Training, Productivity, and Upgrading: Evaluation of Female and Supervisor Training Programs in the Bangladesh Apparel Sector

Lead Research Organisation: University of Warwick
Department Name: Economics


The project studies the role of female managers in firm-level productivity in the ready-made garment (RMG) sector in Bangladesh. The study proposes to evaluate, through a Randomized Control Trial (RCT), the GIZ Female Supervisor Training Program, a scheme that selects and trains female workers to become line supervisors and middle managers. The rapidly growing literature that evaluates training programs on a variety of low-income groups through RCTs has focused on different target populations, e.g., ultra-poor programs in rural areas (see, e.g. Bandiera et al. (2011)) or the extensions of generic financial and accounting training to micro-entrepreneurs (see, e.g., Karlan and Valdivia (2011), Drexler et al. (2011) de Mel et al (2011)). To the best of our knowledge, the project would be the first randomized evaluation of a vocational training program for low income workers employed in larger factories. Furthermore, the project also relates to the recent RCT on the effects of management consultancy on firms productivity by Bloom et al. (2011). The program we propose to evaluate is developed by industry stakeholders and is targeted at the specific needs of the sector. There is a potentially large demand among the target population. Moreover, production worker training programs have been or are being developed by other stakeholders in the RMG sector in Bangladesh and other countries. The lessons learned from the evaluation of the GIZ program thus have broadly applicability.

The RMG sector in Bangladesh provides a uniquely appropriate laboratory to study how increases in productivity translate into job creation and poverty reduction. Most countries begun their process of industrialization by developing an apparel and textile sector. Exports in the apparel sector in low income countries, therefore, provide a unique opportunity for the creation of secure and relatively well paid jobs - particularly among women.

To compete successfully in the global apparel trade, however, firms must upgrade their capabilities and increase productivity. Through this process of upgrading low income countries can move beyond low wages as their exclusive source of competitive advantage and create jobs that pay higher incomes and satisfy more stringent social and environmental standards.

Rising labor costs in China, the world's largest exporter of RMG, have led to an increase in sourcing from other countries, including Bangladesh. However, a lack of skills, particularly at the intermediate managerial and line supervisor levels, poses a critical constraint on growth of the RMG sector in Bangladesh. More importantly, the managerial labour shortages constrain the ability to move up the value chain to produce higher quality, higher value-added products. Due in part to difficulties of retaining employees, firms do not provide formal management training to production workers. Stakeholders in the industry are increasingly focusing on this training as a key issue, establishing programs to provide vocational training program to workers in large factories. The proposed evaluation answers the following questions:
1. What is the impact of training and skills on i) the income, livelihoods and working conditions of production floor workers, particularly young women? and ii) factory productivity, organizational, labour and managerial practices as well as labour relations and conditions inside the factory?
2. What is the current demand and future sustainability of vocational training programs aimed at developing skills for production floor workers in larger factories?

Planned Impact

The project will also have extensive impacts at the industry, firm and individual level.

For the almost 600 women participating in the training program, there are direct impacts from the training obtained. Moreover, if the evaluation demonstrates that the program provides benefits beyond the program costs, we would expect the results to lead to adoption of the program by a larger number of firms. In that case, a much larger number of female production workers would benefit indirectly from the project. Moreover, the detailed analysis of the program will provide insights to those designing and running the program, which may result in program modifications and improvements.

If the training program is found to be effective, producers participating in the project will benefit from increases in productivity either through greater output or through a reduction in output with quality defects. In that event, producers not participating in the project will also benefit. There is the potential for a powerful demonstration effect thereby more factories and buyers learn about the benefit of the program.

Factories and buyers will also benefit from improved understanding of the factors that lead to higher productivity in the factory.

The project will also benefit the international aid community, including government aid agencies (e.g., DFID), intergovernmental agencies (e.g., ILO), and non-governmental organizations. The randomized evaluation of the training program will provide credible evidence of whether this program works. We believe there will be some applicability of the results to similar programs in other countries or industries. The project is designed to increase the relevance to other training programs by studying the specific channels through which the program works.

The effects on the participants will occur during the project and soon after. The broader effects from expansion of the training program will occur soon after the project is completed.
Description The ready-made garment sector is the largest employer in Bangladesh, and has been a key sector for expanding labor market opportunities for women. But while 80 percent of sewing machine operators are female, only around 5 percent of managers are female. We sought to understand constraints to promotion of women to entry-level management positions, line supervisors. During the project, we worked with the German bilateral aid agency (GIZ) to provide training to 286 female and 131 males sewing machine operators from 80 factories to be line supervisors.

The majority of factories with which we worked had at most one female supervisor at the start of the project. The training provided us the opportunity to observe reactions of existing managers and operators to female supervisors, with whom most have very limited experience.

We consider the following three findings to be the most significant arising from the project:

1) Prior to training, employees at all levels of the factory believe that males outperform females as supervisors. We asked employees to rate males and females on eight characteristics which workers believe are the most important traits of managers. Relative to females, males are seen as particularly skilled at technical / machine knowledge and at organizing resources.
2) We conducted extensive skills diagnostic of female and male trainees. Objective measures of skills are not well correlated with the beliefs of workers described above. For example, an extensive diagnostic measuring machine knowledge indicates that males and females are equally skilled.
3) When trainees return to the factory, machine operators working under the trainees initially indicate that the believe the males are better supervisors than the females. But after an exposure of approximately four months, the gender gap disappears.

Thus, perceptions themselves are an important constraint to the promotion of females as supervisors, even when these perceptions do not match reality.

There are two important questions on which the research is still on-going. The first is whether there are productivity differences between the male and female trainees working as supervisors. We have collected very detailed production data from the factories, and are presently processing and harmonizing those data so that an appropriate analysis can be conducted. The second question is whether participating factories changed their hiring and promotion practices after exposure to the project. We are presently contacting the factories to ask about promotion patterns in the period after their participation in the project.

Finally, we would highlight one other output from the project, not directly related to the training. We have collected what we believe is the most detailed data of productivity among manufacturers in a low-income country. The data will be valuable for understanding productivity in the sector more broadly - potentially not only in Bangladesh but in other countries, through related collection elsewhere. Once fully anonymised and harmonized, these data will be made available to the research community more broadly.
Exploitation Route We have produced a report which has been provided to foreign buyers working in the sector and is being delivered to all of the participating factories. We continue to work with many stakeholders in the sector, both to disseminate the results of this project and on follow-on projects. We are encouraged to report that two of the factories we have worked with have now scaled up the promotion of women to supervisory roles in a a large way. One has created an all-female floor with approximately 50 female supervisors and line chiefs and the other has trained over 90 women to be supervisors, promoting (to date) around 35 of them. We have just completed surveys of female-led production lines in these two factories with the aim of producing a document to circulate among stakeholders in the industry - factories, brands and others - documenting the experience of these factories. Both are among the leading factory groups in Bangladesh. We are planning a further workshop in Dhaka around the release of this document in June or July of this year.
Sectors Manufacturing, including Industrial Biotechology,Retail

Description The project has generated both direct and indirect impacts. The direct impacts are on the practices of the ready-made garment factories participating in the project. Among those factories, follow-up interviews reveal an Increased rate of promotion of females to supervisory role in the participant factories in the period beyond the project. Two factories indicated that they had formed all-female production lines, incorporating lessons of initial difficulties female supervisor trainees faced in managing male production workers. With regard to indirect effects, the clearest example is a follow-on project we have conducted with the International Labor Organization (ILO) and the International Finance Corporation (IFC) to develop a module of the Better Work program aimed at developing a scalable intervention leading to increased promotion of females to supervisory roles. This work builds directly on the lessons learned from the ERSC-funded project. Understanding the three main constraints - selection of candidates, building confidence levels of the female candidates, and integrating trainees into the lines - allows us to focus the training to produce a more scalable product. We have provided IFC with a report from the project and they are using that report to develop a plan to scale up the intervention we developed both in Bangladesh and perhaps in one or more other countries as well. The project PI has also participated in a meeting with the Secretary of State for International Development, Justine Greening, to help prepare her for her role in the UN High Level Panel on Women's Economic Empowerment. The discussion in that meeting, and follow-on work reviewing output from a recent conference at the Urban Institute, drew very much on the lessons learned from the ESRC-funded project.
First Year Of Impact 2016
Sector Manufacturing, including Industrial Biotechology
Impact Types Cultural,Economic

Description CSR
Amount £20,000 (GBP)
Organisation Sainsbury's 
Sector Private
Country United Kingdom
Start 11/2013 
End 02/2015
Description Improving Work Advancement POpportunities for Female RMG Workers
Amount $363,147 (USD)
Organisation International Finance Corporation 
Start 07/2016 
End 12/2017
Description Dhaka Dissemination workshop 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Industry/Business
Results and Impact Dissemination of project results to participating factories, industry associations, bilateral and mulitlateral aid agencies, and policy-makers in Dhaka, Bangladesh.
Year(s) Of Engagement Activity 2014
Description Seeds of Change in the Garment Industry: Workshop in Dhaka, Bangladesh 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Industry/Business
Results and Impact We organized a workshop in Dhaka to present the results of this project along with other research being conducted on the garment sector in Bangladesh. The PI presented results from this study, with discussion by an executive from one of the factories that has dramatically expanded the hiring of female supervisors as a result of lessons learned during this project. The audience included managers from other factories, representatives of international brands, academics, policy makers, representatives of NGOs and students. We are planning a similar activity in the summer of 2018.
Year(s) Of Engagement Activity 2017