MONITORING IMPACTS OF COVID-19 ROUND 12 AND OTHER SHOCKS FEB-MAR 2023 Publication Date Aziz Atamanov, Frédéric Cochinard, John Ilukor, Audrey Kemigisha, Talip Kilic, Andrew Mupere, and Giulia Ponzini BACKGROUND In June 2020, the Uganda Bureau of Statistics (UBOS), in collaboration with the World Bank, officially launched the Ugan- da High Frequency Phone Survey (UHFPS) to track the impacts of the COVID-19 pandemic on a regular basis. In June UGANDA 2022, the scope of the survey was expanded to monitor economic sentiments and the socioeconomic impact of other shocks such as the Russia-Ukraine war, Ebola outbreak and extreme weather events. In addition, the survey is being used to collect perceptions on different development policies and programs. 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 con- ducted in June 2020, interviewed 2,227 households and has been followed by eleven subsequent rounds ( Table 1). This brief presents findings from the most recent round (12th) of the UHFPS, conducted in February-March 2023. Table 1. Number of completed interviews by round R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 (June (July/ (Sep/ (Oct/ (Mar/ (Oct/ (June/ (Oct/ (Dec 22/Jan (Feb/Mar (Feb 21) (Aug 22) 20) Aug 20) Oct 20) Nov 20) Apr 21) Nov 21) July 22) Nov 22) 23) 23) Competed interviews 2227 2199 2147 2136 2122 2100 1950 1881 1738 1668 1666 1783 There have been many events affecting livelihoods in Uganda since 2020. The most important shocks in 2022 included inflationary impact from the Russian invasion of Ukraine and outbreak of Ebola Virus Disease (EVD) when residents of two districts (Mubende and Kasandra) were put on a 63-day lockdown which was lifted in December 2022. Population in Uganda also continued to experience extreme weather effects such as droughts, irregular rains, and floods. KEY FINDINGS • Respondents reported more optimistic views on current and future household financial wellbeing and country economic perfor- mance in February/March 2023 compared to October/November 2022. • There was a slight increase in employment and little changes in non-farm family businesses in February/March 2023 compared to December 2022-January 2023. • Food insecurity has improved further, but there were some signs of worsening access to some food products which may be relat- ed to lower farm income reported in February/March 2023. However, it is not clear whether it was just a short -term phenomenon or the start of a more long-term worsening of food insecurity. • Looking 12 months back, more than 85 percent of households experienced at least one shock with the majority facing an increase in prices of food products. • Every third household tried to borrow at least once in the last six months with poor households borrowing mostly from saving groups to finance food and health expenses. Rich households borrowed mostly from banks and Savings and Credit Cooperative Organizations to finance inputs/working capital for non-farm enterprises and education expenses. • Only five percent of households used any type of online retail services before March 2020 and the usage has not increased much during and post-COVID-19 being concentrated among households from the wealthiest quintile. In contrast, usage of mobile mon- ey during the last 30 days was widespread (75 percent of households), but with notably higher incidence in the Central and West- ern regions and among wealthier households. ECONOMIC INDICATORS Current and future economic sentiments Respondents reported more positive views about current and future socio-economic wellbeing in February/March 2023 compared to previous rounds. Figure 1 shows percentage of respondents who shared positive views on current/ future household and country economic wellbeing in February/March 2023 compared to previous rounds. There was a gradual increase in positive views for all indicators. Thus, only 28 percent of respondents viewed the future country eco- nomic situation to be better during next five years in June/July 2022, but this share increased to 41 percent in February/ March 2023. Only 13 percent of respondents felt their household financial wellbeing was better than 12 months ago in June/July 2020, but this share increased to 24 percent in February/March 2023. © 2021 International Bank for Reconstruction and Development/The World Bank MONITORING IMPACTS OF COVID-19 AND There was also a slightly more optimistic view on the chances of being hit by extreme weather events. Respondents were asked about the likelihood of being negatively hit by extreme weather events during the next 12 months. About 48 percent of respondents in February-March 2023 were certain to experience extreme weather events during the next 12 months and which were expected to negatively affect their financial wellbeing. This was slightly lower than what was reported in June/July (57 percent) and October/November 2022 (51 percent). Residents of the Eastern and Northern regions and those from the poorest pre-COVID-19 consumption quintile were more worried of experiencing weather shocks (Figure 2). Figure 2. Share of respondents who think extreme weather Figure 1. Positive economic sentiments across different events are likely to affect their household financial wellbeing rounds, % negatively during next 12 months, % Labor market and household income Figure 3. Working respondents across rounds, % Figure 4. Status of family business, % Note: Data is treated as cross-section. There was little change with employment and nonfarm family business in February/March 2023 compared to the pre- vious two months. Employment rates increased slightly from 75 percent in December 2022/January 2023 to 78 percent in February/March 2023 as some people return to work from being on holiday (Figure 3). The share of households with open non-farm family businesses was at about 45 percent – very close to what was observed in December 2022/January 2023 (Figure 4). 2 MONITORING IMPACTS OF COVID-19 AND OTHER Figure 5. Share of households where income either increased or re- Total household income remained relatively stable mained the same compared to previous period by income source, % with a slight decline in income from family farming. Thus, about 60 percent of households reported either higher or the same level of total income in February/ March 2023 compared to December 2022/January 2023 (Figure 5). This share remained the same as in the previous round. Among wage income earners, 81 percent had either higher or the same level of income. This share was low- er among households who received nonfarm family business income (68 percent) and income from farming (57 percent). Compared to the previous round, there was a slight decline in income obtained from family farming. Thus, in February/March 2023, 57 percent of households reported either higher or the same farm income. This was lower than 63 percent reported in December 2022/January 2023.1 Note: Only those households who received income from a particular source during the last 12 months responded to the question about income. Prices, access to essential products and food security Prices on most products have either not changed or declined since the most recent survey, with fertilizers being a nota- ble exception. Figure 6 shows changes in price indexes for the key products using May/June 2022 prices as a base (see products with a different baseline in the note under Figure 6). For most products, the peak of prices was observed in Octo- ber/November 2022 and then prices did not grow further or even declined. For example, the price index for maize flour fell from 1.4 in October/November 2022 to 1.2 in February/March 2023 accordingly. Prices on fertilizers were an exception as price index increased from 1.03 in December 22/January 23 to 1.23 in February/March 2023. This might be related to the start of the agricultural season. Figure 6. Price index for selected products across rounds (May/June 2022 is a base) Note: For brevity price index for August 2022 is omitted from the figure. Prices on fertilizers, gasoline and diesel have been collected since August 2022 only. Surprisingly, there was a deterioration in access to essential products despite relatively stable prices and little changes in economic indicators. For example, there was an increase in the share of households who needed but were not able to buy beef, sugar, rice, bread, and eggs in February/March 2023 (Figure 7). When asked about the reasons for not being able to buy these products, most respondents mentioned lack of income. Partially, this may be related to the fact that previous round was conducted in December 2022/January 2023 when households had higher than normal purchases due to festivity season. Another potential reason may be related to the slight deterioration of farm income as was shown in the previous paragraph. In addition, by February/March children are returning to schools and the priority is to pay fees. Further rounds of the phone survey will help to check whether it was a short-term phenomenon or a beginning of deteri- orating food security. 1 Farm income might be higher in December because of festivity season and associated with it sell of animals. Farmers were also selling stored harvest during that time. MONITORING IMPACTS OF COVID-19 AND Figure 7. Inability to access essential products when needed and inability to buy desired amounts during last 7 days across different round, % Figure 8. Food insecurity indicators across rounds, % Food insecurity indicators have improved slightly. The share of moderate food insecurity fell from 38 percent in December 2022 -January 2023 to 34 percent in February/March 2023, while the share of severe food insecurity fell from nine to seven percent during the same period (Figure 8). Food insecurity indicators were not affected by worsening access to some products such as bread, rice, sugar and beef reported earlier. This is possible as these goods are not staple and lack of access to them affect nutrition, but not necessarily general food secu- rity levels. Shocks and coping strategies More than 85 percent of households experienced at least one shock during the last 12 months with the majority facing an increase in prices of food products. Figure 9 shows the incidence of different shocks households experienced during the last 12 months. The increase in prices of food products usually consumed by households was foremost the most im- portant shock affecting 73 percent of households followed by weather shocks such as droughts (26 percent) and irregular rains (21 percent). Interestingly that increase in food prices and prices of gasoline were more likely to affect the richest households from the top pre-COVID-19 consumption quintile, while droughts, irregular rains and illness, injury or death of income earning members were more prevalent among households from the poorest quintiles. 4 MONITORING IMPACTS OF COVID-19 AND Figure 9. Incidence of different shocks on households during last 12 months, % Losing jobs or business were typically ranked the highest in terms of severity by respondents. Respondents were asked to rank three more significant shocks. Figure 10 shows the structure of each shock by its severity. Non-farm business closures had the largest share of shocks ranked as the most severe (67 percent), followed by job losses (61 percent) and increase in prices of major food items usually consumed by households (60 percent). Regarding coping strategies, households used multiple ways with relying on savings and reducing food consumption being used the most. Figure 10. Structure of shocks by severity, % Access to credit Every third household tried to borrow money during the last six months and 74 percent among them had an outstand- ing loan to be repaid. Figure 11 shows percentage of households which attempted to take a loan during last six months as reported by February/March 2023. About 35 percent of households tried to take the loan in February/March 2023 which was higher than 31 percent2 reported in June/July 2022. Rural residents were more likely to try to take the loan than urban ones (38 versus 31 percent respectively). Residents in the Northern and Western regions were more likely to apply for loans compared to residents of the Central and Eastern regions. Those households who tried to get the loan were also asked if they had outstanding loan to repay. The majority of loan applicants also had outstanding loans to repay – 74 per- cent at the national level, 86 percent in urban areas and 70 percent in rural areas. ‘No need for loan’ was the most popular reason for not applying for loans and it was mentioned the most among wealthier households. Figure 12 shows the main reasons for not applying for a loan. More than a third of households did not apply for a loan because they did not need it. Households from the poorest pre-COVID-19 consumption quintile were more likely to mention lack of collateral and too much trouble for what you get as main reasons for not applying for a loan. At the same time, households from the richest pre-COVID-19 consumption quintile were more likely to mention no need for taking a loan as a main reason. 2 Recall period in June/July 2022 was 12 months. MONITORING IMPACTS OF COVID-19 AND Figure 11. Attempts to get credit and outstanding loans, % Figure 12. Main reasons for not applying for loan, % Figure 13. Main source for attempted borrowing, % Households borrowed from different sources depending on their welfare status. Figure 13 shows main sources of borrowing across pre-COVID-19 consumption quintiles. Overall, households are more likely to borrow from sav- ings groups and village savings and loan associations. However, this differs across welfare status. Thus, house- holds from the top richest quintile were more likely to borrow from savings and credit cooperative organizations and commercial banks, while households from the poor- est quintile were more likely to borrow from different savings groups and village saving and loan associations. Poor households tried to borrow loans to finance food and health expenses, while the richest households tried to bor- row to get inputs/working capital for non-farm enterprises and education expenses. Households were asked about the purposes of each loan they obtained. Figure 14 shows the distribution of purposes at the national level and among the poorest and the richest quintiles based on pre-COVID-19 consumption per capita. Figure 14. Purposes of loans % Three of the most popular purposes included fi- nancing education expenses (23 percent), purchas- es of inputs/capital for non-farm enterprises (18 percent), and purchases of farm tools, inputs, land, and cattle (14 percent). The picture changes sub- stantially across groups of households with different wellbeing. Thus, financing food purchases and health expenses were the most important for the poorest quintile, while inputs for non-farm enter- prises and education expenses were the most im- portant for the richest quintile. ONLINE RETAIL SERVICES AND MOBILE MONEY Before COVID-19, usage of online retail services was rudimentary, mostly accessible to the richest households with food delivery and e-commerce being used the most. For the first time since the launch of the phone survey, respondents were asked about usage of online retail services. Only five percent of households used any type of online retail services before March 2020. 6 MONITORING IMPACTS OF COVID-19 AND About three percent used food delivery, about one percent e-commerce platforms and less than one percent used tele- health services, financial/insurance, or government services. The usage was distributed very unequally and almost fully concentrated among households from the top welfare distribution measured by pre-COVID-19 consumption per capita. Thus, only one percent of households from the bottom poorest quintile used any type of online retail services compared to 12 percent among households from the top richest quintile. Figure 15. Usage of online retail services before March 2020, % of households Usage of online retail services remained quite low after the pandemic as well. Respondents were asked in the 12th round about usage of food delivery services during the last seven days and usage of any online platform to purchase goods or services other than food during the last 30 days. The incidence of using these services was very low. Only 1.5 percent of households ordered food delivery and the same share of households used any online platform to purchase goods or ser- vices other than food. Consistent with pro-poor distribution before COVID-19, the usage of online retail services remained highly skewed towards the richest households based on pre-COVID-19 consumption. Figure 16. Usage of mobile money during last 30 days, % of households Respondents were very likely to use mobile money, especially in the Central and Western regions and among the wealthiest households. Figure 16 shows the percentage of households who used mobile money account during the last 30 days. Overall, about 75 percent of house- holds used mobile money account with this share being higher in urban than rural areas (82 versus 72 percent respectively). Usage of mobile money accounts was also much higher in more affluent Central and Westerns regions com- pared to poorer Eastern and Northern regions. Data Notes: the UGANDA High Frequency Phone Survey Tenth Round were implemented by the Uganda Bureau of Statistics (UBOS) in October/November 2022. 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 and other shocks. A World Bank team from the Development Data Group and the Poverty and Equity Global Practice provid- ed technical support. This survey is the twelfth of a planned 12 waves of the High Frequency Phone Survey of households in Uganda. 2,421 successfully interview households from the 2019/20 Uganda National Panel Survey were contacted and 1,738 households in the Nineth Round. In the Tenth Round, 1,668 households were successfully interviewed. In the Eleventh Round, 1,666 households were successfully interviewed. In this Round, 1,783 households were interviewed. The same households were and will be contacted in all subsequent waves of the High Frequency Phone Survey. The data are representative 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 https://www.worldbank.org/en/programs/lsms/brief/lsms-launches-high- frequency-phone-surveys