Monitoring the Impact of COVID-19 in Myanmar Agricultural Input Retailers – synopsis of results from five survey rounds through late July 2020

Key findings • Disruptions to input retailers from COVID-19 related restrictions were most severe early in the crisis, stemming primarily from lockdowns and travel restrictions. By late July, these restrictions were affecting less than 3 percent of the retailers interviewed. • The most severe disruptions to input retail shops were sharply lower demand (reported by 82 percent of retailers) and collecting repayment from credit lent out (84 percent). These disruptions in late July still impacted 32 and 38 percent, respectively. • Half had lower fertilizer sales between May and July compared to the same time in 2019. • Adoption of safety practices to mitigate the spread of COVID-19 waned from a peak in mid-May of 90 percent of retailers adopting to 59 percent in late July. • 40 percent of input retailers were more pessimistic about the future of their business in late July than they were in March. • 35 percent plan to reduce the credit they will offer out to farmers in coming seasons.


Introduction
Agricultural input retailers play a key role in Myanmar's agri-food system by supplying farmers with fertilizer, seed, pesticides, and other inputs necessary for successful harvests. Because farm-level input use is an important driver of yields for all major food crops, shocks from the COVID-19 crisis to the input retail sector have major implications for rural household welfare as well as food security.
In this policy note, we present results and analysis on the effects of the COVID-19 crisis on agricultural input retailers from a five-round telephone panel survey of between 150 and 200 retailers in Shan, Kachin, Bago, Ayeyarwady, Sagaing, and Mandalay that was implemented every two weeks from mid-May to late July 2020. 1 The objective of this survey was to provide data and insights to the Ministry of Agriculture, Livestock, and Irrigation (MOALI) and agricultural sector stakeholders so that they better understand the nature of COVID-19 related shocks to Myanmar's agricultural input retailers. Previous policy notes 2 mostly focused on the survey rounds individually, tracking the effects of the COVID-19 crisis on agricultural input retailers as they were happening. In this note, we take a more comprehensive approach by looking back over all five survey rounds to understand how the effects of the COVID-19 crisis evolved over time. In particular, this note presents results from May 2020 through July 2020 across the five survey rounds on (i) disruptions caused by the COVID-19 crisis, (ii) responses to these disruptions, (iii) sales of fertilizer, maize seed, vegetable seed, and pesticides, and (iv) input retailers' employees and hired labor.

Effects of the COVID-19 crisis on input retailers
To understand the COVID-19 effects on input retailers, we asked a series of questions on the different types of disruptions that shops experienced. The first survey round in mid-May captured disruptions from the start of the crisis until the time of the initial survey round. Each subsequent round asked about disruptions in the two-weeks prior to the interview.  Disruptions were most severe early in the crisis -for each category of disruption, the first survey round in mid-May had the highest share of input retailers experiencing the disruption (Figure 1). This is not surprising as there were extensive lockdowns and transportation restrictions put in place for several weeks starting in April. As time progressed, the restrictions were lifted and fewer input retailers reported disruptions. By late July, the direct effects were minimal -mostly related to movement restrictions. Only 10 percent of shops reported in late July supply-chain disruptions in receiving inputs -down from 44 percent at the start of the crisis -while only 3 percent of shops reported a mandatory closure -down from 28 percent -and just 2 percent reported difficulties with employees coming to work -down from 14 percent.
Early in the crisis, demand shocks were the main disruption, experienced by 64 percent of input retailers. While the adverse effects of the COVID-19 crisis lessened over time, about one-third of retailers were still reporting demand disruptions in late July. Demand shocks were one of four disruptions experienced by more than half of our sample of input retailers since the start of the crisis, with 82 percent of shops reporting a demand disruption in at least one survey round ( Figure 2). This highlights how pervasive the crisis shocks were, though some of the effects were more transitory and decreased dramatically in later rounds. Particularly shocks to credit may be more persistent, however.

Figure 2. Respondents reporting specific disruptions in any survey round, percentage share
Source: Input Retailers Phone Survey, late July 2020.
The most common disruption experienced at any time during the COVID-19 crisis was difficulties in collecting repayments from farmers on credit offered to them. Credit repayment difficulties also showed relatively small decreases over survey rounds -38 percent of shops reported the disruption in late July, down only 17 percentage points from 55 percent at the start of the crisis. Input retailers have also experienced challenges obtaining new loans and repaying their creditors, with 15 percent and 14 percent of shops reporting such disruptions in the late July survey, respectively, out of 54 and 45 percent of shops reporting the disruptions in any survey round, respectively.

Input retailer responses to COVID-19 shocks
In each survey, input retailers were asked a series of questions to understand how they have responded to these COVID-19 disruptions. As the disruptions have lessened over time, so too have business responses (Figure 3). Many shops closed or reduced their operating hours due to the initial lockdown periods, as captured in the first survey, but by late July only 3 percent of shops were closed for at least one week due to the crisis and only 1 percent continued to reduce their operating hours. Actions to improve customer and employee safety have been the most common response, but they too have diminished over time, falling from 90 percent of shops adopting some safety practices in mid-May to 59 percent in late July. Shops likely perceived lower health risks from the coronavirus as there were few confirmed cases in June and July, particularly in rural areas. Nonetheless, in late July, 47 percent and 42 percent of shops still were having employees regularly wash hands and wear face coverings, respectively ( Figure 4). However, only 10 percent of shops were maintaining safe distances between customers and employees. Adoption of other safety practices was almost nonexistent. In the late July survey, the second most common COVID-19 response in business practices was changes in the provision of credit (18 percent), almost exclusively in the form of a decrease in the number of farmers to whom retailers offer credit -there is very little variation in the interest rates or terms of the offer. This has direct implications for farmers as they may not be able to access inputs easily. The third most common response in the late July round was adapting business operations, mostly by adopting cellphone technologies to buy or sell inputs over the phone ( Figure 5). Eighteen percent of shops offered delivery services in mid-May, but the share fell over time to 5 percent by late July. Mobile banking service adoption peaked at 5 percent in mid-May but fell to about 1 percent in late July.

Figure 5. Adaptations to business services by input retailers, percentage share reporting by survey round
Source: Input Retailers Phone Surveys, 2020.
Input retailers in our sample do not employ many workers. In the mid-May survey, retailers reported an average of 2 permanent employees and just 1 temporary worker outside of the household. Furthermore, relatively few firms responded to the COVID-19 crisis by hiring fewer workers. There was an apparent lag between the initial lockdowns and labor changes, with the peak of shops changing their number of workers occurring in the second survey round in early June with 14 percent of retailers reporting having done so in that survey round. More detailed questions on labor decisions revealed that most changes were a decrease in temporary workers. We did not observe significant changes in the number of permanent employees.

Changes in input sales over time
To track input sales through the COVID-19 crisis, we asked retailers about their sales volumes in the two weeks prior to each interview for four common inputs -fertilizer, maize seed, vegetable seed, and pesticides. Unsurprisingly, the two most common inputs sold are fertilizer and pesticides, which are applied to crops in all survey areas (Table 1). These two inputs are the most important products for most of the retailers in our sample. Maize seed is also important, though more concentrated in Shan state during the monsoon season. The share of retailers selling fertilizer increased in each survey round and average sales were higher in July than in June. Ayeyarwady and the Dry Zone regions drove July sales of fertilizer as these areas experienced relatively late monsoon rains. In Shan state, fertilizer sales were higher in June and decreased in July as most maize was well established by July. This is also evidenced by the maize seed sales data -less than 10 percent of shops sold maize seed in July.
To get a sense of how input sales changed in 2020 as compared to 2019, we asked retailers in each survey whether their input sales were higher, lower, or about the same as the same two-week period in 2019. In each survey and for each input, the share of retailers reporting lower sales far outweighs the share reporting higher sales (Table 1). There are many factors contributing to the changes, but this presents a troubling picture if we project forward to the monsoon harvests. Declines in fertilizer and pesticide sales imply lower yields and lower crop incomes in 2020 at a time when households are continuing to deal with lost incomes from other sources due to the COVID-19 crisis. The drop in maize and vegetable seed sales suggest a shift to other crops, an increase in the use of saved seed, or a shift at the extensive margin with less crop area planted. The latter two trends would further contribute to farmers realizing lower agricultural incomes in 2020.

Looking forward
To understand future changes in the agricultural input retail sector, we asked input retailers a series of questions about their expectations for coming seasons. The first striking result is that 40 percent of input retailers surveyed were more pessimistic about the prospects for their business in late July than they were prior to the pandemic. Just 11 percent are more optimistic. However, despite the pessimism and decreased sales and revenues in 2020, all but one input retailer in our sample plans to continue operations in the post-monsoon season, though many retailers expect changes in the sector. Most shops expect retail competition to increase in the next year -56 percent expect an increase in competition compared to just 9 percent expecting a decrease ( Figure 6). This is a continuation of trends in the three years prior to the COVID-19 pandemic. Over that time, 79 percent of retailers perceived an increase in input sales competition. Just 7 percent perceived a competition decrease. This is an encouraging trend for farmers in terms of their maintaining access to agricultural inputs, since input retailers do not expect much if any consolidation or market exit in their sector due to the COVID-19 crisis. A troubling change expected in future seasons is that 35 percent of the input retailers surveyed plan to change their offering of inputs on credit to farmers, mostly by offering credit to fewer customers. On a more positive note, 29 percent of interviewed shops plan to make capital investments in their business in the next year. Of those shops planning such investments, 90 percent plan to make improvements to their shop or facilities and 12 percent plan to invest in vehicles for transport. While this is encouraging, there may be shops that are unable to make their planned investments due to large revenue declines. Furthermore, 90 percent of the shops plan to finance those investments through their own savings rather than through loans, perhaps reflecting input retailers having limited access to and engagement with financial institutions.

Recommended actions
Through this analysis of data obtained through five rounds of a telephone survey of agricultural input retailers in Shan, Kachin, Bago, Ayeyarwady, Sagaing, and Mandalay covering mid-May through July 2020, we arrive at six recommended actions for the Government of Myanmar as it plans policy responses to any second wave of COVID-19 cases in Myanmar.
• Government should promote the sustained adoption of safety practices by businesses.
Facebook, particularly the page of the Ministry of Health and Sports, was the primary source of information on COVID-19 safety for input retailers early in the pandemic. This source of information could be leveraged again to promote safety practices, including sustained adoption of face coverings and social distancing.
• Government should plan future lockdowns to minimize disruptions to the agricultural input retail sector, particularly by allowing input retail businesses to remain open, conditional on the adoption of safety practices.
• Government should anticipate that many farm households will realize lower crop incomes than normal at the time of monsoon harvest -stemming from a decrease in input sales -and provide cash or lending support to smallholder farmers (Action 2.1.7 of the COVID 19 Economic Relief Plan (CERP) of the government of Myanmar). 3 • In addition to lower incomes, farmers may have a harder time procuring inputs on credit as input retailers reduce their credit offerings. The Government of Myanmar should promote post-monsoon input purchases through input subsidy programs where feasible.
• Government should support input retailers financially by removing or deferring collection of business taxes and fees (CERP 2.1.3) and by the expanding working capital loan support program (CERP 2.1.1) to include agricultural input retailers.
• Many input retailers adopted cellphone technologies in response to the COVID-19 crisis. Use of mobile payments increased in mid-May but adoption slowed in later survey rounds.