Leveraging Community Insights to Identify High-Return Entrepreneurs in Low-Income Settings: CID Faculty Research Insights

How can community knowledge and incentive-based mechanisms improve the targeting of high-potential microentrepreneurs in low-income settings?

This paper, written by CID Faculty Affiliates Reshmaan Hussam, Natalia Rigol, and Benjamin N. Roth, investigates whether communities can accurately identify high-potential microentrepreneurs in low-income settings and how to elicit this information truthfully. Through a randomized experiment with 1,345 entrepreneurs in peri-urban India, the authors show that community members have valuable insight into which peers have high marginal returns to capital, but that their reports can be strategically biased. Mechanism design tools, such as incentive-compatible payments, can mitigate this distortion and improve the quality of information for targeting interventions. 

Key Findings:

• Community members know who has high returns: Entrepreneurs ranked in the top third by their peers had marginal monthly returns of 24–30%, compared to an average of 9.4%. Using community predictions to allocate cash grants would have tripled the overall return on investment.
• Community information adds value beyond observables: Even after controlling for demographic and business characteristics, peer reports predict marginal returns. Rankings based on observables alone identify high-return entrepreneurs less effectively than rankings that combine observables with peer information.
• Strategic misreporting undercuts accuracy: When participants knew their reports would affect grant allocations, predictive accuracy fell by 27–35%. Respondents favored themselves, friends, and family members.
• Mechanism design improves honesty: Small monetary incentives based on prediction accuracy significantly increased the truthfulness of peer reports, especially when reporting privately. Public reporting only improved accuracy in “no stakes” settings, where respondents were told that their reports would only be used for research purposes.
• Top-ranked entrepreneurs use grants more productively: They invested more in inventory and capital and worked more hours. They also demanded higher wages to exit entrepreneurship, suggesting greater entrepreneurial motivation. 

Impact and Relevance:

This study offers a solution to one of development’s central challenges: how to target interventions effectively when formal data are missing or insufficient. By demonstrating that peers possess highly accurate knowledge of each other’s entrepreneurial potential, information that even rich surveys fail to capture, the paper highlights the untapped value of community insights. More importantly, it shows how simple, incentive-compatible mechanisms from mechanism design can transform this social knowledge into a powerful tool for economic targeting, helping to identify which entrepreneurs can most productively use capital. 
 
In a world where millions receive cash transfers and microloans, improving the efficiency of capital allocation could enhance impact at scale. Governments, NGOs, and microfinance institutions typically allocate resources based on crude proxies—like assets or education—that poorly predict business success. This study suggests that community-based targeting, when properly designed, can outperform these approaches, and do so at very low cost. Moreover, by diagnosing and correcting for strategic misreporting, the authors offer a practical roadmap for real-world implementation without falling prey to favoritism or elite capture. 
 
More broadly, the research speaks to an important shift in development economics: from focusing on what people lack (such as capital, training, or formal records) to leveraging what they know. It aligns with a growing recognition that local actors often have deep contextual insight, and that the challenge lies in surfacing that insight credibly. By bridging behavioral economics, social learning, and mechanism design, the paper provides a scalable method for better policy: one that could improve targeting not just in entrepreneurship, but in areas like social protection, job placement, and agricultural extension.