Monitoring Social and Economic Impacts of COVID-19 on Refugees in Uganda: Results from the High-Frequency Phone Survey First round (October/November 2020) World Bank Poverty and Equity Practice February 24, 2021 Contents Contents ........................................................................................................................................................ 2 List of figures ................................................................................................................................................. 2 List of tables .................................................................................................................................................. 3 Acknowledgments......................................................................................................................................... 4 Key messages and findings ........................................................................................................................... 6 A. Background ........................................................................................................................................... 8 B. Economic activities................................................................................................................................ 8 a. Labor market and farm activities ...................................................................................................... 8 b. Changes in income since lockdown ................................................................................................ 12 c. Impact on poverty and food security .............................................................................................. 13 d. Social assistance .............................................................................................................................. 14 C. Knowledge and behavior .................................................................................................................... 14 D. Access to food, health, and education................................................................................................ 17 E. Coping strategies to socioeconomic shocks ....................................................................................... 19 F. Concerns ............................................................................................................................................. 21 G. Bibliography ........................................................................................................................................ 24 H. Annex .................................................................................................................................................. 24 a. Sampling.......................................................................................................................................... 24 b. Profile of respondents and households .......................................................................................... 25 List of figures Figure 1. Employment status of respondents by region, (% of all respondents) ......................................... 9 Figure 2. Employment status of respondents by strata, (% of all respondents) .......................................... 9 Figure 3. Work stoppages by reasons related to and not related to COVID-19, (% of respondents who stopped working after March 20, 2020) ..................................................................................................... 10 Figure 4. Work stoppages by sector, (% of respondents who were working and used to work before March 2020) ........................................................................................................................................................... 10 Figure 5. Work stoppages by reasons directly related to COVID-19 by economic sectors (% of respondents who stopped working after March 2020) ................................................................................................... 10 Figure 6. Households engaged in planting during first agricultural season and livestock, since the beginning of 2020 by regions, (% of households) ........................................................................................................ 11 Figure 7. Households engaged in planting during first agricultural season and livestock, since the beginning of 2020 who had to change activities due to COVID-19 by regions, (% of households) ............................ 11 Figure 8. Households that needed to sell farm products since 2020 and among them, households not able to do so, (% of households) ........................................................................................................................ 11 Figure 9. Key income sources during last 12 month across regions, (% of households) ............................ 12 2 Figure 10. Income sources during last 12 months and share of households without any income by source after March 20, 2020, (% of households) ................................................................................................... 13 Figure 11. Food security during last 30 days among adults in the household, (% of households) ............. 13 Figure 12. Share of households without social assistance since March 2020, (% of households) ............. 14 Figure 13. Type of assistance received since March 2020, (% of households) ........................................... 14 Figure 14: Knowledge of selected COVID-19 symptoms across countries of origin, (% of respondents) .. 15 Figure 15: Knowledge of preventive measures among refugees and nationals, (% of respondents) ........ 15 Figure 16: Knowledge of preventive measures among refugees by region, (% of respondents) ............... 15 Figure 17: Frequency of mask wearing when in public, (% of respondents who were in public) .............. 16 Figure 18: False beliefs about COVID-19 among refugees and Ugandans, (% of respondents) ................. 16 Figure 19: False beliefs about COVID-19 among refugees, (% of respondents) ......................................... 16 Figure 20: No access to selected basic goods and services when needed, nationals and refugees, (% of households)................................................................................................................................................. 17 Figure 21: No access to selected basic goods and services when needed, refugees by region, (% of households)................................................................................................................................................. 17 Figure 22: Engagement in any learning activities after schools’ closure and plans to return to schools when they reopen, (% of households with any member attending school before March 2020) ........................ 18 Figure 23: Main reasons for not going back to school, (% of households with any member not returning to school) ......................................................................................................................................................... 18 Figure 24: Learning and education activities after closures, (% of households with any member engaged in any learning activity after March 2020) .................................................................................................. 18 Figure 25: Information needs, (% of respondents) ..................................................................................... 19 Figure 26: Number of shocks since March 2020, (% of households) .......................................................... 20 Figure 27: Types of shocks, (% of households) ........................................................................................... 20 Figure 28: Selected coping strategies to socioeconomic shocks, (% of households) ................................. 21 Figure 29: Degree of threat from coronavirus outbreak to household’s finances ..................................... 21 Figure 30: Perception of safety at home since the COVID-19 outbreak, (% of respondents) .................... 22 Figure 31: Perception of safety in the community since the COVID-19 outbreak, (% of respondents) ..... 22 Figure 32: Main reason why felt unsafe in community since the onset of COVID-19 ................................ 23 List of tables Table H-1. Distribution of observation and population across strata ........................................................ 24 Table H-2. Characteristics of respondents and households ....................................................................... 25 3 © 2021 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy, completeness, or currency of the data included in this work and does not assume responsibility for any errors, omissions, or discrepancies in the information, or liability with respect to the use of or failure to use the information, methods, processes, or conclusions set forth. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be construed or considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions The material in this work is subject to copyright. Because the World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522- 2625; e-mail: pubrights@worldbank.org. 4 Acknowledgments This project is possible due to collaboration between the World Bank, Uganda Bureau of Statistics (UBOS) and United Nations High Commissioner for Refugees supported by the Government of Uganda (GoU), particularly the Office of the Prime Minister (OPM). Phone survey data was collected by Laterite. This brief was written by a core team led by Aziz Atamanov and Nobuo Yoshida (World Bank), including Theresa Beltramo (UNHCR), Laura Abril Rios Rivera (World Bank), Ibrahima Sarr (UNHCR), Peter Waita (UNHCR), and Kazusa Yoshimura (World Bank). The report was prepared under the supervision of Antony Thompson (World Bank, Country Manager for Uganda) and Pierella Paci (World Bank, Practice Manager). Implementation of the survey was guided by Joel Boutroue (UNHCR Representative, Uganda) and Margaret Atieno (UNHCR Assistant Representative Protection, Uganda) from UNHCR. The authors are grateful to Stephen Baryahirwa, James Muwonge and Vincent Fred Ssennono from the Uganda Bureau of Statistics for collaboration. We are thankful to Carolina Mejia-Mantilla who started this work. We are also thankful for excellent comments from Zewditu Banteyehun Haile, John Ilukor, Cara Ann Myers, Benjamin Christopher Reese, Shinya Takamatsu, Bo Hurkmans, Lilian Achieng and Charles Alemi. Special thanks to Ally Said and Ibrahima Sarr who drew the sample from the registration data and provided excellent comments. Xiaomeng Chen and Dixita Gupta provided support in imputing consumption and estimating poverty among refugees. Kexin Zhang produced weights for the survey. Martin Buchara provided excellent administrative assistance. Finally, we would also like to thank all respondents of the survey. The State and Peacebuilding Fund (SPF) is a global fund administered by the World Bank to finance critical development operations and analysis in situations of fragility, conflict, and violence. The SPF is kindly supported by: Australia, Denmark, France, Germany, The Netherlands, Norway, Sweden, Switzerland, The United Kingdom, as well as IBRD. 5 Key messages and findings Labour Market and family farming: ✓ The employment rate among refugees was low, reaching 43 percent in October/November 2020. This was significantly lower than the employment rate among Ugandans which reached 90 percent in September/October 2020. Refugees in Kampala had the lowest employment rate (27 percent). ✓ The sector and type of employment among refugees differed across regions. Refugees in Kampala were mostly engaged in services (81 percent) and more than half of the respondents worked in their own business (55 percent). About 46 percent of refugees in the South West worked in the agricultural sector and more than half worked as wage employees. While fewer refugees worked in the agricultural sector in West Nile (36 percent), those that did, mainly worked in their own farms (19 percent). ✓ As reported in October/November 2020, about 13 percent of refugee respondents stopped working since March. Work stoppages were more likely to happen in Kampala (26 percent) followed by South West (24 percent). Refugees in the West Nile region were the least likely to stop working after the lockdown (5 percent). Work stoppages were least pronounced in strata with higher shares of employment in own farms. ✓ Many refugee households were engaged in family farming (69 percent) and livestock (37 percent) since 2020. About 10 percent among them indicated that they had to change their agricultural activities because of COVID-19 during the first agricultural season. Among those who needed to sell agricultural produce since 2020 (20 percent) almost 40 percent were not able to do so in October/November and households in South West were affected the most. Poverty and income: ✓ According to preliminary estimation based on Survey of Well-being via Instant and Frequent Tracking (SWIFT) methodology, poverty among refugees increased after lockdown by eight percentage points from 44 to 52 percent. ✓ Total income either declined or was lost for 89 percent of households since March 2020. Food Security: ✓ Food security seems to have deteriorated significantly compared to the situation in 2018. The share of households which ran out of food because of a lack of money or other resources increased from 61 percent in 2018 to 85 percent in October/November 2020. Access to Services: ✓ Among access to basic needs, access to food was the lowest. Nearly 30 percent of households were not able to buy the main staple food in the week preceding the interview and access to food was the lowest in Kampala (60 percent) followed by South West (41 percent). Access to medical treatment was relatively high, but still 20 percent could not access it when needed since March 2020. ✓ Only 58 percent of households, where any member attended a school before March 2020, had a member engaged in education or learning activities after the closure of schools. The likelihood of participating in learning activities after closures was the highest among refugee households in West Nile (67 percent), followed by South West (47 percent) and Kampala (32 percent). Households in Kampala were the most likely to have members who would not return to school when schools reopened (more than 20 percent). 6 ✓ Among those who engaged in learning, the most common activity was to study alone – the type of learning with the least education and human capital building potential if not combined with any other activity provided by education institutions or education professionals. Thus, in about 21 percent of households, the learning activities included only studying alone, doing homework provided by parents and learning agricultural activities. Coping Strategies, socio Economic Shocks and social assistance: ✓ Every refugee household suffered at least one socioeconomic shock since March 2020. The households from the poorest quintile, based on pre-COVID-19 imputed consumption, were more likely to have more than one shock and were more likely to not implement any strategy to cope with shock(s). ✓ Refugee households in South West and West Nile were covered almost universally by different types of social assistance. Refugees in the West Nile region were more likely to report getting food and other in-kind assistance, while refugees in the South West region were more likely to report getting cash transfers. Refugees in Kampala were the least likely to get social assistance. Knowledge and behavior: ✓ Knowledge of COVID-19 symptoms among refugee households was far from universal and varied by country of origin, but with no significant differences recorded by level of education which may signal about differences in access to information. ✓ Knowledge of preventive measures to reduce the risk of contracting COVID-19 was much lower among refugees than among Ugandans, while refugees in South West reported the highest knowledge about most preventive measures. ✓ False beliefs were slightly more common among refugees than among Ugandans and some were especially prevalent in South Sudanese respondents residing in West Nile, as well as female respondents and those without formal education. ✓ Self-reported adherence to safe practices such as handwashing and avoiding handshakes/physical greetings was very high among refugee households. Mask wearing all or most of the time while in public was reported to be almost universal in Kampala but less frequent in South West and West Nile. Concerns ✓ Many respondents were concerned about the possibility of becoming seriously ill due to COVID- 19 (69 percent) and considered the outbreak to be a substantial financial threat (73 percent). Respondents in West Nile were concerned the least. ✓ Burundian and Congolese respondents in South West felt less safe at home since the COVID-19 outbreak than South Sudanese and Somali respondents. ✓ Perceived safety at home was highly correlated with perceived safety in the community, as Somali respondents tended to feel safer. The main reasons for feeling unsafe in the community were related to living in an unsafe area without safe spaces and having no access to basic services. Female respondents and respondents from female headed households were more likely to feel unsafe because they lived in unsafe area. ✓ Whilst most refugees sought information about education, food, health and employment, resettlement to third countries remained a key information need among refugees in Kampala. 7 A. Background 1. The High-Frequency Phone Survey for refugees in Uganda (URHFPS) tracks the socioeconomic impacts of COVID-19 on refugees. The World Bank (WB) in collaboration with the Uganda Bureau of Statistics (UBOS) and United Nations High Commissioner for Refugees (UNCHR) launched and conducted the URHFPS. The URHFPS tracks the impacts of the pandemic on a monthly basis for a period of three months. Data collection for the first round of URHFPS took place between October 22 – November 25, 2020. This brief discusses results from the first round of the URHFPS. Where possible and appropriate, results for refugees are compared to nationals by using the national High-Frequency Phone Survey on COVID-19 (HFPS).1 2. The survey sample includes respondents with active phone numbers that were selected randomly from the Profile Global Registration System (ProGres) of UNHCR, and the refugee household survey carried out by UBOS and the World Bank in 2018.2 The targeted sample includes 2,100 households and is representative at seven strata constructed as a combination of regions and different countries of origin: Kampala-Somalia, Kampala-other (Burundi, DRC, South Sudan), South West-Burundi (SW-Burundi), South West-DRC (SW-DRC), South West-South Sudan (SW-South Sudan), South West-Somalia (SW- Somalia), and West Nile-South Sudan (WN-South Sudan).3 In terms of population size, the largest strata are South West-DRC and West Nile-South Sudan. Both strata account for more than 85 percent of refugees in Uganda in 2020. The realized sample after the first round was 2,010 households. In order to reduce the bias related to only interviewing households with phone numbers and non-response, the data from the 2018 representative refugee household survey was used to produce and calibrate the weights for the phone survey. B. Economic activities a. Labor market and farm activities 3. The employment rate among refugees was about 43 percent in October/November, compared to 90 percent among Ugandans in September/October, and varied greatly across regions and strata. Respondents were asked if they were engaged in any income generating activity during the last week. At the national level, about 43 percent of respondents reported that they were working (Figure 1). The employment rate was the lowest in Kampala (27 percent) and the highest in West Nile (45 percent). The employment rate was not very different across Burundi, South Sudan and DRC refugees in different strata, but was extraordinarily low among Somali refugees in South West – around 5 percent (Figure 2). Compared to the employment rate among Ugandans, which reached almost 90 percent in September/October (World Bank 2020c), the employment rate among refugee respondents was quite low. 4. The sector of employment and type of activities among refugees differed across regions. Refugees in Kampala were mostly engaged in services (81 percent) and more than half of respondents worked in their own business (55 percent). About 46 percent of refugees in South West were working in the agricultural sector and more than half of refugees were employees. While fewer refugees worked in 1 World Bank (2020a, 2020b, 2020c). 2 World Bank (2019). 3 Selected country of origins accounted for about 97 percent of all refugees in Uganda in 2020. 8 the agricultural sector in West Nile (36 percent), those that did, mainly worked in their own farms at a higher share compared to other regions (19 percent). 5. Many refugees stopped working after the lockdown in March 2020, in particular in Kampala and South West regions. About 13 percent of refugees stopped working after the introduction of the lockdown in March 20, 2020 (Figure 2). Work stoppages were significantly higher among refugees in Kampala (26 percent) and South West (24 percent) compared to West Nile (5 percent). There was substantial variation in the incidence of work stoppages within each region depending on country of origin. For example, work stoppages were large in the South West region among Burundian refugees (31 percent), but not as pronounced among South Sudanese in the same region (4 percent). Such a differential impact may be partially related to the structure of employment in each region and stratum. For example, more respondents worked on their own farms in SW-South Sudan and West Nile-South Sudan strata while work stoppages were also the lowest there. Figure 1. Employment status of respondents by region, Figure 2. Employment status of respondents by strata, (% of all respondents) (% of all respondents) 100 100 90 90 80 80 70 70 60 60 50 50 40 40 30 30 20 20 10 10 0 0 Overall Kampala South West West Nile Kampala - Kampala - SW-Burundi SW-DRC SW-Somalia SW-South West Nile- Somalia others Sudan South Sudanse Respondent not working & no work pre-March Respondent not working & no work pre-March Respondent stopped working (worked pre-March) Respondent stopped working (worked pre-March) Respondent working Respondent working Source: URHFPS first round, authors’ calculation. 6. Work stoppages were mostly related to COVID-19 lockdown restrictions in Kampala, whereas other reasons played a bigger role in West Nile. Not all work stoppages can be attributed to COVID-19 (Figure 3). Reasons directly related to COVID-19, such as “Business/government closed due to COVID-19 legal restriction�, “taking care of children since they are not going to school�, “movement restrictions�, “being ill or quarantined� and “laid off while business continues� accounted for half of all work stoppages at the national level and varied across regions. About 70 percent of work stoppages in Kampala were directly linked to COVID-19, whereas about 59 percent were in the South West region and only 17 percent were in West Nile. 7. Work stoppages were more likely to happen in the services sector and more likely due to COVID- 19. The incidence of work stoppages was the highest in the services sector with no significant difference between those who work/worked in industry and agriculture (Figure 4). Work stoppages were more likely to be related to COVID-19 for those who worked in services, signaling that this sector was most affected by the pandemic (Figure 5). 9 Figure 3. Work stoppages by reasons related to and not related to COVID-19, (% of respondents who stopped working after March 20, 2020) Kampala South West West Nile 17 29 40 59 70 75 Directly related to COVID-19 Not directly related to COVID-19 Source: URHFPS first round, authors’ calculation. Figure 4. Work stoppages by sector, (% of respondents Figure 5. Work stoppages by reasons directly related who were working and used to work before March to COVID-19 by economic sectors (% of respondents 2020) who stopped working after March 2020) Source: URHFPS first round, authors’ calculation. Notes: Not classified sectors are not shown in these figures. 8. A relatively low share of refugees had to change their planting and livestock activities on household farms in the first agricultural season due to COVID-19. Respondents were asked whether they or any household member were engaged in planting activities during the first agricultural season and whether they have been raising any livestock since the beginning of 2020. Refugees in the West Nile region were more likely to engage in agriculture compared to other regions (Figure 6). Respondents were also asked whether they had to change farm activities due to COVID-19. On average, about 10 percent of respondents had to change activities due to COVID-19 with a much larger share in Kampala where, at the same time, only few households were engaged in planting or livestock activities on their own farms (Figure 7). This was close to what was reported by Ugandans in July/August 2020. Among Ugandans about 6 percent had to change planting activities and about 12 percent had to change livestock activities due to COVID-19. 10 Figure 6. Households engaged in planting during first Figure 7. Households engaged in planting during first agricultural season and livestock, since the beginning agricultural season and livestock, since the beginning of 2020 by regions, (% of households) of 2020 who had to change activities due to COVID-19 by regions, (% of households) 90 40 80 35 70 30 60 25 50 20 40 15 30 10 20 5 10 0 0 Overall Kampala South West West Nile Overall Kampala South West West Nile Changed planting activities in first season due to COVID-19 Worked on household farm in first agricultural season Raised livestock Changed livestock activities due to COVID-19 Source: URHFPS first round, authors’ calculation. Note: Small number of observations for Kampala 9. Households changed planting activities due to COVID-19 mainly by abandoning crop farming, reducing the planted areas and planting less variety of crops. Households in West Nile were more likely to abandon crop farming (70 percent), while households in South West were more likely to reduce planted areas (50 percent), planted fewer number of crops (32 percent) and delayed planting (23 percent). The way planting activities were affected mostly included restrictions on movement, delay or inability to get inputs/transport and reduced availability of hired labor. 10. Many refugees reported difficulties Figure 8. Households that needed to sell farm products since in selling their farm products, especially in 2020 and among them, households not able to do so, (% of the South West region. Refugees were asked households) 40 whether they needed to sell family farm products since 2020 and whether they 20 managed to do so last week. At the national level about 20 percent of households needed 0 to sell family farm products and among them Overall Kampala South West West Nile 40 percent were not able to do so during the -20 last week (Figure 8). This share reached a very high 70 percent in South West probably -40 because main market for selling products was in Kampala which was under strict -60 lockdown. -80 % of HH that needed to sell products last week % of HH who were not able to sell products when needed Source: URHFPS first round, authors’ calculation. 11 b. Changes in income since lockdown 11. Assistance from non-government humanitarian agencies was a key income source of income for refugees in all regions, except in Kampala. Respondents were asked about sources of livelihood in the last 12 months. Almost 90 percent of refugees reported receiving assistance from humanitarian agencies such as UNHCR and WFP (Figure 9). The second most frequent source of livelihood was income generated from family farming. Key income sources were relatively similar in South West and West Nile, with slightly higher shares of income from family farming in West Nile. Refugees in Kampala, however, had very different sources of livelihood. Income from non-farm family business, wage employment, remittances and assistance from family and non-family members were more important in Kampala than in other regions. Figure 9. Key income sources during last 12 month across regions, (% of households) 100 90 80 70 60 50 40 30 20 10 0 Assistance Family farming, Non-farm family Wage Assistance from Assistance from Remittances from received from livestock or fishing business employment of family within the other non-family abroad non-government household country individuals agencies eg members UNHCR, WFP etc Overall Kampala South West West Nile Source: URHFPS first round, authors’ calculation. 12. The majority of refugee households reported either a decline or a complete loss of total income, with refugees in Kampala being affected the most. Respondents were asked about changes in household income from each source since March 2020. About 75 percent of respondents reported that their total income has declined since March. About 14 percent of respondents reported a complete loss of income. In total, income either declined or was lost for 89 percent of households. This share was the highest for refugees residing in Kampala (95 percent) and South West (91 percent), and significantly lower than in Kampala in West Nile (87 percent). 13. The heavy reliance on individual transfers, income from nonfarm family business and wage employment was related to the larger income shock in Kampala. Figure 10 presents shares of households with a complete loss of income and juxtaposes it to the shares of households receiving a particular income source. For almost all income sources, higher shares of households in Kampala reported a complete loss of income compared to other regions. Similar to Kampala, although to a lesser extent, households in South West faced loss of assistance from non-family individuals as well as from nonfarm family business. In contrast, only about 1 in 10 households in West Nile reported a complete loss of income from non-farm family businesses as well as family farming. 12 Figure 10. Income sources during last 12 months and share of households without any income by source after March 20, 2020, (% of households) Kampala South West West Nile NHCR, WFP etc Assistance: non-family individuals Assistance: family in the country Remittances from abroad Unemployment benefits Wage employment of HH Non-farm family business, Family farming -100-80 -60 -40 -20 0 20 40 60 80 100 -100-80 -60 -40 -20 0 20 40 60 80 100 -100-80 -60 -40 -20 0 20 40 60 80 100 % of households with no or less income % of households receiving income during last month Source: URHFPS first round, authors’ calculation. c. Impact on poverty and food security 14. Poverty among refugees is estimated to increase after lockdown by eight percentage points from 44 to 52 percent. The URHFPS used a consumption model which was built by using the representative household survey of refugees and host communities conducted in 2018 to identify the strongest correlates of consumption. During the phone survey, information on the correlates were collected. This allowed imputing consumption in the phone survey using the Survey of Well-being via Instant and Frequent Tracking (SWIFT) methodology developed by Yoshida et al. (2015). Given jobs losses, a reduction of income after March 2020 and higher exposure to floods in 2020, strong negative impact on monetary wellbeing of the refugee population was expected. According to the preliminary SWIFT estimation, poverty among refugees was estimated to increase after the lockdown by eight percentage points from 44 to 52 percent. 15. The food security situation of refugee Figure 11. Food security during last 30 days among households has deteriorated significantly. adults in the household, (% of households) Respondents were asked about their food security went without eating for a whole day situation among adults in the household during hungry but did not eat the 30 days prior to the day of interview. Very large shares of households faced constrained ran out of food access to food. One question in the phone survey ate less than they should (“household ran out of food because of a lack of skipped meal money or other resources�) was strictly ate only a few kinds of foods comparable to the question in the representative unable to eat healthy and nutritious… refugee survey conducted in 2018. Comparative worried about not having enough food results from both surveys show a substantial 0 20 40 60 80 100 deterioration in household’s food security 2018 survey October/November 2020 situation in October/November 2020 compared to 2018 (Figure 11). The share of households which Source: URHFPS first round and 2018 survey of refugees and host communities, authors’ calculation. ran out of food because of a lack of money or other resources increased from 61 to 85 percent. 13 d. Social assistance 16. Social assistance played a very important role for refugees, but its incidence and type varied substantially across different refugee hosting regions. Respondents were asked if they received any social assistance since March 2020. The majority of refugees did get some type of social assistance. Only seven percent of refugees reported not getting anything and this share was much higher in Kampala (32 percent; Figure 12). This finding is consistent with the fact that refugees and asylum seekers in Kampala do not benefit from humanitarian assistance except for a one-time cash assistance related to COVID-19. Refugees in different regions report getting different types of assistance (Figure 13). Refugees in West Nile were more likely to get in-kind transfers such as food, while cash transfers were widespread among refugee in South West. Figure 12. Share of households without social Figure 13. Type of assistance received since March assistance since March 2020, (% of households) 2020, (% of households) 35 100 30 80 25 60 20 40 15 20 10 5 0 Overall Kampala South West West Nile 0 Cash Transfers Overall Kampala South West West Nile Other in-kind transfers (excluding food) In-kind transfer, Food Source: URHFPS first round, authors’ calculation. C. Knowledge and behavior 17. Knowledge of COVID-19 symptoms among refugee households was far from universal and varied by country of origin, with no significant differences by level of education. Symptoms such as dry cough (84 percent) and fever (68 percent) were known among most respondents, while fatigue symptoms were known by only 18 percent of respondents and loss of smell was known by only 3 percent (Figure 144). This was similar to the level of knowledge about COVID-19 symptoms among national households (World Bank 2020a). Interestingly, the level of education of household members was not correlated with knowledge of COVID-19 symptoms. 14 18. Nearly all Somali Figure 14: Knowledge of selected COVID-19 symptoms across respondents knew about fever while countries of origin, (% of respondents) 100 less than 6 out of 10 South Sudanese respondents knew about fever as 84 one of the symptoms of COVID-19. 80 Somali respondents were the most 69 likely to know about loss of smell or 60 58 taste, sore throat and shortness of breath. Differences by area of 40 residence were only relevant for 27 knowledge about loss of smell or 20 23 18 taste, a symptom that was more often known by respondents in 5 0 Kampala. Differences in knowledge Dry cough Fever Stuffy or Shortness of Sore throat Fatigue Loss of smell runny nose breath or taste may be related to differences in South Sudan DRC Burundi Somalia Overall information campaigns and access to Source: URHFPS, first round. information. 19. Knowledge of preventive measures to reduce the risk of contracting COVID-19 was much lower among refugees than among Ugandans, while refugees in South West reported the highest knowledge about most preventive measures. Compared to refugee households, most Ugandan households knew about preventive measures, with the use of gloves being the least known measure (Figure 15). Despite having better knowledge about COVID-19 symptoms, refugees in Kampala reported lower knowledge of preventive measures than refugees residing in other areas (Figure 16). Figure 15: Knowledge of preventive measures Figure 16: Knowledge of preventive measures among among refugees and nationals, (% of respondents) refugees by region, (% of respondents) Avoid travel Avoid travel Avoiding touching your face Avoiding touching your face Use of sanitizer Use of sanitizer Staying at home and avoid going Avoid crowded places or out unless necessary gatherings with many people Avoid crowded places or gatherings No Handshake / physical with many people greetings No Handshake / physical greetings Maintain enough distance of at Maintain enough distance of at least 1 meter least 1 meter Use of mask Use of mask 0 20 40 60 80 100 0 20 40 60 80 100 Nationals (June) Refugees (October/November) West Nile South West Kampala Source: URHFPS and HFPS, first round. Source: URHFPS first round, authors’ calculation. Note: Time period is different across surveys. HFPS first round was conducted in June, while URHFPS in October/November 2020. 15 20. Avoiding groups of more than 10 people was the least practiced safe behavior while wearing masks all the time when in public was practiced in 60 percent of refugee households. Safe practices such as handwashing and Figure 17: Frequency of mask wearing when in public, (% of avoiding handshakes and physical greetings respondents who were in public) were practiced among most respondents a 100 week prior to the day of interview. However, avoiding groups of more than 10 people was 80 less often practiced especially among 60 households in South West (82 percent) and West Nile (84 percent) compared to Kampala 40 (94 percent). Mask wearing all or most of the time while in public was reported as almost 20 universal in Kampala but less frequent in South West and West Nile (Figure 17). The 0 Overall Kampala South West West Nile higher frequency of mask wearing in Kampala All of the time Most of the time Less ofthen may be due to stricter COVID-19 containment Source: URHFPS first round, authors’ calculation. measures in that area. Figure 18: False beliefs about COVID-19 among Figure 19: False beliefs about COVID-19 among refugees, (% of refugees and Ugandans, (% of respondents) respondents) Local herbs can be Local herbs can be used to Female used to treat treat corona patients corona patients Gender Taking alcohol can make one Taking alcohol can immune to contracting Male make one immune corona to contracting corona COVID-19 is common flu COVID-19 is Completed secondary & + common flu Education Coronavirus cannot survive in Coronavirus cannot warm weather No formal education survive in warm weather Coronavirus does not affect children Coronavirus does West Nile not affect children Africans are immune to Coronavirus Africans are Residence Lemon/alcohol can be used South West immune to as sanitizers against Coronavirus Coronavirus Lemon and alcohol 0 20 40 60 80 Kampala can be used as Refugees (October/November) sanitizers against coronavirus. Nationals (June) 0 20 40 60 80 Source: URHFPS and HFPS, first round. Source: URHFPS first round, authors’ calculation. 21. False beliefs were more common among refugees than Ugandans and were especially more prevalent in South Sudanese respondents residing in West Nile, as well as female respondents and those without formal education. Respondents were asked if false statements about COVID-19 were true or not. False beliefs, measured by those who believed or did not know whether the statements were true or not, were more common among refugee respondents than among Ugandans (Figure 18). The most 16 common false belief for both communities was that lemon and alcohol can be used as a sanitizer (Figure 18 and Figure 19). The use of local herbs to cure COVID-19 was also a common false belief, which was particularly prevalent among refugee women, refugees without formal education as well as refugees residing in Kampala. Overall, higher education levels were associated with lower levels of false beliefs among refugees. D. Access to food, health, and education 22. Nearly 3 in 10 refugee households were not able to buy main staple food in the week preceding the interview and 2 in 10 could not access medical treatment since March 2020. More than 60 percent of refugee households in Kampala had members who were not able to buy staple food when needed while that was the case for 41 percent of households in South West and 33 percent in West Nile (Figure 21). Notably, only 16 percent of national households reported inability to buy staple food (Figure 20 Bank 2020a). For both refugees and nationals, 2 in 10 households with members who needed medical treatment during the month preceding the interview were not able to access it. The most common reason for not being able to access medical treatment was the lack of financial resources. Inability to access food and medical treatment can result in longer term health impacts linked to malnutrition and complication of diseases. 23. Following the closure of schools, there was a substantial drop in the participation in education activities with a sizable share of households whose members did not plan to return to school once it re- opened, particularly in Kampala. Only 58 percent of households, where any member attended a school before March 2020, had a member engaged in any education or learning activities after the closure of schools (Figure 22). Even though not strictly comparable, this was very close to what was reported among Ugandan households in June 2020 (World Bank 2020a). Likelihood of participating in learning activities after closures was the highest among refugee households in West Nile (67 percent), followed by South West (47 percent) and Kampala (32 percent). Figure 20: No access to selected basic goods and services Figure 21: No access to selected basic goods and when needed, nationals and refugees, (% of households) services when needed, refugees by region, (% of households) Mask West Nile Water South West Soap Kampala Medical treatment Staple food Overall 0 10 20 30 40 50 60 0 10 20 30 40 50 60 Nationals (September/October) Refugees (October/November) Mask Water Soap Medical treatment Staple food Source: URHFPS and HFPS, first round. Source: URHFPS first round, authors’ calculation. Respondents were also asked if any child or household member who attended schools before March 2020 would not return to schools after schools were open. Households in Kampala were the most likely to have 17 members who would not return to school when schools reopened. The main reason for not going back to school was inability to pay fees, which was more common in Kampala, followed by care taking needs, mainly in West Nile (Figure 23). The lower attendance after closures and the higher number of students not returning school among households in Kampala may be linked with the work of non-governmental organizations in South West and West Nile where most refugee settlements are located. Such organizations cover refugee education needs at a low cost, whereas in urban settings refugees must cover their educational needs with minimal assistance. Figure 22: Engagement in any learning activities after Figure 23: Main reasons for not going back to school, schools’ closure and plans to return to schools when (% of households with any member not returning to they reopen, (% of households with any member school) attending school before March 2020) 100 Need to help with taking care of family 80 members 60 Female member will or is now married 40 20 Can't afford fees 0 0 20 40 60 80 100 Overall Kampala South West West Nile % of households % of HHs with members engaged in learning activities after closure % of HHs with members not returning to school once they reopen West Nile South West Kampala Overall Source: URHFPS first round, authors’ calculation. Source: URHFPS first round, authors’ calculation. Figure 24: Learning and education activities after closures, (% of households with any member engaged in any learning activity after March 2020) Studying alone Used reading materials provided by government Session/meeting with lesson teacher or tutor Completed assignments provided by the teacher Listened to educational programs on the radio Doing homework given by parents/relatives Watched educational TV programs 0 10 20 30 40 50 60 West Nile South West Kampala Overall Source: URHFPS first round, authors’ calculation. Note: Multiple answers are allowed 18 24. Among those who were still engaged in learning, the most common activity was to study alone. Respondents were asked about types of education or learning activities children or other family members had been engaged in over the last week. The majority mentioned studying alone, which if not combined with other activities, has the least actual learning potential. Thus, in about 21 percent of households, the only learning activities included studying alone, doing homework provided by parents and learning agricultural activities. In Kampala, the use of reading materials provided by the government and meeting with teachers or tutors was notably lower than among households in South West and West Nile (Figure 24). At the same time, it was only Kampala where watching educational TV programs was mentioned. 25. While most refugees sought information about education, food, health and employment, resettlement to third countries remained a key information need, particularly among refugees in Kampala. In 6 out of 10 households, Figure 25: Information needs, (% of respondents) refugees needed information about access to education and food (Figure 25). It was mainly refugee households in the lowest wealth quintile who were interested in education and food information. Remarkably, only refugees in 3 out of 10 households needed information about employment, although in Kampala such a need was more frequent (41 percent). Overall, 24 percent of respondents reported a need to information about resettlement to third countries with 63 percent among households in Kampala. Source: URHFPS first round, authors’ calculation. 26. The most reliable sources of information on COVID-19 for refugees were radio, community leaders, NGO’s, heath centers/health workers and friends/neighbors with substantial variation across regions. Refugees were asked to select the most trusted sources of information about COVID-19. For refugees in Kampala the most trusted sources were TV (31 percent), followed by friends/neighbors (28 percent) and radio (21 percent). For refugees in South West the most trusted sources were radio (43 percent), friends/neighbors (20 percent) and NGOs (19 percent). For refugees in West Nile the most trusted sources were radio (53 percent), community leader (22 percent) and health centers/health workers (13 percent). E. Coping strategies to socioeconomic shocks 27. Every refugee household suffered at least one socioeconomic shock since March 2020 with variations by area of residence and pre-COVID consumption quintiles. Overall, 62 percent of households suffered a shock. In South West, about 46 percent of households experienced two shocks (Figure 26). The 19 most common shock, was an increase in the price of frequently consumed food items (Figure 27), although such a shock was more common in South West and West Nile than in Kampala. Disruption of farming, livestock, and fishing activities was more common in West Nile and among the poorest households than in South West, Kampala and richer households. Job losses and closure of non-farm businesses Figure 26: Number of shocks since March 2020, (% of were more common in Kampala and South households) West as well as for the richer refugee quintiles Richest Wealth households. Notably, households affected by an increase in the price of farming and Poorest business inputs were mainly those in West Education Secondary & + Nile. No formal education West Nile Residence South West Kampala Overall 0 20 40 60 80 100 2+ 1 shock Source: URHFPS first round, authors’ calculation. Figure 27: Types of shocks, (% of households) 70 60 50 40 30 20 10 0 Overall Kampala South West West Nile Poorest Richest Residence Wealth quintiles Increase in price of major food items consumed Disruption of farming, livestock, fishing activities Illness, injury, or death of income earning member of household Job loss Nonfarm business closure Theft/looting of cash and other property Increase in price of farming/business inputs Source: URHFPS first round, authors’ calculation. 28. The most used coping strategy was to reduce food consumption, followed by receiving humanitarian assistance and engaging in additional income generating activities. Refugees used different strategies to cope with shocks. Food consumption was reduced by almost 35 percent of households, although in South West, such coping strategy was implemented by 44 percent of households 20 and more often by richer than by poorer households (Figure 28). Households in South West as richer households more often used humanitarian assistance from NGOs as a coping strategy than poorer households as well as those in Kampala and West Nile. Engagement in additional income generating activities was more frequent in Kampala and West Nile than in South West. Notably, almost half of the households from the poorest pre-COVID-19 quintile did not use any strategy to cope with shocks probably because their levels of food and non-food consumption were already very low. Figure 28: Selected coping strategies to socioeconomic shocks, (% of households) 60 50 40 30 20 10 0 Overall Kampala South West West Nile No formal Completed Poorest Richest education secondary & + Residence Education level of respondent Consumption quintiles Reduced food consumption Received assistance from NGO Endaged in additional income generating activities Received assistance from friends & family Reduced non-food consumption Did nothing Source: URHFPS first round, authors’ calculation. F. Concerns 29. Most refugee households, especially those in Kampala, had members who worried about the possibility of becoming seriously ill due to COVID-19 and considered the outbreak to be a substantial financial threat. About 70 percent of respondents Figure 29: Degree of threat from coronavirus outbreak to worried about themselves or close household’s finances family members becoming ill (Figure 29). This was especially worrisome for 100 households in Kampala (91 percent 80 versus 84 percent in South West and 56 percent in West Nile). The COVID-19 % of households 60 outbreak was considered to be a financial threat in most refugee 40 households although more severely so in 20 Kampala than in other areas. 0 Overall Kampala South West West Nile A substantial threat A moderate threat Not much of a threat Not a threat at all 21 Source: URHFPS first round, authors’ calculation. 30. Burundian and Congolese households in South West felt safe the least at home since the COVID- 19. Overall, 2 in 10 respondents did not feel safe at home since the COVID-19 outbreak. Perceptions of lack of safety at home were more widespread among Burundian and Congolese households in South West compared to South Sudanese in South West and Somali households in Kampala and South West (Figure 30). Perceived safety at home was highly correlated with perceived safety in the community, as Somali respondents tended to feel safer (Figure 31). Figure 30: Perception of safety at home since the COVID-19 outbreak, (% of respondents) 100 80 60 40 20 0 Kampala - Kampala - SW-Burundi SW-DRC SW-Somalia SW-South WN-South Somalia others Sudan Sudanse Overall Strata Yes, I feel safe at home Yes, but not always No, I don’t feel safe Source: URHFPS first round, authors’ calculation. 31. More than 60 percent of respondents perceived to be safe in their community, however, non- Somali respondents and those not in the West Nile region perceived a lower degree of safety. As noted for perceived safety at Figure 31: Perception of safety in the community since the COVID-19 home, perceived safety in the outbreak, (% of respondents) 100 community was the highest among Somali respondents regardless the 80 place of residence (Figure 31). Security in the community was 60 perceived to be higher among South Sudanese in West Nile compared to 40 South Sudanese in South West. 20 0 Kampala - Kampala - others SW-Burundi SW-DRC SW-Somalia SW-South Sudan WN-South Somalia Sudanse Strata Source: URHFPS first round, authors’ calculation. 22 32. The main reasons for feeling unsafe in the community were related to living in an unsafe area without safe spaces and having no access to basic services. Around 44 percent of refugee households Figure 32: Main reason why felt unsafe in community since with members who felt unsafe in their the onset of COVID-19 community, reported living in an unsafe area as the main reason. In West Nile, more than West Nile half of households reported such a reason (Figure 32). Female headed households were South West almost as twice as more likely to report this reason compared to male headed ones. Lack or limited access to basic services was a Kampala major reason for feeling unsafe, especially in South West, which contrasted greatly with Overall only 3 percent of households reporting such a reason in West Nile. Lack of safe spaces was 0 20 40 60 the third most common reason for feeling There are no safe spaces around me Lack or limited access to basic services unsafe in the community, with Kampala and I live in an unsafe area South West household reporting this reason more often than in West Nile. Source: URHFPS first round, authors’ calculation 23 G. Bibliography Yoshida, N., Munoz, R., Skinner, A., Kyung�eun Lee, C., Brataj, M., Durbin, W. and Sharma, D. (2015). SWIFT Data Collection Guidelines. Version 2. Washington, DC: World Bank. World Bank (2019). Informing the Refugee Policy Response in Uganda: Results from the Uganda Refugee and Host Communities 2018 Household Survey (English). Washington, D.C.: World Bank Group. World Bank. 2020a. COVID-19 Impact Monitoring: Uganda, Round 1. World Bank, Washington, DC. World Bank. 2020b. COVID-19 Impact Monitoring: Uganda, Round 2. World Bank, Washington, DC. World Bank. 2020c. COVID-19 Impact Monitoring: Uganda, Round 3. World Bank, Washington, DC. H. Annex a. Sampling The Profile Global Registration System (ProGres) served as a sample frame for the URHFPS. It was complemented by the data collected for the refugee household survey carried out by UBOS and the World Bank in 2018. The sample was selected from the pool of refugees with phone numbers. The targeted sample included 2,100 observations: 300 observations in each stratum. Four country of origins were targeted in the survey: Burundi, Democratic Republic of Congo (DRC), Somalia and South Sudan. Combination of country of origin and region were used to create seven strata: Kampala-Somalia, Kampala- other (Burundi, DRC, South Sudan), South West-Burundi, South West-DRC, South West-South Sudan, South West-Somalia, and West Nile-South Sudan. The realized sample after the first round of data collection was 2,010 households. All of the information about individual members from ProGres was verified and adjusted if needed. After this, the final distribution of respondents over strata was as shown in table H-1, with the highest number of observations in stratum Kampala-Somalia (N=340) and the lowest number of observations in stratum SW- Somalia (N=238). During the second and third rounds of data collection, the same households will be re- contacted with a possibility of adding new observations to stratum/strata with high attrition or a number of households lower than the quota of 300 observations. Table H-1. Distribution of observation and population across strata number of Population Strata observations weighted share Kampala (Burundi, DRC, South Sudan) 292 2 Kampala-Somalia 340 2 SW-Burundi 284 4 SW-DRC 299 26 SW-South Sudan 312 6 West Nile-South Sudan 245 58 SW-Somalia 238 1 Total 2,010 100 Source: URHFPS first round, authors’ calculation. 24 Given that the refugee population with phone numbers may be fundamentally different from the population without phone numbers (e.g. more educated, affluent, living in urban areas and so forth), and also due to the issue of non-response, respondents in the phone survey may not be representative of the overall refugee population in Uganda. In order to reduce the potential bias and make results nationally representative, we developed a weighting procedure by producing weights using a reference nationally representative survey and calibrating obtained weights to make phone survey nationally representative and resemble the distribution for the key variables of those from the reference survey. The refugee survey from 2018 was used as a reference in creating weights for the phone survey, while post-stratification was conducted to preserve to the extent possible regional population shares from the ProGres dataset as of November 2020. b. Profile of respondents and households Table H-2. Characteristics of respondents and households Residence Overall Kampala South West West Nile Demographics Household size 5.4 5.3 5.0 5.8 Number of children, 0-14 3.0 2.1 2.7 3.3 Dependency ratio, % 54 37 53 57 Age of respondent 38 37 38 37 Male respondent 47 42 59 39 Male head 49 45 65 38 Education of respondent No formal education, % 22 31 27 17 Complete primary and below, % 35 12 36 36 Some secondary, % 30 23 23 36 Completed secondary & +, % 13 34 13 10 Consumption per adult equivalent quintiles Poorest 20 0 6 31 Second 20 0 11 27 Third 20 0 22 20 Fourth 20 0 29 16 Richest 20 100 32 6 Year of arrival Before 2011 12 16 26 1 2011-2017 81 59 62 97 2018-2020 7 24 12 2 Country of origin South Sudan 61 5 16 100 DRC 26 31 60 0 Burundi 6 9 14 0 Somalia 7 54 11 0 Source: URHFPS first round, authors’ calculation. Note: All estimates are household weighted, except for consumption quintiles. 25