ROUND 2 COVID-19 IMPACT MONITORING NOV 2020 Publica on Date BACKGROUND UGANDA This brief presents findings from the second round of the Uganda High-Frequency Phone Survey on COVID-19 (UHFPS) conducted in July/August 2020. In June 2020, the Uganda Bureau of Statistics (UBOS), with the support from the World Bank, officially launched the HFPS to track the impacts of the pandemic on a monthly basis for a period of 12 months. The survey aimed to recontact the entire sample of households that had been interviewed during the Uganda National Panel Survey (UNPS) 2019/20 round and that had phone numbers for at least one household member or a reference individual. The first round (baseline) of the survey was conducted from June 3rd to June 20th and the second round was conducted be- tween July 31 and August 21, 2020. Of the 2,421 households targeted 2,227 households were interviewed in round 1, and of those, 2,199 were interviewed in round 2, representing a 99 percent response rate between rounds. BEHAVIOR AND CONCERNS RELATED TO COVID-19 Safe Practices There was a drastic decline in the reported preva- lence of safe practices in the second round. Compared to Round 1 in June, significantly fewer respondents reported that they reduced the number of times they went to the market; avoided groups of more than 10 people and physical greetings; and washed hands more frequently in the last 7 days prior to the interview in July-August. The change was observed across rural and urban areas and across both male and female respondents. Figure 1. Prevalence of safe practices across two rounds, % of respond- ents The practices of wearing a mask in public and hand- About 80 percent of respondents washed hands all or most washing after being in public are far from being uni- of the times after being in public, while 66 percent wore versal and are highly correlated with the respond- masks all or most of the time while in public. The prevalence ent’s education. Respondents were asked about the fre- of these practices was constantly higher among respondents quency of wearing masks and handwashing after being in in urban areas and those with higher levels of education public in the last 7 days prior to the interview in July-August. (Figures 2 and 3). Figure 2. Frequency of washing hands with soap after being in public by Figure 3. Frequency of wearing masks in public by education level education level in round 2, % of respondents in round 2, % of respondents Awareness and assessment of COVID-19 measures Awareness of easing of lockdown measures an- aware about the policy change. The level of awareness was nounced on June 22 was very high, but at least half of lower among the poorest quintile compared to the richest respondents felt that this move could escalate the quintile and among rural residents compared to urban ones. spread of COVID-19. Respondents were asked if they Respondents were also asked if easing measures would esca- were aware of the revised COVID-19 guidelines that were late the spread of COVID-19. The largest concern was asso- announced on June 22 leading to easing the lockdown ciated with allowing public and private vehicles to operate – measures. On average, 94 percent of respondents were about 67 percent of respondents felt this would escalate the 1 COVID-19 IMPACT MONITORING spread of COVID-19 (Figure 4). The lowest level of concern there was a significant difference in perceptions between re- was related to allowing the food sellers to go home after 21 spondents living in the poorest 20 percent of household pop- days of being in lockdown– 49 percent of respondents felt ulation versus those living in the richest 20 percent of house- this would escalate the spread of the disease. There was no holds, as measured by the pre-COVID-19 household annual significant difference in answers between rural and urban consumption per adult equivalent quintiles. Respondents liv- respondents. ing in households from the poorest quintile were more likely Existing lockdown restrictions were almost universal- to think that lockdown measures would curb the spread of ly perceived as effective measures to curb the spread COVID-19 in comparison to their counterparts living in of disease. Very high shares of respondents in July-August households from the richest quintile. For example, 93 per- thought that existing lockdowns measures were effective in cent of respondents from the poorest quintile be- curbing the spread of COVID-19 (Figure 5). At least 85 per- lieved that suspension of communal prayers or closure of cent of respondents believed that any of the listed measures churches would curb the spread of COVID-19 compared would curb the spread of disease. Despite very high numbers, to 82 percent of respondents among the richest quintile. Figure 4. Share of respondents who felt that easing measures would Figure 5. Share of respondents who felt that existing lockdown measures escalate the spread of COVID-19 in round 2, (%) would curb the spread of COVID-19 in round 2, (%) Many respondents experienced security risks and felt that COVID-19 emergency measures could limit their rights. About 47 percent of respondents agreed that they experienced greater security risks and vulnerability to crime and violence during the COVID-19 crisis (Figure 6). About 57 percent agreed that the response to the COVID- 19 emergency would limit their rights and freedoms. Security risks and concerns were higher among urban residents. Several respondents shared concerns that elites and the authorities might misuse money and supplies al- located for the COVID-19 response. Respondents were asked in July-August if they agreed that money and supplies allocated for the COVID-19 response would be misused and captured by powerful people in the country and if they agreed that corruption in the government had lowered the Figure 6. Share of respondents who agreed with a set of statements and quality of medical supplies and care. Overall, concerns were concerns related to the COVID-19 crisis, (%) widespread (Figure 6). For example, 74 percent agreed that money for COVID-19 response would be captured by pow- erful people and 81 percent agreed that corruption had low- ered the quality of medical supplies and care. ACCESS TO BASIC NEEDS Basic goods and services The incidence of lack of access to essential needs in the last 7 days prior to the interview did not change much across rounds of the phone survey. Among households that needed to access medicine, 32 percent were not able to meet their medicinal needs (Figure 7). The preva- lence of not being able to meet medical treatment needs (conditional on need) was 11 percent - representing a 7- Figure 7. Share of households without access to selected basic needs percentage point improvement vis-à-vis the first round of the when needed by residence and consumption per adult equivalent quintiles survey. The prevalence of not being able to meet the soap in round 2, (%) 2 COVID-19 IMPACT MONITORING needs (conditional on need) was also lower in the second Finally, the prevalence of not being able to meet drinking round. However, it continued to be the lowest among the water needs (conditional on need) was 5 percent. This esti- poorest households and those in rural areas – the main rea- mate was higher among the poorest households and those in son for unmet need was stated as the inability to afford it. rural areas. Health About two percent of households had at least one and this can be partially attributed to malaria and other com- member who received laboratory testing for COVID- mon infections. Respondents were also asked if any family 19. Respondents were asked about different COVID-19 re- member received a laboratory test for COVID-19. On the lated symptoms household members experienced in the last whole, two percent of households got this test, however the 7 days prior to the interview conducted in July-August. As share of households with at least one member having tested shown in Figure 8, the most widespread symptoms were for COVID-19 was higher among sub-samples of households fever (28%), headache (27%), and cough (27%). All three with at least one member with a COVID-19 symptom symptoms were more pronounced among rural residents (Figure 9). Figure 8. Household-level incidence of having at least one member with Figure 9. Share of households, where any member got COVID-19 test a specific symptom during last week in round 2, (%) grouped by experienced symptoms, % Education comparable indicator to the first round – participation of any The rural-urban gap in participation in any education child (3-18) in any learning activities at the household level. or learning activities widened according to the find- The lack of change in this variable at the national-level be- ings in the second survey round. Respondents were tween the two survey rounds masks differences between asked in July-August about each child’s participation in any households residing in rural versus urban areas. The share of education or learning activities in the past 7 days – this rep- households with at least one child participating in any educa- resented a richer body of information on education vis-à-vis tion or learning activities declined slightly in rural areas (56 the first round of the survey conducted in June. In turn, the versus 54%) and increased in urban areas (67 versus 73%). information from the second round was used to construct a This further widened the existing gap between rural and ur- ban areas, from 11 percentage points in the first round to 19 percentage points in the second round. Further, according to the data from the second round, only 42 percent of chil- dren (3-18) living in households in the poorest quintile en- gaged in any education or learning activities. The comparable estimate was 69 percent for children living in households in the richest quintile. The most frequent reasons for children not partici- pating in any education or learning activities were lack of learning materials, low student interest, no access to radio/tv and increased household chores. Respondents were asked for each child the reason why he or she was not participating in learning activities at home. Not receiving materials from school (53%) or government (43%) were the key reasons that were mentioned, particularly among children in rural areas and those living in households Figure 10. Share of households with at least one child (3-18) engaged in in the poorest quintile. No interest from student was report- any education or learning activities (conditional on having at least one ed to be the reason for 35 percent of children and this rea- child) by survey round and rural/urban residence, (%) son was more likely to be mentioned for boys than girls. 3 COVID-19 IMPACT MONITORING No access to radio and tv was mentioned to be the reason for 31 percent of children, particularly among those living in rural areas and in households in the poorest quintile. Finally, increased household chores were reported to be the reason for 13 percent of children, with much large shares in rural areas and older children of secondary age. Only 16 percent of children did not face any problem while learning at home, while many children faced multiple difficulties, especially those in rural areas. Respondents were asked about the key challenges that school-age children were facing while learning at home. Lack of skilled instructors was mentioned in 39 percent of cases and was more important in rural areas. Limited access to learning materials from government and from school was Figure 11. Main reasons why a child was not engaged in education or mentioned in 31 and 26 percent of cases and was again more learning activities, by rural/urban residence, region and consumption per important for rural areas. Among other important reasons adult equivalent quintiles, (%) mentioned were increased household chores (especially among the poorest) and limited access to tv/radio. Food Security In Round 2, the survey included eight questions for construc- to 3 percent for severe food insecure) than in rural area (42 tion of the Food Insecurity Experience Scale for all Ugan- to 35 percent for moderate food insecure and 9 to 6 per- dans. We break down the selected food security indicators cent for severe food insecure). In Central Uganda the decline across rural and urban areas, by pre-COVID-19 household in the share of food insecure across rounds was found signifi- per adult equivalent consumption quintile and regions. Over- cantly higher than in other regions. Thus, moderate food all, in the last 30 days preceding the interview in July-August, insecure declined from 38 to 23 percent and severe food the prevalence of severe food insecurity was 5 percent insecure declined from 8 to 3 percent. among all Ugandans, and the prevalence of moderate food security stood at 33 percent. Ugandans in rural areas were more likely to experience both moderate and severe food insecurity, while Ugandans in the bottom 40 percent of the pre-COVID-19 consumption distribution were more likely to experience moderate food insecurity. Moderate and se- vere food insecurity were significantly higher in Northern and Eastern Uganda compared to other regions. In Western Uganda the incidence of moderate and severe food insecure was the lowest. In addition, we are able to compute prevalence estimates specifically for adults in a way that is comparable to round 1. Among adults, the prevalence of severe food insecurity de- clined from 9 to 5 percent, and the prevalence of moderate food insecurity declined from 42 to 33 percent. The inci- Figure 12. Prevalence of severe and moderate or severe food insecurity dence of moderate and severe food insecurity declined more among all Ugandans, by rural/urban residence and pre-COVID-19 house- in urban (44 to 29 percent for moderate food insecure and 8 hold per adult equivalent consumption quintiles, (%) ECONOMIC SITUATION Credit and safety nets Every fifth household had to borrow money to face assistance from family or neighbors. Several reasons were the COVID-19 emergency since March 2020. Findings related to the situation in the labor market and overall econ- showed that slightly more than 20 percent of households had omy. About 18 percent of respondents mentioned closure of to borrow irrespective of whether they were living in urban business, 24 percent inability to sell produce and 25 percent or rural areas. Households from the poorest quintile were mentioned a reduction in sales. A reduction in sales was a least likely to borrow money. Households from the Eastern particularly important reason in the Central region, among (36%) and Central (27%) regions were more likely to borrow the richest top quintile and urban residents. The inability to money. The households in Western region were least likely sell produce was a prominent reason in the Northern region, to borrow money (8%) (Figure 13). The main reasons for among rural residents and the poorest households. Business borrowing ranged from employment shocks to loss closure was most important in the Western region, among of assistance from family or neighbors. Those who had urban residents and the richest households. In the Eastern to borrow to face COVID-19 emergency situations were region the major reason was due to failure to obtain money asked why they had to do so (Figure 14). The largest share of from family or neighbors. households had to borrow (27%) because they could not get 4 COVID-19 IMPACT MONITORING Figure 13. Share of households who had to borrow money since March 2020 to face the COVID-19 emergency, (%) Figure 14. Reasons why households had borrowed since March 2020, by rural/urban residence, region and pre-COVID-19 household per adult Borrowing from friends, relatives and neighbors was equivalent consumption quintiles, (%) the main source of credit after March 2020, in partic- used this source and access was much higher among urban ular among rural residents and among the poorest residents and those from the richest quintile. In contrast, households. Access to credit from formal financial institu- rural households and the poorest were more likely to bor- tions such as commercial banks and credit institutions re- row from saving groups/loan associations and from friends mained quite low (Figure 15). Only 12 percent of households and neighbors. Figure 15. Selected sources of borrowing to face the COVID-19 emer- Figure 16. How worried households were about repayment of money gency by rural/urban residence, region and pre-COVID-19 household per within repayment period by rural/urban residence and pre-COVID-19 adult equivalent consumption quintiles, (%) household per adult equivalent consumption quintiles, (%) Two third of households were worried about the re- payment of borrowed money within the repayment period, with higher levels of concern among the poorest and rural residents. Respondents were asked if they worried about being unable to pay money within the repayment period (Figure 16). Almost 70 percent of respond- ents were very worried or worried about repaying money on time. This share was slightly higher among rural households (70%) compared to urban ones (62%); and among the poor- est households from the bottom 20 percent of the popula- tion (79%) compared to the richest households from the top quintile (62%). Safety net support has changed across rounds from prevalent food aid in the first round to in-kind non- food transfers in the second. Many urban households Figure 17. Share of households that received different types of social reported getting food aid during the first round following the assistance since the last interview in round 2 by rural/urban residence, regions and pre-COVID-19 household per adult equivalent consumption food distributed after March 20 mainly targeting Kampala. quintiles, (%) About 9 percent of households reported getting food aid. During the second round conducted in July-August (Figure 17), when questions were asked about getting social assis- getting food aid dropped to about 2 percent. However, many tance since the last interview in June, the share of households households reported getting in-kind non-food assistance in 5 COVID-19 IMPACT MONITORING the form of masks, mosquito nets and soap (16%). The high- observed in rural areas (18%), among the poorest quintile est share of households getting in-kind non-food aid were (20%), and in the Northern region (26%). Employment and livelihood Based on the second round of interviews, (July/ August 2020) the employment situation was picking up back to pre-COVID-19 levels. Overall, the share of respondents reporting to have worked in the week preced- ing the second round of interviews (87%) was marginally higher than the share of respondent working before March 2020, when the outbreak occurred. However, there were significant differences across areas of residence and welfare quintiles. In rural areas the share of working respondents was 4 percentage points higher than the share pre COVID-19; while in urban areas, the share of working respondents was 8 percentage points lower with respect to March 2020 levels. Figure 18. Share of respondents working before the COVID-19 outbreak At the same time the share of working respondents in- and during the last 7 days preceding each round of interview, by rural/ creased to levels higher than March 2020 in the three lowest urban residence and pre-COVID-19 household per adult equivalent con- sumption quintiles, (%) quintiles, while the upper two quintiles still lagged behind the pre-COVID19 levels. Although the shares of working re- spondents were still behind the pre-COVID-19 levels in ur- Figure 19 presents the distribution of working and non- ban areas and in the richest quintiles, there was a fast im- working respondents in July-August interview across their provement. Indeed, there was a 19 percentage points in- employment status in June. The vast majority of respond- crease in working respondents with respect to Round 1 both ents working in August were already working in June (65%), in urban areas and in the richest quintile. and in particular in rural areas where about two thirds of the respondents never quit working since the COVID-19 out- break. About 12 percent of working respondents in August, were working before the COVID-19 outbreak, but were not working during the first round conducted after lockdown. The recovery of working activities was especially pro- nounced in urban area (18%) and in the richest quintile (18%). About 10 percent of the respondents started working in August for the first time. Significant differences for these respondents were found across the poorest and richest quintiles and across rural and urban areas: 19 percent in the poorest quintile versus 5 percent for the richest quintile, and 12 percent in rural areas versus 6 percent in urban ones. Among 14 percent of respondents who did not work during the week before the second interview, around 37 percent (corresponding to 5 percent of all respondents) stopped Figure 19. Distribution of respondents’ working status, by rural/urban working with the COVID-19 outbreak. The share of re- residence and pre-COVID-19 household per adult equivalent consumption spondents that were not working since the COVID-19 out- quintiles break was relatively higher in urban areas and in the two richest quintiles. Agriculture was the sector in which the majority of respond- ents worked prior to, as well as following the COVID-19 outbreak (Figure 20). Agriculture appears to have played an important role as a livelihood and risk management strategy following the outbreak. While the majority of the sectors show a contraction in terms of relative respondent participa- tion, the share of respondents employed in agricultural activi- ties has significantly increased since the start of the pandemic. During the first round in June, 44 percent of respondents reported that they or any other member of their households were operating a family business. 74 percent of these busi- nesses were still open at the time of the second interview in August, while 22 percent were closed temporarily and 4 per- Figure 20. Main industry of the respondents working before and after cent were closed permanently. the COVID-19 outbreak, (%) 6 COVID-19 IMPACT MONITORING Figure 21 shows that the mobility restrictions that were put in place in response to the COVID-19 outbreak were most frequently reported as the main reasons why the family busi- ness had to close temporarily (55%). Among respondents whose businesses were permanently closed, COVID-19 re- lated mobility restrictions (32%), fewer customers (31%), and the inability to get inputs (25%) were cited as the main rea- sons for business closures. All these reasons for closure can be attributed to COVID-19. Figure 21. Reasons for family business closures in Round 2, (%) Figure 22 shows the top five sectors where non-farm family businesses operated across the two survey rounds. In Round 1 in June, commerce was found to be the predomi- nant sector (68%), followed by service (10%), mining & man- ufacturing (9%), transport (6%) and agriculture sectors (5%). One month later and with the advance of the pandemic, we found a relative reduction in the share of businesses oper- ating in the two sectors entailing the most interaction with the public: commerce (dropping to 62%) and services (dropping to 8.5%). On the other hand, the share of family businesses in the mining & manufacturing sector increased its share to 13 percent and the share of family businesses in the transport sector increased by almost 2 percentage Figure 22. Family business sectors by survey round, (%) points, likely benefitting from the easing of the lockdown Note: Only sectors with shares higher than 5% are reported in the graph. measures. Respondents whose households were operating a business or held a temporarily closed family business in August inter- view were asked to report the main challenges that the busi- ness faced due to COVID-19. The results, reported in Figure 23, show that the difficulty of selling goods and services to customers was indicated by most of the respondents (58%), followed by the difficulty of raising money for the business (45%) and in receiving supplies and inputs (42%). Significant differences were found between rural and urban areas in reported difficulty in paying rent for the business location (27% in urban versus 8% in rural areas) and reported difficul- Figure 23. Challenges faced by family businesses due to coronavirus (%) ty in paying workers (10% in urban versus 4% in rural areas). One quarter of households operating a family business re- ported that they implemented or planned to implement changes in doing business due to COVID-19. Overall, the most frequently reported measures include the requirement for customers to maintain distance (63%); wearing masks (46%) and reducing the number of customers at the business location (38%). For most of the options, significant differ- ences were found across rural and urban areas. While re- quiring masks and reducing the number of customers, were measures that were more frequently reported in rural areas than urban areas, switching products or services was oc- curred more frequently in urban areas than rural ones. Figure 24. Changes implemented or planned in the way of conducting business due to coronavirus. 7 COVID-19 IMPACT MONITORING Figure 25 shows the changes in family business revenues in the month prior to the interview for the first and the second rounds of the survey. Only firms operating in both rounds have been reported. In June, the overwhelming majority of businesses in each sector reported partial or total losses. However, the share of businesses reporting no or lower rev- enues in the prior month showed signs of improvement in August . The highest reported share of businesses reporting no or lower revenues was in the service sector (56%), fol- lowed by those in commerce (44%) and mining and utilities (41%). The sector with the greatest share of businesses re- porting higher revenues in the prior month was the construc- tion, transport and professional services sector (42%). Figure 25. Change in family business revenues with respect to the prior month, by sector and survey round Income changes Figure 27. Changes in income by selected sources and across rounds, % Figure 26. Share of households receiving income from specific sources of households during last 12 month across two rounds, (%) By August, the share of households receiving income from views, respondents reported mostly losses or no earnings in each income source was higher than in June. The share in- all income generating activities since the onset of the pan- creased from 42 to 47 percent for non-farm family business demic. In the second round, most of the respondents report- income, from 71 to 76 percent in agricultural income, and 16 ed a partial or total loss of income between June and July- to 19 percent concerning assistance from family within the August tied to (i) assistance from family within the country country (Figure 26). Figure 27 shows the changes in income (65%) and (ii) non-farm family businesses (62%). by source and over time. During the first round of inter- AGRICULTURE Impact of COVID 19 on harvest activities Only 6 percent of respondents reported COVID-19 to have affected their crop harvesting decisions in July-August. Most of these respondents reside in the Northern and Eastern regions, where COVID-19 cases have been rising (in addition to Kampala). Among these respond- ents, the following changes in their harvesting activities were most common: delaying the harvest (mainly in the Central and Western regions); harvesting while undertaking COVID- 19 related safety measures (mainly in the Northern region), and failure to hire labor (main in the Northern and Western regions). The main COVID-19 safety measures taken during the harvest period were no handshakes (75%), keeping social distance (96%) and use of masks (54%). And the main reasons for not hiring labor were the high cost of labor (45%), low labor supply due to COVID-19 (25%), and fear of contracting Figure 28. Reported impacts of COVID-19 on harvest activities, by region COVID-19 (19%). 8 COVID-19 IMPACT MONITORING Impact of COVID-19 on livestock production Overall, the proportion of households whose live- the first round, the proportion of households reporting re- stock activities were affected by COVID-19 increased duced access to animal feed declined, while those reporting from 8 percent in round one to 12 percent in round reduced access to veterinary services and animal markets two. The highest increase in the proportion of households increased. While the easing of lockdown measures may have whose livestock activities were affected occurred in the Cen- increased access to animal feed, the increase in cases and tral region followed by the Northern and Eastern regions. In continued closure of markets may have affected access to Western Uganda, the proportion of households whose live- veterinary services and markets. Presenting the results by stock activities were affected by COVID-19 pandemic de- region, the main effects of the COVID-19 pandemic on live- clined (Figure 29). This can be attributed to the fact that stock production were reported to be reduced access animal Western Uganda, so far, appears to have low infection rates feeds and veterinary services in the Central region; reduced and easing of the lockdown measures could have also helped. access to markets and veterinary services in the Eastern re- Overall, the main effects of COVID-19 on livestock produc- gion; and reduced access to veterinary services, followed by tion activities were reported to be reduced access to animal reduced access to markets in the Northern and Western feed, veterinary services and markets. Overall, compared to regions (Figure 30). Figure 29. Share of livestock-keeping households whose livestock activi- ties were affected by COVID-19 by region across rounds, (%) Figure 30. Share of livestock-keeping households with reduced access to animal feeds, veterinary services and markets (conditional on their live- stock activities being impacted by COVID-19) by region across rounds, (%) Impact of COVID-19 on agricultural sales Figure 31 provides an overview of the share of farming regions could be linked to an increase in non-farm income households that needed to sell agricultural produce by region opportunities and dependency on perennial crops that are and over the two survey rounds. The decline in the propor- continuously harvested. Moreover, the share of farming tion of farming households that needed to sell agricultural households that could not sell their produce despite the produce was in Central and Western Uganda, while the pro- need to sell declined at the national level by 3 percentage portion of households that needed to sell produce increased points, and in Central and Western Uganda. Conversely, the in Eastern and Northern Uganda. The latter finding could be share of farming households that could not sell their produce associated with the fact that farmers harvested mostly sea- despite their need to sell increased in Eastern and Northern sonal crops, while the decline in the Western and Central Uganda (Figure 32). Figure 31. Proportion of farming households that needed to sell agricul- Figure 32. Proportion of farming households that needed to sell agricultur- tural produce across rounds, (%) al produce but could across rounds, (%) 9 COVID-19 IMPACT MONITORING Respondents in livestock-producing households were also spondents reporting a decline in egg sales was highest in asked about how the COVID-19 pandemic has impacted the Northern Uganda (87%), followed by those in Central (67%), sales of livestock products, namely milk, eggs, and meat. Western (23%) and Eastern (17%) regions. On the other About 43 percent of these respondents reported that the hand, 56 percent of respondents reported an increase in sales of milk and eggs had declined (Figure 33). The share of meat sales, while 14 percent reported no change and 30 per- respondents reporting a decline in milk sales was highest in cent reported a decline (Figure 32). Much of the decline in Northern Uganda (52%), followed by those in Eastern (49%), meat sales was in Northern Uganda (87%), central (35%), Western (42%) and Central (38%) regions. The share of re- Northern (29%) and Western (8%). Figure 33. Share of respondents who felt that COVID-19 had increased, decreased or not changed livestock product sales, (%) Figure 34. Share of respondents who felt that COVID-19 had increased, decreased or not changed livestock product prices, (%) Finally, respondents in livestock-producing households were Western (52%), Central (38%) and Eastern (23%) regions. In asked if the COVID-19 pandemic had increased or decreased the case of eggs, the share of respondents that reported a sales, and/or changed the prices for the aforementioned live- decline in prices was highest in Northern Uganda (87%), fol- stock products. Nearly half (49%) of the respondents report- lowed by Central (74%), Western (27%) and Eastern (26%) ed eggs prices to have declined, and the comparable figure regions. The majority of respondents (59%) reported an in- regarding the decline in milk prices was 45 percent (Figure crease in meat prices; 29 percent reported a decline; and 13 34). The share of respondents that reported a decline in milk percent reported no change due to COVID-19. prices was highest in Northern Uganda (74%), followed by Data Notes: the UGANDA COVID-19 Impact Survey First Round was imple- mented by the Uganda Bureau of Statistics (UBOS) in June 2020. This survey is part of a World Bank global effort to support countries in their data collection efforts to monitor the impact of COVID-19. A World Bank team from the Devel- opment Data Group and the Poverty and Equity Global Practice provided tech- nical support. This survey is the first of a planned 12 waves of the COVID-19 IM- PACT SURVEY of households in Uganda. 2,421 were successfully interview households from the 2019/20 Uganda National Panel Survey were contacted and 2259 households were fully interviewed. These same households will be contact- ed in subsequent waves of the COVID-19 IMPACT SURVEY. The data are repre- sentative at the regional and national level and survey weights were calculated to adjust for non-response and undercoverage. For further details on the data, visit http://www.worldbank.org/lsms-covid19 10