The Failure of Models that Predict Failure: Distance, Incentives and Defaults

46 Pages Posted: 10 Nov 2008 Last revised: 15 Aug 2010

Uday Rajan

Stephen M. Ross School of Business, University of Michigan

Amit Seru

Stanford University

Vikrant Vig

London Business School

Date Written: August 1, 2010

Abstract

Statistical default models, widely used to assess default risk, are subject to a Lucas critique. We demonstrate this phenomenon using data on securitized subprime mortgages issued in the period 1997--2006. As the level of securitization increases, lenders have an incentive to originate loans that rate high based on characteristics that are reported to investors, even if other unreported variables imply a lower borrower quality. Consistent with this behavior, we find that over time lenders set interest rates only on the basis of variables that are reported to investors, ignoring other credit-relevant information. The change in lender behavior alters the data generating process by transforming the mapping from observables to loan defaults. To illustrate this effect, we show that a statistical default model estimated in a low securitization period breaks down in a high securitization period in a systematic manner: it underpredicts defaults among borrowers for whom soft information is more valuable. Regulations that rely on such models to assess default risk may therefore be undermined by the actions of market participants.

Keywords: Securitization, screening, incentives, subprime, defaults, mortgages, disintermediation, models, lucas critique, soft information, hard information, failure, predictability

JEL Classification: G21

Suggested Citation

Rajan, Uday and Seru, Amit and Vig, Vikrant, The Failure of Models that Predict Failure: Distance, Incentives and Defaults (August 1, 2010). Chicago GSB Research Paper No. 08-19; Ross School of Business Paper No. 1122; EFA 2009 Bergen Meetings Paper. Available at SSRN: https://ssrn.com/abstract=1296982 or http://dx.doi.org/10.2139/ssrn.1296982

Uday Rajan

Stephen M. Ross School of Business, University of Michigan ( email )

701 Tappan Street
Ann Arbor, MI 48109
United States
734-764-2310 (Phone)

HOME PAGE: http://webuser.bus.umich.edu/urajan

Amit Seru (Contact Author)

Stanford University ( email )

650 Knight Management
Stanford, CA 94305
United States

Vikrant Vig

London Business School ( email )

Sussex Place
Regent's Park
London, London NW1 4SA
United Kingdom

Paper statistics

Downloads
5,200
Rank
1,017
Abstract Views
20,079