The Economic Effects of COVID-19 on Ready-made Garment Factories in Bangladesh

C-19 Note
Published on 23 July 2021
COVID-19 poses a considerable challenge to the future of Bangladesh’s apparel sector and to Bangladesh’s economy more broadly. In light of COVID-19’s continued economic and health impacts, this research provides systematic evidence on COVID-19’s economic effects from a survey with a representative random sample of Bangladesh Garment Manufacturers and Exporters Association (BGMEA) member factories. This evidence can be used to inform an appropriate policy response and to facilitate planning for recovery from the pandemic.

Our survey finds that order cancellations and renegotiations by buyers were widespread during the pandemic’s early stages. We find strong evidence that individual buyers’ responses to the pandemic, though, varied substantially. On average, BGMEA member factories experienced a decline in revenue of almost 17.4% in 2020 compared to 2019. Revenue losses varied widely across factories: Those that were smaller, older, less well-managed, and selling to more different buyers experienced larger revenue losses in 2020. Turning to employment, the impact appears to be moderate: Employment declined by 7.4% in the second half of 2020 compared to pre-COVID-19 levels, although it started to recover toward the end of 2020. In terms of capital investment, COVID-19 caused some factories to delay or to decrease planned investment, including in increasing automation.   Finally, in terms of workers’ health, as of late 2020/early 2021, on average, 3.6% of factories’ workers had experienced symptoms of or tested positive for COVID-19. This is likely a lower bound on the true rates. The remainder of this brief presents our survey methodology, details these and our other main findings, and discusses policy implications.

Data and Methodology

In collaboration with the International Growth Centre (IGC)-Bangladesh, the BRAC Institute for Governance and Development (BIGD), BRAC University, and with financial support from the Private Enterprise Development in Low-Income Countries Initiative (PEDL), we conducted a phone-based survey of BGMEA member factories in order to generate evidence on how COVID-19 is affecting their businesses. The BGMEA provided a list of its member factories and supported the research team to contact factories to invite them to participate.

In order to obtain a representative view of COVID-19’s impacts, we conducted the survey with a randomly selected sample of BGMEA member factories; 241 factories participated. Appendix I provides more information on the sample selection and the response rates. The survey was conducted from late November 2020-March 2021. Table 1 presents basic summary statistics for respondent factories.

Table 1: Summary Statistics


On average, BGMEA members sales revenues declined by 17.4% in 2020 compared to 2019. Figure 1 shows the distribution of revenue changes for sample factories; most factories experienced revenue declines in 2020 compared to 2019. Despite the wide variation in losses, on average, the reported revenue loss mirrors the aggregate 16.9% decline in RMG export earnings that Bangladesh experienced in calendar year 2020 (USD 27.47 bn) compared to 2019 (USD33.07 bn). Our survey indicates that revenue losses were larger for smaller factories, older factories, less well-managed factories, and factories selling to more different buyers, controlling for several other of factories’ characteristics.

Figure 1: Percent change in sample factories’ sales revenue, 2019-2020

On the whole, the pandemic did not lead to widespread, sustained factory shutdowns.  Excluding the government-declared public holiday, 25% of factories report closing for at least one day due to the pandemic. Closures peaked during the second quarter of 2020, when factories were closed for an average of 4 days outside of the public holiday that was in place due to COVID-19 from late March to late April 2020. Five percent of factories report closing since the beginning of the pandemic and still being closed at the time of the survey; among these, managers at 64% of these factories thought that it was likely that their factory would reopen. Smaller factories and factories selling to more different buyers were more likely to remain closed, controlling for several other of factories’ characteristics.  

Factories’ total employment fell by 7.4% in the second half of 2020 compared to pre-COVID-19 levels (Figure 2), controlling for factories’ fixed characteristics. While a meaningful decline, this statistic masks heterogeneity in employment declines across factories: Factories employing relatively higher shares of women (above the median share in the survey data) were more negatively affected by the pandemic. For these factories, employment fell by 12.9%, while for factories employing relatively lower shares of women, employment fell by 3.5%.  Importantly, factories did not change their mix of female and male workers, which means that factories were not treating their female and male employees differently. This pattern of impacts suggests that they may be driven by the demand shock to factories, as opposed to a differential shock to women and men’s labour supply. Consistent with this possibility, factories producing woven goods employed higher shares of women and also experienced greater revenue declines during the pandemic. The fact that job losses were larger for factories with relatively higher shares of female workers suggests that female garment workers may have been more adversely impacted by job loss. 

Figure 2: Average Total Employment, Sample Factories

16.3% of factories report temporarily or permanently laying off workers during the first through third quarters of 2020. The incidence of permanent layoffs appears to have peaked in April-June 2020 but remained low, with 3.5% of workers being permanently laid-off with pay and 1.3% of workers being permanently laid-off without pay. Factories also report taking other measures to curb labour costs in response to COVID-19: About 60% report reducing overtime hours, 43% report reducing regular hours, 36% report delaying payment, and 29% report cutting wages during the second quarter of 2020. Many continued working hours reductions during the third quarter of 2020, although wage delays and cuts fell substantially compared to the second quarter (14.4% and 6.8%, respectively). Factories also report strong uptake of the stimulus support for wage payments, with nearly 72% availing of these benefits.

COVID-19 caused factories to delay or to decrease planned investment in increasing automation. Prior to COVID-19, 58.2% of factories reported plans to make investments to increase automation in 2020. Among these factories, 38% report delaying and a further 26% report decreasing investments due to COVID-19. While in the medium-term, this may increase their reliance on labour, in the longer-term, it may delay or erode Bangladesh’s ability to move into higher value-added segments of the global apparel market, translating into fewer and potentially lower quality jobs for workers.  

COVID-19 caused smaller factories to sell land, buildings, or equipment, but not larger factories. About 10% of smaller factories report selling physical capital due to the effects of COVID-19. In contrast, less than 1% of larger factories report selling physical capital due to the effects of COVID-19. We define small factories as those with below the median pre-COVID number of employees in the survey data, which is 1050 or fewer employees.

Factories’ main COVID-19-related challenge from the first through third quarters of 2020 was loss in demand (Table 2). The share of respondents experiencing a decline in demand peaked at 60.1% of respondents during the second quarter of 2020 but remained high, at 44.9%, during the third quarter. Respondents also report facing many other types of challenges during this time, including supply chain disruptions, production shutdowns, difficulties with absenteeism, and challenges to maintain COVID-19 health and safety measures.

Table 2: Reported Challenges faced by firms (Percentage of factories experiencing challenge, by Quarter)

Factories report substantial increases in the share of their output that became stuck in Bangladesh. Surplus finished apparel items are commonly known as stock lots in the apparel sector. Stock lots can occur due to a variety of reasons, with the main ones being rejection due to quality issues, cancellation of shipment, and delays in shipment, either due to buyers’ being unwilling to accept the goods or delays with export processing.   As of April 2020, about 17% of factories’ output was stuck as stock lots in Bangladesh.

Figure 3: Average Share of Output Stuck as Stock lots in Bangladesh

51% of factories report experiencing a shortage of raw materials due to delayed or cancelled import shipments. 25% of factories report experiencing a shortage of raw materials due to delayed or cancelled material shipments from within Bangladesh. The most common sourcing country for raw materials is China; 81% of respondents report that China is one of their main sourcing destinations; of these, 50% report experiencing shipment delays or cancellations. Interestingly, 37% of respondents report switching from importing to sourcing locally for one or more of their input materials due to COVID-19.    

On average, factories report that 3.6% of their workforces were suspected to or tested positive for COVID-19 (as of late 2020-early 2021). Controlling for factories’ size and other characteristics, factories with better management practices and factories selling to more different buyers reported higher COVID-19 rates. While it may be surprising that factories with better management practices report higher COVID-19 rates, it may be the case that these factories were more likely to measure and to track COVID-19 symptoms and viral tests among their workforces, which would lead to higher rates of detection.

Figure 4: Pct Worker Had COVID based on April 2020 employee count

Factories’ take-up of the government’s stimulus package support program was high: 87.4% of respondents report receiving some type of support (Table 3). By far, the most common type of support that factories report receiving was the loan program to cover employees’ wages, which 71.5% of factories participated in. Factories report some challenges to accessing the funds, in particular, slow arrival speeds, burdensome administrative systems to access the support, and workers’ not having mobile financial services (MFS) to receive transfers.

Table 3: Support from Government and Banks

Buyers matter for explaining COVID-19’s impacts on factories. We find strong evidence that buyers’ responses to the pandemic, in terms of whether they cancelled or renegotiated orders, varied substantially across buyers. In Appendix II, we present a table that summarises the responses of 50 largest buyers identified in survey, based on supplier factories’ 2019 revenues and a criterion of buying from two or more factories in the survey sample. We find that order cancellations and re-negotiations are an outcome determined by factories’ specific buyers, as opposed to their destination markets, which are not strong predictors. We asked factories about their largest buyer, second largest buyer, newest buyer, and oldest buyer’s responses to the pandemic; 17.2% of these relationships experienced at least one order cancellation and 46.5% of them experienced at least one order renegotiation due to the pandemic. Among buyers that cancelled or renegotiated orders, only 18.5% of buyers agreed to pay for any of the raw materials already purchased and only 13.2% of buyers agreed to pay for any labour already used.

Looking ahead to late 2021, factories expected business with 55.4% of their buyers to increase, 26.0% of buyers to remain the same, and 18.5% of buyers to decrease. Factories reported that the pandemic had no effect on their perception of the value of their supplying relationship with 36.9% of their buyers, positively affected their perception of their relationship with 37.0% of their buyers, and negatively affected the perception of their relationship with 30.6% of their buyers (Figure 5). 30.6% of factories also identified buyers that they were now interested in supplying to due to the buyer’s response to the pandemic.

Figure 5: Impact of buyers' response to pandemic on factories’ perceived value of supplying relationship

Discussion and Policy Implications

Aggregate statistics, while informative, mask substantial variation in COVID-19’s economic effects on BGMEA members. COVID-19 caused greater revenue losses for smaller factories, older factories, less well-managed factories, and factories selling to more different buyers, controlling for several other of factories’ characteristics. In terms of employment, factories employing relatively higher shares of women, which are also more likely to produce woven apparel, experienced significantly larger declines. In terms of capital divestment, smaller factories were more likely to sell physical capital, while larger factories were not forced to divest.

These heterogeneous patterns of effects suggest substantial scope for targeting of government-provided support to factories and to workers that are differentially more affected. Targeting needs to consider both allocative efficiency and workers’ short-term welfare. By allocative efficiency, we mean that the government should aim to provide resources to factories that have the ability to recover competitiveness with the government’s helping hand. In other words, it is natural for struggling firms to exit, and the government does not want to prop up inefficient firms. But it does want to support struggling firms that have the potential to become competitive again. By workers’ short-term welfare, we mean that the government should consider that, at least in the short-term, due to the industry-wide downturn, it would likely be difficult for laid-off workers to find employment. This provides a social rationale for providing laid off workers with temporary income support.

Buyers matter in determining the economic impacts on BGMEA members. Different buyers adopted very different tactics in terms of whether they cancelled or renegotiated existing orders; these buyer-level differences were much more important in explaining factories’ outcomes than the destination country of the exports. This suggests that Bangladesh may benefit not only from trying to attract demand by appealing to destinations such as the US or the European Union, but also appealing to the types of buyers that develop stronger relationships with their suppliers.

Factories that sell to fewer buyers, which may be those factories that have deeper and longer-term relationships with their buyers, had higher revenues in 2020, controlling for their pre-COVID revenues and several other of their characteristics. The fact that factories selling to fewer buyers also had lower COVID-19 infection rates among their workers suggests an important direction for research to understand the source of this correlation. For example, are factories that employ better practices overall the ones that are able to develop deeper, longer-term relationships with their buyers? Or do buyers that value better working conditions also source more intensively from their suppliers? 

The survey also reveals some of COVID-19’s subtler effects: The pandemic has delayed or deterred factories’ planned capital investments, deterred some firms from reopening existing factories, and disrupted supply chains. These disruptions suggest some of the possible channels through which COVID-19 may have negative effects on the sector in the longer-term even if demand were to fully recover and the public health impacts to cease. In the longer-term, it will also be important to understand whether the sector’s increased debt burden due to, for example, taking on stimulus loans to pay employees’ wages, hampers its recovery.

On more positive notes, the fact that 37% of factories report beginning to source inputs in Bangladesh in response to the pandemic suggests that there may be upsides to the development of domestic markets for raw materials. This possibility warrants further investigation. Further, about 30.6% of factories report that they have identified one or more buyers they aim to develop business with due to the buyer’s response to the pandemic. Coupled with factories’ efforts to fill capacity during low demand periods, these activities may draw new buyers to Bangladesh’s apparel sector. 

The very-high take-up of the government’s stimulus package support program, despite some challenges to access it, suggests that the support program may have played an important role in assisting firms to tide over a few very difficult months. As all BGMEA members had access to the support, however, it is impossible for us to determine its causal impact on their outcomes.

Appendix I and II can be found in the PDF document.

This note is based on research conducted as a part of PEDL ERG 7849.


Laura Boudreau

Columbia University

Farria Naeem

International Growth Centre