Social Norms and Firm Productivity: Evidence from Bangladeshi Knitwear Factories

The study aims to test whether social norms, in particular norms around communication, hinder information transmission in firms in developing countries. The study also tests whether gender norms hinder information transmission inside firms. An incentivized survey experiment is conducted to examine workers’ willingness to participate in information sharing sessions with workers of the same versus the opposite sex (versus an anonymous condition). We also implement a field experiment to study how gender affects information diffusion across workers and quantify the downstream effects on trained and untrained workers’ productivity. Finally, we explore methods to alleviate the impact of gender norms on information transmission in the factory.

Researchers: Laura Boudreau; Sakib Mahmood; Oren Reshef; Afsana Binte Khaleque

Partners: International Growth Centre (IGC); Weiss Family Program Fund; The Abdul Latif Jameel Poverty Action Lab

Timeline: 15 January 2022–15 January 2023

Status: Ongoing

Method: Quantitative

Contact: Sakib Mahmood;


Economic research documents large productivity dispersion across firms in industrialized and developing countries; these gaps are mirrored by large productivity dispersion within countries and industries, particularly in developing countries (Hsieh and Klenow, 2009; Syverson, 2011). Existing economic literature largely focuses on the effects of managers’ and workers’ suboptimal behaviours on firm productivity (Atkin et al, 2016; Bloom et al, 2013; Macchiavello et al., 2015(a)). In contrast, this study asks why certain suboptimal behaviours arise. We hypothesize that specific social norms or cultural informal institutions hinder the adoption of productivity-enhancing behaviours. We focus on the extent to which social stratification (Saunders, 2006) constrains low-status groups’ participation in decision-making processes inside the firm. 


The study aims to (1) provide systematic evidence on the role of social norms in the adaptation of productivity-enhancing practises in developing countries; (2) explore the power dynamics between high social status individuals and low social status individuals and their social stratification in terms of receiving productivity improving information; and (3) identify the influence of gender norms containing the inclusive and equitable workplace for women.

This study is relevant to SDG 5 (Gender Equality), particularly to achieving gender equality and empower all women and girls.


The research design consists of three major parts. 

Part 1 is the survey of target participants—knitting and linking workers—who will take part in an information sharing session. We identify three dimensions of social (in)compatibility: gender (men are high status, women are low status) (Burke and Major, 2014); religion (Muslims are high status, non-Muslims are low status) (Tausch and Hewstone, 2008); and age (older workers are high status, younger workers are low status) (Crystal and Shea 1990; O’Rand 1996). Socially incompatible matches occur when the social inferior must provide productivity-enhancing information to the social superior, a situation that breaks normative behaviour.  

Part 2 consists of skills training. The research team will hire an industrial engineer (IE) to identify inefficient practises in the factory and to provide worker-level skill training to address these weaknesses. 

Part 3 is observational data analysis. The research team will use administrative data collected by the factories to measure the effect on workers’ productivity, assuming stratified, cluster randomized control trials with 10 factories.

Findings and Recommendations

Study ongoing.