'Cream-Skimming' in Subprime Mortgage Securitizations: Which Subprime Mortgage Loans Were Sold by Depository Institutions Prior to the Crisis of 2007?

Abstract

Depository institutions may use information advantages along dimensions not observed or considered by outside parties to "cream-skim," meaning to transfer risk to naïve, uninformed, or unconcerned investors through the sale or securitization process. This paper examines whether "creamskimming" behavior was common practice in the subprime mortgage securitization market prior to its collapse in 2007. Using Home Mortgage Disclosure Act data merged with data on subprime loan delinquency by ZIP code, we examine the bank decision to sell (securitize) subprime mortgages originated in 2005 and 2006. We find that the likelihood of sale increases with risk along dimensions observable to banks but not likely observed or considered by investors. Thus, in the context of the subprime lending boom, the evidence supports the cream-skimming view.

Securitization and Bank Performance

“…Calem, Henderson, and Liles (2010) also confirm this view with findings of evidence of "cream-skimming" behavior during the subprime lending boom. Krainer and Laderman (2011) explore the factors that determine which loans are securitized.…”

Trend growth expectations and U.S. house prices before and after the crisis

“…This flat "average loan-to-value" ratio masks substantial redistribution from seasoned home owners, whose LTVs fell as their home values appreciated, to new mortgage borrowers, especially those in the subprime segment, whose LTVs rose during the boom (Mayer et al, 2009). 7 Nonetheless, it is striking that the enormous increase in household indebtedness evident from the solid line was on average not associated with an increase in the ratio of debt outstanding to collateral value. Our model is different from Iacoviello's in that we abstract from nominal rigidities as we focus on the effects of low-frequency movements in perceived trend growth.…”

“…Our model is different from Iacoviello's in that we abstract from nominal rigidities as we focus on the effects of low-frequency movements in perceived trend growth. On the other hand, to account for the secular trend in real house prices, we employ a two-sector model in which technology in the housing sector is assumed to grow more slowly than 7 See also MacGee (2009). 8 One aspect that has received substantial attention is the structure of housing finance, especially innovations related to the subprime mortgage market.…”

“…8 One aspect that has received substantial attention is the structure of housing finance, especially innovations related to the subprime mortgage market. There is some evidence supporting the view that the incentives of mortgage originators under the "originate-to-distribute" model to properly screen mortgage applicants were severely impaired, facilitating the origination of loans that were more likely to default than might have been apparent to purchasers of these loans in secondary markets (Keys et al, 2010;Calem et al, 2010). Alternatively, certain aspects of automated loan applications may not have been correctly priced by investors (Bubb and Kaufman, 2011).…”

The economics of securitization: evidence from the European markets

“…Asymmetric information problems can come from (1) the information advantage of the originator with respect to the quality of borrowers and the historical performance of individual asset exposures -adverse selection; and (2) the complex security design of securitized assets, which suggests the superior information of arrangers concerning the true valuation of issued securities. Empirically, Downing et al 2012), which argue that banks tend to retain higher quality assets 14 .…”