Implementing behavioral concepts into banking theory : the impact of loss acersion on collateralization

In standard bank theoretic models agents are assumed to be fully rational expected utility maximizers. This fact ignores the huge amount of evidence for anomalies in human behavior found by psychologists. In this paper we argue that the implementation of behavioral concepts into banking theory might increase the predictive power of the models. As an example we consider a loan market and discuss the impact of loss aversion on the degree of collateralization in equilibrium. The very well established concept loss aversion predicts entrepreneurs to pay much more attention to the potential loss of some of their initial wealth due to a collateralized loan than they would do as expected utility maximizers. This results in a higher effort choice which in turn increases the success probability of the loan financed project. Optimal levels of collateralization are derived for different degrees of loss aversion and the problem of private information about the degree of loss aversion is addressed. It is shown that in specific situations banks can offer self selecting pairs of contracts that costlessly eliminate the private information problem.