Microfinance Games

This paper aims to demonstrate that the tendency towards free riding, seen in group-based lending mechanisms, also emerges in experimental settings that simulate microfinance transactions.The paper argues that:

• The inefficiencies of group lending are allayed when customers are allowed to form their own groups voluntarily;
• The endogenous formation of groups allows customers to sort into relatively homogenous pairings that limit the degree to which groups induce free riding and tax-safe behavior;
• Endogenous group formation reduces moral hazard.

The authors describe the methodology as follows:

• The simulations were implemented as a series of microfinance related experiments (or games) in Lima, Peru;
• They took place in a local market populated with small-scale entrepreneurs;
• The participants chose hypothetical risk projects, received loans and managed the risks of defaulting;
• 491 participants played an average of eleven games over the course of seven months;
• The approach involved modifying important parts of credit contracts in order to isolate specific features that drive successes and create tensions;
• The controlled setting allowed identification of participants' choices as "risky" or "safe", enabling precise testing of hypotheses.

The paper concludes that:

• Group-mechanisms reduce defaults;
• Allowing participants to form groups by themselves helps to mitigate moral hazard and establishes the positive features of group banking;
• The microfinance games played showed how strategic behavior and social concerns interact to yield effective contracts that work for both customers and lenders.