Monitoring COVID-19 Monitoring COVID-19 impacts REPORT NO. 7 REPORT NO. 7 on households in Ethiopia 21 / December / 2020 21 / December / 2020 Impacts on Households in Ethiopia Results from Six Rounds of High-Frequency Household Phone Surveys Authors: Alemayehu A. Ambel, Lina Marcela Cardona Sosa, Asmelash Haile Tsegay and Christina Wieser The COVID-19 pandemic and its economic and social effects on households have created an urgent need for timely data to help monitor and relieve the impacts and protect the welfare of the least-well-off Ethiopians. To monitor how the pandemic is affecting Ethiopia’s economy and people and to inform interventions and policy responses, the World Bank is conducting a tailored High-Frequency Phone Survey of Households (HFPS-HH). The HFPS-HH builds on the national longitudinal Ethiopia Socioeconomic Survey (ESS) that the Central Statistical Agency (CSA) and the World Bank carried out in 2019. The HFPS-HH drew a subsample of the ESS sample that was representative of households with access to a working phone. The same households are being tracked over 12 months, with selected respondents, typically the household head, completing phone-based interviews every three to four weeks. Such prompt follow-up allows for better understanding of the household effects of and responses to the pandemic in near real time to support immediate evidence-based responses. This survey brief summarizes the results of the first six rounds of the HFPS-HH, covering the period from April to September 2020. The brief is based on a sample of both urban and rural households in all regions of Ethiopia (Box 1).1 The 15–20-minute questionnaire covers such topics as knowledge of COVID mitigation measures and behavior changes, access to basic staple food items and medicines, educational activities during school closures, access to health care services, employment dynamics, household income and livelihoods, income loss and coping strategies, food security, and assistance received. Respondents are aware of ways to avoid infections and curb the spread of the virus. Recent survey rounds show that fewer people are practicing such preventive measures as frequent handwashing and physical distancing, but more people are wearing face masks. Moreover, fewer households are worried about falling ill from COVID-19 or that the pandemic will negatively affect their finances than they were when the pandemic began. If the services were available at no cost, respondents were quite willing to get tested for and vaccinated against COVID- 19; nationwide, about 98 percent of respondents indicated that they would get tested, and 98 percent would get vaccinated if an approved vaccine were available at no cost Though 8 percent of respondents had lost their jobs at the onset of the pandemic, employment recovered quickly and by September 2020 had reached pre-COVID levels. However, many respondents previously in more stable types of employment now had more vulnerable jobs. When the pandemic began, in April 2020, only about 8 percent of households reported receiving any assistance from either government or nongovernment sources, and by September the number was down to 4 percent. For households that reported receiving assistance, it was mainly food aid, and government was the most important provider. Behavior changes are essential to prevent the spread of COVID-19. From April to September 2020, the survey monitored such practices as handwashing, gatherings, and physical greetings, collecting Information on these practices at separate times. In late April/early May, about a month into the pandemic, 98 percent of respondents reported washing their hands, 96 percent were avoiding handshakes and physical greetings, and 83 percent were avoiding gatherings (Figure 1). However, six months later, in September fewer respondents were still adhering to these behaviors. For example, only 43 percent reported that they avoided gatherings, and those avoiding handshakes and physical greetings were down to 73 percent. There has also been a slight decline in frequent handwashing. 1 Laterite BV Ltd. collected the data. When round 6 (R6) began on September 21, 2020, Ethiopia had 69,709 confirmed COVID-19 cases. By October 13, when R6 ended, confirmed cases had risen to 85,718. Monitoring COVID-19 impacts REPORT NO. 7 on households in Ethiopia 21 / December / 2020 Figure 1: Trends in Behavior Changes to Prevent COVID-19 Infections, R1&R6, Percent R-1 R-6 99 98 98 97 96 96 88 87 87 87 83 81 76 73 71 45 43 40 Rural Urban National Rural Urban National Rural Urban National Wash hands more frequently Avoid handshaking and physical Avoid gatherings greetings Respondents were also asked about their behavior in public, such as wearing face masks and handwashing after coming home. Table 1 compares changes over three months, from about June to September 2020—a period when handwashing declined but wearing face masks increased. In June (Round 3), 89 percent of respondents said that they washed their hands all or most of the time after having been in public. In September the number fell to 77 percent. On the other hand, more respondents who reported wearing facemasks all or most of the time rose from 53 to 65 percent. Table 1: Frequency of Handwashing and Facemask Wearing, Percent Handwashing Facemask wearing when in after having been in public public Frequency R3 (Jun) R6 (Sep) R3 (Jun) R6 (Sep) All the time 73.8 52.4 41.1 44.8 Most of the time 15.7 25.2 12.3 20.4 About half of the time 4.6 9.8 5.7 8.5 Some of the time 5.0 11.4 9.6 14.3 None of the time 1.0 1.2 31.2 12.0 Fewer people are now worried about falling ill with COVID-19 than three months ago. In June (Round 3), 71 percent of respondents reported being very or somewhat worried that they or a family member would fall ill with COVID- 19. Three months later in September (Round 6), only 56 percent were concerned (Figure 2). The trend is similar in both rural and urban areas. Figure 2: Concerns about Falling Ill with COVID-19, R3 & R6, Percent Rural Urban National 57 55 54 38 36 35 34 32 28 21 20 20 19 19 18 16 16 14 13 13 12 11 11 8 VERY SOMEWHAT NOT TOO NOT VERY SOMEWHAT NOT TOO NOT WORRIED WORRIED WORRIED WORRIED AT WORRIED WORRIED WORRIED WORRIED AT ALL ALL ROUND-3 ROUND-6 Similarly, fewer people now see the pandemic as a threat to their household finances (Figure 3). From June to September, the proportion of respondents who considered the pandemic to be a substantial financial threat dropped from 60 to 44 percent. Figure 3: Perceptions of the Effect of COVID-19 on Respondent Finances, R 3& R6, Percent Rural Urban National 72 60 53 52 44 40 27 25 25 21 20 19 19 19 14 14 13 12 12 10 9 7 6 5 Substantial Moderate Not much Not at all Substantial Moderate Not much Not at all Round-3 Round-6 Monitoring COVID-19 impacts REPORT NO. 7 on households in Ethiopia 21 / December / 2020 Figure 4 illustrates willingness to take COVID-19 tests and get vaccinated if the services were available at no cost. If testing for COVID-19 were available at no cost; about 86 percent said that they would definitely get tested and 12 percent thought they would likely get tested. Urban respondents are more likely than rural to get tested. As for openness to vaccination, 98 percent said that they would get vaccinated if an approved vaccine were offered at no cost. There is no notable difference between rural and urban respondents. Figure 4: Willingness To Be Tested and Vaccinated if There Is no Cost, Percent Respondents were asked whether their household had been able to buy enough medicine and enough of the most important food items during the seven days preceding the interview.2 When they were not, they were asked why. From April to September 2020, most households were able to buy the items they needed. Of the five items that the HFPS monitored (Figure 5), teff was the most difficult to buy: one-third of households reported not having been able to buy enough. However, the proportion of respondents who said that they were able to buy needed medicines rose from 71 to 95 percent. All other staple food items maintained their baseline profiles. Figure 5: Ability of Households to Buy Certain Items, R1–R6, Percent Medicine Teff Wheat Maize Edible Oil 93 95 91 86 83 85 81 81 78 78 79 78 77 77 75 76 77 76 77 71 72 73 71 68 65 65 67 64 61 R-1 (Apr-2020) R-2 R-3 R-4 R-5 R-6 (Sept-2020) Figure 6 shows reasons for households that could not buy enough food. Among them were less regular income, higher prices of the item, running out of stock, and market and transport restrictions. Affordability, due to a higher item price or less regular income, was a concern for over 90 percent of households. However, changes in item prices became less a factor and lower income became more of a factor in ensuring access to basic items. At baseline, in April 2020, a higher item price was the most important reason respondents cited, followed by less household regular income. The rankings of these two factors changed starting in R2 (about May), with decrease in regular income becoming the main reason. In Round 6 (September–October), 80 percent of those who said that they could not buy enough food mentioned less income as the main reason. Figure 6: Reasons for Inability to Buy Certain Items, R1–R6, Percent R-1 R-2 R-3 R-4 R-5 R-6 80 74 69 66 55 42 36 34 24 23 16 10 8 8 8 7 6 7 6 4 3 2 2 2 2 1 1 1 0 0 outside Limited operati regular Increas run out income restrict Decrea market ortatio transp Shops ion to Local stock price s not se in have e in /no n & ng go of 2 According to the 2018‒19 ESS, the four most important food items are edible oil and teff, wheat, and maize as grain or flour or cooked. Monitoring COVID-19 impacts REPORT NO. 7 on households in Ethiopia 21 / December / 2020 Respondents were asked about their main sources of household income in the previous 4 weeks and whether income from that source increased, stayed the same, fell, or was totally lost.3 Table 2 shows respondents who reported a reduction or total loss of income by source. We see a declining trend of reduced or total loss of income the longer the pandemic lasts, with the first two months showing more households with reduced incomes than later months. Households reporting non-farm business as their main income source saw their incomes reduced more. Remittances, from both within Ethiopia and abroad, were the income source with the largest pandemic- caused reductions. Remittances, have, however, recovered quickly: only 16 percent of households indicated receiving lower remittances from within Ethiopia and 4 percent from abroad. Table 2: Household Income Reduction or Total Loss, Previous 4 Weeks, R1-R6, Percent R-1 R-2 R-3 R-4 R-5 R-6 Farming, livestock, fishing 41.4 29.5 31.7 24.2 23.8 18.7 Non-farm business 85.1 66.7 63.2 60.6 49.6 41.9 Wage employment 34.4 24.7 20.2 15.9 23.9 19.7 Remittances from within Ethiopia 44.9 42.4 9.1 17.0 15.6 16.1 Remittances from abroad 62.5 40.2 17.6 18.3 16.7 3.9 Income from properties, investments and savings 41.8 31.4 23.1 30.1 19.8 13.8 Pension 4.0 1.3 0.1 0.3 0.1 1.0 Government assistance 27.3 2.1 4.7 11.5 11.4 37.0 Assistance from NGO or charitable organization 41.8 3.9 4.8 2.4 4.4 19.6 In response to income loss, households used a variety of coping strategies. While half of the respondents did nothing, some sold assets and others borrowed money or received assistance from friends and relatives. Some of the most worrying coping strategies were reducing food and nonfood consumption. In the early months of the pandemic few households reduced food or nonfood consumption. Later survey rounds show that cutting back on food and nonfood consumption went up, probably because households had exhausted other means of coping with income losses. By R6, however, food and nonfood consumption was back up to levels similar to those when the pandemic began (Figure 7). Figure 7: Reduction of Consumption in Response to Income Reduction or Loss, R1–R6, Percent Food Non-Food 29 26 21 20 19 19 17 17 16 16 16 16 15 15 13 12 11 11 11 11 10 9 9 8 Rural Urban Rural Urban Rural Urban Rural Urban Rural Urban Rural Urban Round-1 Round-2 Round-3 Round-4 Round-5 Round-6 When the pandemic began (R1), 8 percent of the respondents reported having lost their jobs since the outbreak of COVID-19, and about 63 percent of those in this group indicated that COVID-19 had caused their job loss. Job loss was more severe in urban (20 percent) than in rural areas (3 percent). Yet employment recovered quickly: by R6 employment rates were similar to pre-COVID rates (Figure 8). Despite a strong recovery in jobs, however, employment respondents switched from more stable types, such as government and the private sector to more vulnerable types related to self-employment, casual employment, and family work, which had become the dominant and increasing form of employment (Figure 9). And non-farm household businesses have yet to recover: In April, 23 percent of households surveyed owned a business. By September this was down to 19 percent. 3 In round 1, respondents were asked about their income sources for the past 12 months. Monitoring COVID-19 impacts REPORT NO. 7 on households in Ethiopia 21 / December / 2020 Figure 8: Respondents who Worked in the 7 Days Preceding the Interview, R1–R6, Percent 100 90 80 70 60 pre-COVID R1 (Apr) R2 (May) R3 (Jun) R4 (Jul/Aug) R5 (Aug/Sep) R6 (Sep/Oct) Rural Urban National Figure 9: Self-employment, Casual Employment, and Family Work, R1–R6, Percent 84.0 83.0 82.0 81.0 80.0 79.0 78.0 77.0 76.0 R1 R2 R3 R4 R5 R6 At the onset of the pandemic, about 8 percent of households reported receiving any form of assistance. Results from subsequent survey rounds show some fluctuation and in general a decreasing trend in the proportion of households reporting any assistance (Table 3). For those households that reported any assistance, food aid was the largest form of assistance and government the most important provider. Table 8: Assistance to Households since the COVID-19 Outbreak, by Type and Source, Percent R-1 R-2 R-3 R-4 R-5 R-6 Household received assistance: Any source 7.8 5.7 3.8 6.9 4.0 3.4 Assistance type: Free food 47.2 39.3 66.9 60.2 64.0 74.5 Assistance type: Food or cash for work 15.7 16.8 2.5 8.0 22.2 6.4 Assistance type: Direct cash transfer 38.6 39.9 31.8 32.6 17.6 19.2 Assistance source: Government 76.9 70.9 72.4 70.6 75.2 57.1 Assistance source: NGO 11.8 16.9 5.6 12.0 17.8 38.2 Assistance source: Religious organization 4.2 5.7 10.4 12.7 6.3 1.6 Assistance source: Volunteer or youth organization 7.0 2.1 3.4 2.4 2.6 2.5 Note: Assistance source and assistance type conditional on household receiving assistance. This survey brief is the sixth in a series reporting on the findings of the HFPS-HH. It reports results from the first 6 rounds on the effects of and responses to the COVID-19 pandemic in Ethiopia. Data collection will continue by following up with the same households every four weeks. For each round, the survey brief, table of indicators, and microdata will be available at https://www.worldbank.org/en/country/ethiopia/brief/phone-survey-data- monitoring-covid-19-impact-on-firms-and-households-in-ethiopia. Monitoring COVID-19 impacts REPORT NO. 7 on households in Ethiopia 21 / December / 2020 The HFPS-HH sample is a subsample—households with access to a phone—of those interviewed in 2019 for the Ethiopia Socioeconomic Survey (ESS), which covers urban and rural areas in all regions of Ethiopia. Phone penetration in rural Ethiopia is low; about 40 percent of rural households have access to a phone compared to over 90 percent of urban households. This not only means that the rural sample is smaller but there is also a systematic difference between households that have access to a phone and those that have not. Phone-owning households are better off in terms of total consumption, educational attainment, access to improved water and sanitation, access to assets, and access to electricity. The sample of the HFMP-HH is therefore representative only of households that have access to phones in urban and rural Ethiopia. The respondent is typically the household head; where that person cannot be reached despite numerous call- backs, another knowledgeable household member is selected as the respondent. In Round 1, in April 2020, the HFPS-HH called all the 5,374 households that in the ESS had provided a valid phone number; of these, 3,249 households consented to be interviewed. The subsequent rounds attempted to reach the same 3,249 households. Table 4 shows the number of interviews completed in rural and urban areas. Table 4: Number of Completed Interviews by Round Round 1 Round 2 Round 3 Round 4 Round 5 Round 6 (Apr 22–May 13) (May 14–June 3) (June 4–26) (July 27–Aug 14) (Aug 24–Sep 17) (Sep 21–Oct 13) Rural 978 940 934 838 775 760 Urban 2,271 2,167 2,124 2,040 1,995 1,944 National 3,249 3,107 3,058 2,878 2,770 2,704