theft. Having had both his car and bike stolen, Harrison made the following comparison: ‘My car cost me next to nothing, my bike is worth AU$2000, and as soon as I reported the stolen car the police leapt into action, but they never got back to me about my bike’. The spatial and legal protocols which govern how platform work is performed in urban space have thus proven inadequate.

The discussion so far suggests that in each case, the risk and responsibility associated with platform work is borne by platform workers themselves. While digital labour platforms rely on public infrastructure to maintain their operations, they do so without adequately mitigating the challenges faced by platform workers in their everyday lives.

The social potential of public infrastructure

Office and factory workers typically have designated areas to eat or socialise, whereas for delivery riders the workplace is the street. Lacking designated spaces for interaction or for work, delivery riders tend to spend less time together than other workers, which hinders their ability to organise effectively or form meaningful relationships with each other. Social media has proven valuable in organising different kinds of platform workers, particularly those who are geographically dispersed; however without central meeting places, workers can only build solidarity and shared identities to a limited extent (Heiland, 2020). Just as shared amenities are critical to sustaining platform workers, shared spaces are critical to the formation of platform worker communities.

Taking the regional train to Melbourne from their homes in the outer suburbs, Hassan and other delivery riders are told by train personnel that they can’t board with their e-bikes at certain times because of the inconvenience it causes other passengers. Hassan is a 26-year-old delivery rider from Pakistan, who came to Australia 4 months ago after completing a Bachelor of Computer Science. He and his friends were told that only four bikes were allowed on the train because of security reasons. ‘But in the last day’, says Hassan, ‘we argue[d] with them that how we are going home because it's the last train and then they let us to take the train’. This kind of negotiation is required of delivery riders throughout the working day.

Earning less than minimum wage, many delivery riders live far from where they work, further undermining their capacity to socialise with fellow workers. Though not exclusive to the food delivery economy, this disjunction between where people work and where they socialise is particularly deleterious for a group who tend to work alone. Of 35 delivery workers surveyed in 2023, the median distance travelled from their home to their preferred work area was 11 kilometres. Of these participants, one quarter travelled 20 or more kilometres (Figure 2) to reach their preferred place of work, typically taking their e-bikes into the city on the train. Generally unable to afford to live in the neighbourhoods they service, these riders live across greater Melbourne; however, they tended to work in one of four hotspots: Central Business District; Inner North; St Kilda; Chapel Street. The everyday interactions which helped to sustain worker identities and produce solidarity in the past (e.g. meeting at the local school or greengrocer) are now exceedingly rare for those who cannot afford to live near central activity clusters (Hyman, 2004).

Figure 2.

Map of Melbourne showing home and work areas of 35 surveyed delivery workers.

While food delivery riders in Melbourne meet other riders while commuting by train, most Uber drivers in Dhaka meet other ridesharing drivers while drinking tea from local tea shops. Tea shops are very popular place in Dhaka to hang around and chat with other people. Most participants said that before the establishment of DRDU, they mostly met other drivers either at teashops or via social media like Facebook. As Anil, a DRDU member mentioned:

When I write from my Facebook account and others write from their accounts, these writings come to our Facebook home. I write in my post about the issues I face. Other Uber drivers might have experienced the same issues, and they comment on my post about their experience. Through that post, I give my number and later organise a meeting. There are two other ways we meet Uber drivers. When we gather at a place [like a tea shop or other public spaces], we approach each other and ask, ‘brother, are you an Uber driver?’ This way, we introduce ourselves to each other. Another option is when we visit Uber office to solve various issues at different times, we can see a dissatisfaction [they are not happy the way Uber treats them and the money they earn driving Uber] among people. We then share our Facebook account or mobile number with other Uber drivers and talk to them later.

This quote demonstrates the importance of public infrastructure not only for accessing public infrastructure but also for creating networks among platform workers.

Negotiating urban infrastructures

The relationship between digital labour platforms and cities is a parasitic one. These platforms take advantage of publicly available infrastructure without reinvesting into them. Shapiro (2021: 106) refers to this as ‘infrastructural appropriation’, a form of exploitation which ‘derives from the inexhaustible utility and accessibility of infrastructure – from infrastructure's publicness’. This excess of value which is inherent to collective urban resources is termed by Shapiro as ‘infrastructural surplus’. Rideshare and delivery platforms make use of roads, but they also use other forms of infrastructure, such as water sources, toilets and electrical outlets. The reason the independent contractors working on digital platforms use these facilities is that the platforms do not recognise their workers as ‘employees’, instead designating them as ‘independent contractors’. As these workers are not recognised as employees, platforms are not obliged to provide them employment benefits, or other resources commensurate with the work required such as office space, toilets, kitchens or Wi-Fi.

Delivery riders, as an extension of the platforms themselves, appropriate urban resources in ways which complicate distinctions between public and privately owned resources. At Hungry Jack's, a fast-food restaurant in Melbourne's Central Business District, for example, riders wait for their orders at benches provided for customers. Here, until the end of 2023, the manager would also let riders charge their phones using the built-in charging stations (Figure 3).

Figure 3.

Delivery rider Hassan charging his smartphone at Hungry Jack’s in Melbourne, 28 October 2023.

Those interviewed described other restaurants in which managers dissuade riders from using their toilets, and others still where staff demand that riders wait outside. As privately owned public spaces, shopping centres present another type of resource. At the restaurant, the toilets only stay open until 9pm, whereas the toilets at the shopping centre across the road are frequented by delivery riders as they stay open until midnight. In each case, the publicness of these infrastructures exceeds the designation of private ownership. Platform urbanism presents a shift whereby publicness is no longer defined by ownership (e.g. the state), but by being accessible for collective use (Richardson, 2018; Shapiro, 2021). Yet this publicness is contested, whether by disgruntled pedestrians who are upset with riders using the footpath, cyclists irritated by the high speeds of e-bikes in bike lanes, restaurant workers making riders wait outside, or residential building managers using posters to dissuade riders from entering their lobbies.

Developing relationships with other kinds of night workers allows riders to negotiate greater access to public infrastructure, and it also makes the job more enjoyable. Hassan describes the Hungry Jack's staff as being nice to him and other riders. ‘And some of them start recognising me so they sometimes give me some free stuff’. Access to facilities is negotiated, and the dynamics of the restaurants themselves influence these negotiations. Hassan had the following to say about the Hungry Jack's manager:

Manager is from Pakistan and he's a nice person. […] Whenever he has the visit from some seniors, he asked us to please go out so they don't say anything. […] We were sitting here and he came to us, he said that if you don't have any orders, please order at least a drink. Then you can sit here because my seniors are visiting, otherwise you have to leave.

Through their behaviour, delivery riders undermine established protocols for city use, and given that there is no sign of delivery services abating, cities will need to adapt to the needs of this new constituency.

Similarly, in Dhaka, Uber drivers negotiate with local petrol pumps to access toilets and park their cars in petrol stations if needed. They also park their cars in front of local restaurants while having breakfast, lunch or dinner.

The body as infrastructure

Platform companies operate as middlemen, connecting service providers with consumers. This connective service relies on infrastructure, in part digital – the apps themselves, and in part physical. If the infrastructural networks defining global trade are comprised of container ships, ports and intermodal terminals, the infrastructure defining food delivery networks are the workers themselves. The body of the worker is the key infrastructure, along with equipment for which they themselves are responsible. In other words, the vehicle, gear, energy, smart phone and any other equipment used by these workers is paid for by the workers themselves, often through predatory rental agreements. The kinetic energy produced by the body of the e-bike rider is essential to the circulation of delivery goods. These bodies must therefore be maintained much like other components of the infrastructural networks which constitute digital platforms. In this respect, digital labour platforms turn delivery riders and ridesharing drivers themselves into resources to be appropriated.

The knowledge that platform workers obtain regarding how to navigate the city, and where to locate essential infrastructures (a knowledge which is in many cases socially informed), is itself a resource which platforms appropriate. For example, in Dhaka, neither Uber nor the government provides any safety networks for Uber drivers. To protect them from hijackers and ensure their safety at night, DRDU members use Telegram – a cloud-based mobile messaging app to provide support to each other. Uber drivers share their live locations in Telegram and post a coded message in the Telegram group if they experience any safety issues. Once the coded message is seen by other drivers, the nearest drivers gather in that location to help the driver. Uber drivers told the first author about two cases where Telegram group played an important role in rescuing a driver and their car from hijackers who ordered an Uber trip and then they were trying to steal the car. Similarly, in Melbourne, migrant platform workers often rely on their social networks to navigate the gig economy. For example, Noresh, an Indian food delivery rider, mentioned how they supported a rider to track a stolen bike:

Some of the guy's bikes have been stolen, you know? One incident happened in Hungry Jack's. The guy's bike got stolen. So, I think there was a tracker on the bike. So, we helped them, we tracked them. But we couldn’t find that bike but the police arrived. I think he reported them, and we all helped him. We went to the location point [which the tracker showed], but we couldn’t find the bike, sadly.

Platform workers also share information using their social networks to help other workers in their community. Aslam, an Indonesian food delivery rider, mentioned how the Indonesian rider community use their social media groups to get information on second-hand bikes or busy area for receiving orders:

I guess some Indonesian people who were already working as food delivery riders, have some social media groups. They use these groups to share everything including information on how to get cheap direct maintenance [for repairing bikes] or how to buy the new bike or the second hand bike or where the busy area is.

As indicated by Aslam's quote, information on how to navigate the city and find busy areas is often shared between platform workers. This socially informed knowledge indicates that not only does the body function as infrastructure, but so too do social groups. Following Simone (2004: 420), the ‘complex relationships of mutual dependence’ that characterise urban life in much of the world, themselves constitute an infrastructure. With infrastructure defined by Simone as ‘a platform providing for and reproducing life in the city’ (2004: 408). In each instance, infrastructure sustains. And in the platform economies of Dhaka and Melbourne, bodily and social infrastructures sustain both the lives of the platform workers, and the platforms themselves.

Conclusion and policy recommendations

This paper demonstrates how gig workers in Bangladesh and Australia navigate the gig economy using existing public infrastructure. This paper also provides new insights into how digital labour platform companies are transforming urban space. The findings provide examples of strategies platform workers utilise to access public infrastructures for parking, water, toilets and battery charging. They often experience numerous challenges when accessing these resources. However, they have to rely on these services as platforms do not provide these services for their workers.

The paper makes three contributions. First, this paper expands our understanding of platforms’ operations in cities and the way they extract capital using a pool of ‘precarious freelancers who are shuffled from gig to gig’ (Sadowski, 2020: 450). In addition to using the knowledge and bodies of precarious workers, our findings demonstrate that digital labour platforms like Uber, Pathao, Deliveroo, Menulog, UberEats and DoorDash rely heavily on public infrastructure to maintain their operations. The presence of digital platforms in urban space has changed how people live and work in cities. This trend, known as ‘platform urbanism’, has transformed the narratives of smart city development by integrating not only data and space but also labour, mobilities, consumption, governance, civic citizenship, urban infrastructure and a range of urban services (Leszczynski, 2020).

Second, this paper contributes to platform and labour studies literature (Barns, 2019, 2020; Sadowski, 2020; Veen et al., 2020; Pollio, 2021; Lata et al., 2023) demonstrating that due to the parasitic nature of platforms, the risk and responsibility associated with platform work is borne by platform workers themselves. While digital labour platforms rely on public infrastructure to maintain their operations, they do so without adequately mitigating the challenges faced by platform workers in their everyday lives. As we have discussed, this can be seen in the way Uber drivers in Dhaka face police harassment due to not having any allocated parking space for Uber drivers. Our findings further point out the importance of access to parking, water points and public toilets, especially for platform workers at night.

Finally, we contribute to existing infrastructure and platform studies literature by introducing the concept of ‘parasitic platform urbanism’. Building on literature on infrastructure and urban infrastructure, platform urbanism, platform capitalism and philosophy (Serres, 2013), we develop the concept ‘parasitic platform urbanism’ to refer to platforms as ‘parasites’. Just as parasites benefit at the expense of the other, or the ‘host’, digital labour platforms rely on cities’ existing public infrastructure without acknowledging this reliance. Because it is indirect, platforms do not acknowledge this reliance, and yet without this infrastructure, platform workers would be unable to perform platform work. This is true not only in Dhaka and Melbourne, but across the globe.

To ensure that platforms take greater responsibility to improve the working conditions of platform workers, the occupational health and safety legislations which protect the rights of employees in Australia and Bangladesh could be extended to apply to platform workers. For example, protections which ensure employees access to clean drinking water ought to be applied to ridershare and delivery riders. Such policy responses would help to ensure that platforms take greater responsibility for the conditions faced by their workers, and that greater investment was placed into urban amenities for platform workers. Bolstering access to key amenities in these ways would help to mitigate ‘parasitic platform urbanism’ and to flatten the differences between platform workers and consumers.

In order to recognise the growing number of platform workers who contribute to the night-time economy, further development of Melbourne's night-time economy initiatives is required, including extending opening hours of public toilets and other amenities. Our findings demonstrate that platforms often deny their responsibilities to provide parking for workers. Local governments are often responsible for allocating parking spaces. However, we argue that digital labour platforms have responsibilities to provide these amenities if they are not already accessible to platform workers. Our interviews with Melbourne City Council's transport department have shown that local government staff are interested in understanding the parking needs of both ridesharing drivers and food delivery riders. However, in terms of allocating parking space for Dhaka's ridesharing and food delivery drivers, we have not seen any initiatives taken by the local authorities or BRTA. As Uber drivers in Dhaka are often penalised due to dropping off their customers in different places, it would be helpful for them if the local authorities provide some parking spaces for rideshare drivers and food delivery riders. To foster a mutually beneficial relationship between digital labour platforms, cities and workers, we recommend that platforms provide financial support to local authorities to assist in developing the necessary infrastructure for platform workers.

Delivery and rideshare platforms have a responsibility to care for those using their platforms for work; however, this responsibility will not be upheld unless governments and unions hold platforms accountable. The difficulties faced by platform workers demonstrate a broader trend of divestment from public infrastructure. Reliance on privately owned and operated delivery and rideshare services reduces the impetus for governments to invest in public versions of these services. To encourage platforms to take responsibility, public and worker-led institutions can develop their own platform worker amenities.

In Dhaka, the provision of collective resources such as public toilets is minimal. A greater offering of services for the general public would also help to meet the needs of rideshare drivers. Similarly, the development of a comprehensive network of bike lanes within central Melbourne would support the safety of delivery riders and fellow cyclists.

Paying greater attention to how platform workers use infrastructure and collective resources provides researchers with greater insights into how both platform workers and platform companies influence the development of contemporary urban space. We also recommend that public service staff and policy makers pay greater attention to how platform workers use infrastructure and collective resources so that they can better design cities that accommodate the infrastructural needs of platform workers.