Monitoring COVID-19 Monitoring COVID-19 impacts on firms in Ethiopia REPORT REPORT NO. 44 NO. 21 / July / 2020 21 / July / 2020 impacts on firms in Ethiopia Does the COVID-19 pandemic affect women-owned firms differently? Authors: Girum Abebe, Tom Bundervoet, Christina Wieser Further to the unprecedented levels of health crisis, the COVID-19 pandemic has caused severe disruptions to allocation of resources and economic activities globally. A series of on-going studies including the World Bank’s briefing series show that the pandemic has negatively affected many businesses but to a differing extent. Firms’ exposure to the pandemic and the manner with which they respond is shaped by both external factors, such as the economy’s stage of development, availability of bailout packages and national economic resilience, as well as internal factors, such as scale, export- orientation and sector of operation. Whereas there is no immediate parallel to the pandemic, as in any economic shock firms periodically face, entrepreneurial characteristics, such as sex, skills, experience and embeddedness in rich business network, are key pre-existing firm idiosyncrasies that potentially affect business’ vulnerability and ability to absorb shocks of this nature. Women-operated businesses particularly face intersecting difficulties of scale and gendered challenges that may amplify their vulnerability to the pandemic—women are more likely to run informal, micro and small businesses in low-return sectors than men. A World Bank gender diagnostic, for example, finds that women managed businesses in Ethiopia have lower sales turnover, limited access to formal credit, and are less likely to use hired labor. In Ethiopia, women-owned businesses generally have limited access to credit and savings tend to be paltry even before the pandemic (World Bank 2019). In times of uncertainty, a major interruption in cash flow can thus seriously strain their liquidity position and intensify the effect of preexisting constraints making survival more challenging. Further, with the COVID-19 pandemic and related confinement measures, including school closures, the rising demand for child and family care, and, the sick, household responsibilities disproportionally fall on women, possibly diverting their limited resources and time away from the businesses. There is, for example, evidence that quarantine and confinement measures associated with the Ebola virus has considerably impaired women’s economic and livelihood activities in Liberia ( UN 2020). The COVID-19 pandemic can thus potentially exacerbate existing structural inequalities between men- and women-owned firms further. It is thus crucial to understand how the pandemic and containment efforts affects women-owned businesses and what gender-sensitive measures might constitute effective response to attenuate the potential negative impacts. We employ sex-disaggregated data from a high-frequency phone survey of firms (HFPS-F) to explore the gender-specific effects of the COVID-19 pandemic. The data is primarily collected to monitor the impacts of the pandemic and related containment measures on formal firms in Ethiopia.1 The HFPS-F interviews a random sample of firms in Addis Ababa every three weeks for a total of eight survey rounds, and an additional sample of firms in four other cities in Ethiopia (Adama, Bahir Dar, Hawassa, and Mekelle) for a total of seven rounds. The survey in Addis Ababa started with an original sample of 645 firms in April 2020.2 For the current study, three rounds of the HFPS-F survey consisting of a balanced panel of 454 firms from Addis Ababa is used.3 Occasionally results from pooled data of roughly 1,362 observations are discussed. The three rounds are roughly three weeks apart and were implemented between April 15 and June 18, 2020 in Addis Ababa.4 The sampling strategy is explained in detail in a companion technical note. This high-frequency follow-up employs a survey instrument that collects detailed information on firm 1 The results from the survey is hoped to inform the policy response to monitor and mitigate the social and economic impacts of the crisis brought about by the pandemic. The World Bank, in collaboration with the Jobs Creation Commission (JCC) lead the project and a consultancy firm, Laterite (Ethiopia) Ltd., collected the data. 2 Though we use the term “firm” for ease of understanding, our sample consists of establishments. An establishment is an economic unit that produces one predominant activity, typically at a single physical location. A firm, on the other hand, can consist of one or more than one establishment. We use the term firm and business interchangeably throughout the report. 3 We are not able to combine the Addis Ababa and the regional cities sample as the sampling procedure is very different. We will examine the whether the COVID-19 pandemic affected female businesses differently in regional cities in a separate analysis. 4 At the beginning of R1 on April 15th, the total confirmed COVID-19 cases were 85. On R2 on May 6, the country’s total confirmed COVID-19 cases were 162. By the end of R3, on June 18th, the confirmed cases had increased to 3,954. Monitoring COVID-19 impacts REPORT NO. 4 on firms in Ethiopia 21 / July / 2020 operations, hiring and firing, and expectations of future operations and labor demand, which help compile key indicators in order to better tailor and implement interventions and policy responses and monitor their effects over time. Three survey briefs highlighting the impact of the pandemic on firms’ operations and labor dynamics are already produced. The purpose of this paper is thus to address a key open question on whether and how the pandemic affects women-owned firms differently from men-owned ones. KEY HIGHLIGHTS – GENDER BRIEF In comparison to men-owned businesses, women-owned firms are relatively small—55 percent of women-owned businesses had no employee in contrast to 44 percent of men-owned businesses. Further, only 4 percent of women- owned businesses generated sales turnover exceeding 1 million ETB in the last fiscal year (2011 EC). By contrast, 14.5 percent of men operated business that exceeded the 1 million ETB mark in turnover the same period. A large share of women-owned firms is engaged in trade, tourism and hospitality: sectors that are considered immediate- risk industries for business disruptions due to the COVID-19 pandemic. Perhaps reflecting the entrepreneurs’ resilience, women-owned businesses are no more likely to remain closed compared to men-owned businesses. The COVID-19 pandemic has further widened the gender gap in business earnings. While all firms have experienced a drastic decline in sales turnover, the dip appears to be more severe in women-owned businesses — they generated less than 20 percent of the sales revenue they had earned the same month the previous year. The COVID-19 pandemic has thrown firm profitability into a tailspin. Women-owned businesses made about ETB 10,000 per month in gross profit a year before the on-set of the COVID-19 pandemic. Since the pandemic, they have experienced a significant drop in profit and an acceleration of losses: losses jumped from ETB 786 in April, 2020 to ETB 6,000 in June 2020. While worker layoffs are limited in scale, women employees and those with temporary contract appear to be affected the most. When the pandemic hit in April 2020, 64 percent of laid off workers were women and 54 percent were on temporary contracts. Layoffs have tapered off in recent weeks. However, despite making up only 42 percent of the workforce, 57 percent of workers that were laid off in June 2020 are women. While women-owned businesses are disproportionally affected by the COVID-19 pandemic, less than one percent have received any type of government support. Across the three survey rounds, a total of 18 firms report to have accessed such support services: only two of them are women-owned businesses. The majority of women as well as men entrepreneurs identify waiving of taxes payment (corporate income tax, VAT, private pension contribution, excise tax) as the main policy lever the government should exercise to help them overcome their current business woes. Recent improvements notwithstanding, firms’ outlook about their businesses pro spect in the next several weeks remain subdued. Both men- and women-owned businesses are growing pessimistic about how soon they can return to pre- COVID 19 operation levels. In April 2020, they anticipated that it would take on average 11 weeks. Reflecting growing concern with the spread of the virus and the uncertainty surrounding containment efforts, the number of weeks drastically increased to 24. One in three businesses in our sample are owned by women entrepreneurs. Women and men entrepreneurs are largely similar in basic observable characteristics. Women (men) entrepreneurs are on average 37 (39) years of age, attended 11 (12) years of schooling and 70 percent (70 percent) are married. The sample of firms includes registered firms operating in the industry (13.5 percent) and services (86.5 percent) sectors. Women-owned businesses are over-represented in the service sector with 96.2 percent Monitoring COVID-19 impacts REPORT NO. 4 on firms in Ethiopia 21 / July / 2020 operating in service compared to 81.8 percent for men (Figure 1). About half of women-owned businesses are in wholesale and retail trade and a quarter are in tourism and hospitality, mainly including food vending, restaurants and bars. Women entrepreneurs are thus disproportionally active in high-contact sectors that may be exposed to heightened infection risks and business disruptions. By contrast, men businesses are concentrated in wholesale (33.0 percent), manufacturing (7.7 percent), construction (10.4 percent) and transport and storage (24.1 percent). About 6 percent of women-owned firms export in comparison with 4 percent for men. Out of 7 women- owned firms that export, five are wholesalers, one is in manufacturing and one is in information technology. The sectoral distribution of exporting pattern is also similar in men-owned firms — 6 are wholesalers, two are in manufacturing and construction. Figure 1: Sectoral distribution of firms Services Industry Education and Health Professional and administrative services Information and financial services Tourism and Hospitality (incl. restaurants, bars, etc.) Transportation and storage Wholesale and retail trade Construction Manufacturing All Women Men 0 20 40 60 80 100 120 Women entrepreneurs operate small and young businesses. 55 percent of women-owned businesses are owner-operated (own-account) with no other paid employee (Table 1). By contrast, 44 percent of men- owned businesses are owner-operated. While 13 percent of men-owned businesses can be considered as Small, Medium, and Large (SML), the corresponding figure for women-owned businesses is 6 percent. Taken together, the first two rows of Table 1 indicate that women-owned firms are preponderant in the smallest enterprise category; about 94 percent of the firms are either own-account workers or are micro enterprises. Men-owned firms on average employ 4 workers, twice the average employment by women-owned firms. The median number of workers is 0 in women-owned businesses and 1 in men-owned businesses. Even after excluding own account firms, the mean (7) and median (3) employment in men-owned firms is larger than the corresponding mean (4) and median (2) of women-owned businesses among firms with paid workers (not shown in the table). When firms use paid workers, Table 1 shows that women-owned firms are more likely to employ female workers than men-owned businesses. About one in three workers in men- owned firms are women, while the corresponding figure jumps to two third in women-owned businesses. Overall, about 42 percent of the workforce in our sample of firms are women. Tourism and hospitality, professional services as well as education and health are sectors with the highest concentration of women workers; i.e., women employees constitute more than half of the entire workforce. The heavy presence of female workers in more susceptible sectors, such as tourism and hospitality, suggests that women’s welfare might be disproportionally affected by the pandemic. Close to a quarter of the firms’ workforce is also on temporary contract and relatively a large share is observed in men-owned firms: 26 percent in men and 21 percent in women-owned firms. The sample firms pay an average monthly wage of ETB 2,200 birr for low skilled workers, a pay scale that is roughly equivalent to the average monthly wage in Addis Ababa in 2018. Low skilled workers in men- owned firms earn a 10 percent premium over low skilled workers in women-owned firms. Wage for high- skilled workers in women-owned firms, however, constitutes only 62.5 percent of the corresponding wages in their men counterpart. The absence of a competitive pay structure in women-owned firms might make it difficult for these firms to attract and retain talent. In fact, as will be shown shortly, women-owned firms tend to be prone to worker quits more than men-owned firms. Table 1 also indicates that most of the firms in our sample are young with 12 percent and 60 percent less than one and five years old respectively. There does not appear to be notable difference in firm age between women- and men-owned businesses across the age groups. Monitoring COVID-19 impacts REPORT NO. 4 on firms in Ethiopia 21 / July / 2020 Table 1: Size and age of firms by sex Men Women All Firm size Own-account firm 43.6 55.3 47.4 Micro-enterprise (1-5) 43.1 38.8 41.7 Small, medium or large enterprise (>5) 13.30 5.9 10.9 Employment Average number of workers 4 2 3 Median number of workers 1 0 1 Share of Women workers (%) 32.5 65.7 41.8 Share of temporary workers (%) 26.2 20.9 24.7 Wages (in thousands of ETB)a Monthly wage (low skilled worker) 2.2 2.0 2.2 Monthly wage (High skilled worker) 6.4 4.0 5.8 Years of operation 0 years 13.1 12.3 12.8 1-2 years 30.1 29.4 29.9 3-4 years 17.0 21.8 18.6 5-9 years 23.6 24.6 23.9 10 years or more 16.2 12.0 14.8 Number of firms 305 149 454 Note: The wage data is remarkably stable across survey rounds and the figures in the table represents monthly averages pooled across the three waves. a skill levels are defined based on the self-evaluation of the respondent as what constitutes high skill and low skill requirements in production or service delivery in his/her line of business. Mirroring the visible difference in firm size between the two groups, sales turnover also appears to be much lower in women-owned businesses. In 2011 Ethiopian Fiscal Year (EFY)5, one in three women-owned businesses earned less than ETB 50,000 per year in sales revenue and 64 percent earned revenue that is less than half a million ETB (Table 2). In contrast, about 15 percent of men-owned businesses earned less than ETB 50,000 and about 52 percent had sales revenue below half a million ETB. Men-owned firms are also more represented in the highest revenue bracket. Only 4 percent of women-owned businesses generated sales turnover exceeding 1 million ETB in the 2011 EC fiscal year. By contrast, 14.5 percent of men operated business exceeded the 1 million ETB mark in the same period. Table 2: Share of firms by last fiscal year’s (EFY 2011) annual revenue (%) Men Women All Less than 50,000 Birr 15.7 33.7 21.7 50,00-100,000 Birr 14.7 17.0 15.4 100,000-500,000 Birr 16.4 13.4 15.4 500,000 - 1 million Birr 5.3 5.6 5.4 1 million -10 million Birr 14.5 3.9 11.0 Over 10 million Birr 5 0.4 3.4 Don't Know 11.1 13.8 12.0 Refused 13.3 9.1 11.9 Firm Not Open / Not Yet Generating Revenue / Not Applicable 4.1 3.2 3.8 5 Ethiopian fiscal year runs from July 1 to June 30. Ethiopia follows its own calendar, which is 7 to 8 years behind the Gregorian calendar. The 2011 EC fiscal year, thus ran from July 1, 2018 to June 30, 2019. Monitoring COVID-19 impacts REPORT NO. 4 on firms in Ethiopia 21 / July / 2020 Evidence suggests that containment measures following the COVID-19 pandemic have caused a perceptible decline in demand, severe disruptions to supply chains and fall in equity markets have affected current operations as well as prospects of firms across the globe. Firms in Ethiopia are no exception. The survey confirms that firms’ operations have been heavily affected by the consequences of COVID-19. The immediate impact of the pandemic and containment measures were manifested in closure of businesses in Addis Ababa. In the first-round survey (April 15- May 5, 2020), 39 percent and 40 percent of women- and men-owned businesses respectively were closed (Figure 2). Figure 2: Share of businesses closed in three rounds 39.9 39.5 38.8 SHARE OF CLOSED BUSINESSES 27.1 26.9 26.6 25.1 24.9 24.4 ROUND 1 ROUND 2 ROUND 3 Men Women All Note: The figure shows the share of businesses that were not operational when calls were placed during each of the survey rounds. In the course of the 9 weeks period, a gradual opening of both types of businesses was observed. By the third round in June 2020, the share of all businesses closed declined by 15 percentage points; about a quarter of both men- and women-owned businesses remained closed at the end of June. Further, a comparable share of men- and women-owned businesses are fully operational in R1 and R2 (Table 3).6 In R3, 51 percent of men-owned businesses were fully operational, while the corresponding figure is 55 percent for women-owned firms. Moreover, among businesses that were operational, both men and women entrepreneurs tend to keep them open for 40 percent, 68 percent and 72 percent of all available days in R1, R2 and R3 respectively (not shown in the table). There is also no evidence that women entrepreneurs, compared to men, spend less hours of work on their businesses; for open businesses both men and women business owners spend 8 hours per day on the businesses.7 Table 3: Share of firms by days operational the three rounds, percent Men Women All R1 R2 R3 R1 R2 R3 R1 R2 R3 0 days 39.9 27.1 25.1 38.8 26.6 24.4 39.5 26.9 24.9 1-9 days 30.9 16.5 13.3 30.8 20.4 12.7 30.8 17.8 13.1 10-14 days 29.2 12.6 10.6 30.5 11.4 8.0 29.6 12.2 9.8 15-21 days - 43.9 51 - 41.5 54.9 - 43.1 52.3 Note: In the first round, firms were asked about days of operation in two weeks window (10-14 days), thus information on 15-21 days period is missing. With firm closure and scaling down of operations, workers’ tenure security is a key a public policy concern. It is for this reason that the Ethiopian government’s state of emergency stipulated a key provision banning firms from laying off workers amid the pandemic. Notwithstanding such legal obstacles, firms still laid off workers across the three rounds, albeit with limited scale. About 4 percent (9.3 percent) and 5 percent (4 percent) of men- and women-owned firms respectively had laid off workers in R1 (R2). There is, however, striking difference in the nature of workers being let go. Women and temporary workers appear to be mostly affected by layoffs, particularly in the first round. In R1, about two thirds of the workers laid off are women, and the share of women workers laid off by women-owned firms in R3 is 14 percentage points higher than in men-owned firms. Despite making up only 42 percent of the workforce, 57 percent of workers that are laid off in June 2020 are women. Perhaps reflecting the early shedding of workers, layoffs 6 Fully operational is defined as firms that was operational for 10-14 days in round 1 and 15-21 days in rounds 2 and 3. 7 These also constitute hours that a business is open, but the owner is sitting idle and has nothing to do because customers or buyers do not turn up. Monitoring COVID-19 impacts REPORT NO. 4 on firms in Ethiopia 21 / July / 2020 seem to be tapering off in recent weeks, with 4 percent and 3 percent of men- and women-owned firms respectively reporting the laying off workers in R3. In interpreting the layoff figures an important caveat on sample size is in order: across the three survey rounds a total 244 workers were laid off from 39 firms, and hence sample size concerns pervade inference. Further to layoffs, firms resort to granting leaves as an immediate-run solution to business crises. Across the survey rounds, 30 percent to 20 percent of firms have granted leave to one or more of their workers.8 More than 95 percent of workers, who were granted leave, had been given paid leave. The share of women businesses that granted leave to workers is slightly less than men businesses. This is partly because women businesses are more likely to experience worker quits. About one in ten women-owned businesses reported worker quit in R1, while only 3 percent of men-owned firms lost workers the same way. Given difficult conditions and lack of full-time tasks at the job (capacity utilization was about 35 percent in R3), it is not surprising that workers vote with their feet deciding to quit the firm even when very few firms have reduced the salary of the remaining workers (Table 4). Given an expected slow-down in economic activities related with the pandemic, the vast majority of workers who leave for any reason (quit or contract termination) are less likely to find alternative employment opportunities. In Hawassa Industrial Park, for example, more than 90 percent of the female workers who left employment in the park in the post-COVID- 19 era did not find any other jobs, and about 80 percent would like to go back to their previous employment (Demeke et al. 2020). This is mainly because firms are not hiring new workers. This dovetails well with the observation in our sample that shows that hiring of new workers has also generally remained low with steady improvement among men-owned businesses from 1 percent in R1 to 2.5 percent in R3 (Table 4, bottom row). Table 4. Changes in worker turnover and salary in the three rounds, percent Men Women All R1 R2 R3 R1 R2 R3 R1 R2 R3 Share of firms that laid off workers 4.1 9.3 4.1 5.3 3.9 2.6 4.4 7.8 3.7 Share of laid off workers who are women 64 34 55 63 20 69 64 32 57 Share of laid off workers who are temporary 68 62 79 23 40 31 54 59 70 Granted leave to workers 29.6 20.5 21.7 25.6 18.1 16.4 28.5 19.9 20.4 Workers quit 3.3 1.5 2.6 10.4 4.7 4.9 5.1 2.4 3.2 Reduced salary of workers 1.6 2.7 1.9 0 2.4 0.9 1.2 2.6 1.7 Hired workers 1 1.5 2.5 1.4 2.8 1.4 1.1 1.9 2.1 Note: Overall few firms have experienced worker layoff, granting of leave, reducing salary and hiring of workers. A perceptible decline in commodity demand triggered by the COVID-19 pandemic is by far the single most important challenge businesses are facing today. Collapse in demand combined with containment measures that affect production and services would make serious dent in firms earning potential. Indeed, since the onset of the pandemic, firms have experienced unprecedented decline in sales revenue. Firms in the sample are earning about 31 percent of their last year’s sales revenue in the same month (Table 5). In the previous year, monthly sales revenue in men-owned (women-owned) firms, adjusted for inflation and averaged across the three months, was ETB 343,000 birr (ETB 48,000). The sample average monthly revenue during this period was valued at ETB 237,000. The current mean monthly sales revenues ranged from ETB 150,000 in men-owned firms in R1 to ETB 8,000 in women-owned firms in R2 with the sample average of ETB 72,000 birr across the three rounds. The share of firms with zero revenues increased in men-owned firms from 32.6 percent in R1 to 38.5 percent in R2 and 38.2 percent in R3. In women-owned businesses, this share increased from 33.7 percent in R1 to 37.4 percent in R2 and declined to 31.9 percent in R3. While most of the zero revenue observations is driven by firms that are closed, a significant share of firms still reported to earn no sales revenues even if the businesses were kept open during this period. About 18 percent of women with open businesses reported zero revenue in R1 and R2, in R3 the proportion reporting zero revenues declined to 12.4 percent. On the other hand, the share of men-owned businesses that report zero revenues more or less remained steady between 18 percent and 20 percent across the three rounds. In general, the fact that about 17 percent of firms in the sample have failed to 8 Demeke et al. (2020) report that about a quarter of female workers in Hawassa Industrial Park have been on paid leave since the start of the pandemic. Monitoring COVID-19 impacts REPORT NO. 4 on firms in Ethiopia 21 / July / 2020 generate sales revenue while still being fully or partly operational depicts a worrying picture for business sustainability. Table 5: Monthly sales revenue in the current and previous year Men Women All R1 R2 R3 R1 R2 R3 R1 R2 R3 Sales Previous year same month revenue (in 000 Birr) 283.2 373.4 372.1 51.9 42 49.9 199 251.9 259.8 Current month sales revenue (in 000 Birr) 150.1 85.3 86.3 9.0 8.0 8.8 98.5 57.0 59.1 Share of firms with no revenue (%) 32.6 38.5 38.2 33.7 37.4 31.9 33 38.1 36.1 Profit Previous year same month profit (in 000 Birr) 33.8 37.1 33.3 9.4 9.4 10.6 25.7 28 26 Current month profit (in 000 Birr) -2.7 -6 -9.8 -0.8 -2.2 -6.3 -2.1 -4.7 -8.6 Share of firms with no or negative profit (%) 73.3 71.4 72.5 75.8 73.7 76.7 74.1 72.1 73.9 Note: Revenue and profit figures in the last year same month (in April and May 2019) are adjusted for inflation using the Consumer Price Index as an inflator. Across the three waves, the decline in sales revenue appears to be more severe in women-owned businesses—they generated less than 20 percent of the sales revenue they had earned the same month the previous year, before the COVID-19 pandemic outbreak. Further, in the previous year, sales revenue in women-owned businesses were about 11 percent to 20 percent of men-owned businesses (Figure 3). After the COVID-19 pandemic, the average sales of women-owned firms was about 6 percent of men-owned firms in R1 (April 2020) and slightly picked up to 10 percent in R3 (June 2020). Firms were running profitable businesses before the onset of the pandemic (Table 5). Men-owned firms made about ETB 33,000 to 37,000 in monthly profits between April and June 2019, while women-owned businesses earned about a third of this amount. After the onset of the pandemic, however, firms are operating with significant losses. Losses are particularly piling up on women-owned businesses at an alarming rate. In April 2020, women-owned firms recorded a loss of ETB 786, which is about 30 percent of losses incurred by men-owned firms. In the subsequent periods, however, losses have accelerated from ETB 2,200 in May, 2020 to ETB 6,300 in June 2020. The magnitude of losses incurred by women-owned businesses is equivalent to 64 percent of the losses reported by men-owned businesses. This is particularly alarming as women-owned firms’ sales revenue constitute only 10 percent of their men-counterparts in the same period. While the share of women-owned businesses that remain operational is comparable to that of men, the significant decline in sales revenue and the rapid deterioration of profitability is a compelling sign that the pandemic might be hitting women-owned businesses the hardest, exacerbating pre-existing inequalities further. Figure 3: Share of women-owned firms’ sales revenue to their men counterparts, percent 20 18.3 Women to men sales 15 13.4 11.2 10.2 9.4 ratio (%) 10 6 5 0 Round 1 Round 2 Round 3 Before Covid-19 After Covid-19 Note. Revenue figures in the last year same month (in April and May 2019) is adjusted for inflation using the Consumer Price Index (CPI) as an inflator. Monitoring COVID-19 impacts REPORT NO. 4 on firms in Ethiopia 21 / July / 2020 To guard against attribution problems and unwarranted prompts, except in the last module dubbed ‘COVID’, the survey did not directly mention the COVID-19 pandemic when posing questions. In this section, firms were asked whether and how they are affected by the pandemic and the potential policy responses that they deem necessary to alleviate its impacts. At the start of the survey in April 2020, 94 percent and 96 percent of men and women entrepreneurs respectively reported that their businesses were affected by the pandemic in one or another way (Table 6). In subsequent months, nearly all firms that were interviewed stated that their businesses are affected. Consistent with the preceding observation of the substantial deterioration of sales and profitability of firms, firms confirm that the pandemic is the main reasons for a slump in business. Women entrepreneurs attribute a large share of lost sales revenue directly to the pandemic compared to their men counterparts; in June 2020, women-owned firms report to have lost 80 percent of their sales revenue due to the pandemic (Table 6). Further, about a quarter of women- owned and 30 percent of men-owned firms have cancelled investment or asset purchase owing to the pandemic, a pattern that largely remained unchanged across survey waves. Not surprisingly, both men- and women-owned firms are currently utilizing about 36 percent of their production or service capacity. Table 6. Share of firms affected by COVID-19 pandemic and capacity utilization rate, percent Men Women All R1 R2 R3 R1 R2 R3 R1 R2 R3 Firm affected by COVID-19 93.7 97.6 98.4 95.8 99 100 94.4 98.1 98.9 Share of sales lost due to COVID-19 55.5 74.1 76 66.3 80.4 79.9 59.1 76.2 77.3 Cancelled investment or asset purchase 29.7 29.6 30.1 19 23.2 23.1 26.2 27.5 27.8 Current Capacity Utilization rate 35.6 35.5 35.6 Note: Data on capacity utilization rate is available only in round 3. In April 2020, about 60 percent and 67 percent men and women entrepreneurs respectively identified low demand for their products or services to be the main way that the pandemic affected the firm (Table 7). In subsequent rounds, this figure jumped significantly, and, by June 2020, more than 80 percent of all firms suffered from low demand. As businesses attempt to reopen and resume production and services, access to workers and raw materials become imperative. Restricted movement of workers continued to be the second COVID-19 induced constraint that affected firms. Reflective of their larger share of paid workforce, slightly more men-owned firms are affected by restrictions on worker mobility. Table 7 also shows that firms are increasingly worried about the rise in price of raw materials and intermediate goods. At the on- set of the pandemic in April 2020, about 7 percent and 9 percent of men- and women-owned businesses reported higher raw material prices to be a concern. In June 2020, the share of firms identifying this as a problem has more than doubled. Table 7: Ways in which firms are affected by COVID-19, Percent Men Women All R1 R2 R3 R1 R2 R3 R1 R2 R3 Low demand for products/services 60 79.6 80.3 67.1 86.7 80.9 62.3 82 80.5 Lower supply of raw materials and intermediate goods 9.6 6.8 9.3 6.6 6.3 11.0 8.6 6.6 9.8 Restricted movement of workers 27.9 25.3 28.7 22.2 16.8 21.1 26 22.5 26.2 Forced closure of business 23.7 18.3 21.6 33.6 19.2 25.3 26.9 18.6 22.8 Closure of marketplace/shops 11.6 6.1 9.0 17.3 11.5 8.0 13.4 7.9 8.7 Higher price of raw materials and intermediate goods 6.8 12 15.6 8.6 11.2 18.2 7.4 11.7 16.4 With more than a third of firms generating no revenue, sales slumping, and profitability plummeting, it is inevitable that many firms will find themselves in financial distress. Difficulties in making rent payment is the main financial concern faced by both men- and women-owned businesses in April and May 2020 (R1 and R2 in Table 8). Nearly half of the women entrepreneurs in April and 54 percent in May stated that rent constitute their biggest financial burden in running the business. In June 2020 (R3), more firms identified a non-descript category of “other expenses” as their main financial problem. Comparable share of men and women entrepreneurs struggle with payment of invoices. About a quarter of men entrepreneurs and one in five women entrepreneurs are also exposed to financial pain due to staff wages and social security charges, such as pension contributions. Overall, the nature of financial burden faced by businesses does Monitoring COVID-19 impacts REPORT NO. 4 on firms in Ethiopia 21 / July / 2020 not appear to differ by the sex of the business owner and have largely remained stable across survey rounds. Table 8: Main Financial problems currently faced by firms, (%) Men Women All R1 R2 R3 R1 R2 R3 R1 R2 R3 Staff wages and social security charges 18 26 23 19 19 19 18 24 22 Rent payment 39 41 38 50 54 48 42 46 41 Repayment of loans 9 11 11 10 12 9 9 11 10 Payments of invoices 29 37 36 29 39 39 29 38 37 Other expenses 27 32 52 33 41 54 29 35 52 No specific problem 17 7 10 7 2 2 13 5 7 To facilitate recovery, the government has publicly announced that a rescue package to support both established as well as small and vulnerable firms is on its way. While across the three survey rounds a total of 18 firms report to have accessed such support services, only two of them are women-owned businesses. The support accessed by firms in our sample include reduction or freeze of operational costs such as costs for sheds and working spaces, tax payment rescheduling or waivers, debt relief, interventions on rent (protection against eviction), export support, the provision of personal protective equipment (PPE), COVID-19 testing and medical guidance. The two women-owned businesses each received support on rent and COVID-19 testing services. While women-owned businesses are disproportionally affected by the COVID-19 pandemic, less than one percent have received any support. Further to assessing their access to government support, firms were requested to name preferred policy responses that they would like to immediately obtain from the government. More than 40 percent of firms elected waiving taxes payment (corporate income tax, VAT, private pension contribution, excise tax) to be the main policy lever the government should exercise to help them overcome their current financial troubles (Table 9). This policy option has particularly been embraced by women-owned businesses and appear to be growing in popularity across survey rounds. In May and June 2020, for example, more than half of the women entrepreneurs in the sample voted for tax waiver as the main policy tool that the government should use. Consistent with rent being one of the main financial problems, about a quarter of women entrepreneurs and 20 percent of men entrepreneurs identify covering, reducing, or freezing of operational costs such as costs for sheds and working spaces as key area of intervention by the government. Policy measures that facilitate access to capital grant and zero-interest loans are also identified as appropriate measure by about 36 percent and 42 percent of men and women entrepreneurs respectively in R3 to help firms cope with an interruption in cash flow. Similarly, comparable number of men and women entrepreneurs (8 percent to 13 percent) suggested freeze of loan repayment, extension of loan terms, or partial debt relief as a policy action to alleviate their financial burden. By contrast, only 3 percent to 4 percent of both types of firms identified the reduction or deferral of payroll taxes or providing wage subsidies as main policy responses to help them weather the COVID-19 pandemic induced crisis. Monitoring COVID-19 impacts REPORT NO. 4 on firms in Ethiopia 21 / July / 2020 Table 9: Suggested policy measures by firms, percent Men Women All R1 R2 R3 R1 R2 R3 R1 R2 R3 Covering, reducing, or freezing of operational costs such as costs for 19.9 16.8 20.7 31.1 27.3 25.3 23.6 20.3 22.2 sheds and working spaces Reduction or deferral of payroll taxes or providing 3.4 4.3 3.3 2.8 4.1 3.8 3.2 4.2 3.4 wage subsidies Waiving taxes payment (corporate income tax, 39.7 48.3 46.8 43.6 56.8 54.8 40.9 51.1 49.4 VAT, private pension contribution, excise tax) Freeze of loan repayment, extension of 13.4 12.3 12.7 13.8 8.7 8.1 13.6 11.1 11.2 loan terms, or partial debt relief Providing access to capital through financial 10.8 11.5 13.5 14.2 18.7 21.1 11.9 13.9 16 grants Providing access to capital through access to 9.8 18.3 20.9 6.6 17.3 20.9 8.8 18 20.9 zero-interest loans Note. Other potential policy measures were marginal in terms of importance and are not shown in the table. Monitoring COVID-19 impacts REPORT NO. 4 on firms in Ethiopia 21 / July / 2020 COVID-19 does not only seriously impact firms’ financial prospects, it also changes how firms operate. In April, about 35 percent of firms claimed to have enforced strict social distancing requirements (Table 10). In June, the number increased to 43 percent in men- and 44 percent in women-owned businesses. Reflecting the difficulty of enforcement in confined and overcrowded spaces of production and service delivery, many firms typically operate in, more than half of the businesses concede that they have not placed strict social distancing requirements. On the other hand, more than 70 percent of firms (79 percent in R3) claim to offer alcohol, soap or sanitizers in the workplace from the start of the pandemic. A significant improvement across survey rounds is also observed in the provision of protective gears, such as masks and hand gloves, to workers. Among women-owned businesses, for example, about 37 percent of the firms supplied protective gears to their worker in R1, a number that jumped to 80 in R3. Table 10. Safety and protection of workers against the pandemic Men Women All R1 R2 R3 R1 R2 R3 R1 R2 R3 Place strict social distancing requirements in the workplace 35.6 41.5 42.5 35.3 48.8 45.7 35.5 43.9 43.6 Place alcohol and/or sanitizer at the workplace 72.4 77.9 79.2 71.3 75.8 79.2 72 77.2 79.2 Provide protective gears, such as masks and hand gloves, to workers 44.9 70.1 80.6 36.7 61.9 79.7 42.2 67.4 80.3 The uncertainty of the trajectory of the COVID-19 pandemic in infection rates spread, its duration as well as the ensuing severity of containment efforts renders estimating the scale and speed of economic recovery with reasonable accuracy cumbersome. The survey indicates that both men- and women-owned firms are growing equally pessimistic about the possibility of resuming businesses as in the pre-COVID 19 state soon (Figure 4). In R1 and R2, Women-owned firms anticipated it would take on average 11 and 17 weeks respectively to go back to pre-crises business mode. Reflecting growing concern with the sweeping spread of the virus and the uncertainty surrounding containment efforts, the number of ‘weeks to normality’ drastically increased to 24 in R3, a 40 percent increase over R2. If the businesses environment, especially demand for product and services, does not improve in the next few weeks, it is not clear whether many businesses, particularly those owned by female entrepreneurs, can survive for 24 weeks. Even with significantly scaled down operations, rent obligations and mounting losses can potentially weaken their survival likelihood. Figure 4: Average number of weeks firms expect it would take to resume normal business operation 24.6 30.0 24.3 23.5 GET BACK TO PRE-COVID NUMBER OF WEEKS TO 25.0 17.5 17.3 16.9 20.0 11.1 11.1 11.1 19 STATE 15.0 10.0 5.0 0.0 ROUND 1 ROUND 2 ROUND 3 Men Women All Note: In each survey round, firms were asked how many weeks they anticipate it would take for businesses to go back to pre-COVID-19 levels. The figure indicates the average number of weeks for each round by sex of the business owner. In the same vein, firms have cancelled investment (24.3 percent), sold assets to keep business afloat (6.6 percent) and used personal savings to finance business operations (55.1 percent) (Figure 5). These trends are broadly similar between men- and women-owned businesses. Further, in a separate inquiry in the survey, firms were asked whether they have cancelled investment, asset purchase or renovation plans exclusively due to the COVID-19 pandemic. A strikingly similar share of firms to those indicated in Figure 5 stated that they shelved plans to invest in their firm (see Table 6). Monitoring COVID-19 impacts REPORT NO. 4 on firms in Ethiopia 21 / July / 2020 Figure 5. Savings and investment intentions, percent 24.3 Cancelled investment 21.9 25.5 6.6 Sold asset to keep business afloat 6.9 6.4 55.1 Used personal savings for business 57.1 54.1 0 10 20 30 40 50 60 All Women Men Note: Firms were asked about their investment intentions and whether they have used assets and savings to cover businesses expenses the three weeks period before the interview. This data is available only for the third round. The negative outlook is also confirmed by large layoff expectations (plans) and limited hiring intentions. Despite a recent dip, layoff expectations have generally remained high (Table 11). This negative outlook on employment is not explained by the common churning phenomenon in the labor market, which characterized by entry of new workers and exit of old ones in job relocation processes. In the same labor market, annual hiring and separation rates are estimated to be 21 percent and 15 percent respectively (Abebe et al., 2016).9 In another study, Abebe et al. (2019) find that eight weeks separation rates are between 1 percent and 2.2 percent. Indeed, the last row of Table 11 shows that layoff expectations compared to hiring expectations is at least three-fold larger in men-owned firms and 7 (R1) to 2.7 (R3) fold larger in women-owned businesses. Table 11: Layoff and hiring expectations, percent Men Women All R1 R2 R3 R1 R2 R3 R1 R2 R3 Firm expects to lay-off workers in coming 12.6 5.4 7.2 17.5 5.1 8.2 13.9 5.3 7.5 2 weeks Firm expects to lay-off workers in coming 21.1 13 12.3 19.6 15.8 17.1 20.7 13.8 13.5 3 months Firms expect to hire in the next two weeks 3.6 1.2 2.3 2.5 1.7 3 3.2 1.4 2.5 Note: The numbers in the table concern firms that were interviewed in R1, R2 and R3. Numbers in the table show the share of firms that expect to let go workers in the next 2 weeks and 3 months. Further, firms expect a significant drop in year-end profitability compared to the previous year; i.e., 2011 EFY. Women-owned businesses are more likely to report a ‘much lower’ profit expectations in 2012 EFY than men-owned businesses (Figure 6). For example, in the last round, 75 percent of women entrepreneurs anticipated the worst decline in profit compared to 68 percent for men-owned businesses. In general, 93 percent of women-owned businesses expect a significant dip in profit, the corresponding figure for men- owned businesses is 88.2 percent. 9 Separation rate is defined us the share of workers from the total workforce who left the firm for any reason, including quits, fires and contract termination. Separation rate is thus measured by the ratio of number of workers who left the firm in 12 months period and the average of the total number of workers in the same period. Monitoring COVID-19 impacts REPORT NO. 4 on firms in Ethiopia 21 / July / 2020 Figure 6: Expectations on 2012 EFY profit compared to 2011 EFY profit 80 Share of firms 60 40 20 0 Men Women All Men Women All Men Women All Round 1 Round 2 Round 3 Don't Know Not Applicable Much higher Higher About the same Lower Much lower This summary brief is the fourth in a series of briefs reporting on the findings of the high-frequency phone survey of firms and the first one focusing on results disaggregated by gender. This brief reports on findings from the gender-based analysis of the three rounds of the phone survey for which firm owners or managers in Addis Ababa were interviewed about the effects of and responses to the COVID-19 pandemic between April 15 and June 18, 2020. Data collection continues in the coming months by following the same firms every three weeks. Round 4 of data collection started on May 28, 2020. Each round’s summary brief and table of indicators will be made available via the following website: https://www.worldbank.org/en/country/ethiopia/brief/phone-survey-data-monitoring-COVID-19-impact- on-firms-and-households-in-ethiopia. Abebe, G.; Caria, S.; Fafchamps, M.; Falco, P.; Franklin, S.; Quinn, S.R.; and Shilpi, F.J. (2017). Job fairs: matching firms and workers in a field experiment in Ethiopia. Policy Research Working Paper Series 8092, The World Bank. Abebe, G.; Caria, S.; and Ortiz-Ospina, E. (2019). The selection of talent: experimental and structural evidence from Ethiopia. Available at https://7e6b9778-7cbf-4bbf-96bb- a806c5d9b900.filesusr.com/ugd/e9b89f_665ed15de031417697b0c8dfb5e7c725.pdf. Demeke, E.; Hardy, M., Kagy, G., Meyer, C.J., and Witte, M. (2020). The Impact of COVID-19 on the Lives of Women in the Garment Industry: Evidence from Ethiopia. Living Paper Version 1 (May 31, 2020). Available at https://osf.io/e2ukt/. United Nations (2020). Policy Brief: The Impact of COVID-19 on Women. Available at https://www.unwomen.org/en/digital-library/publications/2020/04/policy-brief-the-impact-of-covid-19- on-women. World Bank (2019). Ethiopia Gender Diagnostic Report: Priorities for Promoting Equity. World Bank, Washington, DC. Available at https://openknowledge.worldbank.org/handle/10986/31420. Box: Survey Methodology The high-frequency phone monitoring survey monitors the economic impacts of and responses to the COVID-19 pandemic on firms with a focus on the effects on firm operations, revenues, and jobs. We call a sample of firms every three weeks between mid-April and mid- September 2020 for a total of eight survey rounds in Addis Ababa. The objective is to collect a final dataset that would consist of a panel of approximately 500 firms in Addis Ababa. The sampling procedure was undertaken in three steps. First, the team cleaned the list of registered firms in number in Addis Ababa, received from the Ministry of Trade and Industry (MoTI), by removing firms with missing or invalid phone numbers. Second, all phone numbers of the cleaned list of firms were shared with EthioTelecom and only active phone numbers were kept constituting the sampling frame. Third, the team drew a random sample of firms without replacement, stratified by firm size (proxied by capital) and sector (industry and services). The sample size was set by a tight budget envelope, with a panel of 500 firms in Addis Ababa after round 8 of the survey operation. Expecting a high non-response rate, we drew a sample of 1,550 firms for Addis Ababa.