Offline contexts of online jobs: Platform drivers, decent work, and informality in Lagos, Nigeria

Funding information Katarzyna Cieslik acknowledges the support of Philomathia Social Science fellowship and grant. Summary Motivation: The Sustainable Development Goals targets include decent work for all by 2030 but progress in subSaharan Africa has been slow. Over the past five years, the platform work sector (ehailing platforms in particular) has expanded considerably on the continent, providing work opportunities to the growing urban populations. The quality of this work, however, is heavily contested. Purpose: We deepen and extend our understanding of the gig economy in subSaharan Africa and assess its potential for creating decent work. Our unique contribution lies in looking at the gig economy in comparison to the local threshold (the informal economy) as opposed to wage employment. Methods and approach: We first review the literature on the gig economy, the informal sector, and decent work in lowand lowermiddleincome countries. We then draw on a case study of ehailing in Lagos, Nigeria, to assess the sector’s potential to create decent work. We discuss our findings against the four pillars of the International Labour Organization’s Decent Work Agenda: productive employment, labour standards, social protection, and social dialogue. Findings: We find that ehailing platforms may offer certain gains compared to regular taxiing in the informal sector but these are contextand platformspecific. Examples include access to insurance, credit, and cash transfers. Gig work remains embedded in the informal economy: it is unregulated and does not abide by labour standards nor offer social protections. We also find evidence of wellorganized labour movements. Policy implications: Ehailing may be productive work, as platform intermediation facilitates outreach and matching, translating into increased wages. This happens with the added cost of longer working hours in the absence of labour standards. We propose exploring the role of digital records kept by platforms (e.g. income volume and regularity, or customer ratings data) to facilitate the progressive transition of transport workers towards formality. We also recommend fostering industry standards (such as fair work certification) to protect workers’ rights.


| INTRODUC TI ON
The gig economy-a free market system in which employers contract independent workers for short-term engagements through virtual platforms-is the most common form of online labour (Collier et al., 2017;Heeks, 2017a).
Often associated with the casualization of the workforce, gig work has recently captured the attention of both the researchers and the policy-makers, following the escalating protests of Uber and Deliveroo drivers in both Europe and North America (Aloisi, 2016;Fleming, 2017;Marcano, 2018;Prassl, 2018). Despite some recent literature, we still know little about gig workers and employers in low-and lower-middle-income countries (L&LMICs), 1 where numbers of registered users are steadily growing (Kässi & Lehdonvirta, 2018). In Europe and North America, the gig economy is believed to disadvantage its workers, as it lacks the security and benefits of wage employment (Berg et al., 2018;De Stefano, 2016). In sub-Saharan Africa (SSA), however, the growing unemployment and underemployment, coupled with the rapid spread of information and communication technology (ICT), puts some of these downsides into perspective (Monteith & Giesbert, 2017). Compared with the uncertainty and inefficiency of the informal sector, which currently supports up to 90% of the working population across SSA, gig platforms may provide a valuable source of income (Kuek et al., 2015). Against this background, we ask: what is the potential of the gig sector to create work opportunities in sub-Saharan Africa? Can gig work qualify as "decent work" against the local benchmark? Is gig work a "virtual sweatshop" or a "freelancer's paradise"?
In this article, we review the existing literature on the gig economy focusing in particular on the quality of work (Berg et al., 2018;D'Cruz, 2017;Heeks et al., 2020;Wood et al., 2019). Our unique contribution is looking at gig work against the backdrop of the informal economy, which is where the majority of the working population in L&LMICs earn their wages (Chen & Carré, 2020;Dell'Anno & Adu, 2020). We use an illustrative case study of the e-hailing sector in Lagos, Nigeria, based on a semi-structured survey of 138 e-hailing drivers. E-hailing platforms are online matching platforms that connect contract drivers to prospective customers via the means of a smartphone-based app, set up the conditions of transaction, and collect customer ratings. The case of Nigeria is particularly relevant in view of the recent attempt of the national government to regulate the e-hailing sector (Olowogboyega, 2020). We compare and contrast the views and opinions as well as the lived experiences of the e-hailing drivers with the existing research on workers in the informal sector in Nigeria, highlighting the relative advantages and disadvantages of platform affiliation.
We find that gig work has the potential to provide a satisfactory source of main and supplementary income to e-hailing drivers by facilitating outreach to clients and formalizing minimum fares. At times, platform affiliation offers access to benefits, such as insurance, credit, and cash transfers. However, as part of the informal economy, gig work remains unregulated and does not abide by labour standards or offer social protections (Collier et al., 2017). The oversupply of labour tends to drive down the minimum service fees, progressively lowering wages. Importantly, we also find substantial evidence of workers' agency, especially in relation to collective organizing and industrial action. This is consistent with the previous research (Beerepoot & Lambregts, 2015;Enwukwe, 2021;Lehdonvirta et al., 2019;Wood et al., 2018, p. 19). Our contribution is as follows: First, we establish conceptual clarity over the existing definitions of geographically tethered and untethered forms of platform work, distinguishing between gig jobs that require varying levels of skill and computer literacy. We then narrow our focus to on-demand transport workers as the most common form of gig labour in SSA.
Second, we look at the gig economy's potential to create "decent work" against the four strategic pillars of the International Labour Organization (ILO) Decent Work Agenda (Berg et al., 2018;Burchell, 2014). These are employment creation, labour standards, social protection, and social dialogue (ILO, 1999). Importantly, we assess gig work against the local threshold, i.e. the informal economy, which is where the majority of the working population 1 The notable exceptions are recent publications by the researchers from the Oxford Internet Institute and Manchester University Centre for Digital Development, most of which are quoted in this article.
of Nigeria find work. In so doing, we respond to the problem of "universalism," as posited by Heeks (2017a) andD'Cruz andNoronha (2016) who argued that labour markets and labour market institutions in L&LMICs are typically quite different to those in developed economies.
Third, we discuss the potential role of platforms to facilitate the progressive transition of transport workers towards formality. We propose exploring the role of digital data records kept by platforms (such as income volume and regularity or customer ratings) to facilitate transition towards formality while introducing industry standards (such as fair work certification) might translate into a higher probability of value creation and sustained growth.
With this article, we aim to bridge the insights relating to gig work from labour studies with the broad scholarship on work and employment in development studies. Even though more locally owned platforms are emerging in Africa (Johnson, Bester et al., 2020), the dominant pattern of workflow in the gig sector entails outsourcing labour from Europe and North America to Africa, Latin America, and Asia, in a process that some scholars call labour appropriation (Gandini, 2019). Online labour platforms allow employers and clients to access the largely unregulated oversupply of labour, in a set-up that benefits the platforms first, then the clients, and only then the workers in relative terms (De Stefano, 2016). This reinforces the long-standing inequalities "between capital and labor and, given the pattern of trade, between global North and global South" (Heeks, 2017b, p. 17; see also Casilli, 2017).
Considering that both the United Nations Development Programme and the World Bank (2016) actively support gig work as a strategy to boost employment and foster inclusive growth in L&LMICs, it is important to consider the implications of these emerging patterns for the political economy of labour (Roy-Mukherjee et al., 2020).
The article is structured as follows: in Section 2, the theoretical part of the article, we define and explain the different types of gig work, providing examples of relevant platforms. In Section 3, we focus on e-hailing as the most dynamic gig sector in present-day Africa, and present the case study in Section 4. Based on both our data and the literature reviewed, we discuss the pros and cons of platform work and consider the policy options for the sector in Section 5. Section 6 concludes.

| G I G WORK , ONLINE WORK , PL ATFORM WORK-DEFINITIONAL QUANDARIE S
The emergence of platform work has brought about transformations of labour relationships and dissolution of work boundaries. In this section, we briefly explain the different types of platform work, outlining the core features, attributes, and main occupational groups.
The term "gig economy" refers to short-term labour market activities that are co-ordinated via digital platforms (Heeks, 2017a(Heeks, , 2017b, known as "platform work." Geographically tethered jobs are largely manual and include occupational groups such as personal and commodity drivers, domestic workers, and delivery couriers (Marcano, 2018) (Figure 1). Recently, more on-demand platforms specialize in service matching for highly skilled freelance workers such as models, photographers, make-up, and nail artists and hairstylists (Berg et al., 2018).
Untethered jobs are also highly diverse in terms of skills required. We distinguish between online freelancers (or e-lancers) trained in highly specialized tasks (Wood et al., 2018) and "micro-workers," ordinarily performing menial tasks such as tagging or image recognition (Mtsweni & Burge, 2014;Irani, 2015). A number of researchers pointed out that microwork is the most fragmented, deskilled, and commodified type of gig work (Bergvall-Kåreborn & Howcroft, 2014;D'Cruz & Noronha, 2016). The category of e-lancers is closely related to outsourcing (also called "microproviders" by Lehdonvirta et al., 2019) and covers highly skilled professionals, often specializing in software development, copywriting, content writing, translation, sales, and marketing support (Kässi & Lehdonvirta, 2018). Finally, "crowdwork" ("crowdsourced work"), a type of gig work that evades categorization, refers to highly labour-intense tasks carried out by an undefined network of online labourers (Ford et al., 2015) ( Table 1).
It is crucial to note that both the quality and the quantity of platform work vary greatly between different types of gig and occupational groups. Some enjoy better opportunities, while others come with the burden of higher risk. For this reason, when assessing the quality and weighing the trade-offs of gig work it is necessary to consider the distinctions outlined above. In this article, we use a study of e-hailing (defined as taxi services mediated by an online platform: considered a geographically tethered, low-skilled job) as the most common form of gig work. Before presenting the case study, we discuss the characteristics of the local context of informal labour, as well as the current youth employment crisis and its impact on both the quality and quantity of work in the context of SSA.

| Africa's youth employment challenge and the promise of ICT
According to the estimates, the populations of many countries across SSA are forecast to continue to grow until at least 2100 (Ezeh et al., 2020). While demographic growth may translate into economic development, this is contingent on the availability of productive employment (Cieslik et al., 2021). Unfortunately, at present, of the 420 million young adults in Africa more than 140 million are unemployed and another 130 million are underemployed (African Development Bank, 2016). Each year, more than 12 million young people seek formal employment but fewer than 3 million wage jobs are created, a crisis that is likely to deepen as a result of the COVID-19 pandemic (ILO, 2020).
While "full and productive employment and decent work for all" have been entrenched in the Sustainable Development Agenda 2030 (Goal 8), increasing youth employment requires substantial economic restructuring (Sumberg et al., 2021). In the meantime, governments are investigating the potential of virtual technologies to create, improve, and sustain labour market growth . The gig sector is seen as an opportunity to harness the "digital dividends" (AUC & OECD, 2021; Kuek et al., 2015;World Bank, 2016).
Even though it is still relatively small in size, it has been growing rapidly in terms of volumes of transactions and worker numbers (Heeks, 2017a;Kässi & Lehdonvirta, 2018).   Samasource, CrowdFlower, oDesk, and Elance (the latter two having merged in 2015 to form Upwork), and the Microwork for Jobs Creation, as well as Kenya's Ajira Digital Programme. In the next sections, we first describe the context of these initiatives and then discuss the criteria for assessing whether "gig work" can count as "decent work" against the local benchmark, namely the informal sector.

| The local benchmark: Work in the informal economy
Currently, the informal economy employs 60% to 90% of the working population in Africa (Chen & Carré, 2020).
While definitions of informality vary, ILO sees it as "all economic activities by workers and economic units that are by law or in practice, not covered or insufficiently covered by formal arrangements" (ILO, 2014). The term covers both: enterprises that are not legally regulated (registered) and employment relationships that are not legally regulated or protected (without contracts) (Webster et al., 2016). Typical employer-employee relationships are not common in the informal sector, as contracts are casual, transactions may include barter, and social capital often serves as collateral. As a result, the informal economy is unregulated and untaxed, with no labour standards or social protection. This translates into increased vulnerability and precarity of labour (Meagher, 2016). At the same time, the lack of bureaucratic and hierarchical structures and the flexibility of social contracts make the informal sector cost-effective, with low entry barriers and fast returns on investment, providing vital income opportunities to the urban poor (Otekhile & Matthew, 2017).
Informality still dominates in SSA, and includes street vendors and family firms, as well as services and smallholder farming (ILO, n.d.). In the next section, we present "decent work" as a core concept and an active agenda of the ILO, and explain how it applies to informal economy settings.

| What counts as decent work? Productive employment, labour standards, social protection, and social dialogue
In 1999, the ILO set up the Decent Work Agenda, promising to strive for "decent work for all." The programme aspires to transform all forms of work, including self-employment and the informal sector (ILO, 1999, see also Vosko, 2002). The ILO defines "decent work" as: work that is productive and delivers a fair income, security in the workplace and social protection for families, better prospects for personal development and social integration, freedom for people to express their concerns, organize and participate in the decisions that affect their lives and equal- Notwithstanding such concerns, at the 90th Session of the International Labour Conference in 2002, the ILO adopted a resolution containing Conclusions Concerning Decent Work and the Informal Economy (Trebilcock, 2005), which builds on four foundational pillars: productive employment creation, labour standards, social protection, and social dialogue (Deranty & MacMillan, 2012).
Productive employment is the kind of work that yields sufficient returns to labour to allow workers a stable livelihood that is above the poverty line. "Adequate" living wages are an important feature of productive employment, but interpretation varies between contexts. Productive employment is also about access to inputs that improve workers' capacity to find and benefit from productive opportunities, such as skills training, leading to career advancement (Szirmai et al., 2013). Both formal and informal work can count as productive employment, but earnings tend to be lower in the informal sector (ILO, 2009).
Labour standards are regulations that safeguard the decent conditions of work, including workers' health and safety, working hours, fair treatment, and non-discrimination (Locke, 2013). Even though the core characteristic of the informal sector is that workers' rights are not protected by law, some informal employers may be coerced into honouring labour standards by economic forces (such as competition for labour) or cultural codes of conduct (such as moral economies) (Bonner & Spooner, 2011).
Workers in the informal sector also tend to not have access to social protection. Social protection includes both social insurance (contributory schemes protecting workers against life-course shocks such as illness and old age, covering formal workers) and social assistance (policy instruments designed to address poverty and vulnerability, covering all citizens). Actors involved in social protection include governments and employers, but also workers themselves (through workers' credit and insurance associations, also known as informal social protection) Finally, social dialogue refers to the workers' right to express themselves in the workplace concerning matters which affect them directly. Social dialogue includes the right to information, but also representation,

| E-hailing in Lagos, Nigeria
In order to illustrate some of the above-mentioned processes in context, we use the recent case study of e-hailing gig workers in Lagos, Nigeria (geographically tethered, on-demand taxi drivers). We build on the findings from our recent research collaboration with insight2impact 2 and Lagos Business School, carried out between March and April 2020. The case study comprised desk research, followed by a random survey of 138 e-hailing drivers. The objective of the study was to investigate the nature of worker engagement on e-hailing platforms and to understand how platform intermediation affects work opportunities for drivers in Nigeria.
We chose Nigeria as a strategic site as one of the largest e-hailing markets in Africa. E-hailing platforms first entered the Nigerian market in 2014, with Uber starting its operations in Lagos, and Bolt (formerly known as Taxify) in 2016. Uber extended its operations to the national capital, Abuja, in the same year. Since then, the coverage of the e-hailing platforms has expanded across major cities in Nigeria, taking advantage of the country's 2 Insight2impact was a joint initiative of FinMark and the Center for Financial Regulation and Inclusion (Cefri) supported by the Mastercard Foundation and the Bill & Melinda Gates Foundation. The programme is no longer running: https://cenfri.org/topic/ insig ht2im pact/ relatively underdeveloped public transport system, high population, and rapid urbanization (Meagher, 2018). The market now comprises a few big international players, as well as several smaller home-grown platforms (Table 2).
Against the background of the platform ecosystems in Africa at large, Nigeria had 21 e-hailing platforms operating by 2019, outscored only by Kenya in terms of number of operators in a single market.
We selected a random sample of 138 e-hailing drivers, working for different-and often multiple-platforms.
The majority of drivers in our sample were under the age of 35 and self-employed, suggesting that this avenue of work has been an attractive option for young adults. This is largely due to the promise of flexible working hours and low barriers to entry. Among older drivers, a small number of respondents reported being unemployed prior to e-hailing, which suggests that e-hailing platforms are also playing a role in bringing unemployed individuals back into the productive workforce. More than half of the drivers in our sample held a tertiary qualification certificate, while 30% completed secondary school and only 3% were educated just up to a primary-school level. This suggests that the industry offers an alternative work opportunity to educated youth who are unable to secure formal sector employment.
Contrary to the e-hailing drivers in advanced economies who typically own their vehicles, over 50% of drivers in our sample were driving a leased car (Berger et al., 2019;Meagher, 2018). The participation of women is very low, suggesting that the industry still has some way to go to become gender-inclusive. Only 1% of the drivers in our sample were women, which is consistent with existing estimations (Berger et al., 2019;Meagher, 2018 income-earning merchants are women. Driving careers in Nigeria are considered unsafe for women which explains low uptake, but gender stereotypes also play a role. While there is some evidence that online gig work has amplified opportunities for women, most scholars agree that gender discrimination transcends online/ offline barriers (Galperin, 2019).
All the respondents in our study expressed an informed consent to participate. The nature of the information gathered was non-sensitive and all data was anonymized and coded. Below we present our findings regarding our respondents' perceptions on the quality of work in the gig sector.

| Gig work as decent work
In relation to the four pillars of the Decent Work Agenda, our survey brought mixed results.
In terms of productive employment, our respondents' declared wages up to five times the national minimum wage (NGN 30,000 per month for Nigeria), with the average monthly income of NGN 159,978 (after platform deductions), equivalent to around USD 420. This is also higher than the earnings of regular taxi drivers, who typically only earn a monthly average of NGN 112,000 (around USD 270) (SalaryExplorer, n.d. ). It appears that joining the platform facilitates income generation through increasing efficacy of reaching more potential customers. This is consistent with existing research: Cramer and Krueger (2016) find that Uber uses a more efficient driver-passenger matching technology than taxis. Uber's is based on mobile internet technology and smartphones, and thereby achieves significantly higher capacity utilization rates (Aiyegbajeje, 2019). From this standpoint, e-hailing platforms may be considered productive work and, indeed, 66.5% of our drivers depend on digital labour platforms as their primary source of income. These findings are consistent with the study conducted by Heeks, Graham et al.
(2020) who report that e-hailing drivers often manage to make a decent living, compared to other platform workers. At the same time, costs remain a challenge: e-hailing platforms charge an average commission of 20% (higher than any other type of gig platform). These come on top of rental fees paid to vehicle owners (most e-hailing drivers do not own the vehicles they drive), the cost of fuel and vehicle maintenance, and the cost of data (1 GB of data costs up to NGN 1,264, about 2% of monthly earnings).
When considering labour standards, it is important to note that no formal employment contracts were found to be in place between the e-hailing platforms and the drivers we surveyed, nor are the drivers themselves registered as business owners, meaning they are, in fact, informal workers. 3 Also, by nature, e-hailing represents a task that is difficult to standardize, as drivers can spend vastly different amounts of time driving the same distances depending on the traffic situation in Lagos. Sixty-seven per cent of our respondents reported e-hailing full-time, averaging around 8.5 hours of work on the platform per day; 62% of respondents work more than eight hours per day, while 30% reported working more than 12 hours per day-which Heeks, Graham et al. (2020) see as the reason behind the relatively high wages. A number of experts criticized the e-hailing work model for encouraging extended work hours which may lead to health and safety risks. However, the non-platform-affiliated informal-sector drivers in Nigeria report equally long working hours (Agbiboa, 2017). The opportunities for career advancement on the platforms are uncertain: on the one hand, the satisfaction scores left by customers can be considered a merit-based progression, on the other hand, these are highly subjective, never verifiable, and may reflect direct and indirect biases (Scasserra, 2020). As noted by other researchers, algorithms calculating driver ratings are opaque, with poor recourse to appeal (Wood et al., 2019). The lack of employment contracts means that benefits typically associated with formal work arrangements, such as overtime wages and paid leave, are not offered to drivers. The same applies to protections against unfair dismissal, access to unemployment benefits, and pensions. While this lack of clear regulation leaves the workers open to exploitation, their situation is no different from the plight of all informal drivers (Agbiboa, 2017).
3 Upon completing their online registration, the drivers are required to "accept the terms and conditions" of the platform, which serves as a legal marker of a legal relationships. However, proper presentation and acceptance tools-which are rarely in place-are key to enforceability.
In terms of social protection, our survey responses varied. From the strictly legal perspective, e-hailing drivers in Nigeria are not employees; as such they are not accorded the same rights and privileges as formal workers and have limited recourse channels. Mpofu et al. (2020) documented numerous risks facing e-hailing drivers and argued that kinship networks and informal associations can be leveraged to mitigate these risks.
However, our results indicate that, although it is not prevalent, a number of workers receive some form of social protection from the e-hailing platforms of which medical benefits were the most frequent. At times, access to these is merit-based: for instance, Bolt-one of the major e-hailing operators in Nigeria-offers drivers access to health insurance coverage, with further driver incentives, should ride targets be met. In this sense, e-hailing drivers appear to have an advantage compared to regular drivers in the informal sector, who have no access to such benefits. At the same time, merit-based schemes increase vulnerability: for instance, missing ride targets due to health problems might result in losing access to health care. The COVID-19 pandemic has emphasized the precariousness of gig work, and since most locally founded African platforms have been in operation for less than three years ( In conclusion, while e-hailing brings vital work opportunities as well as tangible benefits to workers, the downsides are also substantial. Platform affiliation does increase workers' earnings, but the degree to which labour standards and social protections are respected is platform-and context-specific. Compared to non-platform-affiliated colleagues in the informal sector, platform drivers do enjoy certain benefits (e.g. health insurance or injury assistance) (Al-Hasan et al., 2015;Ogunrinola, 2011) (Table 3). While having an income source in the precarious informal economy of Lagos is in itself an asset, the quality of work on e-hailing platforms requires further consideration.

| Problematizing formalization
Gig workers in Nigeria (and a number of other countries) are legally classified as independent contractors. This means that the relationship between workers and platforms is governed as a commercial rather than employment relationship, and none of the existing labour regulations (such as minimum wages, social security, and the law against unfair dismissal) apply.
In August 2020, the Nigerian authorities set out to regulate the sector by means of collecting taxes, establishing licensing fees to platforms along with a percentage levy from every trip completed by platform drivers (Enwukwe, 2021;Otieno et al., 2020). The Guidelines for on-line hailing business operation of taxi in Lagos State document also declares platforms to be "service entities," or app developers, and thus not employers but simply intermediaries (Lagos State Ministry of Transportation, 2020; Paul, 2020). These initial attempts at regulation proved contentious, leading to tensions between drivers and law-enforcement agencies in Nigeria (Omilana, 2020), with the president of the ride-hail drivers' union, calling the regulation "not only exploitative but also inconsiderate to the plight of app-based workers" (Ongweso, 2020). In August 2020, e-hailing drivers operating in Lagos held a week-long protest against the proposed service tax and the mobilization of drivers was astounding, including both the Professional E-Hailing Drivers and Private-Owners Association and the National Union of Professional Appbased Transport Workers (Kanife, 2020;Ongweso, 2020). The protesters formulated concrete demands while also declaring their willingness to work with the regulators and decide jointly about the best measures for the sector (Table 4).
As a result of the protests, as well as the concerns expressed by some of the larger platforms, the initial plan was amended, with platforms being charged lower fees depending on the number of drivers registered with them-a 20% reduction compared to the original scheme (Paul, 2020). In addition, the service tax was lowered to 10% of every transaction completed with a passenger, as opposed to the planned 20% (Olowogboyega, 2020).
Achieving lower licence-renewal fees and a 90-day grace period for the drivers to obtain all the required documents were other significant achievements of the unions. All of these have important consequences for gig workers. Differentiating between small-, medium-, and large-scale platforms eases the entry barriers for new (potentially local) platforms, and lowering the service taxes preserves drivers' incomes (Otieno et al., 2020).
Moreover, the substantial achievements of the unions in the negotiations are a testimony to the growing bargaining power of the workers. At the same time, it is important to note that none of the demands pertaining to the regulation/formalization of the platforms as de facto employers was met.
In order to improve the workers' bargaining power it is necessary to look at the data privacy regulation (Webster, 2020). The Government of Nigeria introduced the Nigeria Data Protection Regulation in 2019 (Government of TA B L E 3 Decent work pillars: Comparisons between e-hailers and regular informal sector drivers

Decent work pillars (ILO) E-hailers (platform workers) Regular informal sector drivers
Productive employment • Relatively high wages, up to five times the minimum wage • Irregular (fluctuating) earnings • High costs (platform intermediation fees, data costs, compulsory vehicle check-ups) • The average wage is slightly higher than the minimum wage but still low (USD 100) • Irregular (fluctuating earnings)

Labour standards
• Standard fares for distances established by the platforms, regardless of time units • Client-ranking based career progression (algorithmic control) • Self-established fares, adjusted to real-time traffic situation • At times a portfolio of regular clients (often family and acquaintances) Social protection • Cash transfers, insurance, and credit available to some e-hailers (platform specific) • Family and kinship networks • Informal workers associations

Social dialogue
• Information asymmetry between platforms and workers • Informal networks, unions and support groups (WhatsApp) but unions not formally recognized • Some formal channels for the resolution grievances (often inefficient) • Informal workers' unions and associations, workers boards and support groups (WhatsApp) • The third sector: nongovernmental organizations and foundations promoting the workers' rights in the informal sector Nigeria, 2019), but implementation is proving laggardly. Currently, platforms have a tremendous data advantage and employ various techniques to analyse, score, and predict their workers' performance. Data offers an unexplored advantage: a detailed and comprehensive formal record, documenting e-hailers' work history, including frequency of work, amount and consistency of earnings, reliability, and customer satisfaction scores. If digital data records belonged to the users, they could help overcome a number of the obstacles inherent in the informal sector (e.g. lack of employment history). Drivers could capitalize on their platform scores and use the digital records of their performance as references to seek employment on other platforms or in support of loan application processes. Unfortunately, platforms consider this data proprietary, limiting its public use.
Data-based measures towards formality can be effective only in the context of the improved digital literacy of workers. Improving workers' understanding of platform-scoring mechanisms, data harvesting, and algorithmic control is as important as encouraging coalitions, informal associations, and unionizing. Skilling workers about the platform-scoring mechanisms and the value of the data that is collected from their devices could increase their bargaining power (Wood et al., 2018). Similar solutions are already underway in a number of countries in Europe and North America, with data trusts (such as Worker Info Exchange 4 ) set up to build collective bargaining power (Bacchi & Asher-Schapiro, 2020). 4 Worker Info Exchange is a non-profit, non-governmental organization whose mission is to support workers in accessing, managing and controlling Productive employment

Lagos
State government should streamline their tax and documentation and drivers will pay because government must collect taxes and revenues. 6. We want to be on the side of government because as citizens, it is only our government that have our interest at heart.

Verification of riders identity with *BVN* [Bank Verification Number, based on finger-print and face recognition software] & other Bio metric means. In other to reduce hoodlums from using the apps on a negative mission.
Labour standards (worker protection: health and safety) Minimum trip fare on the apps shouldn't be less than 1k. Due to the current economic situation of the country.

Reduction of percentage or commission from the apps companies which has further enslaves the drivers. It should be subject to review every year. For benefit of all parties.
Labour standards (fair wages)

| D ISCUSS I ON AND P OLI C Y RECOMMENDATI ON S
When discussing the future policy options for the gig sector it is necessary to recognize that, in many ways, gig work in Lagos is embedded in the informal economy (Aiyegbajeje, 2019). Informality in Nigeria is substantial and still growing, with estimates pointing to 38.83% to 57.55% of the country's gross domestic product coming from informal work (Dell'Anno & Adu, 2020). As observed by Onwe (2013, p. 60), "the informal sector has the potential to provide the needed impetus for employment generation" though elsewhere it has been observed that the "existing policy responses to growth of the informal sector have not been encouraging" (Meagher, 2011). Adopting strategies that address the multi-rooted problems inherent to the informal sector should run parallel to designing policies that open the space for the economic activities of gig workers, make use of their assets, and expand their markets. Following the ILO's informal economy recommendations: "to promote decent work, there needs to be a comprehensive and integrated strategy cutting across a range of policy areas that eliminates the negative aspects of informality, while preserving the significant job creation and income-generation potential, and that promotes the protection and incorporation of workers and economic units in the informal economy into the mainstream economy" (ILO, 2007). Effective policy recommendations should cover both the labour demand side and labour supply side of the gig sector, as well as labour market functioning.
In terms of labour demand policies, the job-creation potential of a relatively small but rapidly developing gig sector should not be underestimated. While technology creates efficiencies that generate economic opportunities, it is the platforms (often headquartered abroad) that reap most of the benefits. We draw attention to homegrown platforms (80% of platforms surveyed) which have recently started emerging, in particular in West and South Africa (Johnson, Bester et al., 2020). While their workers are still informal labourers, home-grown platforms are recognized as local businesses that generate revenue for the national economies. Recent studies also found that they are more responsive to market fluctuations, innovate faster, and are more accommodating of workers' needs. In response to the recent COVID-19 pandemic, a number of local platforms responded by rapidly readjusting their offering, leveraging last-mile delivery networks to pivot into logistics and courier services as a response to the lockdowns .
In terms of labour supply, over the past decade, gig work platforms have exploited the legal loophole, associated with the novelty of their business model. While platform firms exert monitoring, supervision, and algorithmic control at arguably greater capacity than many traditional employers, they refuse to recognize gig workers as employees (Marcano, 2018). Worker misclassification arguments have been brought to courts worldwide, recently including Africa (see Hogan Lovells, 2018). If gig workers are not employees, then gig work does not represent a new form of labour relationship but simply capitalizes on the intermediation efficiency of technology while remaining embedded in the informal sector. From the point of view of furthering the Decent Work Agenda, reclassifying e-hailing drivers as formal employees contracted for short-term jobs by businesses and/or individuals would certainly yield best results (Collier, 2017;Ford et al., 2015). At the same time, from the perspective of labour demand, enforcing formalization could trigger platforms to relocate to more lenient regimes, a mechanism well known from offshoring and outsourcing literatures (Peck, 2017, p. 203). Pushing for formalization might also result in negative externalities. In a number of countries, informal workers are eligible for certain forms of universal social welfare.
If gig workers were to count as formal workers (either employees or lawfully self-employed) they might, paradoxically, lose their entitlements to these basic social protections.
Advocating for industry standards can provide a promising middle ground. Graham et al. (2020) propose decent work certification schemes that, though voluntary, could pressure platforms into improving the working conditions for e-hailers. Oxford Internet Institute's Fairwork Foundation is leading the way here with a code of good practices for platforms and five principles of "decent platform work" (fair play, fair conditions, fair contracts, fair management, and fair representation) forming the baseline for an ethical "seal of approval" (Fredmand et al.
Some legal scholars, like Todolí-Signes (2017), argue that there is a need for a new special labour regulation and a new category of a platform worker: neither an employee nor self-employed, with particular social protections. At the same time, he warns that such a new employment model could monopolize the market, squeezing out traditional employers.
In terms of labour market functioning, Heeks et al. (2021) find that digital platforms have the potential to make markets more efficient, effective, complete, and formalized. They also observe that they make markets more unequal by both exacerbating existing power imbalances and creating new ones. For this reason, Kleibert and Mann (2020) argue that the "opportunities" created by such technologies are both contingent and potentially limiting.
Rebalancing the information and power asymmetries between platforms and workers is possible by increasing the bargaining power of the latter (Webster, 2020). Despite the challenges (such as the spatial disconnection of the gig transport workers, the competitive set-up of the industry, and the lack of formal recognition by labour unions), networks of drivers have been found to operate via social media and to effectively pressure policy-makers (Vandaele, 2018). Bottom-up organizing through connective action in the gig sector certainly requires further research (Bennet & Sagerberg, 2012).

| CON CLUS ION
In this article, we have discussed the potential of the gig sector to generate "decent work" in the context of sub-Saharan Africa. The region is of particular relevance due to the persistent rise in youth populations, which is expected to continue until at least 2100. The potential of this demographic dividend is contingent upon meeting the challenge of finding work that enables people to make a decent living but, to date, the employment crisis has deepened in a number of African countries. We used the case study of platform drivers in Lagos, Nigeria, to explore the gig workers' perspective on the conditions and relations of work in the sector and juxtapose it with the existing research on the informal transport sector. Our findings suggest that gig work may help boost youth employment in the country, in particular in consideration of the extremely high youth unemployment rate of 20% in Nigeria 5 (Statista, 2021, see also Nwokoma, 2021). We believe that our casestudy results from Nigeria are generalizable to a number of other countries in SSA. All the economies in the region are affected by the youth employment crisis, and they have all been making strides in ICT availability and affordability (though arguably at vastly different degrees). In a number of these countries, the regulators currently lack a good understanding of tech-enabled work and livelihoods and how these actors help the youth to achieve their professional aspirations.
While e-hailing represents a particular, tethered form of online work, it shares a number of features with other forms of platform work. In most national legal codes, "gigs" are not jobs, but rather tasks or projects, performed by workers who are not employees but simply individuals . From the perspective of labour relations, this has important implications for both social security and welfare provisioning. According to Wood et al. (2019), gig work falls close to a market-mediated, open employment relationship, where market mechanisms determine job outcomes. Most work platforms employ reputation scoring as means of career advancement and thus maintaining high customer satisfaction rates gets priority over negotiating working conditions or fair rates (Aloisi, 2015).  explain how the workers' lack of bargaining power translates into economic exclusion, and how intermediated value chains misappropriate and disempower new entrants to platforms. As is the case with all open employment relations, platform work also transfers both the economic risks, and the costs of upskilling, onto workers (Wood et al., 2019). According to Gandini (2019Gandini ( , p. 1040, in the gig sector the long-standing two-directional labour-capital relationship of supply and demand becomes triangular, with the platform playing the role of a market intermediary or a "shadow employer." Based on our survey, we suggest that e-hailing can provide the source of productive employment, as platform intermediation facilitates outreach and matching, translating into increased wages. This often happens at the expense of longer working hours: as labour standards are lacking, the drivers work long shifts, but the oversupply of labour drives down the fares. While social protection such as insurance, pension, and sick pay are certainly insufficient, some platforms recently started offering cash insurance and credit schemes which help shield the gig workers from shocks and smooth consumption in the wake of the COVID-19 crisis. Achieving social dialogue has been difficult because of information asymmetry and lack of bargaining power on the part of the workers, but the number of informal associations is growing, and the relative success of the recent protests is testimony to the growing power of formal unions. Importantly, when compared to the local benchmark (i.e. the regular, non-affiliated cab drivers in Lagos's informal sector), the platform drivers appear to be closer to achieving "decent work" standards.
Against this background, we propose that e-hailing may provide a potential transition stage between the informal sector and formality. While not recognized as "employees" by the industry, e-hailing drivers have a documented digital track record of earnings, including regularity and volume, as well as performance ratings. These could be used to assess their creditworthiness and document work experience.
As demonstrated by our case study, different levels of stakeholders have important roles to play in assuring the productive and fair development of the gig sector. Labour authorities' transport regulators and data-studies specialists should work together with e-hailing platforms and gig workers themselves to monitor and shape the developments in the sector. Given the platforms' power to set faring fees and the rules of interaction in the market, labour-protection authorities have a role to play in ensuring that principles for fair work are upheld for drivers.
When assessing the pros and cons of gig work, it is important to look beyond the standard employment relations binaries of employer and employed, formal and informal, full-time and part-time. Globally, alternative forms of labour organizations have been on the rise in the context of what is referred to as the "demise" or "deconstruction" of unemployment as a conceptual framework (Benanav, 2019). "Not enough attention has been paid to how labour law standards, fashioned for the 'employee' paradigm, should be reshaped to meet the needs of platform workers regardless of their employment status" . For the sake of both gig workers and employers, the new frameworks should be designed in a process that is as efficient and swift as it is just and inclusive.

ACK N OWLED G EM ENTS
The authors are thankful for the inspiring conversations held with the experts from the Centre for Financial Regulation and Inclusion (Cenfri) in Cape Town, South Africa, in particular, Ms. Chernay Johnson. The case study included in this paper is based on the insight2impact research project conducted by Cenfri and FinTec in 2019/2020 with the support of the Bill & Melinda Gates Foundation. The authors are also indebted to the collaborators from the Lagos Business School, who made the data collection possible, shared their insights, and provided useful comments. Last but not least, we would like to thank all the e-hailing drivers who participated in the study for allowing us to gain insight into their working lives.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available on request from the authors.