Introduction and Background

Since the 1990s, social protection has been an important facet of the Millennium Development Goals (MDGs) and has been promoted by the World Bank in the furtherance of international poverty reduction, i.e. social risk management. In the SADC region, the MDGs (replaced by the Sustainable Development Goals (SDGs) in 2015), have been vernacularised through the SADCs social and human development targets which constitute the region’s Code of Social Security (Dafuleya, 2020). Whereas means-tested non-contributory social assistance programmes were used to cushion vulnerable groups such as the long-term unemployed/underemployed, single mothers, the elderly, orphaned, and vulnerable children (Lavinas, 2018), these categories were expanded by the new risks induced by the global pandemic in 2020. Although there had always been a need to ensure a basic set of guarantees for all, the pandemic (and the ensuing stagnation of economic activity owing to lockdowns), re-energised the ILOs 1950s clarion call for social risk management predicated on breaking the indistinguishability between income and individual/social well-being. In Zimbabwe, the realisation of this in urban areas was, however, made impossible due to limited public assistance from the government (as well as the donor community which was more active in the rural areas). Decades of principally focusing on agricultural subsidies for subsistent farmers in the rural areas of the country meant a neglect of urban areas where urbanites struggled to prevail over the social risks conterminous with a free market economy.

The delicate ecosystem within which the informally employed urbanites were expected to exist deserves particular mention, i.e., Zimbabwe’s hyper-inflationary environment and other negative macro-economic indicators not only meant increasingly reduced livelihood outcomes but also an absence of income security for the millions surviving on hand to mouth. The implications of this on the informal sector were far-reaching not only for those in Zimbabwe but most low- to middle-income countries (LMICs), e.g. the COVID-19 lockdowns put roughly 400 million informal workers out of work in India (Kelley et al., 2020). For Zimbabwe, this situation, along with a reduced cereal production due to climatic shocks (FAO, 2019), constituted a multiplicity of risks and directly impacted on the total number of food-insecure Zimbabweans, spiralling both poverty and vulnerability statistics. It is in cognisance of these contextual factors that precarity for Zimbabwe’s informally employed urbanites mainly stemmed from a limited state presence in social protection along with the expectation for citizens to increasingly create their own mechanisms for survival, i.e. without support from the institutions (social and structural), traditionally expected to do so (Allison, 2012; Seekings, 2021).

A number of challenges have beleagured the informal economy in Zimbabwe and limited the ease of conducting business for the informally employed urbanites. While unemployment was always a problem in Zimbabwe, a 2020 month-on-month inflation rate of 27% and a year-on-year inflation of over 676% (RBZ, 2020) meant that both the formally and informally employed were now exposed to in-work poverty as their paltry incomes failed to keep up with inflation. According to Pikovskaia (2022), the informal sector has traditionally been marginalised and stigmatised by the government and local authorities. Moreover, the lack of clearly defined legal and regulatory frameworks (see Matsongoni & Mutambara, 2021) has led to running battles between the state and informal workers. This is also corroborated by Wadzanai et al. (2018:20) who argue that ‘the informal economy has received little attention from the authorities with an absence of an enabling policy environment to mainstream it into the formal structures of the state.’ In the wake of this, when the pandemic hit, the degree to which the urban-based survivalist entrepreneurs were unprotected from shock became apparent, e.g. 13 million of the 16 million people requiring food aid in Southern Africa were urbanites (WFP, 2020). In 2019 (1 year before the pandemic), a total of 7.7 million Zimbabweans (in a country with a total population of 15.6 million) were already food insecure, i.e. 5.5 million in rural areas and a further 2.2 million in urban areas. A year later, in 2020 (during the pandemic), this national figure of those food insecure had incrementally gone up to a total of 8 million (see Moyo, 2022).

For the reasons stated above that this study has a special interest in the survivalist entrepreneurs who constitute the social fabric of the informal sector. Survivalist retail entrepreneurs typically include street corner hawkers and vendors who sell all manner of products or offer a range of services for the purpose of daily survival within the informal sector (Iwu & Opute, 2019). However, given the fluidity of the term (depending on the socio-economic context at hand), this study conceptualises survivalists as not only including the vendors but also those in the Informal Cross Borders Traders (ICBT) sector who import commodities for quick re-sell at cheaper prices in informal markets than retail outlets. The survivalists are also conceptualised herein as including the millions of unemployed youths who graduate from the country’s universities and thereafter find themselves on the streets vending and fighting against several odds which include a hyper inflationary economy, anti-vending bylaws, and ever-shrinking livelihood outcomes. Needless to say, informal employment is not just a common phenomenon in Zimbabwe but a common preoccupation for most youths across Sub-Saharan Africa, i.e. with the ILO’s (2015) school-to-work transition survey (SWTS) finding out that at least 80% of young workers in Sub-Saharan Africa, fall into the category of informal employment.

Despite the proven disinterest by most African governments in providing formal channels of shock-responsive support (with direct implications on several human rights discussed later in the paper) and ensuring the resilience of survivalist businesses (UNICEF, 2019) — our study desists from humanitarianizing these economic actors. Instead, we are not averse to their standing as economic actors within much broader configurations of political economy. Hence, our focus is purposively centred on their use of not only endogenous solutions to risk (proceeds from their own entrepreneurial ventures, networks, and their role in influencing economic outcomes) but also their utility of and access to exogenous forms of social risk reduction, e.g. emergency relief from donors (Dafuleya et al., 2021).

Although non-state interventions constitute a key part of the endogenous networks to social risk reduction (Tippens, 2020), there has been little research on how the knowledge of other forms of aid is localised and utilised (through homophilous interactions), to the benefit of not only survivalists immediate needs but also the resilience of their businesses. For this, our study explores urban-based survivalists (those involved in ICBT, street corner hawkers, vendors), use of not only endogenous solutions to risk, i.e. networks and their role in influencing economic outcomes — but also their utility of and access to exogenous forms of social risk reduction, e.g. emergency relief from donors during the pandemic amidst spiralling levels of urban hunger and poverty. The study will thus initially discuss the nexus between limited state-led social relief and heightened vulnerability for urbanites. Thereafter, the paper will explore the concept of indigenous social security, i.e. networks in the context of reciprocated assistance and emergency relief in Zimbabwe/neighbouring countries. Finally, the paper will briefly discuss the methodology used before presenting the findings.

Heightened Urban Vulnerability During the Pandemic — Limited State-Led Interventions

The declaration of the COVID-19 pandemic as a national disaster by the Zimbabwean government on 19 March 2020 subsequently led to the pronouncement of a 21-day nationwide lockdown (Nhapi & Dhemba, 2020). In a bid to restrict the spread of the virus, social distancing, and stay-home measures were adopted with social gatherings being restricted to specific numbers. Given the highly informal nature of the economy, many survivalists were adversely affected by this given how they did not possess savings or any form of income replacement. In the absence of social safety nets or income replacement to cushion the informal sector against the loss of income, abiding by the government restrictions thus proved to be economically destructive with many informal traders risking infection just so that they could work and sustain their livelihoods (Mackworth-Young et al., 2021). By the second quarter of 2020, the number of food-insecure Zimbabweans was forecasted to have reached 8.6 million (60% of the population), and by the end of 2020, frustrations were mounting on why infection control measures (such as lockdowns) were not complemented by support packages for families and communities (Mackworth-Young et al., 2021). Unfortunately, the social protection system was already underfunded with the Ministry of Public Service Labour and Social Welfare (MoPSLSW) being allocated a meagre 3.4% of the total 2019 national budget (Chipenda & Tom, 2021). This can be said to have reflected the government’s lack of prioritisation of social protection, i.e. with the many social protection programmes the ministry oversees and is responsible for implementing being financially incapacitated before commencement.

When the government finally instituted a number of measures to ease the effects of the pandemic, they were not shock-responsive nor risk-informed measures aimed at increasing the resilience of vulnerable people’s livelihoods.Footnote 1 Rather, they were for the most part underwhelming and often just an adoption of templates in use in South Africa and elsewhere in the region, e.g. employment-related measures such as state compensation schemes and economic stimulus measures. Other measures, such as the deferment of rent and mortgage payments from 1 April 2020 to 30 June 2020Footnote 2 as well as NSSA’s intention to double pensioners’ regular monthly pay-out for April in a once-off discretionary bonus — were piecemeal and of little benefit to the informally employed (Mashingaidze, 2020). Cash transfers were mostly designated as a response for peri-urban to urban areas while food packages, voucher, and cash-based support were for rural areas. Subsequently, the Zimbabwean government set aside resources for an ambitious Cash Transfer Programme, i.e. targeting a million beneficiaries to receive US$3 per month (Nhapi & Dhemba, 2020). That being said, while the Zimbabwean cash transfer programme was only to last for 3 months, the regional average was for 6 months. Furthermore, the initial amount disbursed of ZWL 180 per month (later revised to ZWL 300 or approx. US$3), was also in pale comparison to regional trends (see World Bank, 2020; Chipenda & Tom, 2021).

For Zimbabwe’s rural population (77% of the country) — relying on subsistent farming, in the period leading up to the pandemic, they too experienced challenges in meeting their cereal requirements with a 30–40% reduction in total maize production (FEWSNET, 2021). However, the disbursement of 7114 metric tonnes of grain to reduce food poverty in these rural areas — made them better off than the urban areas where cash transfers of US$3 were the main intervention, i.e. against a monthly food basket of approximately US$120 (Chipenda & Tom, 2021). This explains why we witnessed the spike in urban hunger and poverty — depicted in Fig. 1 (also see Vinga, 2020). As shown in Fig. 1, there was not only a 12% increase in the number of urban families unable to meet their cereal requirements but also a 6% increase in the number of urban dwellers unable to meet their basic food basket (BFB) (WFP, 2021).

Fig. 1
figure 1

Urban hunger spikes in Zimbabwe (pre and during COVID-19)

Full size image

Although identifying potential beneficiaries in social protection programmes involves setting criteria for inclusion, and then locating the individuals who meet those criteria — in the event of a crisis such as COVID-19, the dynamics of this seemingly simple exercise, i.e. selection and identification, can be very complex. This is particularly so with COVID-19 and the large segments of the population (who were previously self-reliant) that the pandemic drove into poverty. It is for this reason that inclusion and exclusion criteria (coupled with targeting), became highly contested areas in the design and implementation of social protection interventions. Coming up with a list of eligible cash transfer beneficiaries thus theoretically entailed making use of the Ministry of Public Service, Labour and Social Welfare (MPSLSW) databases, means-tests and compiled lists from the Minister of Women Affairs, local authorities and informal traders (Chipenda & Tom, 2021: 8). The subsequent departure from this initial arrangement led to the purported use of ‘sophisticated algorithms’ by the Ministry of Finance and Economic Developments to determine beneficiary eligibility, i.e. analysing bank balances and mobile wallets to not only select but electronically pay beneficiaries (Nhapi & Dhemba, 2020). Without the option of in-person registration, solely utilising information on people’s bank/ mobile wallet balances to include or exclude those below/above a certain income threshold, theoretically ran the risk of technologically excluding those operating outside digital money platforms — particularly so in a country where out of over 14 million, only 5.3 million have active mobile money accounts (Mautsa, 2020). However, notwithstanding the ethical implications in utilising information on people’s bank/mobile wallet balances, this decision was only announced by the finance ministry with there being no actual evidence of its actual implementation. Although there is also no data on the extent to which this would have been a reliable targeting mechanism — we do know that by February 2021 a social protection mapping of the country showed a meagre 2% of total population coverage in selected social cash transfers (see Bilo et al., 2021). Further to this, doubts on the ministry’s use of remote beneficiary enrolment methods are also substantiated by the government’s lack of the required technological capacity to implement such ‘sophisticated’ algorithms.

In the wake of the abovementioned lack of clarity on beneficiary selection criteria, fears arose over the opaqueness of the exercise and a potential failure to position urban poverty as an urgent human rights concern. While this was one way that the survivalist urbanites were disfranchised, the registration of SMEs for rescue packages (designed to cushion their businesses from collapsing), was also marred with its own share of controversies, i.e. the lists often included individuals who were not involved in the sector (The Standard, 2020). Unlike in some countries, where the pandemic forced them to expand the distribution of social grants, while enhancing the level of innovation in selecting the most vulnerable members of the population etc., Zimbabwe performed dismally in protecting its urban population. As shown in this section, the country had no pre-COVID-19 risk-informed non-contributory social assistance programmes designed to address the issues most common amongst its youth and adult population, i.e. unemployment, underemployed and in-work poverty (Bhaiseni, 2020; Calef et al., 2016; Nhapi & Dhemba, 2020). There was also a lack of robust Management Information Systems (MIS), which negatively affected the overall reach and efficacy of available interventions (see Table 1). It is in the wake of this that the following sections of this study will explore the role of mutual assistance networks in mitigating the risks arising from an absence/limited reach of centralised social protection programmes, i.e. how networks of reciprocated assistance were used in lieu of (and in combination with) emergency relief from donors during the pandemic.

Indigenous Social Security (ISS) — an Overview

During colonial times, formal social security in Zimbabwe was limited to the white population. The indigenous population was left to depend on indigenous forms of social security. After attaining Independence, Zimbabwe did not witness a de-racialisation of the colonial non-contributory social security system, rather, the government choose to abolish non-contributory (social) pensions altogether. The same conservative approach applied to social welfare programmes. The government categorised social welfare programs based on the intended recipient population (the disabled, the elderly, orphans, etc.). This was in great contrast to countries such as Botswana where post-independence, the government instituted broad-based social protection programmes such as nutrition and supplementary feeding programmes for vulnerable groups, public work schemes, family support programmes, active labour market programmes, and social pensions (see World Bank, 2013).

Zimbabwe’s 1988 Social Welfare Assistance Act only provides limited public assistance to destitute people (incapable of work) and those aged 65 or older or with a disability. This is done through four broad categories of Social Protection, i.e. social safety nets, social insurance, labour programmes and social care services. Although in the event of a halt or loss of income, the assumption has always been that working age groups would benefit from the abovementioned broad categories available for indigents — in practice, there are actually very few institutionalised programmes for such populations. Within most social protection programmes, there also exists an overlap in the ‘children’ and ‘youth’ category, i.e. the design and targeting of most social protection programmes does not distinguish between ‘youth’ and ‘adults’, rather they do so along the children vs adults dichotomy. As a result, several countries such as Namibia and Lesotho, have a child support grant which covers children up to the age of 18, i.e. during the first 3 years of their ‘youth’ and hence the overlap. In Zimbabwe however, the 2013 National Youth Policy designates all persons between 15 and 35 years of age as youth with there being no social protection programmes that target this group separately from the rest of the adult population (Masuku & Macheka, 2021). Given how the youth are the demographic group that dominates the informal sector — a sector that accounts for more than 60% of Zimbabwe’s economic productivity (Dzawanda et al., 2023), this was the population most affected by the loss/halt in income during the pandemic, i.e. a situation only worsened by an absence of measures to protect livelihoods and build long-term resilience.

It is in the context of not only the historical factors (stretching as far back as pre-independence) but also the more recent socio-economic crises noted in the previous section that traditional/cultural systems (as well as informal social arrangements of protection) have mitigated the risks arising from an absence/limited reach of centralised social protection programmes in Zimbabwe. In dealing with the abovementioned protection gaps, households have basically had to look beyond the state for support and collectively engage in community welfare systems — which although having existed for the greater part of Zimbabwe’s history, have through the passage of time been vernacularised and evolved into new/more robust systems of mutual assistance, e.g. burial societies, rotating/income savings and lending schemes/credit associations (Mushunje & Kaseke, 2018). By definition, rotating savings and credit associations (known as stokvels in South Africa, metshelo in Botswana, and round in Zimbabwe), are informal group saving schemes where through the voluntary contribution of a fixed amount at a regular period, e.g. monthly, quarterly, and so forth, members can democratically decide on the use of the funds from the common pool (Matuku & Kaseke, 2014).

Nhede’s study (2016) of Zimbabwe has illustrated how the abovementioned mutual assistance networks, have also included support from friends, extended family/relatives, voluntary schemes, kinship-based solidarity networks/groups, community membership/rotating schemes, etc. This is however not to imply that informal social support systems only thrive in contexts such as Zimbabwe where there is a protection vacuum. For instance, although Botswana has a very generous and extensive social protection framework, traditional primary institutions of social support such as the extended family, the chieftaincy system, benevolent neighbours and so forth have influenced the salience of mutual aid associations and community support networks in modern-day Botswana (Mokomane, 2018). In fact, in contexts where formal social protection schemes are in place, informal schemes can remain a common phenomenon, i.e. due to the inadequacy of the available forms of (formal) protection. Such complementarity between the formal and informal can also be seen in other contexts where traditional social security systems have not only been a forerunner but also a blueprint upon which other state-led interventions, i.e. national programmes for eliminating poverty, are based (Ibid). This is important to note because it stands as an antithesis to other submissions that suggest a disjuncture between the informal (traditional) and formal systems of social protection, e.g. Mabundza and Dlamini's (2018)  argument on (i) the reduced efficacy of indigenous/traditional forms of social security due to urbanisation, migration, and the disintegration of the family systems in postcolonial Africa and (ii) the limited co-existence of formal and informal forms of social assistance, with the formal system being conceptualised as both competitive and playing a substitutive role against informal protection. In contexts such as Zimbabwe, it is also worth noting that the aforementioned competitiveness of formal systems is not a variable worth considering, with the limitations of the formal system rationalising the notable pre-eminence of informal social welfare arrangements.

Mushunje and Kaseke (2018), have also critiqued the resilience of mutual assistance forms of support in contemporary times — arguing that extended families (due to increased individualism), are no longer able to adequately offer material forms of support with the main utility of such institutions now predominantly becoming instrumental, e.g. helping in the physical care of orphans, the aged and sick family members. While the authors attribute reasons for this to difficult economic conditions, cultural erosion, and the pandemic-induced deaths that considerably divided the family unit, Tippens study (2020) of Tanzania, accredits the diminishing importance of familial support systems to popular perceptions of how these networks of support guarantee a lesser reciprocation of rendered support. As a result, the authors note the pre-eminence of non-familial age-dependent forms of recourse, e.g. fictive kin and religious networks for material assistance (such as money and food), while the extended family networks, friends, and NGOs are perceived to be more useful for instrumental forms of support (such as housing, childcare and so forth).

Intersection Between Social Welfare and Indigenous Social Security (ISS) — Contribution to Current Debates

The right to social protection is observed in the Universal Declaration of Human Rights (1948), the International Covenant on Economic, Social and Cultural Rights (ICESCR) as well as several other human rights instruments. Social protection has therefore been particularly accepted as something key in not only promoting social inclusion but also averting the risks synonymous with poverty by ensuring support/income security for all. For instance, the ICESCR’s Article 9 not only recognises everyone’s right to social security but also everyone’s right to social welfare (also see Articles 7 and 13 of the ICESCR). While social welfare is internationally understood as the services provided by the state or private organisations to assist the sick, old, or indigents — there has been a notable disregard for the non-formal or traditional systems of welfare, i.e. what Seepamore (2018) terms the ‘complex, and sometimes culturally institutionalised practices that characterise social welfare and social security in many African countries’. Elsewhere, although the significance of these non-formal mutual assistance networks has also been evidenced by their increasingly pivotal role in the African community welfare system — they have still remained an under-researched area (Patel et al., 2012). The rationale of examining the abovementioned mutual assistance networks is further corroborated by the arising protection gaps most common within formal systems of welfare. For instance, although we have witnessed a fall in the global extreme poverty rate, i.e. commensurate to the expansion of social welfare programmes (first across Southern Africa, East Africa, and then most parts of West Africa), the low value of benefits relative to obligations of state-led interventions/national welfare programmes (particularly in the global South), has been their most notable criticism (see Seekings, 2019). In the case of Zimbabwe, although the UDHR and ICESCR’s universal rights have been vernacularised through the 1998 Social Welfare Assistance Act — the act has (amongst other issues) been most criticised for its limited reach (see the previous section). It is in cognisance of the limitations of such formal (exogenous) welfare programmes that this study makes significant contributions to not only the literature on human rights but also social welfare, i.e. by exploring the agency of communities in utilising the non-formal or traditional (endogenous) systems of welfare as a means to retaining their dignity — not only in the midst of chronic or transient vulnerabilities but also amidst a near absence of state-led welfare programmes.

There are also only a few studies in existence that focus on household networks in the context of reciprocated assistance and emergency relief. The studies discussed in the previous section are therefore instructive on the trajectory of the current debates on the subject not only in Zimbabwe but a few other countries in the region. In Zimbabwe, recent empirical studies have contributed to the discourse on the role of networks in facilitating household access to external assistance. Whereas in Tanzania, the complementarity between informal mutual assistance networks and other forms of support from donor agencies was noted as being high (see Tippens, 2020), Zimbabwe is different with not only a low degree of complementarity between the two but a seemingly higher reliance on informal mutual assistance over both state- and donor-funded relief. For instance, although Dafuleya et al.’s (2021) study notes a higher reliance on informal mutual assistance groups over donor agencies in Zimbabwe, there is no account given of the sociologies influencing such patterns. Our study therefore attempts to explore some of those factors and how they influence the said choices.

While employment within the formal sector is essentially for the minority in Zimbabwe, it is also important to note that the poor are often too poor to be unemployed, i.e., without any form of income replacement, the poor are often faced with a survival-based decision to either seek respite through entrepreneurial ventures or risk starvation (even if this demands for a shift in moral legitimacies). For this reason, over 85% of Zimbabwe’s national population is eeks out a living in the informal sector (Kede, 2020; Masiyiwa, 2020). The majority of studies focusing on reciprocated assistance and emergency relief (as examined herein), have however humanitarianised this population, often examining the utility of the said mutual assistance networks as something autonomous to the economic agency of those who are survivalists/entrepreneurs — as well as their potential to extract more value from such networks for purposes transcending their immediate humanitarian needs. Hence, this study not only explored survivalists’ use of various endogenous solutions to risk (proceeds from their own entrepreneurial ventures, mutual assistance networks etc.) but also their utility of other exogenous solutions to risk in realising positive economic outcomes.

While there certainly is an interplay between public health emergencies and socio-economic emergencies, the agency of survivalist retail entrepreneurs in exploring endogenous solutions to evading poverty was conceptualised in this study as something that did not arise out of the pandemic but rather, as an outcome of decades of economic distress in the country. In other words, a limited state presence in social protection was examined as something tantamount to an expectation for Zimbabwean citizens to increasingly create their own mechanisms for survival, i.e. without support from the institutions (social and structural), traditionally expected to do so (Allison, 2012). It is for this reason that our paper focused on the period from 2020 to 2021, i.e. when the toughest test for survivalists’ resilience and agency was at its peak with the pandemic/ensuing lockdowns commencing on 30 March 2020. These findings therefore not only constitute a meaningful contribution to the ongoing debates on reciprocated assistance and emergency relief, but they are also an important datum for policy initiatives that intend to integrate formal and informal approaches to social assistance.

Methods

In helping the study achieve the abovementioned objectives, a qualitative approach was used. We delimited the study to the capital city — Harare’s high-density suburbs (as well as the CBD area) which proved to be a hive of survivalist entrepreneurial activities, i.e. Ruwa, Harare CBD, and Hopley (Harare South). One of the researchers had previously worked within the said areas and this facilitated easier entry. These areas were mostly poverty-stricken and thus commonly targeted by aid and development agencies for short-term relief and livelihoods programmes — with some of them having programmes that had been running for decades before the COVID-19 pandemic. We therefore carried out in-depth interviews with a purposively selected sample of urban informal workers, community organisations/associations of survivalist entrepreneurs within urban areas, as well as aid and development partners. The study was therefore qualitative and used a sample of 35 participants comprising of:

  • One community association of survivalist entrepreneurs in Harare (Vendors Initiative for Social and Economic Transformation -VISET)

  • One development partner

  • One youth association

  • Twenty nine informally employed urbanites

VISET was purposively sampled given how the organisation works with vendors who ply their trade in most major cities in Zimbabwe. The VISET representative then referred us to the first vendor in each area such as Ruwa. These vendors then referred us to other vendors in the areas and from thereon, we snowballed the rest of the participants. The development partner (Plan International) was mentioned by most survivalist entrepreneurs as having assisted them. We then sought to interview them on the role they played in assisting informal workers during COVID-19. The youth association was sampled after being cited as a key organisation which engaged the unemployed youth in a bid to secure funding for their small scale businesses. Interviews with vendors were held in the streets as most participants did not want to leave their stalls unattended for fear of loss of business.

Discussion of Findings

Targeting, Selection, Identification Criteria

The government’s failure to regularise the informal sector and a corresponding limited availability of crucial data on the generality of the country’s urban poor, incapacitated the MPSLSW targeting exercise (Arruda, 2018). Save for generic information from press statements and government updates, parliamentary debates, and budget presentations, there were also deliberate attempts to conceal the urban populations’ vulnerability and response data on the part of the government (Chipenda & Tom, 2021). Elsewhere in Africa where there were extensive welfare regimes and thus pre-existing population databases, it was easier to cut down on the time factor and expedite the enrolment campaigns to include new beneficiaries (using multi-dimensional approaches) and expand programmes (Calef et al., 2016). There were also much more efficient methods adopted in successfully targeting informal workers during COVID, for instance South Africa used pre-existing high-quality and relatively up-to-date databases which then helped the authorities to target everyone not covered. Botswana also offered a P1000 once-off grant specifically to informal workers who registered on a new database by the Local Enterprise Authority, thus creating the country’s first ever registry of informal workers (and SMMEs). In Zimbabwe however, although there was data from pre-existing large-scale systems, e.g. the Harmonised Social Cash Transfers (HSCT), it could not really be used to inform any additional enrolment campaigns given how the HSCT was not subjected to the standard retargeting exercises, i.e. every 2 years to eliminate households who may have undergone structural changes. Without such retargeting exercises, HSCTs could thus not be relied upon as an accurate source for vulnerability data — even within areas where the programme was being implemented for years. In the case of those survivalists in the urban area informal sector who for many years were delicately surviving on the margins of the protection framework but nonetheless getting by through entrepreneurship — the loss of those meagre incomes necessitated the re-definition of set criteria/parameters to accommodate new needs.Footnote 3

As we have already discussed, the multi-dimensional poverty index was then used in selecting and enrolling new beneficiaries into social protection programmes. In our interviews, we also noted how predetermined Multidimensional Poverty Indicators were also used in assessing need/poverty by donor agencies with health, education, and living standards being assessed — beginning at the level of the person or household. Living standards were therefore a particularly important determinant of both household and individual deprivation scores, e.g. a certain development partner asked beneficiaries questions such as whether they owned a television set, the food they ate, the frequency they had meals, or whether the family owned a bed. The field workers then used the Orange Sheet whereby people who scored higher in meeting the conditions on the score sheet qualified. It was imagined that through this (a series of selection methods), they could compile the most deserving beneficiaries and scale up existing protection mechanisms to cover more people. However, in a climate where the numbers of people driven into increased vulnerabilities were continuously spiralling, targeting, selection, and identification gaps often emerged.Footnote 4

We noted that Plan International used a provincial approach or geographical targeting to select areas to implement its programmes. At a provincial level, the use of the Zimbabwe Vulnerability Assessment Committee (ZIMVAC) report informed the decision to target Ruwa within metropolitan Harare. The vulnerability map helped the development player narrow down the potential target areas in which to implement their programmes. Working with the local government, the organisation agreed on ‘hotspots’ in Ruwa, thereafter, they conducted door-to-door interviews targeting persons with disabilities, child-headed families, and female-headed families as well as poverty/low-income households. While this was a far cry from the Ministry of Finance’s suggested complex algorithms (Nhapi & Dhemba, 2020), the selection criteria saved its purpose expediently. However, the main issue was the lack of a pre-existing database of informal workers, combined with a lack of other (accurate and up to date) databases which could have been used to target the informal sector workers (by means of triangulation/exclusion, as it was done in South Africa), and then roll out income replacement programmes for them, i.e. given how they were now out of work due to lockdowns.

Antagonism: Survivalists and the Punitive Actions of the State

The International Covenant on Economic, Social and Cultural Rights of 1966 (ICESCR), recognises the right to work as a basic human right, i.e. everyone’s right to gain his/her living by work which he freely chooses or accepts, with signatory states being mandated to take appropriate steps to safeguard this right (see Article 6). Although almost all of the survivalist entrepreneur participants in our study were in their 30–40 s (thus relatively young men and women at the prime of their productive years), their contribution to national development was nonetheless minimal. Partly contributing to this was the government’s failure to regularise/mainstream the informal sector, choosing rather to criminalise it by perennially destroying trading sites and enforcing a host of other punitive bylaws, i.e. which was counterintuitive to safeguarding everyone’s right to gain his/her living by work (Mwonzora, 2022; Vaziralli, 2020). This hostile operational environment for survivalists, the poverty and economic shocks, etc. were however realities that pre-dated the pandemic. In 2017, for instance, 70% of the national population was living below the Total Consumption Poverty Line (TCPL) with the country indicators only worsening during the pandemic, i.e. in terms of both the food poverty datum line and total consumption poverty line (UNICEF, 2020). As already mentioned, due to how Zimbabwe’s 1988 Social Welfare Assistance Act is designed, the majority of young and middle-aged people (whether destitute, indigent, or survivalists enduring disrupted incomes due to the pandemic) — were excluded from state-based systems of protection with the situation not being much different elsewhere in the region. For instance, only seven out of sixteen SADC member states had pre-COVID-19 social assistance programmes targeting the adult demographic group (UNDP, 2019. The remainder of countries in the region only catered for children, the elderly (with special provisions for those who are also war veterans), and the disabled. In essence, our data confirmed the reality of existential risks, arising from the absence of any workfare programmes nor institutionalised feeding programmes specifically targeting the youth and adult population, i.e. who constitute the bulk of the long-term unemployed, underemployed, and survivalists.

Subsequent to the outbreak of the COVID-19 pandemic, we then saw a majority of low-income wage earners (predominantly survivalists in the informal sector), being reduced to near destitution due to a loss of livelihoods. For instance, the ICBT sector (the biggest informal sector in the country), was overlooked from the government stimulus recovery packages, sinking the millions in this sector into urban poverty when their migratory practices came to a halt, i.e. as borders closed for several months due to the March 2020 lockdowns (Kede, 2020). So, not only did the ICBT benefit the traders, but it also helped reduce urban hunger, i.e. the imported goods were sold at much cheaper prices in the streets and townships than available in the retail outlets — thus contributing to greater levels of health and nutrition at the household level. The border closures therefore came with a complete loss of income and exhaustion of financial capital for cross-border survivalists (which also impacted our sample of survivalists who survived by vending the imported commodities). Further to this, when the borders eventually did open, the requirement for COVID test certificates (pegged between USD $60 and $80 on both sides of the border), severely reduced their livelihood outcomes.

‘Lockdowns affected my business in that I could not even pay rent’, ‘things were so bad that we spent all our capital to buy food’, ‘no one wanted to hire us for maricho as people were afraid to contract COVID’

Our interviews also revealed how poverty was more pronounced amongst survivalist urban informal workers (the majority of whom were living on hand to mouth even prior to the pandemic). The interviewees thus told of how the lockdowns and associated restrictions, such as the destruction of their vending stalls and cabins on April 2020 and 19 June 2021 by the municipal police in Mbare, Highfield, and many other high-density areas — particularly affected them. Given how the survivalist entrepreneurs businesses mostly thrived in congested public spaces — this put them at loggerheads with the public health interests of city authorities, with the June 2021 demolitions proceeding despite a High Court order barring the destruction of stalls and in some cases resulting in the loss of valuable stock (see also Mwonzora, 2022). Our interviewees narrated how due to the banning of street vending, and the introduction of social distancing amongst other restrictions, they could no longer conduct their businesses in the more viable undesignated spaces, e.g. the street pavements in the central business district. Although this painted a rather hopeless humanitarian picture, our follow-up inquiries later revealed the resilience and adaptability of these survivalists even in the face of the pandemic, punitive actions of the state, and associated socio-economic distress.

Survivalist Entrepreneurs During the COVID-19 Pandemic: Entrepreneurial Coping Strategies

Noting the challenges in the previous section, our interviews set out to inquire how the urban-based survivalists managed to meet their basic needs. We discovered that there was a diversification into other ‘endogenous’ means of income generation which were independent of ‘exogenous’ assistance from donor agencies, e.g. some diversified into different sources of livelihoods such as poultry and backyard gardening for domestic consumption. While we did not test gender and its role in determining membership to mutual assistance networks, we did find data that corroborated previous studies with male-headed homes being barely targeted for emergency relief (see Mokomane, 2018; Dafuleya et al., 2021). Based on our interviews with a development partner, around 65% of their selected beneficiaries were female (which was a deliberate move as women are mainly responsible for food preparation). However, we also found that adult male’s exclusion from aid programmes and limited membership in support networks did not necessarily translate to limited recourse — rather, this was often offset by their use of the township house and yard as an important asset in sustaining urban livelihoods as well as other ‘endogenous’ means of income generation, independent of ‘exogenous’ assistance from donor agencies. The use of backyards to engage in income-generating projects was not peculiar to our interviewees, other studies have proved how the house and yard can become an important asset in sustaining urban livelihoods (see Mususa, 2012; Olivier, 2019). While the Mususa (2012) study of the Copperbelt identified this practice as being more common in the more spacious low-density area of the mining town, with an argument that those in the high-density residential areas were constrained from exercising the same agency due to the smaller nature of their yards — our study noted something completely opposite.

It must be said that the majority of our interviewees could only afford to rent out a single room in the townships, a handful (mostly the first-born male who would be heir to the family patriarch), actually did own houses, i.e. mostly inherited from deceased or retired parents now living in the rural areas. In the case of the later sample, they gave some impressive accounts of how they were using the houses to generate income in more than one way. Their responses also contradicted the Mususa (2012) study where the utility of the house was limited to being either a hedge against poverty/inflation for absentee landlords in the rural areas or an asset in sustaining urban livelihoods for resident landlords. Our findings provided evidence of the cross-generational utility of the township house in benefitting:

  1. (i)

    The offspring of pensioners — through subsidised accommodation plus a space to undertake income-generating activities such as poultry and backyard gardening. The houses also supplemented profits from such ventures by being partially rented out to tenants.

  2. (ii)

    The rural area-based pensioners — who would then receive a fraction of the revenue generated from their town property, i.e. in the form of a domestic remittance. This would then contribute to increased spending on food and healthcare emergencies for them in the rural areas.

Given the above, the high-density areas (where most houses measure below 300 m2), were a hive of survivalist ventures with the house standing out as a cross-generational asset, key in sustaining both urban and to a lesser extent, rural lives. That being said, however, unlike international remittances, such urban-to-rural domestic remittances were typically small amounts and thus had limited impacts on durables for rural recipients (Gwatidzo & Mupingashato, 2021). Our data showed some of the feasibility challenges in stretching the revenue extracted from the house, with its cross-generational capacity becoming obsolete particularly during the peak of the pandemic. For instance, the total annual income from a combined effort of renting off part of the house (at a monthly average of $30 a room and $360 over 12 months) and poultry projects as shown in Table 1 (which rarely exceeded 100 chickens per batch given the size of backyards), these activities earned survivalists just enough to survive. Given the US$35 dollars per person poverty datum line (PDL) — calculated as the monthly national food basket of seventeen food items (Chivandire, 2019), this came up to an average annual rate of $1680 for a family of four. This therefore meant that our interviewees were making just enough for their subsistent needs with no surplus for savings, schooling, and emergencies nor the urban to rural domestic remittances.

Table 1 Average annual income from the use of houses to sustain urban livelihoods
Full size table

Survivalist Entrepreneurs During the COVID-19 Pandemic: Mutual Assistance Networks

For our interviewees who rented single rooms in informal settlements and thus did not have space to engage in backyard income-generating projects, they also exercised some creative use of mutual assistance networks, e.g. internal saving and lending schemes were used in lieu of public assistance programmes. Saving groups were a particularly key form of micro-finance whereby community members arranged themselves into small groups (15 to 25 members) for the purposes of saving money with the accumulated savings being lent out to each other (also see Karlan et al., 2017). However, while some of these internal saving schemes were entirely endogenous and born out of a local initiative, some were also exogenous and organised by non-state donor agencies such as Plan International (Fig. 2). In understanding one of the ways that the endogenous saving schemes were being financed, two factors were noted: (i) survivalists used proceeds from their irregular incomes and (ii) remittances were redirected to such networks, i.e. Zimbabwe has a thriving diaspora and subsequent remittance record. In 2021, Zimbabwe’s annual remittances stood at US$1.430 billion — which was a 43% increase from the previous year and during the pandemic (the period under inquiry), remittances surpassed the 2015 all-time high of US$939 million (RBZ, 2022).

From the vending that I do, I manage to save USD$1 per day for the whole month then I give that money to a person whose turn it is to receive the round…

Fig. 2
figure 2

List of non-state organisations most active during the pandemic

Full size image

In the case of both endogenous (see interview excerpt above) and exogenous internal lending and saving schemes, however, the common goal was to facilitate a responsible use of group savings, borrowing and allow for social insurance services while also helping group members to finance income-generating activities (at a minimal interest). However, our interviews revealed how the amounts available for lending were often too little to provide any meaningful capital towards a sustainable survivalist entrepreneurial venture or, in some cases, repayment became a challenge resulting in the network collapsing. This is corroborated by previous studies (see Mushunje & Kaseke, 2018) where it was found that, although informal social assistance systems are more inclusive than formal social security systems, they often offer only the most rudimentary social protection while also being prone to a high default rate in the collection of contributions, i.e. due to the high mobility of those in the informal sector. However, our findings also revealed that internal saving schemes that usually pooled resources towards the purchase of food stuffs were more resilient, often lasting longer than the aforementioned schemes with a lending portfolio, i.e. something that could be tied to the generalised risk of doing business in Zimbabwe (as well as small profits, see Table 1), with most start-ups failing to recoup loans from borrowers.

I switched to maize cobs because I could go into residential communities calling out as market stalls had been banned...

Given the challenges articulated in the excerpt above, interviewees from VISET said that their more enterprising members adopted the digital marketing strategy and utilised Facebook (marketplace), WhatsApp (business), and others. These mediums helped them to reach out to a wider network of potential buyers despite the tough regulations controlling movement and social distancing. Some survivalists in the high-density areas also adopted a “door step approach” by putting their merchandise directly in front of their houses. The challenge in this instance was the police who often raided them, hence the resort by some to the use of using their car boots to display and market their merchandise (a new model which emerged during COVID-19). While other studies on mutual assistance networks in Zimbabwe argue that households in a community group (as opposed to smaller groups such as burial societies) have lower ties between them and lower information-sharing opportunities/incentives (see Dafuleya et al., 2021) — we noted something different. Our findings showed that knowledge of requisite grants or other forms of aid was being circulated through community groups (such as Internal saving schemes), which usually pooled resources towards the purchase of foodstuffs. Although previous studies (Ibid) have asserted that larger ex-post bereavement mutual assistance groups have weaker ties and thus limited utility, often being confined to the circulation/sharing of cash, food, and credit with little impact on members’ knowledge/receipt of emergency relief — our findings proved otherwise. We saw the adaptive and responsive qualities of mutual assistance networks to the needs at hand, i.e. information was the main currency exchanged during this time. Such information mainly pertained to where it was safe to trade (as the survivalists engaged in a cat-and-mouse game with the authorities), which in itself was proof of mutual assistance networks’ potential to go beyond the simple circulation of cash, food, and credit amongst members.

In the 3rd phase of lockdowns, some survivalists trading in fruit and vegetables were eventually exempted and could operate from their designated places of residence. Some literally syphoned the food items they bought through the COVID-19 assistance and channelled the money they got to join internal savings and lending schemes or other ventures. This data disproved other studies findings (see Mabundza & Dlamini, 2018; Arrey, 2018), where the formal and informal systems of social assistance were found to rarely co-exist - with the formal system being both competitive against and often substituting the informal. That being said, in furthering our understanding of the survivalists’ utility of and access to exogenous forms of social risk reduction, we found that for the youth who had received the government fund, very little could be achieved from this as the Relief fund was in RTGs (the local currency) and yet most goods in the country were priced in USD. Furthermore, according to the Anti-corruption Trust Southern Africa report, politicians in the Midlands province were the majority of beneficiaries of the COVID-19 relief fund. The fund was hijacked and looted by the politicians. So the fund did not reach the intended beneficiaries (New Zimbabwe, 2020).

Coverage of Social Protection Response and Its Adequacy

Whereas elsewhere social risk was still managed despite a limited state and financial capacity, e.g. we saw the expansion of basic non-contributory social protection with special focus on urban poor, peri-urban, and populations in the border areas of Mozambique (de Lima Vieira & Andrés, 2020) — for Zimbabwe, there remained a limited state presence in this regard. The Zimbabwe Youth Council and VISET confirmed that the government had set aside the ZWL 17 million fund to finance youth entrepreneurial activities during the pandemic. However, a VISET executive member revealed how only a meagre 158 VISET members in the whole of Masvingo benefited from the government cash fund, i.e. out of over 3000 who had applied.

In mitigating against the arising protection gaps, some development players effectively rolled out some COVID-19-specific social protection programmes in the capital city. Within the Harare Metropolitan area (as covered in our study), these players mainly including the four listed below.

According to a Plan International Programmes Officer — programmes on offer were the COVID-19 assistance and Urban Social Assistance Programme. On average, these were active for one to one and a half years. Plan International also covered other towns, such as Chiredzi, Chipinge, Kwekwe/Redcliff, and parts of border towns in Manicaland. For the COVID-19 assistance, Plan International’s target was 10,000 beneficiaries for Harare and Ruwa. However, only 2053 were assisted and this was again indicative of how despite there being many who needed assistance during the pandemic, only a few received any form of intervention. This emphasises the vulnerability and precarious existence of survivalist urban populations who had also lost their sources of livelihoods due to the COVID-19 restrictions and regulations.

The majority of our participants confirmed having received assistance from non-state players from 2020 to the end of 2021 (an average of 1½ years). For the COVID-19 assistance programme, beneficiaries received USD$12 per person/household. The focus was the provision of food (and detergents) redeemable at selected retailers such as N-Richards, Pick n Pay, Cash, and Carry. Food feedback spelt out that this assistance improved the quality of life. However, without conscious synergies amongst the development players, targeting, coverage of assistance, and the nature of assistance were disjointed. Although the Zimbabwe Vulnerability Assessment Committee (ZimVAC) was instituted as a technical advisory committee comprised of government representatives, development partners, UN agencies such as the WFP, NGOs, technical agencies, and academia, we argue that its inefficacy as a conveyor of inter-organisational exchanges during the COVID-19 crisis contributed to the said absence of conscious synergies amongst stakeholders. This resulted in social protection interventions being confined to small pockets of the vulnerable populations (Mashingaidze, 2020). Plan International for instance gave groups training on complimentary skills as well as knowledge on how to secure funding for resilience building for the post-COVID era. Concurrently, twenty-four internal savings and lending schemes (ISAL) along with a GBV programme were operating in Ruwa. However, basic assistance was insufficient to make the beneficiaries self-sustaining, creating more dependency on the donors with some beneficiaries also abusing the funds (see interview excerpt below):

‘It wasn’t enough to pay school fees’, ‘it covers a little gap’, ‘the assistance was inadequate’. I was taken off the beneficiary list after I took all the money and spent it on girlfriends so my wife reported me’.

As a result of the inadequacy of the USD$12 per person/household — some felt that ‘this is money we just receive without working so it’s enough. We can’t complain’. The majority of the beneficiaries professed that they needed to supplement this assistance with their non-existent income. Once again, we were reminded of the enterprising spirit of the survivalists when a Plan International Programmes Officer told of how they had to remove several people from the beneficiary list as they were found guilty of reselling the food items that they had received. One participant felt that the donors’ strict lists of what to buy as well as how the commodities were utilised restricted their entrepreneurial inventiveness. Thus, apart from the obvious challenges with complementarity, the fact that the cash transfer from donor agencies could not be converted into any other use but the purchase of stipulated commodities — meant that the $12 had little help in either complementing income-generating efforts or improving greater levels of education/schooling, health, and nutrition. While previous studies have noted that there is a higher reliance on informal mutual assistance groups over donor agencies in Zimbabwe (see Dafuleya et al., 2021), our data therefore helped addressed the ‘why’ component. In other words, informal mutual assistance networks were the preferred option by survivalists given how they were more inclusive (no means test), more adaptive (the ready choice of joining either cash or grocery-based networks with no guidelines on the usage of rendered relief) and more responsive, i.e. relief was rendered as and when the need arose.

While in this study, we had initially hypothesised emergency relief from donors as something that could be utilised concurrently with endogenous solutions (albeit on a short-term basis), to supplement income loss for the informally employed survivalists — we noted that complementarity was very poor. Where income-generating projects were donor funded (which was minimal and far in between), creativity was also dismal, resulting in a flooded market where most beneficiaries were competing to trade the same goods/products. For this reason, the cessation of the COVID-19 assistance entailed a slump into poverty for many as they had also dipped into the capital meant to resuscitate their informal livelihoods. Generally, the beneficiaries of the COVID-19 assistance aid had no economic recovery plan post the pandemic period.

Conclusion

This study has shown how during COVID-19 the impact and scale of the interventions (both from the state and non-state players) was limited. For instance, the WFP’s urban assistance scheme delivered monthly cash transfers to 326,000 people in Zimbabwe’s 23 urban areas, which was a drop in the ocean given the country’s pre-existing socio-economic challenges and the exponential rise of vulnerable people during COVID-19 (Vaziralli, 2020). Another factor that constrained the impact and scale of donor agency interventions was funding gaps, e.g. UNICEF had a US$70.9 million funding gap to cover multiple hazards in Zimbabwe (Zimbabwe Humanitarian Situation Report, July 2020). The World Food Programmes Urban Social Assistance and Resilience Building Programme, designed to “support long-term development through urban agriculture, setting up savings and lending groups, entrepreneurship training, etc. to better equip vulnerable communities to withstand shocks, also reported a US$180 million shortfall.

That being said, the study also identified how the urban-based survivalists exercised remarkable agency in not only establishing an informal community welfare system (such as rotating/income savings, lending schemes, and credit associations) but also using the said informal systems to retain their rights to social security, i.e. as enshrined in the ICESCR (Articles 9, 7, and 13) and the UDHR (Article 22). Although the development players’ insistence on a strict adherence to food lists robbed beneficiaries of a chance of complimenting the donations with proceeds from their irregular survivalist incomes, we witnessed the urbanites’ creative use of networks (and other entrepreneurial ventures) to survive poverty with little to no comprehensive state-led social protection responses. The study also found that:

  • Although offering a mostly rudimentary form of social protection, informal mutual assistance networks were more adaptive, responsive, and inclusive and thus more preferred over exogenous relief — with complementarity between the two being low.

  • Contrary to other studies, we found that ex-post bereavement mutual assistance networks have stronger ties with a more versatile, knowledge sharing centred utility that transcends the circulation/sharing of cash, food, and credit.

  • Adult male’s exclusion from aid programmes and their limited membership in support networks did not necessarily translate to limited recourse and was rather counteracted by other endogenous means of income generation, independent of exogenous forms of assistance from donor agencies.

  • Internal saving schemes that pool resources towards the purchase of food-stuffs (financed through irregular incomes and remittances), were more resilient than schemes with a lending portfolio.

In contradiction of the ICESCR’s (Article 6) and the UDHR (Article 23), recognition of the right to work/gainful employment as a basic human right, the study also identified an existential case of antagonism, with the Zimbabwean government not complementing or facilitating the livelihoods efforts of urban-based survivalists — but rather assuming the role of a punitive state and undoing those very bottom-up initiatives to not only survive but thrive in a harsh economic and political environment. For instance, the criminalisation of survivalist business enterprises, banning of street vending, coupled with the destruction of stalls, can be argued as being punitive and counterintuitive. Nonetheless, this also necessitated a shift in moral legitimacies, i.e. the adopting of new moralities by the urban-based survivalists to justify and explain economic activities that fall outside the remit of the formally legal (see Mususa, 2010). Where following the law meant starving one’s family indoors, survivalist entrepreneurs resorted to the doorstep approach (or in some cases, legitimate alternatives such as digital marketing). While the latter was plagued with challenges such as raids from the police, hefty fines, or even arrests, the end seemed to justify the means with citizens continuously imagining new (even illegal) ways to not only survive but thrive under very challenging circumstances.