1. Introduction

There is an increasing tendency among development actors to highlight social protection (SP) policies and instruments as the preferred means of addressing poverty and inequality and enhancing inclusion in the global South (Deacon 2007; Hickey and Seekings 2017; Hickey et al. 2020). However, the expansion of SP in sub-Saharan Africa is taking place in a context where informal labour markets are the norm, and where most workers have uncertain and precarious livelihoods and lack access to formal SP. To fill this gap, a myriad of collective initiatives formed voluntarily by groups of informal workers seek to provide alternative forms of SP, such as cushioning in the event of illness or death in the family or help in accessing formal SP (Riisgaard, Mitullah and Torm 2022). In this article, we refer to these groups as informal worker associations (IWAs). Although they are considered informal because their members operate informally, some groups are highly organized structures, while others are more loosely formed. Most IWAs are sector-specific rather than family/kinship networks or other forms of community-based Indigenous systems. Moreover, they may fulfil some of the same functions as formal worker unions, including providing advocacy and representing members in their dealings with employers or authorities (namely, lobbying for voice and representation). Examples of matters on which IWAs negotiate include worker registration, access to jobs and social insurance contributions (to formal insurance schemes). Smaller, less organized IWAs often provide SP measures tailored to workers’ needs, such as support during periods of sickness and bereavement.

During the COVID-19 pandemic – an example of a widespread and multifaceted crisis – informal workers suffered disproportionately from the combined effects of the virus and containment strategies owing to the precarious nature of their livelihoods and their limited access to formal SP. IWAs have been shown to play an important role both as providers of informal SP and in facilitating access to formal health and social insurance to protect workers facing idiosyncratic shocks linked to individuals, households or lifecycle events, such as eviction, illness or large-scale events whose effects are limited in time but disproportionate in their impact. However, little is known about what happens to IWAs and their ability to cushion informal workers when, in addition to idiosyncratic shocks, they face a covariate crisis that affects a whole neighbourhood, community or country, as in the case of the COVID-19 pandemic (Bhattamishra and Barrett 2010). Initial assessments indicate that the impact of public health measures against COVID-19 – in particular, full and partial lockdowns – on informal workers’ livelihoods was sudden and severe and, in many cases, led to a permanent loss of income (WIEGO 2020; Wangari et al. 2021). However, most informal workers were forced to continue working despite the lockdowns and risks of contagion, since the relief measures taken by government were limited or inadequate to cushion them (Boatang-Pobee et al. 2021; Kugler et al. 2023).

The first cases of COVID-19 were confirmed in both Kenya and Tanzania in mid-March 2020. Although the countries adopted different coping strategies, the pandemic was a severe social and economic shock for the informal economy in both and tested the ability of IWAs to cushion workers. This article examines the hypothesis that SP delivered by IWAs is more fragile in the context of covariate crisis than when dealing with idiosyncratic shocks. More specifically, we investigate the following two-part question: What were the challenges faced by informal workers during the COVID-19 pandemic and to what extent are IWAs able to cushion workers during multiple crises?

To answer this question, we used survey and interview data from informal workers engaged in the informality-prone sectors of construction, micro-trade and passenger transport in the capital cities of Kenya and Tanzania.¹ The data were collected as part of a larger project on SP and informal workers in urban areas of Kenya and Tanzania (SPIWORK).² The article focuses on the challenges facing workers and the capacity of IWAs to provide and uphold SP in times of simultaneous idiosyncratic and covariate crises, using the example of the COVID-19 pandemic. In order to examine the situation before and after the outbreak, we supplement quantitative data from 2018 with a follow-up survey and interviews undertaken in the latter half of 2020. Comparing Kenya, where COVID-19 restrictions were strictly implemented, and Tanzania, where COVID-19 was largely downplayed, allows us to assess country-level differences in tackling the pandemic.

Prior to the pandemic, the characteristics of IWAs and the kinds of assistance they offered were found to vary across sectors (Riisgaard, Mitullah and Torm 2022). A comparison of different sectors thus allows us to assess how sector-specific characteristics might influence both the challenges faced by IWAs during the pandemic and their responses. Even though the pandemic was a covariate crisis, it presented workers in different sectors and situations with widely varying challenges.

Our findings reveal country- and sector-level similarities, but also some differences in the responses to COVID-19. In both countries, micro-traders largely continued with their daily activities, while transport workers were significantly affected by the lockdown restrictions, even though many continued working at lower capacity. In construction – which was not considered to be an essential service – many lost their jobs, especially those working on larger projects, and restrictions on movement across counties in Kenya and regions in Tanzania meant that many workers were unable to find work. However, some small-scale construction work did continue, subject to mobility restrictions, whereby workers avoided leaving restricted areas for fear of being unable to return.

As regards IWAs, the picture is mixed across both countries. Some of the more established associations remained fully operational, especially in the Kenyan transport sector; others restricted their activities, providing limited financial assistance and social cushioning, or even redirected their focus to other more achievable services, such as providing training on COVID-related health and safety measures. Differences emerge when considering the interrelation between government measures and the IWAs’ ability to cushion workers. For instance, in Kenya, most informal assistance came from the IWAs’ own initiatives, while in Tanzania, measures were often the result of government directives. Notwithstanding the sectoral and country differences in both the reactions to and implications of the COVID-19 pandemic, most of the IWAs that we surveyed did not collapse; however, many became inactive and others functioned at a reduced rate. The article concludes that, while IWAs respond well to one-off individual shocks, they are often not resilient enough to carry workers through nationwide health and economic crises. Our findings thus show a need to strengthen the capacity of formal protection systems to handle covariate crises and provide informal workers with adequate protection, in turn preventing informal support structures from collapsing.

This article is structured as follows. In the second section, we provide a brief overview of the setting of our study, followed by a presentation of our methodology and data in the third section. In the fourth section, we outline the literature relevant to, and the theoretical framing of, our analysis, which is presented in the fifth section. We examine (i) the challenges faced by informal workers during the COVID-19 pandemic and (ii) the ways in which the pandemic has affected IWAs, including their role in addressing the needs of their members. In the sixth section, we provide some concluding remarks, including a discussion of the relevance of our findings for policy.

2. COVID-19 in Kenya and Tanzania

In Kenya, as in many other African countries, various measures were put in place to contain the spread of COVID-19 at the start of the pandemic. These measures, which can be characterized as a partial lockdown, included a stay-at-home order, requirements to social distance, wear masks and sanitize hands, the closure of educational institutions, suspension of international flights, a dusk-to-dawn curfew, a ban on public gatherings, a mandatory quarantine, the closure of bars and restrictions on restaurant opening hours. The lockdown was lifted in early July 2020, but Nairobi remained under a dusk-to-dawn curfew until October 2021. Measures such as requiring people to wash their hands, social distance and wear masks, bans on public gatherings and encouraging workers – especially those in the public sector – to work remotely remained in place. In order to cushion workers from the effects of the pandemic, the Government introduced measures such as tax exemptions for low-income earners, tax cuts for those with higher incomes and a reduction in value added tax (VAT). However, these measures were largely removed in January 2021 when the pandemic was still at its peak. Furthermore, apart from isolated cash transfers and food subsidies given to some workers, informal workers were largely left to fend for themselves, since they were not specifically targeted by any benefits or cushioning programmes. Existing social registries only account for the poorest and the most vulnerable, with registration focusing on households and individuals targeted for social assistance programmes. As a result, during the pandemic, many informal workers were left struggling to cope with not only their day-to-day needs but also their arrears in association and healthcare payments, all while incurring debts with their employers and/or IWAs because they could not repay loans.

In Tanzania, the Government’s initial reaction to COVID-19 was to implement swift measures to contain the spread of the virus. However, even during the short period of lockdown (from March to May 2020), most informal workers across multiple sectors continued with their normal activities. In May 2020, the Government reversed its directives, began lifting containment measures and stopped reporting the numbers of COVID-19 cases and deaths. The late President Magufuli directed people to resume their activities and removed quarantine requirements for international arrivals. He said that tourists would have their temperature checked upon arrival; “If they have no signs of corona, let them go see the animals” (SABC News 2020). In June 2020, the Government of Tanzania declared that the immediate COVID-19 pandemic was over and all schools were allowed to reopen, leading most informal workers to believe that the pandemic had ended. This change of strategy came from concerns about the economic and social impacts of the lockdown, not least in relation to the vulnerable parts of the population. The Government addressed these concerns, for example, by introducing an economic stimulus programme that reduced VAT from 16 to 14 per cent, thereby lowering fuel costs and directly benefiting transport workers. Although there was an acknowledgement that COVID-19 would be present for an unforeseeable amount of time, the public was told to treat it like other common viruses, such as the flu. Accordingly, people were to continue with their work, while observing the precautions set out by the Ministry of Health.

The COVID-19 containment measures impacted society as a whole, but informal workers in particular, the implications being exacerbated by inconsistency in application and effect. For example, in both Kenya and Tanzania, food market vendors were considered essential workers and could largely continue their activities, yet some markets were closed and adhered to COVID-19 guidelines throughout (Boatang-Pobee et al. 2021). New ad hoc street markets emerged in some urban areas and became popular, especially among those who had lost their jobs as a result of the pandemic (see Kiaka et al. 2021). Passenger transport workers were also allowed to continue operations in both countries, albeit with reduced capacity, and some public transport vehicles were used to transport food and other commodities between rural and urban areas.

3. Methodology

The analysis in this article is based on (i) baseline data on informal workers in the construction, micro-trade and transport sectors gathered in 2018 as part of the SPIWORK project,³ and (ii) data from a follow-up survey of informal workers and key informant interviews collected in two phases (July–August 2020 and October–November 2020) in the capital cities of Nairobi and Dar es Salaam. As indicated in the previous section, lockdown measures were lifted relatively quickly in both countries. COVID-19 was still rampant when we started collecting data in mid-July 2020, but informal workers were already referring to the pandemic in the past tense. Accordingly, in this article we use “post-COVID” to refer to the 2020 data collection period. All research was authorized and complied with the ethical requirements in each country.

During the first data collection phase, in July–August 2020, workers in the trade, construction and passenger transport sectors were interviewed using a semi-structured questionnaire with both open and closed questions to ascertain how the crisis had affected their existing work–life challenges and what kind of new challenges IWAs had experienced. For ethical reasons, and in order to protect both researchers and respondents from potentially contracting COVID-19, interviews were conducted by telephone at a time convenient to respondents. When respondents requested physical interviews (mostly during the second data collection phase), precautionary measures, such as wearing masks, keeping the required distance and washing hands after interviews, were taken to ensure the safety of researchers and respondents. The subjects explored in the survey included membership contributions, association meeting frequency and coping and support mechanisms. All the respondents (and their associations) had participated in the baseline survey in 2018, and in both countries the questionnaires were administered by trained research assistants.

In order to ensure a broader base of information, we selected approximately ten associations per sector, per site (Nairobi/Dar es Salaam), totalling around 70 IWAs (35 in each country), which varied in size, type (key purpose) and robustness. A few of the IWAs covered are formal trade unions, such as the Transport Workers Union of Kenya (TAWU), the Tanzania Road Transport Workers Union (TAROTWU) and the Tanzania Union of Industrial and Commercial Workers (TUICO). Other types of IWAs in our sample include savings and credit cooperative organizations (SACCOs), which are common in the transport sector in Kenya, and village community banking associations (VICOBAs) in Tanzania, as well as large umbrella associations for small businesses, such as the association of groups of small businesses and producers, Jumuia ya Vikundi vya Wenye Viwanda na Biashara Ndogondogo (VIBINDO) in Tanzania, and the Nairobi Informal Sector Confederation (NISCOF) and Kenya National Alliance of Street Vendors and Informal Traders (KENASVIT) in Kenya. Lastly, we also surveyed many loosely formed smaller informal welfare groups, known as chamas.

Seeking to capture different views within the IWAs, we interviewed both ordinary members and association leaders from each association. In Kenya, we conducted 71 interviews with informal workers (34 leaders and 37 members), of whom 23 were with micro-traders, 21 with construction workers and 27 with transport workers. In Tanzania, we conducted 65 interviews (38 leaders and 27 members), of which 20 were with micro-traders, 12 with construction workers and 33 with transport workers. Across the two countries, the average age of the workers in our sample was 35 and the average daily earnings were about US$10. In terms of educational attainment, around 48 per cent of the workers in our sample had completed secondary school in Kenya, compared with 39 per cent in Tanzania. Lastly, men made up about 70 per cent of the sample in both countries.⁴ We acknowledge that the small sample does not allow for any quantitative analysis beyond descriptive statistics.

The second data collection phase, between October and November 2020, involved conducting key informant interviews with selected decision-makers and practitioners dealing with informal workers. In Kenya, this included the Micro and Small Enterprises Authority, the National Construction Authority, the National Health Insurance Fund (NHIF) and the Boda Boda Safety Association of Kenya (BAK). In Tanzania, key informant interviews were conducted with the Dar es Salaam Commuter Bus Owners Association (DARCOBOA) and VIBINDO. The purpose was to obtain their views on the effects of COVID-19 on informal workers. The interviews were recorded with the participants’ permission and detailed notes were also taken. The qualitative data were coded and analysed using Nvivo, while the quantitative data were analysed using STATA.

4. Literature and theoretical framing

Although there are signs that this is starting to change, the role of informal support structures for informal workers is not yet fully acknowledged globally, being overlooked in current formal SP extension strategies and the conceptual thinking behind them (Riisgaard, Mitullah and Torm 2022; Awortwi and Walter-Drop 2018). Nevertheless, across the global South, informal SP mechanisms and social networks provide important risk management strategies for households and individuals. For instance, IWAs partially fill the gap in formal SP provision, especially where such protection is either non-existent or weak (Awortwi and Walter-Drop 2018; Kamau et al. 2018; Mushunje and Kaseke 2018; Oduro 2010). Although they often offer only the most rudimentary protection, informal SP systems also tend to be more adaptive and responsive to immediate needs, as well as being more inclusive than formal schemes (Kaseke 2003). In Kenya and Tanzania, there are examples of IWAs going beyond risk management to offer SP services such as educational investments/grants, vocational training and representation with authorities (Riisgaard, Mitullah and Torm 2022). In Kenya, Oware (2020) argues for the importance of traditional SP arrangements in smoothing incomes and consumption, yet she also notes that informal SP may create exclusion and have limitations. Similarly, Kamau et al. (2018) find that informal support systems through IWAs are effective during individual shocks.

Studies considering a wider range of developing countries and types of community-based risk management suggest that informal risk-sharing mechanisms are most efficient in dealing with idiosyncratic shocks linked to individuals, households or lifecycle events, such as illness or death, but that they may fall short when it comes to covariate crises that affect a wider geographical area (Bhattamishra and Barrett 2010; OECD/ILO 2019). In fact, covariate crises, namely those related to weather events, economic downturns and public health events, tend to break down informal insurance mechanisms, in turn adversely impacting vulnerable households whose resources are already constrained (Watson 2016; WIEGO 2020). IWAs are often small and have limited resources, designed explicitly to pool resources in order to support workers facing individual one-off shocks, sometimes on a rotating basis. They are therefore not suited to respond to a covariate crisis that affects everyone at the same time. Other limitations include IWAs not necessarily being pro-poor (Calder and Tanhchareun 2014; Dercon and Krishnan 2002) owing to entry barriers and contribution requirements (Riisgaard, Mitullah and Torm 2022). As IWAs are member-based, they tend to be exclusionary and differentiate among members according to status within the association. Given the limited ability of IWAs to insure against covariate crises, Oware (2020), in line with Bhattamishra and Barrett (2010), concludes that a nationwide crisis requires intervention from governments and non-governmental organizations. She emphasizes the need for comprehensive state SP, although in most cases this is unrealistic owing to limited state resources. Oware (2020) also engages with the interaction between formal and informal SP schemes, arguing that the former should supplement the latter. She follows Oduro (2010), who considers whether the addition of formal SP “will displace informal social protection and have an adverse impact on the welfare of the poor and vulnerable” (p. 23). The key here, as emphasized by Oduro (2010), is that the extension of SP should reinforce rather than undermine the positive aspects of informal support mechanisms. Adding to this discussion, Dafuleya (2018 and 2023) acknowledges the fragility of informal systems in the face of covariate risks, yet also cautions that activities in the informal economy cannot be treated uniformly. Our data certainly testify to the diversity of both informal workers and their associations and illustrate the need to take sector-specific differences into consideration. We seek to assess the ability of IWAs to provide SP for informal workers during a covariate crisis, while simultaneously attending to daily idiosyncratic shocks, such as evictions, confiscations of goods and sickness in the family. To this end, we draw on the experiences of informal workers, as well as examples of how IWAs have sought to mitigate crises. As in Riisgaard, Mitullah and Torm (2022), we follow the SP typology elaborated by Devereux and Sabates-Wheeler (2004) and define SP broadly as encompassing preventive (health insurance, pension, etc.), promotive (microfinance, vocational training, etc.) and transformative (e.g. representation and voice) SP.⁵ We elaborate on these types of SP in supplementary online appendix B and make explicit linkages to our data in the analysis in section 5.2.

5. Analysis

The following three subsections present our findings as follows. First, we analyse the main COVID-related challenges faced by informal workers. Second, we examine the support that workers received from IWAs to deal with these challenges. Third, we consider the associational level and the different coping and cushioning mechanisms used by IWAs, organized according to different interlinking, and sometimes overlapping, themes. All three sections draw primarily on our qualitative data, yet these are supplemented, when relevant, with quantitative data to provide a cross-sectoral view and enable comparison with the pre-COVID situation. Such data triangulation allows us to draw up a fuller and more nuanced picture.

6. Concluding discussion

This article has examined the multiple challenges faced by informal workers as a result of the COVID-19 pandemic and the coping mechanisms adopted by IWAs to cushion their members. In both Kenya and Tanzania, many informal workers had no choice but to continue working during the lockdown, and thus the informal sector played an important part in keeping the economy moving during a challenging time. Ironically, though, owing to their limited power and fragmentation, informal workers have received little attention in debates on the economic and social costs of COVID-19.

The pandemic-induced lockdown, the severity of which differed between Kenya and Tanzania, disrupted both formal and informal work as well as the services offered by IWAs, most workers experiencing a loss in earnings. Many IWAs had traditionally cushioned their members from idiosyncratic shocks, including by covering costs associated with sickness, funerals, school fees, weddings and unemployment. During the pandemic, however, they stopped or suspended contribution requirements, which in turn made it difficult for them to continue providing such preventive SP. Some associations did continue to provide promotive SP, such as financial support, often in the form of interest-free loans, which, although inadequate, provided evidence of the innovativeness of IWAs in responding to crisis. Thus, despite their limitations, associations played a critical role in supplementing government emergency aid by delivering necessary resources and supporting vulnerable members who lost their jobs during the pandemic. However, since most informal groups make savings throughout the year and share dividends (savings plus profit) at the end of the year, they did not have a buffer in the event of unexpected crises, such as the COVID-19 pandemic. Indeed, interest-free loans present a temporary solution but may not be effective for long-term crises. One of the lessons from this period would therefore be for IWAs to reserve some of their savings to avoid starting from a blank sheet every year. One way of doing this would be to register as cooperatives with investment portfolios. Moreover, except for an intensification of representational activities in the transport sector (especially among dala-dala workers in Tanzania), this article reveals a missed opportunity by IWAs in exploiting/promoting transformative SP (voice and representation), since they did not attempt to capture the attention of policymakers by advocating for support during the COVID-19 crisis. Overall, although there are examples of informal support structures proving their worth through specific COVID-19 coping strategies, our study confirms the hypothesis that the support delivered by IWAs is more fragile during a covariate crisis than during individual shocks, notwithstanding country- and sector-level variations.

Sectoral evidence reveals that, across the two countries, construction IWAs were hit harder than micro-trade and transport IWAs. The relatively higher resilience of transport associations is due to the fact that they are generally more established and regulated than construction IWAs. Moreover, in Tanzania, transport sector IWAs averted planned strikes and promoted their members’ interests by negotiating a collective bargaining agreement with the Government. The transport sector faced particular challenges, including reduced numbers of passengers and carrying capacity. As a result, most workers earned less or lost their jobs, yet many IWAs continued operating despite a reduction in revenue (from membership contributions and fees) and, in some cases, the loss of members.

The lack of, or inadequate, resources because of partial payments generally restricted IWAs’ capacity to issue loans during the pandemic. In both countries, they were overwhelmed with loan requests from members and, as a result, issued fewer and smaller loans. In Tanzania, the provision of loans decreased particularly in the construction sector, while the delivery of welfare services increased, indicating some re-prioritizing in order to meet immediate needs. Transport sector workers were particularly challenged by the government “level seat” directive, yet for the most part dala-dala workers continued to operate as usual, except for having to enforce health protection measures among passengers (wearing masks and using sanitizer). Moreover, specific regulations, such as allowing boda-boda access to the city centre and reducing VAT on petrol, served to cushion informal workers in the transport sector from pandemic-induced restrictions. Many Tanzanian transport IWAs took on new tasks, association leaders taking part in the health and safety training and the acquisitions of affordable supplies such as sanitizer. Importantly, the more innovative associations were able to survive thanks to initiatives such as the relaxation of associational rules and reducing contributions and penalty charges to support emergent needs. Such flexibility and the redirecting of funds or activities allowed IWAs to respond quickly to the COVID-related challenges facing informal workers. This stands in contrast to the state-level SP, such as that provided by the NHIF, whose rigid rules did not allow for worker contribution (or penalty) waivers, as expected by many members. Formal insurance schemes, therefore, have important policy lessons to learn from the IWAs’ flexibility in adjusting their rules and processes to cushion their members in a crisis.

Even though most of the IWAs surveyed did not collapse, at least during the initial phase of the pandemic, their capacity to support members during a covariate crisis proved to be limited. This suggests that a double-edged approach, involving both associations and the government, is necessary in order to enable IWAs to weather storms and to ease their burden. One approach might be to create a social register that would enable the government to identify individual informal workers and provide them and their associations with targeted support. In Kenya, however, many IWAs are registered with the Department of Social Services, suggesting that lack of registration is not necessarily the reason why cushioning was limited. Moreover, in the Kenyan SP registry, informal workers are not included except in an implicit mention of collecting data for potential beneficiaries (Maintains 2021). Thus, one policy recommendation would be to include IWAs and informal workers more explicitly as a special category in an enhanced SP registry. This should address the social insurance subscription challenges that IWAs face in order to enable quick and effective responses, especially in a crisis. However, registries do not come without their challenges – in particular, the digitalization of social data has been found to facilitate exploitation in certain political settings (Khan and Roy 2019). The question of registration has also been discussed in Tanzania and some small pilot initiatives have been launched to register micro-traders so that they can be easily reached and included in national response strategies in the event of crises of a magnitude similar to the COVID-19 pandemic.

However, despite the authorities’ good intentions, we acknowledge that there will always be a need for IWAs, given that limited government budgets prevent standard formal SP measures from meeting the requirements of all citizens, especially informal workers. There is thus a need for measures aimed at strengthening the resilience of informal cushioning mechanisms, since IWAs remain the primary absorbers of idiosyncratic shocks. In line with the literature (Oware 2020; Oduro 2010), we reiterate the importance of formal SP supplementing and reinforcing, rather than undermining, the positive aspects of informal support mechanisms. As well as encouraging informal workers to subscribe to formal social insurance, IWA representatives should participate (not only on an ad hoc basis) in processes to design insurance schemes and in official decision-making mechanisms relating to policies affecting informal workers to ensure that their needs are given adequate consideration. Policymakers should tap into the latent capacity of IWAs by partnering with them in designing SP programmes that cater to IWAs and individuals during covariate shocks.

In addition to longer-term policy implications, a more immediate and short-term strategy to respond to future nationwide crises would be to extend targeted cash transfers to protect informal workers and enable their associations to weather such storms. In Kenya, the beneficiaries of the top-up cash transfer funds disbursed during the pandemic were mainly individuals who were already recipients of other cash transfer programmes (such as households in informal settlements), whereas there were no mechanisms for many other workers who needed help. The Tanzania Social Action Fund provides cash transfers to those in extreme poverty, yet informal workers are not part of that group and there are no cash transfers specifically for those not poor enough to be covered by social assistance – also referred to as “the missed middle” (Guven, Jain and Joubert 2021).¹² Thus, a system of unconditional cash transfers during a covariate crisis might provide a way to reach the most vulnerable individuals in the urban informal sector.

Acknowledgements

We gratefully acknowledge funding from the Danish International Development Agency (DANIDA). We also thank all the people who kindly took time to share their experiences and knowledge with us.