The state, capital, and worker vulnerability: The case of ride-hailing drivers in Ghana

Abstract

The vulnerability associated with the drudgery of drivers in the ride-hailing enterprise of the platform economy has come under both public scrutiny and scholarly study. There remains, however, a dearth of knowledge around how driver vulnerabilities are produced and maintained, and which actors drive those. This paper contributes to the discourse by unpacking how the political economy of digital capitalism plays out to undermine the fortunes of ride-hailing drivers in Ghana. Using qualitative interviews and focus group discussions, the paper shows that drivers’ vulnerabilities stem primarily from the unbridled control over the means of production, which are the digital platforms and the vehicles, as well as the ensuing unequal power relations between capital owners and drivers as capital producers. The paper also characterises the ambivalent role of the State as a capitalist agent that maintains the status quo, albeit nuanced. The need to interrogate alternatives to augment State regulation is therefore recommended for mediating the relationship between capital owners and drivers as capital producers. Effective alternatives would be reducing capital’s monopoly by replacing private, foreign platforms with local public platforms and strengthening drivers’ collective agency to mediate the excessive power of the capital owners.

Introduction

The tides of digitalisation sweeping across the world are altering the functioning of labour markets and challenging the effectiveness of existing labour market institutions, with far-reaching consequences for labour. Examples of such alterations include the use of online digital platforms or applications to supply both tangible and intangible goods and services (Anwar & Graham 2019; Schmidt 2017) and the use of algorithms that match the labour supply to demand (Drahokoupil & Piasna 2017), captioned digital labour platforms. The coalescing of these digital capital and labour markets, involving owners of the platforms, users or customers, and workers, according to the ILO (2016), constitutes the platform economy. Characteristically, the operating relations in the platform economy are established in this digital space (Prassl & Risak 2016).

The proliferation of platform work has received scholarly attention with mixed reactions. Optimists from a neoliberal lens argue that platform work has provided opportunities for transforming the entire work process towards flexibility. There is enough evidence that digital work platforms have expanded in developing countries with weak economies and are regarded as one of the best options for creating employment avenues for the youth (Graham et al., 2017). It is also argued that the platform economy increases labour supply, especially in weak economies (Vallas 2019). This increase is attributed to wage stagnation and the concomitant decline of standard working arrangements. This is especially the case in Africa, where almost 34 million people were unemployed in 2019, of whom 12.2 million were young people aged 15 to 24 years (ILO 2021). It is also estimated that about 45 million workers registered on various platforms in Western, developed countries and about 25 million registered workers on platforms were based in China (Codagnone et al., 2016; Heeks 2017). In developing countries, estimates indicate that about 25% of workers use platform work to provide or complement their earnings. These benefits notwithstanding, available literature also indicates that, although the growth of platform work provides employment opportunities for the unemployed, the conditions of work provided by these platforms have often been exploitative. The implication of platform work on employment and work is undoubtedly a critical element to consider in assessing whether platform work has been creative or damaging.

While ubiquitous in almost all sectors of the global economy, the transport sector, with its use of ride-hailing applications, stands only second to retailing applications. The transport sector’s ride-hailing service providers in Ghana are among the most prominent platform production sites (Cannon & Summers 2014). This is occasioned by the high uptake facilitated by burgeoning mobile phone use (IWS 2019). The basis of ride-hailing operations is to connect drivers to their customers through digital applications, with all conditions for providing the service determined by the customer’s demands (Cannon & Summers 2014), while the platform owners (POs) determine and build into the operating software, the terms of engagement. The virtual connections of ride-hailing are far from equal. They are laced with economic power relations that emanate from the value placed on the productive resources controlled by some players and desired by others.

Empirical studies that highlight the vulnerabilities of labour ride-hailing drivers, in particular, are replete but generally coalesce around low and insecure income, excessively long hours of work, exclusion from social protection, a representational gap and occupational health, and safety challenges. The common occupational health and safety challenges include road accidents, physical and mental stress with all kinds of body aches and pains, verbal assault, sexual harassment, and criminal attacks from riders (Abraham et al., 2020; Akorsu et al., 2020; Amir & Graham 2020; Bodie 2018; Howard 2017). Akorsu et al. (2020) show how these vulnerabilities were further exacerbated during the government’s COVID-19 management interventions, which could not reach workers without an identifiable employer. Among the factors that drive people to opt for digital work is their existing vulnerability as people with no employment avenues (Vallas 2019). These throw up the questions, such as who are responsible for the production of drivers’ vulnerabilities? How are these vulnerabilities produced? And, what is the ideological base for its maintenance? While some extant literature has contributed to this focus by providing some answers to these questions, these have been mostly based on United States of America and Europe, leaving out the nuanced experiences of Africa (Dubal 2017; Hua & Ray 2018; Rosenblat 2018; Wells et al., 2021). The primary motivation of this paper, therefore, is to add to the discourse by attempting to unearth how vulnerabilities are produced, reproduced, and maintained in a typical African country such as Ghana. In so doing, this paper seeks to go beyond merely highlighting the peculiar vulnerabilities of ride-hailing drivers.

I proceed first, by highlighting the state of the literature that delineates the conceptual positioning of the various players in the digital ride-hailing space. This is followed by a description of the research methods for this study, after which the empirical data are presented and discussed. The concrete emergent issues and implications are presented in the concluding section.

The political economy of the actors in ride-hailing

The emergence of new digital technologies and platform work, particularly, is the new face of the global capitalist system. The current shape and character of the global digital ecosystems bring to the fore the importance of political economy discourses, which have most of their tenets in Marxist perspectives on capitalism. Political economy is useful for unpacking how capitalism operates to determine the players in the digital landscape, as well as the nature of the relationships between and among the various players. Johnston (1988) defines political economy as: who gets what, when, and how? This seemingly simplistic definition raises important questions about (i) how citizens are positioned and/or even considered in policy-making and in the distribution of national wealth, (ii) the role and effects of power and authority on economic choices, (iii) the relationship that exists between the state and the market, and (iv) the relationship that exists between national and international entities. Answers to these questions can be sought in abstracting the social, political, and economic ties that shape control, access, and use of platforms as productive resources and how these produce and preserve domination and subordination.

Historically, we know that control over the means of production does not just enable production but allows the exploitation of the labour of others (Marx 1982). In contemporary capitalism, the means of production are not only just land and machinery but also digital platforms. While digital platforms portend an economic revolution where digitally enabled workers such as ride-hailing drivers will transform themselves into micro-entrepreneurs with more freedom, the resultant production systems are carefully engineered around the private control of the digital platforms, and embedded with values and power relations that limit the freedom of workers. It is reported that the platforms are intentionally configured to prevent these supposed independent contractors from making informed decisions about their work and those who have fewer assets at their disposal (e.g. vehicles) do not benefit from the same flexibility as those with more assets (Akorsu et al., 2022; Holtum et al., 2022). In addition to how control over private assets dictates the extent of precarity, available insights into how social identity markers such as migrant status, class, gender, ethnicity, and even socio-spatial dynamics intersect to worsen further, the precarity of ride-hailing drivers, have been shown (Holtum et al., 2022; Wells et al., 2021). Thus, the traditional socio-economic perspectives for labour market analysis continue to be relevant in the platform economy.

On the one hand, it is the excessive value placed on digital platforms, which Betancourt (2015) describes as having seemingly ‘magical’ properties that gives it the potential to colonise social relationships and valorise social activity and human behaviour. On the other hand, the monopoly power over the aggregation, capitalisation, and colonisation of digital platforms enables what has been called technological hegemony and a reinvention of colonialism in the Global South in the form of monopoly over digital technology by big multinationals, particularly from the United States of America, (Kwet 2019). Such technological hegemony is not only an issue of economic domination but also ideological domination – the worst form of colonialism. This view is also shared by other writers who call it ‘data colonialism’ and argue that it is the foundation of contemporary capitalist accumulation (Fraser 2019; Klerkx et al., 2019). Platform capitalism today asserts a sophisticated system of monetary and social control over physical existence (Srnicek 2017). The need for critical debate on the so-called benefits of the digitalisation agenda to understand further how digitalisation is changing existing business models and affecting almost every sector of the global economy, has been widely recognised (see e.g. Reis et al., 2020).

Another analogue description of data colonialism is ‘platform imperialism’, which is rooted in the argument that such notions of imperialism have gained significant popularity with the rapid growth of platform technologies in the 21^st century (Jin 2015). Historicising the evolution of imperialism in the 21^st century, Jin (2015) cites Lenin’s Pamphlet (Imperialism, the Highest State of Capitalism (1917)) as the focal point of discussion. The above-cited work by Lenin reveals the idea of the domination of monopolies, finance capital, and the export of capital which uses the State machinery to colonise the periphery from the metropole. Just as with Lenin’s analysis, the capitalist uses oppressed peripheral labour to produce cheap commodities and create peripheral elites of the middle class to patronise their commodities and undermine indigenous industries. This approach is reminiscent of the digital platform economy that uses the State as a conduit to exploit drivers who subscribe to these platforms. Hence, digital imperialism plays a significant role in explaining the current power relations between developed and non-Western, less developed countries (Jin 2015). Within the protocols of this new form of colonisation is the commodification process which Prodnik (2015, 233) describes as the ‘crucial preconditions for the general preservation of capitalist-social relations and continuing expansion of capital’. A notable concern is that the process of commodification influences people at the subjective level, altering their social relations while widening economic inequality. The commodification process radically transforms social bonds and values that are off the radar of market exchange, thereby enhancing individualism in societies.

Also characteristic of contemporary capitalist accumulation is the role of nation-states in fuelling their resilience through market-oriented neoliberal policies. According to Xing and Hersh (2006), this occurs in a two-stage process where the State first retreats to allow market liberalisation from political and social control, and in the second stage, returns to political regulation and control of the market. Xing and Hersh (2006), therefore, call regulation pro-capitalist and a reflection of the ‘willingness of the bourgeoisie to accept socio-political and economic reforms’ (38). The reformist nature of regulation does not threaten the existing capitalist status quo. Wedged in this global scheme, national state policies tend to preserve the control of a dominant group amidst economic, social, political, and ideological reforms. State reformism then fuels the resilience of capitalism. The national and international political-economic interests underlying production relations determine the shape and character of the digital ecosystem globally (Munthali et al., 2018).

Consistent with the foregoing is how the influence of globalisation, democratisation, and trade liberalisation has changed the political economy of the local transport system (Htet 2021) and presented obstacles to social progress, including the disrespect for workers’ rights and interest, the disruption of the local economy, and social security of some of the actors of the ride-hailing business (Li et al., 2022). From other perspectives, ride-hailing emerged in Africa as part of the growing economy, but its commodification and informalised nature have resulted in poor job quality (Anwar & Mark 2022; Huws et al., 2017). Technological changes including digital innovations have changed the way people earn, learn, shop, and play, as well as the geography of production and the contours of work. These changes have unfolded in a neoliberal era resulting often in what many see as financial chicanery, unrestrained corporate power, and economic austerity, attributes that have changed the policies of nations, not only allowing capital to escape from even the reformist regulatory oversight, but also restricting the influence of policymakers on how business is run and regulated (Kozul-Wright 2018).

While the above brings to the fore, the importance of regulation to counterbalance the extensive power of the platforms, it also highlights the limitations of State regulation. The need to expand the scope of regulation beyond State standard setting, monitoring, and sanctioning is therefore emphasised. Efforts by the international community (i.e. multilateral accords), self-regulation by private POs (i.e. code of conduct), and workers’ pressure (i.e. collective agreements), have at least the potential to influence the regulatory space (Eyert et al., 2022; ILO 2021). Notwithstanding this, the ILO (2021) cautions that different efforts have different levels of effectiveness, and that codes of conduct cannot prevail over State regulation, for example. Regarding self-regulation, Akorsu (2018) raised concerns that when it comes to the protection of workers, because the same actors who drive the capitalist system of production, which alienates workers, cannot also be responsible for regulation and protection. Thus, workers’ collective action remains pertinent in the regulatory space. As also reported by Akorsu et al. (2022), organising among ride-hailing drivers offers optimism about workers’ agency and the exercise of associational power. The political potential of such efforts, however, is in question without trade union support. Yet, it is alleged that trade unions, in their current state, are incapable of satisfying the needs of the emerging workers of the digital economy (Webster et al., 2021). The subsequent discussions, therefore, attempt to examine the power relations of digitalisation, how different stakeholders are locked in and impacted, and most importantly, how power is or could be more balanced.

Methods

Data for this paper derive from both primary and secondary sources. The overall research design is qualitative, underpinned by constructivist epistemology, and supported by interpretivist analysis of contextually unique experiences. This paradigm emphasises the value of study participants’ ideas as they are expressed and the ability of language to convey meaning (Ritchie & Lewis 2014). Specifically, the study is exploratory, since the issue of interest is just beginning to receive adequate research attention (Swedberg 2018). Data were collected through interviews and focus group discussions (FDGs), using semi-structured guides that allow for in-depth and contextual explanation through probing (Ritchie & Lewis 2014). Three different categories of research participants were purposefully chosen for the study. The first was seven key informants from State-relevant state institutions, the Trade Union Congress of Ghana (TUC), and one platform operating office. These interviews emphasised matters of policy. Second, 17 executives from the associations representing drivers were engaged in two FGDs, eight in Accra and nine in Kumasi, to discuss the needs and working conditions of the drivers. Lastly, 16 individual drivers including one female driver were interviewed about their lived experiences as drivers. There was no predetermined sample size for the individual interviews. Data collection continued till saturation was reached. A total of 40 individuals participated in the study. The fieldwork was undertaken between the 25 and 31 October 2020. This took the form of face-to-face interactions with research participants by the author in Accra. Follow-up data gathering to fill gaps was conducted via telephone interviews between 20 and 23 November 2020. To ensure the trustworthiness of the data, a virtual debriefing and validation workshop involving some of the research participants and peer researchers was held on 30 November 2020.

While the interviews with some drivers were conducted in the Twi language, the English language was predominantly used in the FDGs, interviews with key persons and with some of the individual drivers. The interviews and discussions were digitally recorded, in addition to field notes. Verbatim transcriptions followed translations from Twi to English. The transcripts were subjected to thematic analysis, which refers to identifying themes and then deriving associations, explanations, and relationships to make meaning (Ritchie & Lewis 2014). After familiarisation with the transcripts to understand the content as well as the context of data, using open coding, codes were generated manually to highlight emerging concepts. In the final stage, associations, explanations, and relationships were derived from these themes through an inductive process of drawing on existing literature to establish conceptual claims grounded in the themes as well as drawing from the researchers’ reflexive interpretation of the themes. These are presented and discussed in the next section.

Presentation and discussions of key findings

The data are presented and discussed around the key actors in ride-hailing and how their activities produce driver vulnerabilities. Assumptions include how actor activities or operational structures are not inadvertent but rather deliberate strategic actions to dominate the production process. Ride-hailing brings together five key actors: the State, POs, car owners, riders, and drivers.

The state

As an actor in ride-hailing, the State attracts and maintains the operations of international POs in their respective countries. Despite the disproportionate physical infrastructural base, digital penetration in Ghana is reported to be high. This is not surprising given the government’s adoption of neoliberal orientation, which involves liberalising the mobile telephony subsector and introducing changes to the regulatory framework of the telecom sector as a whole (Afutu-Kotey 2013). The provision of infrastructure to support its operations has therefore been the focus of the government of Ghana (Demuyakor 2020).

When it comes to opening up the local market, the State does so by first neglecting regulation and absolving FDIs of their responsibilities towards local citizens. As one State official indicated,

The Uber introduction was rushed, and they opened the door to other forms of digital platforms, so from the start, we didn’t get it right as a nation so the Ministry of Transport even lags in terms of the policy to regulate them… As for the behaviour of the providers, it is the same nonsense everywhere. If your institutional structures are open, they take advantage of that. (SO 27/10/20).

The drivers on their part expressed cynicism about the government’s commitment, insisting that the behaviour of the platform operators emanates from some official protection they enjoy. The drivers asserted: ‘There is no regulation. Unfortunately, we had a meeting with the ministry of communication, and surprisingly they don’t know what the ride-hailing companies do and it is so unfortunate’ (Fi, 30/11/20). It is important to note, however, that previously, local transport had been regulated at local government levels. Since the ride-hailing companies are private commercial services, regulation has been around revenue collection and fare fixing rather than employment regulations or protection. What is worrying about the current platform relationships is the power dynamics with their associated exploitative tendencies. The second means by which the State has opened up the market has been by giving tax exemptions to foreign investors. For example, even though Uber has been operating in Ghana since 2016, Uber B.V. has now been registered with the Ghana Revenue Authority (GRA), and so, is expected to charge riders 21.9% Value-Added Tax, which will be remitted to GRA from 8 May 2023. It is still not clear what Uber’s tax commitment might be.

Thus, though the State does not invest in the platforms, its role depicts the first of the two-stage process described by Xing and Hersh (2006), and in which the State first retreats to allow liberalisation of the market from political and social control. Though there is no regulation in Ghana yet, Akorsu et al. (2020) report that policy discussions have started in Ghana amid angst that organised labour was not represented. Rather, stakeholders involved include the local government, Driver and Vehicle Licensing Authority (DVLA), the Ministry of Transport, and the Ghana Communications Authority. This delayed reaction to pro-worker regulation typifies the second stage, at which the State returns to political regulation of the market as a reformist solution (Xing & Hersh 2006). The non-inclusion of organised labour in the ongoing policy discussions also confirms the pro-capitalist character of the State. Indeed, the drivers expressed cynicism about the government’s commitment, insisting that the behaviour of the platform operators is backed by official protection. They were also alarmed that the initial State priority was not to regulate the sector but rather collect taxes from drivers through the attempted levying of driving licences by the DVLA.

The platform owners (POs)

The POs are international businesses operating in different parts of the world with global economic interconnectedness that goes well beyond the framework of nation-states. Their business model entails connecting drivers with clients via an online application, with the terms of the service being governed by customer location and demand. Customers feel secure using ride-hailing apps, which increases demand for the service. According to a profile of online ride-hailing services, there are several platform operators in Ghana. Among them are Uber, Bolt, Yango, Dropping, Swift Wheels, Daily Care, and Fenix. The first three named companies are the most popular platforms. Uber Technologies Incorporated, an American company, was to first to be launched in Ghana in 2016; Bolt, formerly Taxify, an Estonian company also started operating in Ghana in 2017 and Yango, originating from Russia, was introduced into the Ghanaian market in 2019. Though they are competitors, they have similar characteristics, typical of capitalists. In terms of interests, the POs prioritise profit maximisation and wealth creation over and above all other factors in the production system. This economic interest is sustained by their control over the most critical productive resource, the platforms or operating software. Kwet (2019) argues that the monopoly over digital technology by big multinationals, particularly from the United States of America, gives them economic domination over those dependent on such digital resources. This power, in turn, enables them to dictate a certain operational structure that benefits them and constrains the drivers and their competitive tendencies. The POs determine and then build the terms of engagement into the operating software. The operational structure, in particular, is the main avenue for creating driver vulnerability. The subsequent subsections provide examples.

Control over employment status

By maintaining an arm’s length relationship with the drivers and relentlessly referring to them as ‘partners’, the POs obfuscate drivers’ employment status. We found that the drivers do not perceive the relationship as a partnership as expressed in these words. ‘Well, they call us partners… but they do not treat us as partners if they did, they will have discussions with us and will not be blocking us’ (Fr, 27/10/20) and:

I don’t even know our status, because if we look at the freedom or the opportunity to decide when to work, we would think we are on our own, but then if you’re denied access, you are not on your own. So I would say we have been employed either by Uber or Bolt indirectly (Eu, 29/10/20).

Consequently, the POs are absolved of all employer responsibilities including the provision of social protection – the fundamental source of drivers’ vulnerability. The lack of clarity about drivers’ employment status also inhibits efforts to secure interest representation, especially when it comes to the cover of institutional backing for claiming rights as workers.

Control of work hours and intensity

The operating platforms have no in-built mechanism for drivers to rest in between rides. By ensuring that the platform is always accessible, the POs exert subtle pressure that ensures a measure of persuasive continuous driving and thus, dictate tacitly, drivers’ work hours. While Uber allows for a maximum of 12 hours of continuous driving, this itself contravenes the legal requirements of both the Ghana Road Traffic Act and Act 683 of 2004, which stipulate a 30-minute rest after every 2 hours of continuous driving and an hour rest after every 4 hours of continuous driving. Like many laws in Ghana, the Road Traffic Act is not enforced and neither is the Labour Act 651 (2003), which specifies a maximum of 8 hours per day. In addition to the disregard for these legal provisions, the POs use the promise of bonuses to pressure drivers to work longer hours. The payment of extra financial incentives is tied to a specified number of trips, within a specified duration and the content of each of such promotions varies. This further intensifies drivers’ work intensity as clearly evident in the words of one drover:

The nature of the work is such that you need to work more to earn more so I sometimes sleep in the car, I have my toothbrush and paste in the car, and a small fan to drive away mosquitoes … Friday, Saturday and Sundays are so busy, you will not even have time to eat (Am, 26/10/20).

The drivers are conscious of the health and safety implications of the excessively long hours of work but are locked in due to the absence of alternatives. This situation is not just found in Ghana but has been reported in the literature by many authors (Abraham et al., 2020; Akorsu et al., 2020; Amir & Graham 2020; Bodie 2018; Howard 2017).

Control of remuneration

Closely tied to the hours and intensity of work is the app-mediated remuneration model that determines drivers’ earnings. The data reveal that, while charging low fares as a way of attracting customers, Uber, Bolt, and Yango retain 25%, 20%, and 15%, respectively. This is completely regardless of the other production costs incurred by drivers such as vehicle renting, fuel, and data bundles. With this model, the standard 8 hours of work does not guarantee the earning of the minimum wage and as indicated by a driver, ‘if you decide to work let’s say eight hours a day, and take days off, you won’t make enough money to cover the sales, data, credit and so to make money, we need to work for longer hours’ (Eu, 29/10/20). The current National Daily Minimum Wage (NDMW) is GH11.82. For drivers to earn anything close to the NDMW, they need to work not less than 15 hours a day and 7 days a week as well as subscribe to the offer of promotions from the POs. The income insecurity among ride-hailing drivers is the product of deliberate profit-seeking behaviour, which is also consistent with the assertion that POs drive down prices almost to a point that generates precarious existence for their drivers (Cunningham-Parmeter 2016).

Control of work management

The app-directed surveillance of drivers’ activities in the course of discharging the service takes away their freedom. Ride requests are veiled in ways that do not give the driver all the information he/she needs to make informed economic or even safety decisions. For example, in a typical request, the rider’s identification and destination are withheld from the driver. This deliberate manipulation forces drivers to accept all requests, even when it is not economically beneficial, a form of forced labour. Drivers have complained that in some instances, the distances travelled to reach riders are longer than the actual trips, yet only the actual trips are charged to the rider at the expense of drivers. Even worse, the drivers’ inability to know the identity of the riders in ‘blind requests’ has been a major source of security risks, according to the drivers. Drivers in Accra all expressed the strong view that their security risks could be minimised if the POs allowed rider identification and location on the one hand, and following from that driver discretion in choosing trips on the other. Another control mechanism that the platforms enable and encourage is the recording of rider assessment of drivers. This puts drivers under excessive pressure and have even led to abuse from riders. As one driver noted, ‘Some riders do not respect the drivers at all, they are very intolerant of our mistakes, use rude language towards us’ (Jo, 26/10/20).

The car owners

The study revealed that many of the drivers utilise vehicles that they do not own. Rather, the owners of the cars used for ride-hailing are usually local entrepreneurs who invest in the supply of the vehicles. As capitalists, they own and control vehicles, another critical resource that drivers need for their work. As such, the car owners are driven by a profit motive, although sometimes via complex paths. Online ride-hailing uses small automobiles with low engine capacities of 1.0 to 1.3 litres. Examples include the Hyundai Getz, KIA Morning, Toyota Yaris, and Toyota Vitz. These cars are frequently chosen because they use less fuel, which implies more earnings and cheaper operating costs. Some local private investors, referred to as car owners, purchase the vehicles for drivers under two types of arrangements, both of which are rarely governed by written contracts. The first arrangement entails renting the vehicles and making weekly payments of about GH400. Drivers refer to this form of payment as the ‘sales’ option. The second arrangement, referred to as ‘work and pay’, calls for weekly payments to the car owner for 2 years, after which the driver receives ownership of the vehicle. The drivers prefer the ‘work and pay’ arrangement for two reasons, even though the prevailing market prices of the vehicles have often doubled, and the ownership is often only acquired after the vehicles have deteriorated. According to the drivers, this arrangement is preferable because they eventually get to own the vehicle, while still making weekly payments. All of the drivers interviewed insisted that ride-hailing can only be profitable with vehicle ownership. ‘We work long hours to enable us to pay for the cars. To cut everything short, the “work and pay” arrangement is killing some of us drivers’ (Mo, 27/10/2020). The unanimous view is that unless all the drivers own the cars they use, long hours of driving cannot be stopped because defaulting in the payment means losing the car. Thus, the profit-seeking behaviour of car owners creates a second level of capitalist exploitation that worsens the plight of drivers.

The riders

In ride-hailing, the riders are the passengers who are transported in the cars to their chosen desitinations. The riders come into close physical contact with the drivers. They are treated as clients by the POs, who share a measure of their power with riders to boost their competitive edge. Riders engage in virtual communication with the platform administrators and gain control over the drivers as a result. The ability of riders to rank or evaluate the level of service provided by drivers is the unique source of riders’ power. Our examination of the data shows that POs depend on user reviews to maintain their competitive edge in the ride-hailing business. Regarding riders, the drivers say:

They call you, you get there and they snatch your phone. I am a victim. My phone was snatched and so I had to get a new one to continue working. In fact, they snatch our phones, cars and money and get away with it because the Bolt allows them (Am, 26/10/20)

They treat the riders better than they treat us…they are more interested in the riders…they give bonuses to the riders, you drive a rider on a trip and at the end it is GH0.00 and the rider would not have to pay anything…they disclose our identity to riders but not theirs [to us] and that is the source of our security issues. (Na, 27/10/20).

According to drivers, verbal abuse from riders makes them feel worthless sometimes, because they feel trapped and unable to respond strongly to difficult situations for fear of being scored poorly by the riders. The fact that these events are caused by the riders’ ability to rate drivers and the potential consequences of blocking from the POs is the source of many of the vulnerabilities regarding verbal abuse, sexual harassment, and criminal attacks. In these cases, driver vulnerabilities ensuing from rider behaviours can nevertheless still be attributed to POs, who bestow much greater power resources on the riders.

The drivers

The ride-hailing drivers are predominantly male. While a few females engage in it, the general view, even for the female driver, is that it is too demanding for females. Of the male drivers, the oldest was 47 years of age and the youngest was 27 years of age. The drivers were mostly secondary school leavers with almost no university graduates. Within the interviewees, there was only one university graduate who had been unemployed for 7 years, whereas most of the drivers were also engaged in other livelihood activities before commencing work in online ride-hailing. The principal reason for engaging in platform work was to earn a living, driven mainly by job losses, particularly after COVID-19’s consequences and layoffs. While few respondents still maintain a second economic activity, most of them indicated that their online ride-hailing was their primary source of income. One driver, who had previously been a traditional taxi driver, intimated that he was quitting ride-hailing because the taxi business was more lucrative given the absence of service charges and compulsory low fares.

So overall, these drivers are workers, the productive forces in the production relations but so obfuscated by the powers and structures of POs, that they lose their identity as workers. The drivers undertake the service provision as a way of earning a living. The conditions under which this is done are determined and carefully controlled by the POs. While one driver admitted categorically that: ‘I don’t even know our status (Eu, 29/10/20), others think of themselves as partners, entrepreneurs, contractors and even as employees as in the case of Am, 26/10/20, who says: “I think I have two employers, the app operators and my car owner”. Yes, I consider myself a worker’. This overt confusion of drivers’ status is arguably a purposive and deliberate tactic of POs, since it absolves them of any responsibility towards drivers, such as would be required if they were employers. It is this deliberate arm’s length relationship created by the POs that leaves drivers in a precarious state. It is a situation which is not helped by the State’s ambivalent role, most notably the absence of state employment protection, which thus maintains the precarity produced by the POs and the car owners.

Overall, then, the relationships among the actors of ride-hailing are hierarchical and unequal (see Figure 1). The ability of the POs to control both drivers’ access to, and continued usage of, the platform, as well as their profits, soundly subjugates drivers to POs. It was discovered that due to the control over another vital resource, the vehicles, car owners are likewise powerful actors, especially in relation to drivers. This highlights the significance of vehicle ownership as a source of power in ride-hailing, creating a second level of exploitation among drivers. Vehicle owners are important stakeholders in ride-hailing operations. The ability of riders, the passengers who come into close physical contact with the drivers, to rank or evaluate the service provided by drivers is the unique source of riders’ power over drivers, bestowed to them by the POs in order to gain competitive advantage. My argument is that riders only serve as an avenue of control for the POs. Though there seems to be no interaction between the two capitalist actors, the POs and car owners, both dominate drivers but in different ways. The POs exert control through the platforms and also through the riders, while the car owners exert pressure using their high rates of hire or purchase, without necessarily controlling the working lives of drivers directly. Moreover, whatever power is yielded in all these relationships is enabled and strengthened because of the pro-capitalist role of the State.

Figure 1. The power relations among ride-hailing actors.

Source: Adapted from Akorsu et al. (2020).

Conclusions and policy implications

Ride-hailing as an economic enterprise is operationally structured to bring together the various actors into a somewhat hierarchical relationship that is not value-free but determined by control over the means of production. The preceding analysis and evidence have shown that the platforms, as a technological resource and the central means of production, which are under the control of private ownership, are fundamentally but not exclusively responsible for the plight of ride-hailing as a site of poor working conditions. Through the platforms, the drivers locate passengers, access directional maps, find the best route with fewer traffic jams, and derive some value from the rides. It is drivers’ desire for, and dependence on, these elements of the platforms that subjects them to manipulations of the POs. They have the capacity for strategically controlling, through the remuneration model, how long drivers work and how much production costs are pushed to drivers and client patronage. The POs can also appropriate profits, making it unnecessary for them to invest in other relatively less profitable means of production, such as cars and cell phones. The POs abjure all investment in the ownership and control over vehicles, since it is not through vehicles that the most profit appropriation is assured. Yet, vehicles remain another important productive resource. Without vehicles, there cannot be rides. Still, with the excessive value placed on the platforms, or rather, through PO’s technological manipulation, vehicles seem a second-order means of production. Thus, the private investors who own and control vehicles also present the typical capitalist profit-seeking tendency in ways that exacerbate drivers’ vulnerability.

I, therefore, conclude that the creation of drivers’ vulnerabilities occurs as part of the capitalist order and perpetuated by agents including the State – though in a more complex and nuanced way. By investing in digital infrastructure and opening up the local market to foreign investors, without a commensurate regulatory framework, the State inadvertently offers an enabling environment for the influx of foreign platforms. More profoundly, however, it is the State’s inertia to provide a regulatory and protective framework that makes it culpable rather than any proactive, explicit, or direct government involvement. Indeed, this is an inertia emanating from the fact that the State itself is subsumed by the global market hegemony and lacks the ability to appreciate ride-hailing platforms for what they truly are – controlled pseudo-markets. For the drivers, however, the State’s inertia appears due to corrupt, self-seeking transactional relations between state officials and the POs. Unlike the State, the POs and the car owners directly plant and water the seeds of driver vulnerabilities. The POs do so through the control of the platforms, while the car owners do so through the control of vehicles, both essential productive resources for the drivers. The belated response of the State in starting policy discussions (and without consultations with organised labour as a viable stakeholder with a countervailing force) hardly threaten the foundations of the economic order and point to the reformism earlier alluded to. With such reformism, the vulnerabilities created by capital are thus maintained by the State, characterising itself as a pro-capitalist institution.

This paper brings to the existing literature some nuanced experiences in the ride-hailing space in Africa, which Ghana typifies, and which signifies most notably, the role of the car owners. Unlike their counterparts in the North, drivers are constrained in their ability to own cars, either through outright purchase or formal hire purchase systems, and this adds to their vulnerability. Also, the paper goes beyond merely reporting the vulnerabilities of ride-hailing drivers to unearth how the vulnerabilities are produced, reproduced, and maintained and by which capitalist agent. This paper has highlighted the reformist character of State regulation and alluded to literature on the limitations of other regulatory forms such as multilateral accords, self-regulation by private capital, and inefficacy of the traditional trade union strategies.

Regardless of the limitations of regulation, I recommend that under the circumstances, there is the need for a coherent policy that regulates the relationship between capital owners and drivers as capital producers. Regulation, however, can serve its purpose of ensuring that the rights of the ride-hailing drivers are protected, only if and when organised labour is involved in the policy discussions. Still, even with regulation, I doubt a transformation in the manner predicted by Marx is possible without alternatives. Alternatives other than rules that simply codify existing power relations would be more promising in empowering marginalised drivers. Examples include the creation of local public platforms to replace private international platforms and drivers’ collective agency, both of which offer enormous prospects and optimism to countervail the excessive power of POs. These suggested alternatives are certainly not sacrosanct. While the former would require a strong political will, the latter would require a strong political potential beyond the expression of associational power to be effective. These, however, remain important at least for further scholarly interrogation.