Foreign multinational enforcement of local workplace safety laws in Bangladesh

Fieldwork by:
84 garment factories
2017 - 2018
Governance Initiative
AEA RCT registration number:

In countries with poorly-enforced labor laws, workers in factories that supply goods to foreign firms often face hazardous workplace conditions and limited labor rights. A researcher partnered with the Alliance for Bangladesh Worker Safety, a consortium formed by US multinational companies, to evaluate the impact of privately enforcing local labor laws on garment factories in Bangladesh. Enforcing the legally-required creation and operation of workplace safety committees increased factories’ compliance with the labor law and some measures of factory safety, without reducing factory efficiency. These effects persisted in the long run and were stronger among factories with good managerial practices.

Policy issue

In low and middle-income countries, workers in domestic firms that supply goods to foreign multinationals often face poor workplace conditions and limited labor rights. Specifically, worker welfare and safety is a challenge: in 2019, work-related mortality accounted for 5 to 7 percent of all global deaths annually1 and at least one in nine workers experienced non-fatal occupational accidents2—a majority of these deaths and accidents took place in low- and middle-income countries. Even when workplace regulations do exist, governments may lack the capacity or the political will to enforce them.2 A 2018 global ranking of labor laws and enforcement found that 65 percent of developing countries systematically violate or provide no guarantee of rights to workers.3 Furthermore, firms may also lack the capacity to comply, perhaps due to poor management, like an inability to build relationships with workers, or prioritization of business considerations instead of safety concerns.4

In response to the poor workplace conditions and weak regulations in some domestic industries, many foreign multinational companies (MNCs) who source from LMICs choose to enforce local labor laws and increase compliance capacity by working with labor unions, supporting the ability for workers to voice safety concerns, and coordinating with local government. However, little is known about whether local suppliers comply with MNC-enforced labor regulations, if the enforcement of these regulations actually improves labor standards for local workers, and what effect enforcement has on wages, employment, and productivity. 

Context of the evaluation

In Bangladesh, domestic firms in supplier industries are tightly connected to foreign MNCs, which rely on the country’s low prices and large production capacity. This is particularly true for the apparel sector; Bangladesh is the second largest exporter of clothing in the world behind China, and the apparel sector has been a major driver of the country’s economic growth.

In the apparel sector, Bangladesh has been perceived as having weak legal protections for workers and poor, unsafe working conditions. A series of high-fatality industrial accidents in 2012-2013 —which observers connected to workers’ inability to form labor unions and advocate for safer conditions—led the government and multinational firms to take steps to improve workers’ safety and labor rights. In 2013, the Government of Bangladesh amended its labor law to improve workplace safety. A key provision of the amendment is the requirement that factories with at least fifty workers must create worker-manager safety committees of 6-12 members. These committees are meant to conduct activities that improve factory safety and worker well-being, including factory risk assessments, worker trainings, and regular disaster management drills.

Also in 2013, a group of 29 US-based multinational retailers (including Gap and Wal-Mart) formed the Alliance for Bangladesh Worker Safety (hereafter, the Alliance). Member companies required all factories in their supplier base to participate in workplace safety measures, such as building safety audits and worker training and empowerment programs. Overall, the Alliance covered between 600-700 garment factories and 1.21 million workers in the country.4 The Alliance formed a key part of the Government’s 2013 action plan to improve safety conditions, with MNC groups sharing the responsibility of overseeing safety in the factories they covered.

However, enforcement of and compliance with workplace safety regulations was low in factories under the government’s purview: as of mid-2017, only 210 of these 1,549 factories had formed safety committees (which were required by law). In addition, only 20 percent of the factories in the sample had complied with the legal requirement to create a safety committee as of mid-2016.

garment workers in a factory
Photo: J-PAL

Details of the intervention

In partnership with the Alliance for Bangladesh Worker Safety, the researcher conducted a randomized evaluation to test the impact of enforcing the 2013 amended labor regulations on domestic supplier firms in Bangladesh. Eighty-four domestic supplier factories under the purview of the Alliance were randomly assigned to either receive a safety committee enforcement program in 2017 (41 factories) or to serve as a comparison group (43 factories, all of which would receive the program after the conclusion of the evaluation).

The Alliance’s safety committee enforcement program focused on existing laws that required domestic factories to have safety committees. The program was designed to help factories establish their committees and ensured that existing committees were running correctly. The committees were to meet regularly and were responsible for arranging worker trainings and safety drills, conducting factory risk assessments, investigating accidents and occupational illness, and making recommendations on safety improvements to the employer, among other related responsibilities. The Alliance also trained committee members and monitored committees’ performance. Overall, the enforcement program involved 2.5 months of support to establish the committee and train workers on its roles and responsibilities, followed by an additional six months of monitoring to track committee activities.

The researcher measured the impact of the safety committee enforcement program on local factories’ compliance with safety committee laws and on committee effectiveness; worker safety, well-being, and satisfaction; firm productivity; employee retention and wages. The researcher collected data through factory visits, surveys on business performance, administrative data from the Alliance, and international export records from Bangladesh. Additionally, to measure management strength, the researcher counted the frequency of production-related meetings that managers held with workers. Data was collected five months after an initial visit established baseline data, which demonstrate the short-run effects. This second visit typically took place after three to four months of factories in the group which received the enforcement program being exposed to the program. The researcher also returned ten months after the initial visit to establish longer-run effects. This third visit took place when the enforcement program factories were no longer being monitored closely.

Results and policy lessons

The safety committee enforcement program increased factories’ compliance with the labor law and improved factory safety conditions. These effects were stronger in factories with better managerial practices. In factories with poorer managerial practices, the program decreased workers’ job satisfaction, possibly because it raised workers’ expectations beyond what the factory was able to achieve.

Compliance with safety committee laws: The safety committee enforcement program increased  compliance with Bangladeshi labor law in factories that were offered the program. Factories that were offered the program outperformed factories in the comparison group by over 0.2 standard deviations on an index measuring adherence to the legal requirements for safety committees, an effect which persisted 3-4 months after the end of the program. In particular, safety committees in factories that received the enforcement program were more likely to fulfill their responsibilities: 56 percent of factories that received the program had conducted at least one risk assessment toward the end of the program, compared with 15 percent of factories in the comparison group. These effects on compliance were stronger for factories that had better managerial practices. However, despite improvements in compliance, it is important to note that the program did not result in full compliance of Bangladeshi law. The researcher suggests that this could be explained by imperfect monitoring by the Alliance—for example, some factories had not conducted a risk assessment at the time of factory visits.

Committee effectiveness: In the longer-run, the program led to a 0.14 standard deviation increase over a comparison group score of 0.11 standard deviations on an index measuring the effectiveness of the committees at improving worker safety, increasing safety awareness, and engendering a culture of safety. These effects were strongest in terms of increasing the amount of factory safety spot checks. The researcher suggests that the overall improvements in worker health and safety may have been driven partly by increased information sharing between workers and managers, primarily via meetings with the safety committees. Committees in participating factories met more often and had longer and more substantive meetings than comparison group factories, as measured by an analysis of the content of the meeting notes. Furthermore, factories with better management practices saw the greatest improvements in committee effectiveness: compliance increased by 0.42 standard deviations in better-managed factories, compared to 0.12 in poorly-managed factories. This indicates that the quality of management in a given factory can have a substantial impact on improving the effectiveness of safety committees.

However, the program did not significantly affect factories’ safety culture or workers’ awareness of the safety committees. Some program and comparison factories already had high levels of committee awareness, perhaps due to their participation in a separate Alliance fire safety and worker training program. In addition, program factories did not outperform comparison factories in terms of committee effectiveness 3-4 months after the program. This was likely due to comparison factories improving the effectiveness of their own committees in an effort to prepare for future enforcement by the Alliance.  

Worker wellbeing and satisfaction: After around five months, the program decreased worker-reported job satisfaction and did not affect worker-reported mental health and well-being. The program reduced an index of worker-reported job satisfaction by 0.15 standard deviations relative to workers in the comparison group. This decrease in job satisfaction was concentrated in factories with poor managerial practices where the program did not improve compliance or safety.

Productivity, employment, and wages: The program did not meaningfully affect factories’ productivity, levels of employment, or wages relative to factories in the comparison group, suggesting that improved working conditions do not necessarily come at the expense of factory productivity.

These findings demonstrate that private enforcement of labor laws can improve compliance and contribute to achieving the law’s objectives, especially in well-managed domestic factories and when there are low levels of government enforcement. However, the different findings for factories with better versus poorer managerial practices highlight the importance for MNCs of taking into account suppliers’ ability and capacity to meet higher labor standards. Further research could provide clarity on how to lower the risk of highly dangerous workplaces, the impact of MNC enforcement on state capacity, and how to improve worker safety in other settings.

Informed by the results of this study, the Alliance incorporated learnings into the design of its safety committee enforcement program. Following direct dissemination of the results by the researcher to the Alliance and member companies, several firms mentioned that these findings influenced their policies and programming with suppliers. In addition, in part in response to the findings that the program effects were concentrated in better-managed factories, the Alliance launched a training program for mid-level managers on improving managerial capabilities for safety. The post-Alliance organization, Nirapon, has maintained these practices.

 ILO. 2019, “Safety and Health at the Heart of the Future of Work.” International Labor Organization.
  ILO, 2019, “World Employment Social Outlook Trends 2019.” International Labor Organization.
International Trade Union Confederation. 2018. “2018 ITUC Global Rights Index: The World’s Worst Countries for Workers.” International Trade Union Confederation, Brussels, Belgium.
Distelhort, Greg, Jens Hainmueller, and Richard M. Locke. 2017. “Does Lean Improve Labor Standards? Management and Social Performance in the Nike Supply Chain.” Management Science, 63(3): 707-728.