Use of referrals in the Blue Collar Employment Economy in Delhi

Familiar strangers: evidence from referral-based hiring experiments in India.

Project Introduction

Across developing countries, labour markets are often characterized by significant search and matching frictions, which contribute to high turnover in entry-level jobs, while large segments of the workforce remain excluded from stable or better-paying employment. Firms frequently rely on referral-based hiring to reduce search frictions, lower recruitment costs, and improve match quality, but referrals also carry the risk of perpetuating existing inequalities — by disproportionately circulating opportunities within specific social networks, referral systems can systematically exclude historically underrepresented groups, reinforcing exclusion and limiting upward mobility. This study examines whether reallocating referral opportunities toward disadvantaged groups can generate improvements in both firm performance and worker welfare, exploring the role of caste in shaping referral dynamics within the Indian labour market.

Partners

Researchers

Project Methodology

A field experiment was conducted in a manufacturing firm in India, in which the probability of referral opportunities being extended to lower-caste incumbent workers was deliberately increased. After an initial session established a baseline for quality standards and worker sentiment toward teammates, each team received an additional worker assigned through a dominant-caste referral, a marginalised-caste referral, or a marginalised-caste worker unconnected to the existing social network. An added layer of teamwork on a rope-making task increased the need for coordination, allowing researchers to measure worker turnover, representation, and productivity gains across the three arms.

Impact

The findings reveal substantial impacts of reallocating referral opportunities toward historically disadvantaged groups, including measurable productivity gains, with overall output improving. Contrary to a large body of literature suggesting that greater representation within teams may undermine cohesion and reduce performance, the study found no significant evidence of adverse effects on team cohesion. A complementary lab-in-field experiment helped clarify the underlying mechanism, suggesting that referrals not only expand access for underrepresented groups but also mitigate potential costs by embedding new hires within existing social networks, sustaining both cohesion and firm productivity.

Documentation

This project is currently in progress, and will be updated with findings as results arise.

Similar Projects