Research Info

Home /Optimal Design of ...
Title Optimal Design of Securitization in a Principal-Agent Relationship Based on Bayesian Inference for Moral Hazard
Type JournalPaper
Keywords MBS Securities, Moral Hazard, Optimal Multi-Task Compensation Scheme, Indian Buffet Process, Nonparametric Bayesian Inference.
Abstract In securitization process, originator, by selling the loans to risk-lover investor, could allocate mortgages’ risk with them. Therefore, he may haven’t any incentive to screen borrowers, resulting moral hazard problem. In this paper, it is attempted to apply insights from principal-agent model to this type of agency problem. Accordingly, investor, in order to mitigate asymmetric information, uses compensation scheme. Further, we consider a situation in which investor, by using Bayes rule, deals with inferring various dimensions of undertaken efforts and puts resulted joint posterior beliefs of data observations, related to credit positions of pooled loans and various dimensions of undertaken efforts, in the contract design problem. It is shown that the contract which is optimal for investor is a function of information content of observation and inferred information. This suggests that the information is of value. Therefore, using additional information prevents originator opportunisms, and originator more likely performs underwriting standards.
Researchers Elham Farzanegan (Third Researcher), mohsen ebrahimi (Second Researcher), Ezatollah Abbasian (First Researcher)