This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elseviers archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

http://www.elsevier.com/copyright

Author's personal copy

Residential mortgage default: Theory works and so does policy q

Allen C. Goodman a, Brent C Smith b,a Department of Economics, Wayne State University, Detroit, MI 48202, USAb Department of Finance, Insurance and Real Estate, Snead School of Business, Virginia Commonwealth University, Richmond, VA 28284, USA

a r t i c l e i n f o

Article history:Received 28 January 2010Available online 23 October 2010

JEL classifications:E44G21G28K11R20

Keywords:Residential mortgage defaultPredatory lendingMortgage pricing

a b s t r a c t

Using a national loan level data set we examine loan default as explained by local demo-graphic characteristics and state level legislation that regulates foreclosure proceduresand predatory lending, using a hierarchical linear model. When controlling for loan andlocal conditions, we observe significant variation in the default rate across states, withlower default levels in states with higher temporal and financial costs to lenders. State levellegislative influences provide a foundation for discussion of national level policy that fur-ther regulates predatory lending and financial institution foreclosure activities.

2010 Elsevier Inc. All rights reserved.

1. Introduction

Residential mortgage default is a complex event triggeredby a host of household and socio-economic events. The re-cent collapse of the subprime market, coupled with theimposing downturn in the housing and broader economicmarkets has resulted in a rapidly expanding rate of mortgagedefaults, many of which end in foreclosure and REO.1 We

know from previous literature that there are trigger events,underwriting issues, economic factors, and interest ratechanges that impact the probability and timing of default(Quercia et al., 2005; Renuart, 2004; Vandell, 1995; Fosterand Van Order, 1984, 1985). We further recognize that varia-tions in state level legislation regulating predatory lendingand foreclosure proceedings exist, and that both legislativedirectives impose varying costs and benefits that affect the va-lue of a mortgage to the lender and borrower. We have onlylimited information, however, on the impact of such legisla-tion on the propensity of a default event and the ultimate fore-closure or REO from the lenders perspective (Ambrose andButtimer, 2000; Capozza and Thomson, 2005, 2006; Cuttsand Merrill, 2008).

We do not argue that the optimal foreclosure rate iszero. While lenders, of course, predicate their loans onthe probability of repayment, those lenders who wouldseek only to minimize either the number of bad loans, orthe volume of dollars foregone due to foreclosure, wouldclearly forego potentially substantial profits. Still massforeclosures can have deleterious neighborhood and na-tional impacts, as has become apparent in the United

1051-1377/$ - see front matter 2010 Elsevier Inc. All rights reserved.doi:10.1016/j.jhe.2010.09.002

q This paper has benefited from helpful conversations with BrentAmbrose, Wayne Archer, Edward Prescott and two anonymous reviewers.Mark Watson of the Federal Reserve Bank of Kansas City providedassistance with the LPS data. We are indebted to the Federal Reserve Bankof Richmond and LPS Applied Analytics for providing access to the datavia a research affiliation between Brent C. Smith and the Federal ReserveBank of Richmond. All views, however, are the responsibility of theauthors and do not reflect those of the Federal Reserve Bank of Richmond.Additionally, all errors are the sole responsibility of the authors. Corresponding author.

E-mail addresses: allen.goodman@wayne.edu (A.C. Goodman),bcsmith@vcu.edu (B.C Smith).

1 The phrase real estate owned property indicates that the property inquestion has been foreclosed on and taken back by the mortgage lender ortrustee. Real estate owned and foreclosures are not synonymous, howeveran REO is the result of a foreclosure that is not cured or ends in a short sale.

Journal of Housing Economics 19 (2010) 280294

Contents lists available at ScienceDirect

Journal of Housing Economics

journal homepage: www.elsevier .com/locate / jhec

Author's personal copy

States since 2007. Policies that would reduce the numbersof foreclosures, and their external impacts, would bewarmly greeted by mortgage lenders and borrowers alike.

The literature on foreclosures generally asserts that, ex-cept for workout efforts on the part of the lender, the ulti-mate decision to default rests with the borrower. Althoughwe do not challenge this proposition, we do posit that thelegislative costs imposed on financial institutions create anincentive for credit rationing by lenders. Underwriting isthe lenders instrument in risk reduction. State policiesthat impose costs to financial institutions for high-cost (of-ten categorized as predatory) lending and foreclosure pro-cessing are designed to motivate restraint on lenders. Byinstituting higher underwriting standards, financial insti-tutions attempt to reduce potential costs to the overallloan portfolio. This would suggest that restrictive lendingpolicies simply result in lenders imposing greater restraintin granting loans via the underwriting mechanisma formof credit rationing. Lenders will attempt to mitigate thehigher cost of instituting foreclosure in states that imposemore stringent legislation by (1) increasing the borrowerscost of funds (Pence, 2006) and (2) instituting differentialunderwriting standards with more rigid benchmarks ap-plied to borrowers in high-cost states. We expect to ob-serve this in a lower default rate for those high-coststates as a result of the higher acceptance threshold.

Utilizing a dataset of over 20 million loans aggregatedat the zip code level, we attempt to determine the politicaland locational drivers of default. We rely on state levelvariations in the foreclosure timeline and estimated cost,as identified in Cutts and Merrill (2008), as proxies forthe costs of default to the lender. We also include variablesfrom the predatory lending literature to test their relation-ship to the default rate (Bostic et al., 2007). The zip codeloan data and the state level variables are supplementedwith static and dynamic demographic information to con-trol for economic variations and location fixed effects.

Our modeling approach assumes that since financialinstitutions use the underwriting process to minimize theirexposure to future defaults, state laws governing lendingand the foreclosure process can affect the pricing and num-ber of loans offered, and hence the numbers and percent-ages of adverse outcomes. In states where lenders absorbfewer costs associated with default, the lenders haveincentives to institute more liberal underwriting practicesthan in higher-cost states. Because borrowers cannot self-select to avoid disparities caused by location, the borrow-ers in different states will face different loan requirements.In theory, public policy that seeks to reduce the potentialfor large scale defaults, yet maintain access to funds as pro-vided by the market, should include provisions recognizingthat (a) laws regulating the time before lenders can exe-cute foreclosures influence the rate of foreclosures acrossthe market; and (b) the more options and/or lower costsfor borrowers to default, the greater the costs to lenders.

Increased incentives on lenders to tighten underwritingstandards would be expected to influence the rate of fore-closures across the market. This conclusion rests on severalassumptions. Foreclosures impose costs on lenders, andthe costs can be estimated and built into the price of amortgage. Because of differences in state laws, lenders face

different foreclosure costs depending upon the state wherethe loan originates. Assuming lenders are rational andwell-informed of the cost differentials, they will have ahigher acceptance bar for borrowers in states with higherforeclosure rates relative to the price that can be legallycharged for loans. This practice reduces the lenders expo-sure in states with higher relative foreclosure costs. In the-ory, lenders will accept fewer high-risk applicants in high-cost states than in low-cost states. One should thereforeexpect to see fewer foreclosures in states with higher coststo the lender.

In the next section, we outline the well developed liter-ature on state legislation governing predatory lending andforeclosure proceedings. We rely on the literature to estab-lish the role of underwriting in reducing lender cost andthe loan pricing (and by default rationing) process. We fol-low with the modeling approach to the analysis and thedata, along with a discussion of the results. The paper con-cludes with a summary and discussion of the implicationsfor state policy.

2. State legislation

2.1. Predatory lending laws

One of the first legislative acts directly addressing pred-atory mortgage lending was the Federal Home Ownershipand Equity Protection Act of 1994 (HOEPA) (Pub. L. 103-325, 108 Stat. 21,600).2 HOEPA defined a class of mortgageloans that can be cla