from the Philadelphia Fed
— this post authored by Ronel Elul and Sebastian Tilson
We use a matched credit bureau and mortgage data set to identify occupancy fraud in residential mortgage originations, that is, borrowers who misrepresented their occupancy status as owner occupants rather than residential real estate investors. In contrast to previous studies, our data set allows us to show that such fraud was broad based, appearing in the government-sponsored enterprise market and in loans held on bank portfolios as well.
Mortgage borrowers who misrepresented their occupancy status performed worse than otherwise similar owner occupants and declared investors, defaulting at nearly twice the rate. In addition, these defaults are significantly more likely to be “strategic” in the sense that their bank card performance is better and utilization is lower.
Introduction
Policymakers and the popular press have cited anecdotal evidence to suggest that one of the contributing causes to the housing bubble was pervasive mortgage fraud.1 Recent academic work has also verified the existence of mortgage fraud along several dimensions. Ben-David (2011) finds evidence of inflated prices. Griffin and Maturana (2015a) examine three dimensions of fraud among securitized nonagency loans: unreported second liens, owner occupancy misreporting, and appraisal overstatements. Piskorski, Seru, and Witkin (2015) study second lien misreporting and occupancy fraud in the nonagency securized market. Mian and Sufi (2015) argue that borrowers misstated their incomes on mortgage applications.
In this paper, we use a matched credit bureau and mortgage dataset to identify occupancy fraud in loans originated between 2005 and 2007. This occurs when mortgage borrowers claim on the mortgage application that they will be the owner occupants of the property, will not rent the property out to another individual or family, and do not intend to sell the property quickly. Borrowers may have an incentive to commit occupancy fraud because the benefits can be substantial: Banks often require declared residential mortgage investors to offer higher down payments and charge them higher interest rates because of the elevated default risk of investor loans (which we also document in this paper). In contrast to previous work, our data allow us to confirm that occupancy fraud was pervasive and did not just affect private-securitized loans. We show that more than half of all investors were fraudulent. And this applied to government-sponsored enterprise (GSE) – guaranteed, private securitized, and portfolio-held loans (by contrast, Federal Housing Administration (FHA) loans exhibited markedly lower fraud rates).
[click on image below to continue reading]
Source: http://www.philadelphiafed.org/research-and-data/publications/working-papers/2015/wp15-45.pdf
