Banks' Incentives and the Quality of Internal Risk Models

41 Pages Posted: 11 Dec 2014

See all articles by Matthew C. Plosser

Matthew C. Plosser

Federal Reserve Banks - Federal Reserve Bank of New York

João A. C. Santos

Federal Reserve Bank of New York

Date Written: December 1, 2014

Abstract

This paper investigates the incentives for banks to bias their internally generated risk estimates. We are able to estimate bank biases at the credit-level by comparing bank generated risk estimates within loan syndicates. The biases are positively correlated with measures of regulatory capital, even in the presence of bank fixed effects, consistent with an effort by low-capital banks to improve regulatory ratios. At the portfolio level, the difference in borrower probability of default is as large as 100bps, which can improve the typical loan portfolio’s Tier 1 capital ratio by as much as 33%. Congruent with a regulatory motive, the sensitivity to capital is greater for larger, riskier, and more opaque credits. In addition, we find that low-capital banks’ risk estimates have less explanatory power than those of high-capital banks with regard to the prices set on loans, indicating that low-capital banks not only have downward biased risk estimates but that they also incorporate less information.

Keywords: banks, regulation, capital, internal ratings, Basel II

JEL Classification: G21, G28

Suggested Citation

Plosser, Matthew C. and Santos, João A. C., Banks' Incentives and the Quality of Internal Risk Models (December 1, 2014). Available at SSRN: https://ssrn.com/abstract=2535856 or http://dx.doi.org/10.2139/ssrn.2535856

Matthew C. Plosser (Contact Author)

Federal Reserve Banks - Federal Reserve Bank of New York ( email )

33 Liberty Street
New York, NY 10045
United States

João A. C. Santos

Federal Reserve Bank of New York ( email )

33 Liberty Street
New York, NY 10045
United States
212-720-5583 (Phone)
212-720-8363 (Fax)

HOME PAGE: HTTP://WWW.NEWYORKFED.ORG/RMAGHOME/ECONOMIST/SANTOS/CONTACT.HTML

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
99
Abstract Views
542
rank
148,047
PlumX Metrics