Title: Incentivizing Lending to Women Entrepreneurs: Evidence from Vietnam
Abstract: While there is a large literature on optimal incentive pay, there is relatively little empirical evidence from high-stakes, field experimental settings with higher-skill workers completing complex tasks. In collaboration with a leading Vietnamese commercial bank, we implement and evaluate the impact of two types of incentives to loan officers and their managers for the recruitment of women-owned or -led small- and medium-sized enterprises (WSMEs) as new borrowing clients. The average loan size in this segment is about $20,000 USD (about 5x GDP per capita), and the largest loans are around $100,000 USD. 50 bank branches employing 550 lending staff are randomly assigned to one of the following treatments for 6 consecutive 1-month periods: (1) a monthly multi-category ("inclusive") contest; (2) piece-rate incentives; or (3) control. The multi-category contest rewards not just overall top performing agents, but also top performers among new agents (“rookies”) and most improved performers. We find that, overall, both interventions cause an increase in WSME lending, and roughly co-move in terms of magnitude of impact. In terms of potential spillovers to sales of other products, there may be moderate negative impacts on non-WSME lending early in the treatment period, however over time there is a significant, positive spillover on non-WSME lending. These results have the potential to inform policies to promote (W)SME lending, and more broadly provide evidence from a high-stakes, real-world setting on optimal incentive pay.
Russell’s research focuses on topics in development with overlapping projects in environmental, agricultural, and behavioral economics. His current work is on development finance, especially small and medium enterprise finance, and digital payments networks.