Much has been written about the subprime mortgage market and its connection to the rash of home foreclosures across the nation. While there is little disagreement that a number of foreclosures involve subprime loans, there is some question of whether the driving force is the ill-suited loans that subprime borrowers received or some other factor acting in concert with subprime loans. This issue as well as other homeownership experiences of subprime borrowers is addressed in a study by Kristopher Gerardi, Adam Hale Shapiro, and Paul S. Willen. The authors also rely on an analytic approach that offers an alternative perspective and treatment of the default decision. What follows is a summary of their findings.1
The authors acknowledge that there is no universally accepted definition of a subprime mortgage. Various definitions rely on loan characteristics, borrower characteristics, or a combination of the two.2 For the authors, who focus on the subprime lending channel, it is a mortgage “originated by a subprime lender, where a subprime lender is identified using HUD’s annual list.”
The authors recognize that subprime lending has been at the center of the recent foreclosure crisis. This, in turn, has generated much public policy debate regarding the regulation of the subprime market and, more fundamentally, whether subprime borrowers should be extended credit to become homeowners. But they hasten to point out that a distinction should be made between those homeowners who use a subprime mortgage for their initial home purchase and those who use a subprime loan to refinance an existing mortgage. Their concern is that the current policy debate improperly groups all subprime borrowers together instead of focusing on borrowers who purchased their homes with a subprime mortgage, a group the authors emphasize in their analysis.
The authors also address the question of why households default on home mortgages. They develop a model that includes the influence of house prices and interest rates on the default decision that is used in standard models in the literature, but the model also incorporates a component of portfolio choice that allows a homeowner’s unique financial situation to be a factor. They use the model to derive results that provide further insight into the relationship between subprime lending and foreclosures.
The authors focus on the homeownership experiences of subprime borrowers in Massachusetts from 1989 to 2007 and their role in the dramatic rise in the state’s foreclosures during 2006 and 2007. They give particular consideration to homeowners who purchased homes with a subprime mortgage.
Marvin M. Smith, Ph.D., Community Development Research Advisor
The analysis was undertaken using a special data set consisting of a “historical registry of deeds records from January 1987 through August 2007 for the entire state of Massachusetts, as well as 2006 and 2007 Massachusetts assessor data.”3 The data cover two housing cycles in Massachusetts, which allows the authors to “document the foreclosure incidence of ownerships financed with a subprime mortgage versus ownerships financed with a prime mortgage.” More specifically, the data permitted the estimate of the determinants of default for the entire duration of ownership, in contrast to relying on loan-level data sets to estimate the determinants of default for single mortgages issued at purchase, which is the traditional approach found in the literature. The latter, according to the authors, can misrepresent the incidence of foreclosure over the entire homeownership, since “most subprime loans are refinances of a previous mortgage of unknown type, so typically, we have no way of knowing whether a subprime loan played any role in the initial transition into homeownership all we know is that the borrower refinanced into a subprime loan at some point.”4
The estimating model used in the analysis is designed to take advantage of the unique data set that allows “tracking the same borrowers across different mortgage instruments for the same residential property.” Thus, the authors are able to “characterize sale and default probabilities across the time horizons of entire â€˜ownership experiences.’”5 The model also deviates from those commonly used that base the decision to default on when the value of the house is less than the value of the mortgage on the house. In addition to the default decision being influenced by house prices and interest rates, the authors’ model also takes into account an individual’s financial situation. Their model allows for the household’s financial circumstances to affect the valuation of the house and the mortgage; hence “individual household valuations of identical assets typically won’t be identical.” This yields the intuitive prediction that fewer financial resources lessen “the value of the house relative to the value of the mortgage,” which increases the likelihood of default.
The authors have two major findings. The first is that “homeowners that begin with a subprime purchase mortgage end up in foreclosure almost 20 percent of the time, or more than six times as often as experiences that begin with prime purchase mortgages.” The converse side is that nearly 80 percent will have a “successful” outcome, which the authors define as either remaining in the house for at least 12 years or selling the house. When the authors focused specifically on the foreclosure crisis of 2006 and 2007 in Massachusetts, they found that subprime mortgages played a prominent role. But they stress that two distinct groups of subprime borrowers contributed to the crisis. While homeowners whose homes were purchased with a subprime mortgage accounted for roughly 30 percent of the foreclosures, “a large factor in the crisis stemmed from borrowers who began their homeownership with a prime mortgage, but subsequently refinanced into a subprime mortgage.”
The second key finding is that “house price appreciation (HPA) is the main driver of foreclosures.” The authors estimate that periods of low or negative HPA increase significantly the probability of default for both subprime and prime borrowers (see Fig. 1 in the study, p. 53). This was demonstrated dramatically with the decline in house prices during 2006 and 2007.
The authors also note the interplay between subprime mortgages and HPA. They indicate that subprime lending created a “class of homeowners who were particularly sensitive to declining house price appreciation, rather than, as is commonly believed, by placing people in inherently problematic mortgages.”
Finally, the authors state that although their analysis focused solely on Massachusetts, they believe the implications of their study could be broadly applied across the nation.