Monitoring COVID-19 Monitoring COVID-19 impacts REPORT NO. 1 REPORT NO. 1 on refugees in Ethiopia 18 / December / 2020 18 / December / 2020 Impact on Refugees in Ethiopia Results from a High-Frequency Phone Survey of Refugees Authors: Christina Wieser, Nfamara K. Dampha, Alemayehu A. Ambel, Asmelash Haile Tsegay, Harriet Mugera, and Jeffery Tanner1 The COVID-19 pandemic and its economic and social effects have created an urgent need for timely data to help monitor and mitigate the social and economic impact of the crisis and protect the welfare of the least well-off in Ethiopian society. To monitor how the COVID-19 pandemic is affecting Ethiopia’s economy and people and to inform interventions and policy responses, the World Bank designed and conducted a series of High-Frequency Phone Surveys of households, firms and refugees. The COVID-19 pandemic started in Ethiopia in March 2020 with the capital Addis Ababa being hit the hardest. Ethiopia reported its first case of COVID-19 on 13 March 2020. The number of reported cases reached 108,930 by 28 November 2020, almost half of which (50,749) were in Addis Ababa. Of these patients, 68,250 have recovered, while the total number of deaths from COVID-19 stands at 1,695. In response to the outbreak of the pandemic, the Government of Ethiopia (GoE) introduced a range of containment policies. The GoE declared a five-month State of Emergency in April 2020 but largely allowed economic activities to continue during the public health crisis induced by the COVID-19 pandemic. On 16 March, all schools were closed and have not reopened at the time of the survey. Both public and private sectors workers were directed to work from home to the extent possible, social gatherings were limited, and cross-regional travel was banned. Though entry into Ethiopia was never limited, the GoE introduced a 14-day quarantine for incoming travelers, citizens, residents, and foreigners alike, disrupting Ethiopia’s large aviation industry. Data on the socioeconomic impacts of COVID-19 are essential to inform timely and adequate policy and program responses. Past pandemics have shown that health and economic consequences, as well as mitigation measures, lead to serious socioeconomic effects that unfold across populations and generations. Moreover, pandemics and measures taken to curb their spread can have a rippling effect on human capital, food security, and social cohesion. Notably, population groups already in precarious circumstances are likely to be more severely affected, as their capacity to cope with shocks is limited. It is therefore essential to better understand the socioeconomic effects of the COVID-19 pandemic on the most vulnerable populations. It is in this context, that the World Bank Group, the Ethiopia Agency for Refugee and Returnee Affairs (ARRA), the United Nations High Commissioner for Refugees (UNHCR), and the World Bank - UNHCR Joint Data Center on Forced Displacement (JDC) collaborated to integrate refugees in the ongoing high-frequency phone survey of households (HFPS). These phone surveys fill socioeconomic data gaps while providing initial evidence to inform targeted policy and program responses. In the context of the COVID-19 pandemic, face-to-face surveys were not feasible due to mobility restrictions and the risk of infection between respondents and enumerators. In contrast, phone surveys are highly suited to the circumstances and allow for rapid and frequent data collection relevant to inform a quickly evolving and uncertain periods.2 The high-frequency phone survey of refugees monitors the economic and social impact of and responses to the COVID-19 pandemic on refugees and nationals, by calling a sample of households every four weeks. Round 1 of data collection for the refugee sample occurs simultaneously with Round 6 of the national HFPS operation. This 1 The team would like to thank the Ethiopian Agency for Refugee and Returnee Affairs (ARRA), the United Nations High Commissioner for Refugees (UNHCR), and the Joint Data Center on Forced Displacement (JDC) for their vision, commitment to this work, and collaboration in the efforts of data collection and reporting. In particular, the team would like to thank Ato Addisu Kebenessa (ARRA), Ato Kiros Kinfe (ARRA), Mathijs le Rutte (UNHCR), Charlotte Ridung (UNHCR), Carolyn Ndawula (UNHCR) and Cleve Massamba (UNHCR) for their contributions and support. Particular mention goes to Theresa Beltramo and Ibrahima Sarr, who contributed to the sampling of this study. 2 See for example Dabalen, A., Etang, A., Hoogeveen, J., Mushi, E., Schipper, Y., von Engelhardt, J. (2016). Mobile Phone Surveys in Developing Countries: A Practical Guide for Microdata Collection. Available at https://openknowledge.worldbank.org/handle/10986/24595 Monitoring COVID-19 impacts REPORT NO. 1 on refugees in Ethiopia 18 / December / 2020 survey brief summarizes the results of the first round of the joint national and refugee HFPS, implemented between 24 September and 17 October 2020.3 The brief is based on a sample of 1,676 refugees and 2,706 national households across three survey domains: (i) Refugees in Addis Ababa; (ii) Eritrean refugees; and (iii) Somali refugees.4 The HFPS of refugees builds on ARRA/UNHCR’s registration database of refugees and is representative of Eritrean and Somali refugees and refugees living in Addis Ababa with a working phone number. These refugee populations were among those with the highest phone penetration rate. The 15-minute questionnaire covers such topics as knowledge of COVID-19 and mitigation measures, access to educational activities during school closures, employment dynamics, household income and livelihood, income loss, as well as coping strategies. Refugee households seem well aware of COVID-19, and the vast majority have knowledge of behaviors necessary to minimize the risk of contracting or spreading the virus. Access to schools for refugee children was a challenge before the pandemic and continues to be a challenge during school closures related to COVID-19 . While far fewer refugee children attended school before the pandemic compared to nationals, those that went to school are more likely to attend distance learning activities. The COVID-19 pandemic has hit total household incomes hard: 27 percent of refugee respondents report that household incomes were either reduced or had totally disappeared. Though very few refugee respondents were working before the pandemic, 32 percent of those that were working before the pandemic reported to have lost their jobs. Government and the international community have maintained assistance throughout the pandemic with one third of refugee households receiving more assistance since the outbreak began. Refugees are favorable towards getting tested and vaccinated for COVID-19. About 93 percent, indicate that they would get vaccinated, if approved COVID-19 vaccine were available at no cost. To prevent the spread of COVID-19 and to ensure that measures to slow it such as mobility restrictions and market closures are effective, it is essential that people are aware of the need to change their behaviors. Virtually every refugee household (99.1 percent) had heard about the coronavirus or COVID-19. The respondents reported being well-informed about how to prevent contracting the virus and actions to reduce its spread. Almost everyone knew about handwashing, avoiding handshakes, using face masks, avoiding crowds, etc. Yet, there are some differences across refugee groups in adopting behaviors to curb the spread of the virus. About 87 percent of refugee respondents indicated to have washed their hands with soap or sanitizer all of the time and an additional 11 percent indicated to have washed their hands most of the time (Figure 1). Yet, Somali refugees seem to be lagging behind in this behavior with only 85 percent of respondents indicating that they adhere to handwashing with soap or sanitizer all or most of the time. Similarly, this self-reporting suggests wearing face masks all or most of the time is practiced less frequently among Somali refugees compared to Eritrean refugees and those refugees living in Addis Ababa (Figure 2). 3The data collection was undertaken by Laterite (Ethiopia) Ltd. 4 More precisely, Eritrean refugees are those served by UNHCR Sub-Office Shire (99.9 percent of whom are of Eritrean origin), and Somali refugees are refugees served by the UNHCR Sub-Office Jijiga (97.7 percent of whom are of Somali origin). Refugees served by the UNHCR Representation in Addis Ababa include all refugees registered in Addis Ababa with 84 percent of Eritrean origin, 7 percent of Yemeni origin, and 2 percent of Somali origin. Monitoring COVID-19 impacts REPORT NO. 1 on refugees in Ethiopia 18 / December / 2020 Figure 1: Respondents who washed hands with soap or sanitizer all or Figure 2: Respondents who wore face masks in public all or most of the most of the time in past 7 days, Percent time in past 7 days, Percent All of the time Most of the time All of the time Most of the time 93.7 81.9 96.3 80.3 89.5 86.9 41.7 46.7 38.3 29.4 15.3 14.9 10.6 9.1 4.0 2.1 Addis Ababa Somali Eritrean Total refugee Addis Ababa Somali Eritrean Total refugee refugees refugees refugees population refugees refugees refugees population Whether and how COVID-19 has affected the availability and prices of medicine and food staples is yet to be fully understood. Combined with loss of income, higher food prices and eventual shortages could erode the welfare of refugee households. The HFPS asked respondents, whether their households were able to buy enough medicine and enough of the most important food items during the week preceding the survey.5 Where they were not, we asked for the reason. Refugee households seem to be able to access most of the essential items more compared to the national population. In particular, refugee households’ ability to purchase food items if needed is higher compared to the national population but lower for medicine. For example, 96 percent of refugees were able to buy oil in the week preceding the survey but only 77 percent of nationals were able to do so (Figure 3). Similarly, for teff, wheat, and maize, we observe that refugee households were able to buy enough of what they needed—teff (95 percent), wheat (96 percent), and maize (91 percent)—with a larger proportion of refugees being able to buy the items. Yet, nationals had higher ability to buy medicine (95 percent) compared to refugees (89 percent). This higher access for food staples is likely a result of assistance refugees receive, which may reduce the necessity to buy certain items. Figure 3: Households’ ability to buy items in past 7 days, Percent Figure 4: Respondents’ concerns about COVID-19, Percent 94.3 89.1 Refugees National population 78.2 76.9 76.3 74.2 73.3 68.7 64.9 95.8 56.4 95.5 95.2 95.1 90.8 88.5 79.0 78.1 76.6 66.8 very or somewhat worried Pandemic is substantial or to fall seriously ill moderate threat to household finances Addis Ababa refugees Somali refugees Eritrean refugees Total refugee population Medicine Teff Wheat Maize Edible Oil National The majority of refugee households are concerned about the impact of the coronavirus on their health and livelihoods. About 73 percent are very or somewhat worried that they or a family member will fall ill with COVID- 19 (Figure 4). Similarly, about 78 percent of refugee households made it clear that the pandemic is a substantial or a moderate threat to their finances. Refugees living in Addis Ababa are least and Somali refugees are most concerned about illness and the effect on finances. Overall, refugees are more concerned than the national population about falling seriously ill and the pandemic posing a threat to household finances. 5 According to the 2018‒19 ESS, the four most important food items are teff, wheat, maize, and edible oil. Monitoring COVID-19 impacts REPORT NO. 1 on refugees in Ethiopia 18 / December / 2020 On 16 March 2020, Ethiopia closed all primary and secondary schools and schools have not reopened at the time of the survey, leading to students losing valuable months of schooling. Temporary school closures may lead to permanent drop-out of children from vulnerable households, for whom early drop-out is rife even in ordinary circumstances. The long-term impacts of lost months of schooling may be particularly severe for children in low- income families, because it will jeopardize their development of human capital and their earning potential. Moreover, schools can serve as a safe space for refugee children as they can support refugee children’s transition to “normalcy” and protect them from further exposure to risks such as gender-based violence, pregnancy, early marriage, and dropout. The HFPS shows that education for refugee children was a challenge before the pandemic and continues to be a challenge thereafter during school closures. The survey asked households how many children were in school before the outbreak began and whether they are now engaged in any learning activities. Most strikingly, very few refugee children went to school, even before schools closed in March 2020. Only 2 of 10 refugee households with primary school children sent their children to school before the outbreak of COVID-19 (Figure 5). This is more than three times less than the rate of national households, among whom almost 7 of 10 primary-school aged children attended school. Few refugee children attended secondary school in Ethiopia pre-pandemic. Moreover, the percentage of Ethiopian households whose children attended secondary school is almost five times larger (24 percent) than for refugee households (5 percent). Though few refugee children went to school before the pandemic, access to learning activities during school closures is encouraging. About 38 percent of households that have children, who had gone to school before the pandemic, were engaged in any form of learning activity at primary school level and 51 percent at secondary level (Figure 6). More soberingly, half of all refugee children, who went to school pre-pandemic, do not participate in any learning activities during school closure. In contrast, many more national children went to school before the pandemic but a lower percentage was engaged in learning activities during school closure. Education was rare for refugee children before the pandemic struck and remains so during school closures. Figure 5: Households with School-Aged Children who Attended School Figure 6: Households with Students who Previously Attended School pre COVID-19, Percent and now Engage in Learning Activities, Percent Refugee population National population Refugee population National population 68.1 50.5 37.9 24.1 24.2 17.4 20.3 5.4 Primary school Secondary school Primary school Secondary school We see large variations in refugee children attending school by refugee group, with Somali refugee children most likely to have attended school before the pandemic. In fact, Somali refugee households are three times more likely to have children attending school (primary or secondary) compared to Eritrean refugees and refugees living in Addis Ababa (Figure 7). Yet, Somali children who attended school before the pandemic are also least able to engage in learning activities during school closure (Figure 8). Considering the types of distance learning activities, by far the most important activity (in primary and secondary) is to “complete assignments provided by the teacher”. This is encouraging, as it shows that the GoE and the international community were reasonably successful in providing learning activities for those (few) refugee children during school closures. In Ethiopia, protracted school closures affect both refugee and national households. Yet, apparent education gaps between refugees and nationals can be expected to increase. Few refugee children are able to learn and lose Monitoring COVID-19 impacts REPORT NO. 1 on refugees in Ethiopia 18 / December / 2020 valuable months of education and may not return, all of which has important implications on foregone human capital accumulation. Figure 7: Households with School-Aged Children who Attended School Figure 8: Households with Students who Previously Attended School pre COVID-19 By Refugee Group, Percent and now Engage in Learning Activities By Refugee Group, Percent Addis Ababa refugees Somali refugees Eritrean refugees Addis Ababa refugees Somali refugees Eritrean refugees 64.8 66.7 63.3 41.6 35.6 39.3 37.7 29.0 18.7 16.2 2.7 2.6 Primary school Secondary school Primary school Secondary school The COVID-19 pandemic is affecting economic activity in Ethiopia. Though the State of Emergency declaration in place from April through September 2020 prohibited firms from laying off workers, lay-offs nevertheless occurred in Ethiopia’s sizable informal sector. The HFPS provides evidence about changes in the employment situation of households resulting from the pandemic. Figure 9: Currently Working, Percent 87.4 Very few refugee respondents were working at the time of the phone interview but job losses were nevertheless severe. Less than 1 in 5 refugee 32.3 respondents was working in October 2020 compared to almost 9 out of 10 18.9 18.7 12.9 for nationals (Figure 9), with Somali refugees being most likely to currently have a job. For refugee respondents not working at the time of the interview, about 28 percent had been working before the outbreak—a Total refugee refugees refugees Addis Ababa population Eritrean Somali National population refugees percentage that rises to 50 percent for Somali refugees. Of those who lost a job since the outbreak, 61 percent attributed their job loss to COVID-19 and 23 percent to seasonal or casual work. About 9 percent indicated that they are temporarily absent, suggesting that they intend to go back to their jobs. Job losses were highest for Somali refugees—the refugee group that was most likely to be employed pre-COVID—and lowest for Eritrean refugees. Notably, very few refugee households operated a household business (3 percent) compared to nationals (19 percent). One of the channels through which households are negatively affected by the pandemic and its associated restrictions of movement and assembly is through reduced income. The survey asked households about their income sources over the last 12 months and followed up by asking whether the income from a particular source has increased or decreased since COVID-19 broke out (Table 1). Income sources for refugees of the last 12 months are diverse and vary significantly by refugee group. The largest differences in household income sources can be attributed to whether refugees live in or out-of-camps6. While few refugee households relied on non-farm businesses, wage employment was the main source of livelihood for those 6 Out-of-camp refugees included in this study can be affected by one of two GoE policies: (1) GoE’s “out-of-camp policy” and are refugees or asylum seekers who are provided with a permit to live at a place of their choice outside of refugee camps or settlements covering all living costs by themselves or through their relatives, friends, or other persons who commit to supporting them. (2) Urban refugees are refugees in need of special medical attention unavailable in camps and/or refugees with serious protection concerns or the inability to stay in camps for security or humanitarian reasons. Urban refugees are provided with monthly cash assistance and other basic and social services free of charge. Monitoring COVID-19 impacts REPORT NO. 1 on refugees in Ethiopia 18 / December / 2020 groups with a large share of out-of-camp refugees, particularly for refugees living in Addis Ababa and Eritrean refugees (about 50 percent out-of-camp). A strikingly large share of refugees relied on remittances as their main source of income with three quarters of refugees in Addis Ababa indicating to have received remittances in the past 12 months. Receiving assistance—either from the government, the international community, or Non- Governmental Organizations (NGO)—is an important source of income, particularly for Somali refugees, almost all of whom are camp-based. Table 1: Household Income Sources, Last 12 months, Percent Addis Ababa Total refugee Somali refugees Eritrean refugees refugees population Non-Farm Business 3.7 4.2 5.3 4.9 Wage-Employment 19.7 4.4 13.0 13.8 Remittances from Within Country 10.1 3.9 7.5 7.8 Remittances from Abroad 74.0 13.2 45.2 49.0 Government Assistance 2.1 4.9 24.1 17.8 Assistance from NGO or Charitable Organization 13.5 76.0 19.8 22.8 Other Sources 1.1 11.3 1.6 2.3 Refugees experienced large losses in income since the COVID-19 outbreak, with non-farm businesses and wage employment particularly affected. About 57 percent of households that cited nonfarm business as a means of livelihood in the past 12 months reported less income from that source (Table 2). Similarly, 55 percent of households with income from wages had lost some or all of their income from this source. Almost half of refugees indicated remittances from abroad among their main sources of income before the pandemic. Of these, more than one third indicated a decline in this income source. Table 2: Reduction of Total Household Income Since the Outbreak by Source, Percent Experienced reduction of income Income source since COVID-19 outbreak Non-Farm Business 4.9 57.0 Wage-Employment 13.8 55.3 Remittances from Within Country 7.8 26.5 Remittances from Abroad 49.0 36.4 Government Assistance 17.8 6.7 Assistance from NGO or Charitable Organization 22.8 7.6 Even when looking at the four (4) weeks preceding the survey interview, refugees are still affected by a steep reduction in labor income. A larger share of refugee households reported a reduction in income from non-farm business and wage employment in the past four (4) weeks compared to nationals (Table 3). On the other hand, refugees are less affected by remittances from abroad and reductions in assistance. In fact, the GoE and the international community seem to have responded well to the negative effects of the pandemic. Assistance from government, international community, and NGOs increased for one-third of refugees7 but reduced for the majority of the national population, as they worked to keep pace with increased needs. Table 3: Reduction or Loss of Income in Past 4 weeks, Percent Refugee population National population Non-Farm Business 54.3 41.9 Wage-Employment 44.6 19.7 Remittances from Within Country 31.1 16.1 Remittances from Abroad 35.3 59.7 Government Assistance 13.0 80.9 Assistance from NGO or Charitable Organization 23.1 56.4 7 For example, UNHCR provided ETB 300 per individual for the provision of extra soap for hygiene purposes related to COVID-19. Monitoring COVID-19 impacts REPORT NO. 1 on refugees in Ethiopia 18 / December / 2020 Asked whether the total income of the household has changed since the outbreak began (Figure 11=0), 27 percent of respondents report either a reduction or a total loss. Nearly half (45 percent) of refugees in Addis Ababa—where the effects of the pandemic were most severely felt- and the majority of refugees does not rely on regular assistance—report a reduction in income. Though we cannot directly compare refugees and nationals in terms of income loss since the outbreak 8, we can look at changes in total household income in the past four (4) weeks. It seems that for the national population, household income losses were more prevalent at the onset of the pandemic but they seem to be recovering better compared to refugees. Nationals were more than twice as likely to report an increase in total household income in the past four (4) weeks (11 percent compared to 5 percent of refugees) (Figure 11). Figure 10: Change in Total Income since the Outbreak, Percent Figure 11: Change in Total Income in Past 4 weeks, Percent Increased Stayed the same Reduced Total (100%) loss Refugee population National population 69.2 2.8 2.7 2.0 2.2 62.4 25.7 18.9 24.5 42.1 34.8 23.0 20.2 73.3 65.9 11.4 52.8 5.4 36.3 3.3 3.2 2.1 4.5 6.5 Addis Ababa Somali Eritrean Total refugee Increased Stayed the Reduced Total (100%) refugees refugees refugees population same loss Households experiencing less or no income in the past four (4) weeks were then asked what coping strategies (if any) they had applied to better manage the income reduction. Applied coping strategies for refugees and nationals are very similar (Figure 12). The most common coping strategy applied was to rely on savings—about one-fifth of refugee and national households relied on savings to cope with reduced income. The second and third most prevalent coping strategies are to reduce food and nonfood consumption. Reducing consumption is concerning because it could affect the long-term health and welfare of household members. The national population is more likely to not adopt a coping strategy. More than half of nationals had not applied a coping strategy to compensate, suggesting that they have sufficient means to cope with income losses. On the other hand, refugees are more likely to sell assets (11 percent) and rely on friends and family (13 percent). Figure 12: Coping Strategies, Percent Refugee population National population 50.9 28.5 22.9 20.8 15.1 14.3 12.8 12.1 11.3 11.1 7.9 5.9 5.6 5.6 5.1 5.0 3.9 3.0 0.7 0.7 0.7 0.2 agricultur… Engaged in purchases assistance assistance nothing al and Non- consumptio consumptio Relied on All others Borrowed from NGO generating friends & (Agricultur additional friends & Credited Received Received savings activities Reduced Reduced nonfood family income family Did Sale of from assets from food n n Refugee households have been hit hard by the COVID-19 pandemic but government and the international community responded well to the pandemic. About 30 percent of refugees have received assistance through any means (government, NGOs, or faith-based institutions), since the outbreak began (Figure 13). Somali refugees— 8 Due to the fact that we asked pre-COVID income losses in round 1 of the national survey which took place in April 2020. Monitoring COVID-19 impacts REPORT NO. 1 on refugees in Ethiopia 18 / December / 2020 all of whom live in camps—are more than twice as likely to receive assistance compared to Eritrean refugees (about half of whom in our sample live in camps). Refugees in Addis Ababa receive the least assistance by nature of their out-of-camp status. Overall, those living in camps seem to be better protected during the pandemic by nature of the assistance provided to them. The largest type of the assistance provided was free food and direct cash transfers (Figure 14). Free food is more common among refugees heavily reliant on camp assistance. Figure 13: Assistance to Households since the COVID-19 Outbreak, by Figure 14: Assistance Type (of those Receiving Assistance), Percent Type and Source, Percent 73.8 Addis Ababa refugees Somali refugees Eritrean refugees 94.1 84.1 76.1 73.8 71.1 30.4 30.3 14.8 21.1 refugees refugees refugees population Eritrean Somali Ababa 8.6 Addis refugee 5.2 Total 0.0 Free food Food or cash for Direct cash work transfer The survey asked households about their likely behavior towards getting tested for free for the COVID-19 virus and getting vaccinated, if an approved COVID-19 vaccine were available at no cost. About 98 percent of refugee respondents indicated that they would get tested for the COVID-19 virus (Figure 15). Similarly, a very high share, 93 percent, indicated that they would get vaccinated, if an approved COVID-19 vaccine were available at no cost. Though there is a small gap to the national population in the likelihood of getting vaccinated (97 percent of nationals would get vaccinated), recent health campaigns to include refugees in the Ethiopia National Extended Program on Immunization seem to be effective in communicating the importance of being vaccinated. For households, which indicated that they would not get vaccinated, among the most important reasons were “worry about safety of vaccine and possible side effects.“ Figure 15: Likelihood of Getting Tested and Vaccinated, Percent Refugees National population 97.9 97.9 97.7 93.0 Would definitely or likely Would get vaccinated if get tested for free for approved COVID-19 COVID-19 virus vaccine were available at no cost This survey brief reports the findings of the HFPS of refugee households. It reports results based on refugee and national households in Ethiopia, who were interviewed between 24 September and 17 October 2020, about the effects of and responses to the COVID-19 pandemic. Data collection continues by following up with the same refugee and national households. Each round’s 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. 1 on refugees in Ethiopia 18 / December / 2020 The high-frequency phone survey of national and refugee households monitors the economic and social impact of and responses to the COVID-19 pandemic on households in terms of such topics as access to food staples, access to educational activities during school closures, employment dynamics, household incomes and livelihoods, income losses and coping strategies, as well as external assistance. The final dataset will cover a panel of about 3,200 Ethiopian households and 1,650 refugees that is representative of national and refugee households with access to a mobile phone. To the extent possible, the same households and respondents will be tracked with selected respondents completing phone-based interviews every four weeks. This high-frequency follow-up allows for a better understanding of the effects of and responses to the COVID- 19 pandemic on households in order to inform interventions and policy responses and monitor their effects. The respondent is typically the head of household; where that person cannot be reached despite numerous call-backs, another knowledgeable household member will be selected as the respondent. The HFPS sample of households consists of a subsample interviewed for the Ethiopia Socioeconomic Survey (ESS) in 2019, namely those households with access to a phone, covering urban and rural areas in all regions of Ethiopia. The HFPS sample of refugees is based on ARRA/UNHCR’s proGRES database, which contains a list of all registered refugees in Ethiopia. Only refugees living in Addis Ababa, Somali refugees served by the UNHCR Sub-Office in Jijiga, and Eritrean refugees served by the UNHCR Sub-Office in Shire have phone penetration rates above 10 percent and are included in the HFPS. Phone owning households are better off in terms of many socioeconomic dimensions (i.e. total consumption, educational attainment, access to services, access to assets). The sample of the HFPS is therefore only representative of households, which have access to a mobile phone for the national and the refugee sample and is therefore likely to represent an upper bound on refugee welfare.