Read working paper
INSEAD Working Paper 2013/78/EFE/FIN
Because many markets for services involve heterogeneous principals and agents, traditional models focusing on one principal and one agent need to be extended to allow heterogeneity. We propose that this extension can be modeled as a dual matching-contracting problem of endogenously repeated trust games. We show that the equilibrium can be achieved by organizational policies that attract an optimal mix of principals with either high or low tolerance to (noisy) signals of misconduct. A critical mass of alert principals suffice to discipline agents by exiting upon misconduct signals, while the rest need to be inert to both achieve optimal matching and avoid the deadweight loss of over-punishment. To test the empirical implications of our model, we take advantage of a mutual fund scandal in 2003 that involved some funds receiving a correct signal of misconduct, while others received an incorrect signal. Our analysis shows that matching according to mutual fund management policies explains differences in fund outflows among investigated fund families. These findings have implications for the theory of incentives in markets and the impact of organizational structure on market efficiency.