G²LM|LIC

Latest News

Improving Health and Safety in the Informal Sector: Evidence from a Randomized Trial in Bangladesh

Workplace accidents and hazards carry significant economic costs and harm to workers’ health and productivity. However, improving occupational health and safety (OHS) measures can enhance operational efficiency, benefiting worker well-being, firm productivity, and profitability. This is particularly important for small and medium-sized enterprises (SMEs) and the informal sector in low- and middle-income countries (LMICs).

To investigate if entrepreneurs’ lack of awareness about OHS acts as a barrier to firm growth and profitability in LMICs, this study conducted a randomized controlled trial in Bangladesh’s light engineering industry. The study partnered with BRAC, a prominent NGO, with two primary objectives: evaluating the impact of OHS training on firms’ business practices and performance, and assessing the complementarity between OHS training, business training, and access to financing. The results showed significant improvements in overall business practices, especially in safety and basic accounting. Firm output increased with the OHS intervention, while OHS training combined with business training led to higher employment rates and increased investment. The interventions also improved firms’ market linkages and owners’ management skills. This study contributes to the literature on enterprise training programs, highlighting the importance of OHS management skills for SME growth in LMICs. OHS training not only enhances practices, output, and productivity but also improves working conditions, benefiting worker health and well-being. While the interventions enhanced business practices and safety awareness, their impact on employment outcomes was modest.

In conclusion, this study demonstrates the effectiveness of enterprise training in improving workplace health and safety for informal firms in developing countries. OHS training positively influenced safety practices and accounting within firms. Additional business training and access to financing did not further improve OHS practices. Along with enhancing firm output and revenue, the interventions fostered better management skills among owners. Promoting OHS practices not only enhances productivity and profitability but also outweighs the costs in low-income settings with subpar working conditions. However, further research is necessary to expand our understanding of creating safer work environments across different industries and developing countries. Read the detailed working paper here.

Is Mobile Money Changing Rural Africa? Evidence from a Field Experiment

A new study has shed light on the potential of mobile money to accelerate urbanisation and structural change in Sub-Saharan Africa. The study, conducted in Mozambique, builds on a growing body of literature that highlights the role of mobile money in reducing transaction costs and improving financial inclusion in developing countries.

The goal of the study was to measure the impact of mobile money on migration, remittances, and welfare among rural households. Following the introduction of mobile money for the first time in rural villages of Mozambique using a randomized field experiment, the authors show that mobile money had a significant positive impact on out-migration and remittances among rural households by reducing transaction costs. The latter effect was particularly strong during times of negative aggregate and idiosyncratic shocks, suggesting that mobile money can help households cope with unexpected events.

Consistently with these results, the study also found that when rural households had access to mobile money and received money from migrants, they increased their welfare and consumption spending. They also suffered less from hunger and improved their health and education outcomes. Overall, these findings have important policy implications for governments and development organizations seeking to promote financial inclusion and reduce poverty in rural areas through mobile money services. By reducing transaction costs and improving risk-sharing, mobile money has the potential to improve welfare outcomes for millions of people living in remote areas across Sub-Saharan Africa. Read the detailed Working Paper here.

Girls’ Night In? Effects of the Kenyan COVID-19 Lockdown on Web Browsing

This study examines gender-based differences in the effects of the COVID-19 pandemic lockdowns on digital time use and activity in Kenya. The authors collected novel data online during the lockdown, combining survey responses from 316 Kenyans with their individual internet browser histories spanning the prior 90 days. This data set includes over 3.9 million unique website visits, providing an objective record of time use that is free from intentional misreporting or recall bias, and which was easier to collect than subjective time use data, which was greatly hindered by the lockdown. By combining browser and survey data, the authors are able to measure gender differences in the impact of the lockdown for different sub-groups. By identifying the key variables – gender and local broadband speed – that they used for their sub-population analyses and their estimates of differential impacts, the researchers were able to characterize the sample population along a wide range of attributes, which could be compared with those of the overall Kenyan population.

An analysis of the lockdown period revealed a significant increase in browser time for the overall sample, with an average of 41 minutes per person-day in levels and 0.15 points on a logarithmic scale. Further examination of the effects of the lockdown on women and those with high-speed broadband, relative to men and those with low-speed broadband, revealed no significant differences in total browser time between the two subgroups. However, there were significant changes in the content consumed. Women experienced a substantial increase in time spent on Netflix, with an average of over 35 minutes per day per person more than men. Additionally, people with faster broadband speed spent more time on LinkedIn, which could suggest future divergent labor market outcomes.

Throughout the March – June 2020 sample period, women and people in areas with high-speed internet spent substantially more time online than men and those in areas with low-speed internet, respectively. This finding is in contrast to the results of Miller et al. (2021), which were derived from a similarly gathered sample of PC users in India, where men spent significantly more time online. Although the women in both of these samples were slightly less likely to be employed, the Kenyan sample revealed that women were twice as likely to be single, which may partially explain the higher time online. They observed a significant increase in overall browser usage after the first Covid-19 lockdown. The detailed data revealed distinct disparities when examining specific domains, despite no discernible difference in overall usage by gender.

Kenya is a nation with high gender disparities, and women around the world have been disproportionately affected by the Covid-19 pandemic. This paper revealed that women engaged more in self-investment activities than men, post-lockdown. This is particularly true for the relatively younger, unmarried women in the sample, which is a positive sign for the advancement of women in Kenya in the future. Read the detailed working paper here.

A small selection of our Projects

Our newest publications

Building a Survey without a Recent Census

Low-income countries are also often data poor, lacking recent cen- suses or surveys to inform policy and programs. Without a recent national population census, countries do not know how to sam- ple for nationally-representative surveys. This brief discusses how countries can generate nationally-representative surveys without a recent census by relying on publicly available high resolution ge- ospatial data. The brief discusses how such geospatial data and sampling were used to create the Sudan Labor Market Panel Sur- vey (SLMPS) 2022. Sudan’s last population census was in 2008, a number of years ago, and before the country was divided. The SLMPS is Sudan’s first nationally representative household survey since 2014 and first labor force survey since 2011. These methods can be used in other contexts that lack a recent official census or for sampling between censuses in areas with substanial changes in population.

Improving Health and Safety in the Informal Sector: Evidence from a Randomized Trial in Bangladesh

Workers in small businesses in low- and middle-income countries are exposed to significant risks of occupational accidents and illnesses. A safe and healthy workplace could improve the productivity and sustainability of the business. In this paper, we conduct a randomized controlled trial in Bangladesh that provides informal firms with information on occupational health and safety (OHS) to improve their workplace practices. The intervention comprised two treatment arms: one focused solely on OHS training (the OHS arm), while the other oered business training and access to financing in addition to OHS training (the OHS+Biz arm). After two years, treated firms showed improvements in business practices, particularly those related to safety and a decent work environment. Moreover, both treatment arms experienced increased output and sales revenue. The OHS+Biz arm generally had no additional impact on firm outcomes compared to the OHS arm, suggesting that OHS information is the primary factor driving safer and healthier workplaces, which consequently can lead to better firm outcomes.

Browsers Don’t Lie.

The authors measure the digital impact of the Indian COVID-19 lockdown using an online survey coupled with consensually provided browser history records from over 1,000 individuals, spanning over 30 million website visits. Browser histories provide the authors with rich and objective historical time-use information that they collected without access to the extensive in-depth and in-person questioning, involving multiple time windows (e.g., past day/week/month), typically required for verification and validation in time-use surveys, while also avoiding the measurement problems of recall bias and misreporting in subjective reports.

Is Mobile Money Changing Rural Africa? Evidence from a Field Experiment

Rural areas in Sub-Saharan Africa are typically underserved by financial services. Mobile money brings a substantial reduction in the transaction costs of remittances. We follow the introduction of mobile money for the first time in rural villages of Mozambique using a randomized field experiment. We find that mobile money increased migration out of these villages, where we observe lower agricultural activity and investment. At the same time, remittances received and welfare of rural households increased, particularly when facing geo-referenced village-level floods and household-level idiosyncratic shocks. Our work suggests that mobile money can accelerate urbanization and structural change in Sub-Saharan Africa.

Encouraging Female Graduates to Enter the Labor Force. Evidence from a Role Model Intervention in Pakistan

In this study, we test if a low-cost, motivational nudge in the form of stories of female role-models can encourage female graduates from low-income households to increase labor force participation. We conducted a randomised control trial with 2500 female undergraduate students in female only public colleges in Lahore, Pakistan. We randomly selected half of the sample to watch videos of successful relatable female role models to encourage students to enter the labor force. The treated students had a significantly higher growth mindset immediately after watching the video. However, this did not translate into significantly higher job search effort or likelihood of working for the first 15 months after the intervention. Eighteen months after the intervention, at the onset of the first nationwide COVID-19 lockdown, treated students were 4.7 percentage points more likely to be working. This result was driven by respondents who belonged to households with lower income and parental education at baseline, possibly due to greater likelihood of a primary earner in their household losing job after the lockdown, and being more stressed about the loss of household income.

See all publications