Choice of Rating Technology and Price Formation in Imperfect Credit Markets

Abstract

Accurate rating systems are of key importance for banks to price and manage their loan portfolios. In this paper we analyze the choice of the rating technology in an oligopolistic banking sector. In our model the rating system estimates the probabilities of default for the individual borrowers and therefore provides important input for the pricing of the loans. We model the technology choice and the pricing as a two-stage game. In the first stage banks choose the rating technology and in the second stage banks choose their pricing policy given the imperfect (oligopolistic) market using a risk-based pricing approach. The presented probabilistic framework and the modeling of the technology choice is novel in the banking literature and can provide important insights. In a comparative static analysis we study the implications for a market with two banks, which can employ two different rating systems (low or high accuracy). We find that in equilibrium the rating technology choice critically depends on the cost structure. If the additional costs for the high accuracy system are large both banks will have no incentive to adopt this technology. If the additional costs are low equilibrium behavior of banks results in the implementation of the accurate technology. In this case credit spreads unambiguously decrease and credit volume increases. The use of the more accurate technology, however, does not necessarily result in higher profits for the banks. Only if the costs are sufficiently small the equilibrium behavior results in a Pareto improvement. This has important implications for banking regulation which aims to provide incentives to use high accuracy rating systems (e.g. Basel II regulation).