What’s the Market-Reach of COVID-19? Evidence from Ethiopia’s Ready-Made Garment Industry
EASST fellow Eyoual Demeke (World Bank) shares evidence from high-frequency phone surveys with a random selection of 4,326 female Ready-Made Garment workers employed before the start of the COVID-19 pandemic. The working paper is co-authored by Christian Johannes Meyer (University of Oxford), Morgan Hardy (New York University Abu Dhabi), Gisella Kagy (Vassar College), and Marc Witte (IZA Institute of Labor Economics). This is part of a series of posts highlighting COVID-19 related research conducted by CEGA affiliated faculty, PhD students, and fellows.
In our globalized world, pandemics transmit impacts through markets. An example of this comes from the ready-made garment (RMG) industry, where the sharp economic downturn due to the COVID-19 pandemic has threatened the livelihood of low-wage production workers. This is especially the case in countries such as Ethiopia, where over the past decade the government has pursued a strategy of labor-intensive industrialization through export-oriented light manufacturing, invested massively in enabling infrastructure, and set up ten special economic zones (“industrial parks”) across the country.
In this blog post, we share the results of a phone survey of a random sample of 3,896 female RMG workers employed before the start of the coronavirus pandemic in Hawassa Industrial Park (HIP), Ethiopia’s flagship economic zone. In January 2020, before the crisis hit, almost 28,000 workers were employed by 20 different firms in HIP. At full capacity, HIP could employ 60,000 workers. Our data collection occurs at a time when COVID-19 is spreading in Ethiopia — though not locally in Hawassa — and in key export markets of the RMG industry, resulting in factories suspending production due to collapsing global demand. This allows us to observe employment changes, coping strategies, and the welfare of workers indirectly impacted by the pandemic.
Our data suggests that even without local COVID-19 spread, there have been significant changes in employment, migration away from urban areas to rural areas for women who are no longer working, and high levels of food insecurity.
These findings indicate the imperative for policy makers to put in place insurance and social protection policies to help firms retain workers, even in places not directly affected by the pandemic. Wage subsidies to firms may represent one potential instrument to protect jobs. Such a wage subsidy can be flexibly combined with a cash transfer to workers should they lose their job. In contexts like Ethiopia’s industrial parks, with a great degree of formalization and availability of reliable administrative data, targeting and implementing such support schemes is highly feasible. Using the contact details available from personnel data, such targeting could even be implemented after workers have returned to rural origin communities.
Job loss: The employment of female workers in Ethiopia’s garment industry has changed dramatically due to a sharp drop in global demand during the COVID-19 pandemic. In our sample, 41 percent of respondents employed in January 2020 were put on leave or terminated by the time of our survey a few months later (see Figure 1).
Challenges in Finding New Employment: The vast majority of respondents not currently working in HIP have not been able to find another job of (91 percent). This is despite the fact that 41 percent of those who lost their job have since tried to find new employment or start a business. We ask respondents to list the most important barriers to finding employment among the following options: lack of money, not having transportation, safety concerns related to health, safety concerns related to violence, and lack of accommodation. The most common barriers are lack of money, safety health concerns, and lack of accommodation. In addition, the barriers are similar among those in Hawassa compared to those not in Hawassa, except for the barrier about safety due to health concerns. Significantly more individuals not in Hawassa report this as a barrier (42 percent of those not in Hawassa compared to only 25 percent of those in Hawassa).
Migration: 28 percent of respondents appear to have migrated away from Hawassa. The vast majority relocated to a rural area. A large share of workers in HIP and in our sample are rural-urban migrants who had originally moved to Hawassa for work. The lack of employment, health risks posed by their small and shared living arrangements, and weak local social safety nets may have prompted these workers to return to their origin communities.
Many who have left the city desire to return if possible, with only 45 percent reporting to be in their desired location. The lack of income is the most common barrier to movement for those not in their desired location and a lack of accommodation is second (see Figure 3).
Food insecurity: Levels of food insecurity are high; and rates are higher for those currently still in the city where garment industry jobs are located. 52 percent of respondents are worried about not having enough food in the seven days prior. There are lower levels of food insecurity among those that have left Hawassa, which could be explained by less food expenditure when staying with friends or family.
