Institutional Design as a Commitment Device in Credit Markets with Asymmetric Information: Experimental Evidence

The aim of this experiment is to test the role of institutional design in credit markets as a commitment device against renegotiation: when there is asymmetric information does a lower degree of centralization enhance efficiency? Does decentralization alleviate the adverse selection problem in credit markets? We run a large‐scale computerized experiment involving 12 different data sets and 3 different uncertainty scenarios on a sample of 120 subjects. The results obtained confirm the superiority of a decentralized institutional framework: the number of poor projects undertaken in a decentralized market was significantly smaller than the number of poor projects undertaken in centralized markets in all the scenarios. This experimental evidence shows that the institutional design is crucial in seeking financial discipline and therefore can shed some light on the debate on ‘Anglo‐Saxon’ versus ‘German–Japanese’ credit practices. (J.E.L.: C90, D82, G21, L10).