ROUND 4 COVID-19 IMPACT MONITORING NOV 2020 Publication Date BACKGROUND This brief presents findings of the fourth round of the Malawi High-Frequency Phone Survey on COVID-19 (HFPS COVID-19) conducted in September 2020. In May 2020, the National Statistical Office (NSO), with support from the MALAWI World Bank, launched the HFPS COVID-19; a monthly survey of a nationally representative sample of households previously interviewed as part of the Malawi Integrated Household Panel Survey to monitor the economic impact of the pandem- ic and other shocks. The first round (baseline) of the survey was conducted in May/June 2020, the second round was con- ducted between July 2-16, 2020, the third round between August 12-27, 2020, and the fourth round between September 14- 29, 2020. KNOWLEDGE, BEHAVIOR AND CONCERNS OF COVID-19 TRANSMISSION In the fourth round of the survey, respondents reported a stayed consistent across Rounds but in Round 4 a significant significantly lower level of concern over COVID-19. 84% of number of respondents reduced from reporting it was a sub- respondents are very worried or somewhat worried stantial thread to only a moderate threat. about themselves or their immediate family member(s) be- In Round 4, 6% of respondents that experienced any coming seriously ill from COVID-19 in comparison to of the common COVID-19 symptoms called the 93% in Rounds 2 and 3, and 94% in Round 1. About 95% of Government Provided Toll Free Number or any oth- respondents consider this crisis as a substantial or moder- er line designated by the Ministry, this is an increase from ate threat to their household’s finance and this has 2% reported in Round 3. In Round 4, there is a drastic drop in the prevalence of re- In the 7 days leading to the interview, although only 47% of spondents washing hands with soap frequently from 75% respondents wore a mask all of the time as shown in in Round 3 to around 49% in Round 4. Figure 4, this is a decrease in mask wearing between August and September. However, there is also a decrease in re- However, more respondents are practicing the avoid- spondents reporting that they never wear a mask, 12% in ance of handshakes/physical greetings, 74% in Round 3 September in comparison to 15% in August. Overall, the to 79% in Round 4. This is promising especially given cultural trends suggest that Malawians are now putting on masks as a practices especially in rural areas. Over time, the least adopt- preventive measure against COVID-19 but are not strict in ed preventive measure against COVID-19 remains avoiding regard to how often they wear the masks whenever they go of groups of people which has consistently stayed below out in public as shown in Figure 4. 50% across rounds as shown in Figure 3. 1 COVID-19 IMPACT MONITORING EDUCATION Schools across Malawi reopened for the new school year in ported that they would not, and 6% remained unsure as September, however the government only allowed candidate shown in Figure 5. classes to resume in-person studies while the other classes In Round 4 (September), 43% of respondents indicated that were scheduled to reopen in October . In Round 3 (August), their children had returned to school, 56% indicated that the survey asked respondents about their intention for their their children would return in another phase of the children to return to school. 85% of respondents indicat- school opening. Less than 1% indicated that their children ed that their children would return to school, 9% re- would not return to school or were uncertain. EMPLOYMENT Employment decreased during the first months of While there are more people working, there is evi- the pandemic, but the proportion of working popula- dence that people have changed jobs which could tion is similar to the pre-COVID levels. Overall, the suggest volatility in the job market. Between August proportion of respondents working has continued to rise and September about 19 % respondents that were working and is slightly higher than the proportion of people who changed jobs, between July and August about 23% respond- were working in mid-March at the time of the COVID-19 ents changed jobs, between July and May/June around 16% of outbreak. Over 79% of respondents reported that they were workers changed jobs. working in September in comparison to approximately 74% The proportions of wage workers reporting the ability to in August, 70% in July and 69% in May/June. In September, work as usual at either their place of employment or re- while the share of the working population has increased, it motely has increased. In September, 92% of wage workers has stagnated in urban areas where 75% were working last were able to work as usual in the last week, in comparison week compared to 82% before the pandemic. The propor- with the 80% reported in between July and May. tion of the working population has been steadily improving across the last 4 rounds. The increase in the rural areas could be attributed to the crop farming activities in the dry season that rural households might be engaged in. Note: In the figure it is assumed that all those who were working at baseline were also working before mid-March. 2 COVID-19 IMPACT MONITORING Even though the overall share of respondents who are working is improving, there continues to be some evidence of people moving in and out of work. About 46% of respondents have been working continuously since the COVID-19 outbreak. 4% of respondents started working between August and September, 9% started working between July and August and 4% just returned to work. 29% have not returned to work since the outbreak, 1% were not working even before the outbreak and 6% have been in and out of work. As shown in Figure 10, the service sector experienced the largest recovery in the share of respondents returning to work; 15% to 42% in August and September respectively. Agriculture remains the biggest sector Malawian households are employed in and while the trends show an overall drop, it is in line with the pre COVID-19 period where not many households engage in crop farming during the dry season. INCOME CHANGES AND LIVELIHOOD SOURCES Most of the households reported to experience an and 18% from wage employment of household mem- income drop. 69% of respondents reported that their bers. However, the share of households experiencing a household’s total income decreased over the last one decline in income was highest for non-farm family busi- month, and this decrease was evident across the three main ness activities at 67% of households, compared to 60% sources of income (agriculture, non-farm enterprises, and for agriculture (household farming, livestock or fishing), wage employment). The sources of income differ across and 48% for wage employment. households, 42% of reported income is from farming, live- as shown in Figure 12. stock or fishing, 23% from non-farm family business 3 COVID-19 IMPACT MONITORING CREDIT Between the mid-march (the outbreak of COVID- 19) and August, 40% of all respondents needed to access credit; of these, 55% were able to get it and 45% of could not get it. This could suggest that much of the population remained vulnerable since they could not fulfill their obligations. Respondents in urban areas were just as likely as respondents in rural areas to fail to acquire credit if they needed it. In Round 3, the survey asked respondents if they needed to still had outstanding loans taken before the COVID-19 borrow money or take a loan since the beginning of the Outbreak. COVID-19 outbreak in mid-March. 40% of households Of the 84% that did not take a new loan since August, responded affirmatively. about 23% still have outstanding loans taken before the In Round 4, 16% of households reported that they COVID-19 outbreak as shown in Figure 14. took a new loan since Round 3 and, of these, about 34% Poorer households were just as likely as richer households to still have outstanding loans taken before the start of the pandemic. The same relationship is observed for newer loans taken since August. This shows that the pandemic has impacted the finances (and livelihoods) of households across the whole income distribution, and not only the most vulnerable. 4 COVID-19 IMPACT MONITORING AGRICULTURE Overall the proportion of households that engaged in crop compared to livestock. The proportion of livestock keeping farming increased from 13 percent in 2019 Dimba season to households increased in South and North but declined in 34 percent in 2020 Dimba season. The highest increase in Central Malawi. Agricultural households were also the proportion of households engaged in farming was in asked if COVID 19 lead to better, normal or less than North(24%), Central(21%) and South(21%) as shown in Fig- normal sales of crop and livestock. Most of the crop and ure 16. Also, the proportion of non agricultural households livestock households report COVID 19 to have led to the during the Dimba season declined from 44 percent in 2019 less than normal sales in crops and livestock. Livestock sales Dimba season to 35 percent in 2020 Dimba season. The de- were affected most compared to crop sales. Over 50% and cline in non agricultural households was highest in South 44% for livestock and crops, respectively as shown in figure (12), Central(8%), and North (4%). The proportion of live- 17. stock keeping households overall increased by one percent suggesting that COVID-19 affected cropping farming most Data Notes: Malawi High-Frequency Phone Survey COVID-19 (HFPS COVID-19) is implemented by the National Statistical Office (NSO) on a monthly basis during the period of May 2020 and June 2021. The survey is part of a World Bank-supported global effort to support countries in their data collection efforts to monitor the impacts of COVID-19. The financing for data collection and technical assistance in support of the Malawi HFPS COVID-19 is provided by the United States Agency for International Development (USAID) and the World Bank. The technical assistance to the Mala- wi HFPS COVID-19 is provided by a World Bank team composed of staff from the Development Data Group —Living Standards Measurement Study (LSMS) program and the Poverty and Equity Global Practice. In Round 1, 2,337 households that had been previously interviewed during the 2019 round of the Integrated Household Panel Survey (IHPS) were contacted, and 1,729 households were successfully interviewed, with the goal of re- interviewing them in the subsequent monthly HFPS COVID-19 rounds. 1,646 of these households were successfully interviewed in Round 2, 1,624 in Round 3 and 1,618 in Round 4. The 2019 IHPS data are representative at the national- and urban/rural-levels and phone survey weights were calculat- ed (i) to counteract selection bias associated with not being able to call IHPS households without phone numbers, and (ii) to mitigate against non- response bias associated with not being able to interview all target IHPS households with phone numbers. For further details on the data, please visit https://www.worldbank.org/lsms-covid19 and email enquiries@statistics.gov.mw. 5