The impact of credit risk mispricing on mortgage lending during the subprime boom

Summary

Focus

Our paper examines the impact of beliefs and pricing mistakes during the boom in subprime lending of the early 2000's. We collect original data on private and FHA mortgage insurance premiums from 19992016. This details the evolution of offerings in their scope as well as in their price. To characterize the overall pricing of mortgage insurance, as borrowers substitute among loan types, we construct chainweighted price indexes of insurance products in four risk categories. These indexes reveal broad changes in the pricing of default risk over time, but cannot distinguish between changes in the underlying credit risk from changes in the accuracy of risk pricing.

To distinguish between changes in the quantity versus pricing of credit risk, we fit a parametric model of default behavior to PMI prices in 2013. This quantifies default risk conditional on borrower's equity, the distribution of house price changes, and borrower credit worthiness. With 2013 PMI premiums as our benchmark, but allowing for differing expectations about house price appreciation, we judge the accuracy of premiums in 2005.

Contribution

There was a large increase in higher risk mortgage lending in the United States in the early 2000s. Expectations of higher house price appreciation shared between borrowers and lenders can explain this by reducing default risk and lowering the user cost of ownership. A shift in the supply of risky loans can also explain this, inducing borrowers to borrow more with lower rates. There is an active and ongoing dispute about the relative contributions of optimism and mispricing in this lending boom. We contribute to this debate by identifying a pattern of mispricing in higher risk mortgage lending and estimating the quantitative effects of this mispricing on the market for risky mortgages.

Findings

Our examination of mortgage credit risk pricing yields two findings: The boom in high-risk mortgages was a response to both the mispricing of risk and to optimistic beliefs about house prices. The mispricing largely stemmed from pooling across widely disparate credit risks, and resulted in adverse selection within the pool. In comparing the, insurance premiums for riskier products in 2005 versus 2013, we find they were not systematically higher in 2013, just differentiated by FICO score, as the errors implied by pooling were eliminated. For the highest-risk products, primarily insured by the government, insurance was underpriced both before and after 2008. But the insurance premiums themselves were substantially higher by 2013, consistent with more pessimistic beliefs about housing markets. We find that the 2005 mispricing contributed in the vicinity of 2 to 15 percentage points of the 24 percentage point difference in the share of high-risk mortgages in 2005 versus 2013, the remainder due to "rationing," i.e. changes in lending practices that virtually eliminated loans to the riskiest borrowers at any price.

Abstract

We provide new evidence that credit supply shifts contributed to the U.S. subprime mortgage boom and bust. We collect original data on both government and private mortgage insurance premiums from 1999-2016, and document that prior to 2008, premiums did not vary across loans with widely different observable characteristics that we show were predictors of default risk. Then, using a set of post-crisis insurance premiums to fit a model of default behavior, and allowing for time-varying expectations about house price appreciation, we quantify the mispricing of default risk in premiums prior to 2008. We show that the flat premium structure, which necessarily resulted in safer mortgages cross-subsidizing riskier ones, produced substantial adverse selection. Government insurance maintained a flatter premium structure even post-crisis, and consequently also suffered from adverse selection. But after 2008 the government reduced its exposure to default risk through a combination of higher premiums and rationing at the extensive margin.

JEL classification: G21, E44, E32

Keywords: financial crisis, mortgage insurance, housing finance, default risk