key: cord-0949190-sw63uhdx authors: OGANDO, Ana Carolina; ROGAN, Michael; MOUSSIÉ, Rachel title: Impacts of the Covid‐19 pandemic and unpaid care work on informal workers’ livelihoods date: 2021-09-22 journal: Int Labour Rev DOI: 10.1111/ilr.12332 sha: 7a7a4ab2b2eb3b67a88b59ddb9e3246421163b2b doc_id: 949190 cord_uid: sw63uhdx The COVID‐19 pandemic has led to a health, economic and care crisis for all workers including those in the informal economy. This article draws on data from the first round of a mixed‐methods longitudinal study conducted (in June 2020) by the research advocacy network Women in Informal Employment Globalizing and Organizing (WIEGO) in partnership with informal workers’ organizations in 12 cities. It assesses the impact of the multidimensional crisis on care responsibilities and the resulting effect on livelihoods and food security. A gendered analysis of paid work and unpaid care work sheds light on the unique features of the current ‘pandemic recession's’ impact on the world's labour force. Over 60 per cent (or 2 billion) of all workers globally are informally employed and this includes about 90 per cent of all employment in developing countries (ILO 2018a) . With respect to the economic and health crisis generated by the COVID-19 pandemic, the ILO has projected that 1.6 billion of these workers would be among the most severely affected (ILO 2020a). By definition (see ILO 2018a), informal employment is any work which is not covered by de jure or de facto legal or social protection. Informal workers therefore have no safety net if they are unable to work and often rely on their daily earnings to support themselves and their households. Further projections (ILO 2020b,c) have suggested that, as a result, the number of working poor was expected to increase, globally, by between 20.1 million and 35 million people during the course of 2020. In low-income countries, poverty levels among informal workers were projected to have increased from 18% to 74% (ILO 2020c). In addition to the drastic impact of the crisis on informal workers, a particular feature of the current crisis is that, unlike the series of recessions and economic downturns since the Great Depression which have had a greater impact on men's employment, the COVID-19 crisis, like other 'pandemic recessions', has resulted in a so-called 'shecession' (Alon et al. 2020; Collins et al. 2020) . Part of the explanation for the gendered impacts of the COVID-19 crisis is that women tend to be concentrated in the services sectors, and particularly retail and hospitality, which have been affected more severely (Alon et al. 2020) . For example, 42% of women in the informal economy are employed in sectors that have been identified as 'high risk' during the current crisis compared with 32% of men in informal employment (ILO 2020c) . However, the gendered structure of the labour market may explain as little as a quarter of the gender differences in employment losses during the COVID crisis (Madgavkar et al. 2020) . The other main factor, discussed below, is the large and rapid increase in the burden of unpaid care work faced by women. The COVID-19 pandemic might best be described as a multi-dimensional crisis characterised by an interrelated economic, health and care crisis. Not only are informal workers more likely to lose their incomes, but they may also be at higher risk of contracting COVID-19 due to occupational health and safety risks at work and having less access to basic infrastructure both at home and the workplace. With the closure of schools, day care centres, public transportation and many formal businesses, women in informal employment have had to balance the sudden increase in unpaid care responsibilities alongside the need to earn enough money to feed themselves and their families. This paper aims to examine the contours of this multidimensional crisis experienced by women in informal employment during the onset of the COVID-19 crisis between March and June 2020. The data we analyse is based on telephonic surveys and qualitative interviews with informal workers from 12 cities in 10 countries. We examine the intersection of the care and economic crises among this sample of workers and identify some of the key features of the care crisis at the base of the economic pyramid. We focus on four types of informal work that are prevalent in cities across the global south -waste picking, domestic work, street vending and home-based work. The remainder of the paper is structured as follows. The next section begins with a brief overview of the gendered nature of unpaid care work, more broadly, before turning to a review of the literature on unpaid care work in periods of crisis. Section 3 outlines the study methodology and describes the approach adopted by the study team to sampling and interviewing informal workers during the first wave of the global COVID-19 pandemic. In Section 4, the results are presented in four parts. First, we outline the size and shape of the increase in care responsibilities across our sample. Next, we analyse the association between increased care responsibilities and reduced work and earnings during the first six months of the crisis. We then turn to an overview of asset depleting strategies that are used by workers who reported an increase in care. Finally, we consider the link between an increased care burden and food insecurity by measuring levels of reported hunger by gender and by the distribution of increases in care work. Section 5 concludes with a discussion of the implications of the interrelated crises experienced by women, in particular, at the base of the economic pyramid. In all countries, women do more unpaid care work than men (ILO 2018b). Even as women's labour force participation increases globally, there have been no significant shifts in the unequal responsibility for unpaid care work (Kabeer 2007; Charmes 2019; ILO 2018b) . Much of this care work is non-monetized and happens within households and between community members. It is referred to in the literature as unpaid care work and includes cooking, cleaning and caring for children, people living with disabilities, the ill and elderly, and other adults in a household as well as unpaid community work (Elson 2000; Esquivel 2014) . Unpaid care work can be further broken down into direct care that involves a personal and emotional engagement, such as caring for a child, and indirect care that refers to all activities required to sustain direct care including cooking and cleaning (Folbre 2006; Esquivel 2014) . Though most countries do not count or value the contribution of unpaid care work, it makes all other forms of paid and unpaid work possible (Razavi 2007; Fraser 2014 ). Feminist research emphasises the interlinkages between social, economic, health and ecological crises in reproducing and deepening the multiple inequalities women face (Fraser 2012; Ghosh 2013; Castañeda and Gammage 2012) . The COVID-19 pandemic reveals similar interlinkages as in past financial and health crises, though these are intensified due to the more pronounced loss of income for women informal workers. At the outset of the crisis in South Africa, for example, women informal workers were more likely to lose income than men in the informal economy and both women and men in formal employment (Rogan and Skinner 2020) . The loss in earnings was most severe for those women in vulnerable informal work suggesting a high risk of extreme poverty, hunger and food insecurity in their households (Rogan and Skinner 2020) . As early as January and February 2020 when China and other South East Asian countries closed their borders, street vendors and market traders could no longer replenish their stocks or buy raw materials, production orders for home-based workers stopped, and the price of recyclables on which waste pickers depend plummeted (WIEGO 2020a). Lockdown measures taken across This article is protected by copyright. All rights reserved. the world in March and April 2020 prevented most informal workers from leaving their homes to engage in paid work. Data from WIEGO's global study shows that among the four sectors, homebased workers (who are predominantly women), somewhat ironically, were the least able to work and the slowest to recover as a result of lack of jobs or work orders from employers and their contractors, fluctuations in demand, supply, prices and other factors in the supply chain (Chen et. al 2021) . In response to lower or no earnings women often act as shock absorbers for their households and communities (Elson 1995; van Staveren 2002; Espey, Harper and Jones 2010) . The loss of income in households is partially offset by drawing on women's unpaid care work to replace goods and services that were previously purchased (Ghosh 2013) . A qualitative review of the 2008 financial crisis across 17 middle and low-income countries showed that women spent more time searching for bargains, collecting firewood and water, and recycling clothes to stretch household budgets (Heltberg et al. 2013) . This increase in unpaid work and indirect care work may save resources but it constrains the time available for paid work and direct care work. Data from 38 countries confirms women's increase in unpaid care work during the pandemic and the fact that more women than men are leaving the workforce (UN Women, 2020). In South Africa, data show that both women and men have increased the time spent on childcare though women continue to report more time spent on childcare than men -this is referred to as a 'childcare shock' (Casale and Shepherd 2020, 17) . The limited research focused on women informal workers shows how they absorb the costs of care provision by increasing their unpaid care work and reducing the time spent on paid or unpaid work (Moussié and Alfers 2018; Horwood et al. 2020) . In focus group discussions with women informal workers prior to the pandemic, child care responsibilities were reported to impact on their income security in the following ways a) women seek work that is more flexible, but more insecure and less well-paid, b) childcare changes work schedules in a way that negatively impacts on incomes, c) when women care for children and work simultaneously, they are distracted and productivity decreases, and d) savings are depleted when women cannot work due to childcare responsibilities (Alfers 2016) . Asset depleting strategies such as the sale of assets and taking out new loans can bring in much needed income during crises. In the wake of the 2008 financial crisis, informal workers This article is protected by copyright. All rights reserved. reported that access to low-interest loans allowed them to maintain basic living standards by paying for rent, utilities, food and children's education. They also used the loans to purchase raw materials, stock, and tools and equipment for their businesses (Horn 2009 ). Small low-interest loans can help to fill immediate gaps in care provision, but the effects of debt repayments on unpaid care work is unclear. The microfinance literature is mixed as to the impacts on women's paid work (see Kabeer 2005) and less is known of the effect on women's unpaid care work. However, there is some evidence that debt repayments can add pressure to limit household expenses and thereby increase women's unpaid care work over the medium to long-term (Floro and Dymski 2000) . Economic crises may dampen the demand for goods and services produced by microfinanced enterprises, while increases in unpaid care work among women beneficiaries act as a supply constraint for these enterprises to grow (Vasudevan and Raghavendran 2019) . The essentialization of unpaid care work as women's responsibility means it is often not made explicit as a coping strategy within households during times of crisis (Heltberg et al. 2013 ). Yet there is evidence of the negative impacts on women and girls as caregivers, and children and other dependents as care receivers. Overextending women and girls' time on unpaid care work leads to poorer health, nutrition and learning outcomes for children, decreases in school attendance among girls, inadequate care for the ill and sick, and emotional and physical exhaustion of primary caregivers (Hossain and McGregor 2011; Rai, Hoskyns, and Thomas 2014; Elson 2012; Espey, Harper and Jones 2010) . During the Zika outbreak in 2015, loss of employment and unequal care responsibilities also led to marital tensions, gender violence and, in some cases, abandonment by partners (Freitas et al. 2020) , further exacerbating the vulnerability experienced by many women. Research identifies similar increases in gender-based violence in households following the food, fuel and financial crisis of 2008 (Heltberg, Hossain and Reva 2012) . In turn, gender-based violence impedes women informal workers' access to paid work (Pillinger 2017; WIEGO 2020b) . It is well established in the literature on unpaid care work that gender-responsive social protection measures and quality social services can reduce and redistribute unpaid care work from women in households to the state and other providers (Elson 2000; Razavi 2007 ). This can free up time for women to engage in paid work and guarantee some income security throughout their lives. With school closures and saturated healthcare services, women informal workers had limited access to existing services and expanding these services in a pandemic context proved challenging. The data from South Africa show that workers with more stable incomes during the lockdown period were more likely to be able to pay and send their children to a childcare provider once these reopened than those with unstable incomes during the same period (Wills et al. 2020) . Low-cost unregistered private childcare services catering to low-paid women workers and their children struggled to survive the lockdown period in South Africa with many unable to reopen as they cannot meet the new COVID-19 health and safety guidelines (BRIDGE et al. 2020) . In June 2020, the research advocacy network WIEGO (Women in Informal Employment: Globalizing and Organizing) undertook the first round of a mixed method longitudinal study of urban informal workers in 12 cities 1 across 10 countries. The survey component interviewed 2,292 respondents through a telephonic interview platform (73% women and 27% men). This baseline round of research captured information on working conditions (including earnings) prior to the onset of the crisis (i.e. in January and February 2020) and measured the impact of the lockdowns and the pandemic itself during the height of the crisis (in April and June 2020). The interviews included a core module for all workers and a set of sector-specific questions. The data were captured in an online survey form to ensure consistency across study sites and to facilitate conversion to a statistical analysis package. All interviews were conducted in the respective local language and responses were coded and translated back to English. Respondents were sampled through networks of membership-based organisations (MBOs) of informal workers in each city. The study was conducted in partnership with these organisations as well as the local researchers with whom they (or WIEGO) have worked before. Respondents were sampled using a purposive quota approach in which the sample in each city was designed to reflect the key characteristics (such as type of work, gender, place of work and type of product made or sold) of each organisation's membership. The design of the sample means that it is not intended to be representative of informal workers in each of the cities. There are likely to be a number of systematic differences between workers who belong to MBOs and informal workers who are not members of worker organisations. Moreover, the inclusion of two sectors, home-based and domestic workers, which are comprised almost entirely of women (see Table 1 ) means that the sample does not reflect the real gender distribution within the informal economies of the selected cities. In other words, the study is likely to have over-sampled women informal workers in some cities. More broadly, however, as a non-probability based sampling approach, the survey should not be considered as representative of any sub-group or location. Rather, the sample is intended to reflect the broad characteristics of each group of workers within the MBO. In other words, given the design of the sample, the study results are intended to be broadly indicative of the way in which the impacts of the pandemic differ by, inter alia, the nature of government restrictions, sector of work, severity of the pandemic and gender. . In each city (see Table 1 ), researchers sampled informal workers predominantly from the four sectors outlined earlier 2 . As illustrated in the table, there are three cities (Ahmedabad, Dar es Salaam and Tiruppur) and two sector (domestic workers and home-based workers) sub-samples which are predominantly women. Where gender comparisons have been made, the analysis was run separately to exclude the sectors which are comprised exclusively of women to ensure that the results presented in the paper are consistent. Thus, while the results in this paper are based on the full sample from the table below, they are not driven by the selection of worker group samples which do not include men and are broadly in line with emerging research on the gendered nature of the impact of the pandemic on employment. Nonetheless, the results should be seen as only illustrative of the ways in which women in informal employment may be impacted by, inter alia, the ongoing multi-dimensional crisis and the unequal burden of care. This article is protected by copyright. All rights reserved. The qualitative data, which includes fifty-two semi-structured interviews with informal worker leaders, most of whom were also surveyed, explores workers' experiences of increased care and their interconnected impacts on livelihoods. Furthermore, the interviews provide insights on how care responsibilities are associated with broader informal workers' concerns regarding increased emotional burden as an element of care work, food insecurity and increased debt, all which point to constraints for livelihood recovery. Interviews were analysed and categorized according to references to: indirect and direct care responsibilities, home-schooling, care impacts on work and earnings, food insecurity and coping strategies at the individual and household levels. Across the 12 cities, and as documented in numerous studies on the gendered impacts of the COVID-19 crisis (see Casale and Posel 2020, UN Women 2020), women reported greater increases in household responsibilities due to the pandemic, relative to men (Figure 1 ). About a third of the women in the study sample and a quarter of men reported an increase in direct care responsibilities in the form of care for children, the elderly or household members who are sick. Even larger percentages of women (roughly 50%) and men (about 44%) reported an increase in indirect care activities such as cooking and cleaning. While these gender differences in increased care responsibilities are perhaps not as large as may have been expected a priori, they are in line with existing research on the COVID-19 care crisis and are likely explained by higher levels of care responsibilities borne by women prior to the crisis. In other words, women are likely to have reported greater increases in care responsibilities than men on top of an already unequal distribution of these responsibilities. This article is protected by copyright. All rights reserved. Source: Authors' calculations from the WIEGO 12-City Study (2020). Interviews with informal worker leaders shed light on the ways in which men and women experienced increases in care activities. For households with children and/or elderly, women often referred to the fact that care responsibilities had become more "strenuous," adding to the mounting pressures experienced as a result of a loss of income and earnings. Women workers shared details on how the time spent at home during lockdown was largely dedicated to care responsibilities. A market trader leader in Accra, Ghana noted how "the house chores took most of [her] time", while a home-based worker in Pleven, Bulgaria stressed that more household members at home led to more tasks: "now you need 4 hours to complete the tasks instead of 2 hours." In Accra, Ghana, Lima, Peru and Mexico City, Mexico, worker leaders reported on the struggles single mothers were facing once they had fewer networks of family members to rely on for care support. Worker leaders also mentioned that with more people in households, not only did care work increase, but so did the level of household tensions. In Ahmedabad, India a woman waste picker leader recounted how women were facing constant pressures by family members: "during the lockdown, when everyone was at home, there were constant altercations in the house and [women] have to pay heed to repeated demands for food." Along these lines, a home-based worker in Pleven explained: "My greatest concern was to protect my family's mental health. This article is protected by copyright. All rights reserved. very much a result of the emotional labour that goes into ensuring the wellbeing of family members and the household, consequently adding to their anxiety and exhaustion. As a domestic worker from Mexico City reported: "the double, triple workloads lead to a physical burden, a stronger physical burden, and mental, psychological one." It is worth noting that several of the men interviewed agreed on the fact most household members had been affected by increased care activities. However, men frequently mentioned how women were shouldering a larger part of this work. The Interviews with workers who continued to work in public spaces during lockdown reveal how indirect care responsibilities may have increased due to workers' adherence to strict hygiene protocols upon returning home as a means of protecting family members, specifically children and the elderly. This was particularly the case for street vendors and waste pickers, who expressed greater fear of spreading the virus in their households. A street vendor in Bangkok, Thailand recounted his daily routine to disinfect his clothes, bills and coins in order to protect his young children, while a waste picker from Mexico City mentioned the fears related to having to work and live with many family members in cramped living arrangements. He stated: "we worked fewer hours and took turns to care for our grandmother and mother", however there was "risk of contagion" especially in terms of not being able "to physically distance from each other" in a small apartment. Notably, street vendors and waste pickers alike referred to the fear of becoming ill, but for many of these workers, work was the only way to "bring a plate of food home", as a street vendor from New York, USA noted. In contexts where lockdown measures were strict, there was more pressure on direct care needs within households. In the initial lockdown period of the COVID-19 pandemic many of the direct care needs have been linked to school and day care closures, placing demands on both men and women workers to support online home-schooling and care of younger children. Women workers frequently discussed the challenges related to online schooling, including how it cut into their work hours. A home-based worker from Ahmedabad described her day as "taking care of children, as well as making them attend online classes and helping them do their homework". Not having internet access, smartphones, a computer or tablet exacerbated the difficulties in managing online home-schooling. The risk of infection and lockdown measures make it difficult to shift these direct care responsibilities to others outside of the household. To a lesser extent, concerns were also raised about possible increases in child labour by a woman home-based worker from Tiruppur, India. In Mexico City, a woman waste picker reported concerns over the psychological long-term impacts on children living in households with violence. The qualitative data highlights how the increase in both indirect and direct care responsibilities undermine the capabilities of both the caregivers and care receivers with potential long-term negative consequences. In April, when most workers from the study were living in cities under 'lockdown', women who reported an increase in either direct or indirect care responsibilities reported a reduction in working days to an average of only 1.4 days a week (down from 5.5 days a week). However, among women who did not report an increase in unpaid care work, their average working days in April decreased to 2.2 days a week. In other words, women who reported an increase in unpaid care work reported working about 33% fewer days in April than their counterparts that did not experience an increase in care responsibilities. Among men, working days decreased from just under 6 days a week to about two days, irrespective of whether there was also an increase in care responsibilities. Source: Authors' calculations from the WIEGO 12-City Study (2020). By June and July when restrictions had lifted in most cities, women who reported increases in care work also seemed less likely to return to their pre-crisis working levels. During this period, working days increased to just over 4 days/week for women without increased care responsibilities but only to three days/week for those with greater care responsibilities. Men with increased care roles were also less likely to get back to full-time work after the lockdown period. Those without care responsibilities reported working, on average, nearly 5 days/week in June/July compared with less than 4 days/week for men who increased their care activities. Therefore, women who experienced an increase in their care responsibilities reported a larger decrease in working days during the lockdown period and less of a return to pre-COVID working days in the June-July period. Among men, an increase in care responsibilities was not associated with fewer working days in April but it did coincide with fewer working days after April-perhaps suggesting a slower return to work coincided with an increase in care responsibilities. Another way to consider the association between unpaid care work and the ability to undertake paid work is to identify the percentage of workers who reported working 'zero days' in a particular period. Figure 3 suggests that, in the period immediately prior to the crisis, less than 3 per cent of both women and men reported not working at all in a typical week. Not surprisingly, in April, most workers reported not working at all as cities implemented lockdown restrictions. However, for women, an increase in care responsibilities coincided with a far greater likelihood of not having worked at all (75%) compared with women who did not report increases in care work (61%). Among men there was a much smaller difference in the percentage who reported working in April across the two groups (care increases vs. not). Among both women and men, however, increases in care work are associated with a substantially higher likelihood of not working after the lockdown period (June/July). In this period, 44% of women and 30% of men who increased their care responsibilities did not work at all during the June/July post-lockdown period. These findings suggest that, while the lockdown restrictions affected most workers, those who reported increases in care work seemed less likely to be able to return to paid work in the months following the government restrictions. Source: Authors' calculations from the WIEGO 12-City Study (2020). This article is protected by copyright. All rights reserved. While the quantitative data cannot determine the direction of causality between unpaid care work and paid work, the qualitative data suggest several direct impacts of care responsibilities on work. Several women leaders emphasized how, on the one hand, they were faced with accepting the risks associated with working during the pandemic or, on the other, losing further earnings. The trade-offs most frequently mentioned include: going to work for fewer hours and leaving children at home alone, taking children to work despite city restrictions and safety guidelines, or choosing not to work. The latter option was deemed quite worrisome for many women informal The broad pattern of a slower return to work post-lockdown, particularly for women workers appears to hold across the four worker groups. Among both the sectors with women and men represented -street vending and waste picking -the average number of days worked in June is lower for women who reported an increase in care work even though pre-COVID working days were roughly equal for all workers. The association between care work and fewer working days is also evident when the sample is disaggregated by sector (Figure 4) . The most consistent finding in the graph is that the recovery of working days after lockdown was slowest for workers who reported an increase in care responsibilities. The reduction in working days in April 2020 does not appear to be associated with increases in care work but the recovery (in June 2020) of pre-lockdown working days for those with increased care responsibilities tends to be more muted. This is the case across all four sector groups despite the vastly different working conditions associated with each type of work. Source: Authors' calculations from the WIEGO 12-City Study (2020) Finally, average working days were considerably lower in April for women who reported increases in direct and indirect care (Figure 5a ). Women who did not report indirect care increases, for example, worked about a third more than women who did report an increase. Among men, there was very little difference in April working days for those with and without increased care responsibilities (both direct and indirect). This article is protected by copyright. All rights reserved. Source: Authors' calculations from the WIEGO 12-City Study (2020) By June/July, working days were much closer to 5 days/week for both women and men who did not report increased care responsibilities (Figure 5b ). Women who reported increased care worked, on average, a day a week less than women who did not increase their care responsibilities. Men worked more days overall than women in June/July (despite working similar days before the crisis) and, when they did report additional care work, this was not associated with as large a decrease in working days in June. This article is protected by copyright. All rights reserved. Source: Authors' calculations from the WIEGO 12-City Study (2020) Given the association between an increase in care responsibilities and a larger decrease in paid working days during and after the lockdown period, it is not surprising that there appears to be a correlation between an increase in care work and a decrease in earnings. Among women, in particular, April earnings were substantially lower (about 20% of pre-COVID earnings) for those who reported an increase in care (compared with about 35% for those who didn't report a care increase). Similarly, earnings in June were far closer (about 70%) to pre-COVID earnings for women who didn't report an increase in care. Among women who did increase care responsibilities, their post-lockdown earnings were only about 50% of their pre-COVID levels. Particularly with respect to increases in indirect care, women's earnings were lower in April -only 20% of pre-COVID levels (Figure 6a ). Increases in direct care among women were also associated with lower earnings among women but by a smaller magnitude. This article is protected by copyright. All rights reserved. Source: Authors' calculations from the WIEGO 12-City Study (2020). In the post-lockdown period, many respondents had returned to work but with earnings far lower than pre-COVID levels. Most men in the sample (and irrespective of whether care responsibilities had increased) reported earnings at about 60% of pre-COVID levels. Women who did not report increased care activities reported earnings levels of more than 60% of February earnings. However, women with increased direct care responsibilities appeared to be less able to return to pre-lockdown earnings levels. This group of women reported earnings at only about 42% of their pre-COVID levels. At the same time, women who reported increases in indirect care earned, on average, only half as much as they did before the crisis. This suggests that both types of increased care responsibilities were associated with substantially lower earnings in June for women. The constraint posed by direct care responsibilities may be due to the lag in some countries between the easing of lockdown measures and the reopening of day care centres and schools. Source: Authors' calculations from the WIEGO 12-City Study (2020). In several cities, women worried about the "destabilizing" effect of the pandemic on their income. The uncertainty over the duration of lockdown measures caused anxiety and mental health issues among many women workers. A waste picker from Mexico City reflected on the intertwined losses of work, earnings and "emotional stability" resulting from the pandemic: "It destabilized things in many ways and I can't work with anything else because I have to take care of my children...now everything is devastated". Ultimately, the "critical situation" of not being able to earn any income in order to care for children, while not being able to foresee any "concrete solutions" in the short term reveals the extent of livelihood insecurity faced by many women informal workers across the cities. Given the severe shock to livelihoods among workers that were working largely full-time prior to the COVID 19 crisis, it is not surprising that a reliance on a number of coping mechanisms to mitigate the loss of earnings were reported. Figure 7 illustrates the prevalence of several asset depleting strategies that can lead to increased debt, the loss of already meagre savings and the erosion of household assets. The findings in the graph suggest that both women and men who reported increased care (of any type) were also more likely to have relied on an asset depleting strategy to replace their lost earnings than those workers who did not report increased care. Men were particularly more likely than women informal workers surveyed to have drawn down on their savings (which could also reflect that they were more likely than women to have savings prior to the crisis). The association is somewhat less clear, however, by sector. Domestic workers and street vendors do not appear to have relied more on these particular asset depletion strategies when care responsibilities increased. Among home-based workers and waste pickers, however, the pattern resembles the overall sample where care increases are associated with asset depletion strategies. Resorting to asset depleting strategies resulted from workers' inability to meet household expenditures, particularly electricity, water and rent, along with school fees and related internet costs. This is reflective of the insufficient and unevenly distributed relief provided across the cities 4 . Only 40% of the workers surveyed reported receiving cash or food in the cities where these 4 Please see Chen et. al relief measures were available (Alfers, Ismail and Valdivia, 2020) . The vulnerability experienced by families was described by a woman waste picker in Lima: "There is hunger and the need to pay for water and electricity --we are really behind on all of these responsibilities for months now. How much will they be charging us for the water and the electricity? And although the government put in place its [cash transfers] to provide some relief for all these months, some [workers] have received them and [others] have not. That is why we feel so aggrieved. We really needed to start working, to obtain whatever we can because we were already struggling even before the pandemic." In this sense, interviews gleaned information on how workers perceived the impacts of increases in care responsibilities, coupled with a loss in earnings and inability to work, on an ensuing cycle of debt. For a home-based worker in Delhi the perception was that many other women home-based workers were barely surviving since their "savings had all been eaten into." Traders and vendors from Accra as well as waste pickers in Lima expressed a similar concern over the fact that many colleagues simply did not even have any savings to fall back on. Borrowing money and taking out loans were additional concerns for women because they had no "assurance in terms of their income to pay imminent instalments." Underscoring this view, a home-based worker leader in Ahmedabad describes how women in the sector were affected: "During the pandemic, their debt increased dramatically. People were compelled to take a loan at a 30% interest rate to cover their food needs. They could not ask for money from their relatives as they had also met the same fate. It was such a difficult time that the women had to run the house by getting into debt." These reflections reveal the severity of the constraints households faced. More so, interviews with waste pickers and home-based workers point to frequent concerns women had about how to break the cycle of debt. In many contexts, the COVID-19 pandemic has led to an acute humanitarian crisis. One indicator which captures the effects of such a crisis is the percentage of workers who reported their own and their household's hunger (Figure 8 ). Just over a third of women workers without increased care responsibilities reported that either an adult or child had recently gone hungry in their household. Among women with care increases, the percentages are notably higher (46% and 42%, respectively). Among men in the sample (not shown in the graph) there does not appear to be an association between care increases and household hunger levels. Source: Authors' calculations from the WIEGO 12-City Study (2020). Reflections on food insecurity and particularly the impacts on children within households was a critical theme in interviews with women workers. A common thread throughout interviews was what women workers, who depend on "work for their food", could do to help "children and family members survive hunger". Some workers reported decreases in food consumption, while women particularly described their efforts to ensure that at least their children's nutritional needs were being met. A woman domestic worker from Delhi recounts how many workers did not have enough food "to feed the entire family, so they would give smaller amounts to each member." Another domestic worker from Delhi reflects on a similar strategy among other households: "The women fed their children first, and then would eat themselves if anything was left over, otherwise they would drink water and sleep". In some situations, women were again left with a difficult trade-off: managing food intake of the elderly versus children. A waste picker leader in Ahmedabad summarized the effects of a loss of work and income on food security: "These women are not getting work due to which they don't get money. In turn, the lack of money leads to a lack of food. They could not buy This article is protected by copyright. All rights reserved. food-grains, but one has to fill one's stomach." Narratives from a number of cities reveal how fundamental worker organizations were in facilitating workers' access to food relief. In Ahmedabad, workers leaders from all four sectors highlighted the efforts by the Self-Employed Women's Association to provide or help distribute food aid to workers. In Lima, communal kitchens proved essential support for young women workers who were faced with drastic losses in earnings. Workers' discussions regarding food insecurity were more frequently linked to reports on loss of work and earnings, rather than to direct mentions of increased care responsibilities. However, it is clear how the responsibility for guaranteeing some level of food security fell heavily on women. In turn, this had strong implications on how women perceived their increased care responsibilities. Data from a 12-city study of urban informal workers has shown that most women reported an increase in either direct or indirect care work during the onset of the COVID-19 pandemic. In the sectors in which there are both women and men (street vending and waste picking), a clear pattern of a widening gender gap in unpaid care work emerged across the cities. Similarly, in a sector such as home-based work, found predominantly in the South Asian countries and employing women almost exclusively, roughly 70% of respondents reported an increase in unpaid care work. These increases in care work, both direct and indirect, coincided with fewer working days, on average, during the peak of government restrictions and in a slower return to full-time work by the middle of the year (June). The association between increased unpaid care work and reduced earnings was also a clear pattern which emerged across the 12 cities. Among men that reported increased unpaid care work, the differences between pre-COVID earnings and earnings in April and June 2020 were marginal. Among women, however, there is a clear association between an increase in care work and lower earnings. In April, women with increased care responsibilities earned only 20% of their pre-COVID income and these earnings recovered to only 50% by mid-year. Considering that women's earnings decreased from an already lower base (relative to men's earnings in the informal This article is protected by copyright. All rights reserved. economy) the implications for these losses in earnings are likely to have put many women and their households in a desperate situation. The prevalence of asset depleting strategies highlights the lengths to which many workers were forced to go in order to survive. The increase in care, particularly among women home-based workers and waste pickers, was associated with increases in household debt as well as a reliance on savings (where available). To a lesser extent, informal workers sold household assets to compensate for lost income and while this was not a common strategy, nearly a fifth of homebased workers with an increased care burden reported selling assets. All three of these asset depleting strategies have important implications for the recovery of livelihoods in the postpandemic period. The erosion of already meagre savings and household assets combined with an increase in debt are likely to impede workers, and women in particular, from re-establishing their livelihoods. The disproportionate impact of the care crisis borne by women is, therefore, likely to set back women in informal employment even further (relative to men) during the forthcoming recovery period. Unlike their counterparts in formal employment, informal workers have not been able to rely on furlough schemes, debt relief or other measures designed to help workers and small businesses survive the crisis, and nor do they have access to social insurance such as unemployment benefits, survivor benefits or paid sick leave. In addition, they have limited access to social protection measures to support increasing care responsibilities such as paid parental leave and benefits and health insurance (UNICEF, ILO and WIEGO 2021). The care services they relied on before the pandemic, such as creches and schools, were closed and not prioritised for reopening -even in those instances where informal workers were recognised as essential workers. For example, in Kenya informal traders selling fresh produce were categorised as essential workers but children were banned from markets while schools and creches were closed from March -December 2020. This forced some women informal traders to act outside of the law to care for their children while working in the markets (Boatang-Pobee et al. 2021) . In most contexts, informal workers have fallen between the gaps in emergency relief systems precisely because they are not registered on government databases, tax systems or through unemployment insurance funds (Rogan and Skinner 2020; Chowdhury 2020; De Schutter 2020; ILO 2020d). In addition to the impact of the crisis of employment losses, increasing care needs within households are impacting negatively on women's earnings and time for paid work. This points to trends seen in previous health and economic crises where many women either took longer to recover their earnings or never regained their pre-crisis earnings. It is estimated that by 2030, the pandemic's effects on the resurgence of poverty will be felt hardest by women, particularly those in their prime reproductive and productive years and those from sub-Saharan Africa and South Asia. Projections point to 118 women in poverty for every 100 poor men globally in 2021 with such projections increasing for poor women by 2030 (Azcona et al. 2020) . Understanding these poverty trends has implications for economic recovery plans and longterm developmental outcomes for women informal workers and their dependents. As in past crises, it highlights the need for a paradigm shift where economic and labour market policies are designed to support women informal workers, alongside an expansion in public services and basic infrastructure to reduce and redistribute unequal care responsibilities (Elson 2012 , Ghosh 2013 , Heintz, Staab and Turquet 2021 . The triple dividend that comes from investing in public care services by creating new decent work opportunities, supporting women to engage more in paid work, and protecting those who require care should drive gender-responsive economic recovery plans (UN Women 2015). Investments in public care services will be more effective if they incorporate informal workers' needs, demands and work conditions in their design and implementation, alongside universal social protection policies that guarantee income security when women and men have more care responsibilities or require care throughout their lives (UNICEF, ILO and WIEGO 2021) . Women informal workers' paid and unpaid labour may act as an immediate shock absorber in the absence of social protection, but it leads to poverty, exhaustion and undermines women's capabilities now and in the long-term. This article is protected by copyright. All rights reserved. Our children don't get the attention they deserve: A synthesis of research findings on women informal workers and child care from six membership based organizations Informal Workers and the Social Protection Response to COVID-19: Who got relief? How? And did it make a difference? This Time It's Different: The Role of Women's Employment in a Pandemic Recession From Insight to Action: Gender Equality in the Wake of Covid-19 Women Informal Traders and Child Care during the COVID-19 Pandemic: Findings from Accra, Durban and Nakuru The Plight of the ECD Workforce: An urgent call for relief in the wake of covid-19 Gender and the early effects of the COVID-19 crisis in the paid and unpaid economies in South Africa The gendered effects of the ongoing lockdown and school closures in South Africa Gender, Global Crises, and Climate Change The Unpaid Care Work and the Labour Market. An analysis of time use data based on the latest World Compilation of Time-use Surveys. Geneva: International Labour Office Covid-19 and informal work: Evidence from 11 cities This article is protected by copyright. All rights reserved Kinjal Sampat and RMKU. 2020 Bengaluru: Rajasthan Mahila Kaamgaar Union and the Indian Institute for Human Settlements COVID-19 and the Gender Gap in Work Hours Looking back to look ahead: A rights-based approach to social protection in the post-COVID-19 economic recovery Gender Awareness in Modelling Structural Adjustment Progress of the World's Women. New York: United Nations Development Fund for Women Social Reproduction in the Global Crisis: Rapid Recovery or Long-Lasting Depletion? Crisis, care and childhood: the impact of economic crisis on care work in poor households in the developing world What is a transformative approach to care and why do we need it Can society be commodities all the way down? Polanyian reflections on capitalist crisis Behind Marx's Hidden Abode: For an Expanded Conception of Capitalism How do mothers feel? Life with children with congenital Zika syndrome Financial Crisis, Gender, and Power: An Analytical Framework Measuring Care: Gender, Empowerment, and the Care Economy Economic Crises and Women's Work Don't Let Another Crisis Go to Waste: The COVID-19 Pandemic and the Imperative for a Paradigm shift Living through Crises: How the Food, Fuel, and Financial Shocks Affect the Poor Coping and resilience during the food, fuel and financial crises Manchester: Women in Informal Employment: Globalizing and This article is protected by copyright Attitudes and perceptions about breastfeeding among female and male informal workers in India and South Africa A 'Lost Generation'? Impacts of Complex Compound Crises on Children and Young People International Labour Office. 2018a. Women and Men in the Informal Economy: A Statistical Picture Care work and care jobs for the future of decent work. Geneva: International Labour Office COVID-19 and the world of work: Updated analysis and estimates COVID-19 and the world of work: Impact and policy responses COVID-19 and the world of work: Updated analysis and estimates Extending social protection to informal workers in the COVID-19 crisis: Country responses and policy considerations Marriage, motherhood and masculinity in the global economy: Reconfigurations of personal and economic life Is Microfinance a 'Magic Bullet' for Women's Empowerment? Analysis of Findings from South Asia COVID-19 and gender equality: Countering the regressive effects Women informal workers demand child care: Shifting narratives on women's economic empowerment in Africa Violence and Harassment Against Women and Men in the World of Work: Trade Union Perspectives and Actions. Geneva: International Labour Office This article is protected by copyright. All rights reserved The Political and Social Economy of Care in a Development Context: Conceptual Issues, Research Questions and Policy Options The Covid-19 crisis and the South African informal economy: 'Locked out' of livelihoods and employment Microfinance and the Care Economy Progress of the World's Women 2015-2016: Transforming Economies, Realizing Rights. New York: UN Women Whose Time to Care? Unpaid Care and Domestic Work During Covid-19 COVID-19 Global Gender Response Tracker Family-friendly Policies for Workers in the Informal economy Informal Workers in the COVID-19 Crisis: A global picture of sudden impact and long-term risk A Sector Hanging in the Balance: ECD and Lockdown in South Africa 2020-A-Sector-Hanging-in-the-Balance-ECD-and-Lockdown