Title: Adverse Selection in Private Mortgage Securitization - Evidence from Loss Given Default

Speaker: Abdullah Yavas, University of Wisconsin-Madison

Time: March 16th, 2018 15:00–16:30

Venue: Conference Room 106B, Zhonghui Building (College of Economics, JNU)

About the speaker:

Abdullah Yavas holds Robert E. Wangard Real Estate Chair, as well as Department Chair of Real Estate and Urban Land Economics at the University of Wisconsin-Madison’s Wisconsin School of Business. Previously, he was the Elliott Professor of Business Administration and the Research Director of the Institute for Real Estate Studies in the Smeal College of Business, Pennsylvania State University, where he received the Fred Brand Award for Outstanding Teaching. His research interests include real estate brokerage, mortgage contracts, real estate auctions, economics of information, and experimental economics. Yavas has authored or co-authored more than 60 articles in peer-reviewed journals on real estate, finance, and economics.Yavas is the Editor of Real Estate Economics, and a Fellow of Real Estate Research Institute. Yavas also serves on the Monetary Policy Committee of the Central Bank of the Republic of Turkey. In 2018, he was elected as the second vice president of the American Real Estate and Urban Economics Association. He will serve as the vice president of that association in 2019, and as the president in 2020.

Abstract:

Loan performance varies in two dimensions: default probability and loss given default. While the relationship between securitization and default probability has been studied intensively, the effect of securitization on loss given default remains unexplored. Using a unique dataset containing over forty thousand mortgage liquidations, this paper studies whether, given liquidation, privately securitized loans have different loan losses from portfolio loans. After controlling observable loan characteristics and servicer effect, we find that securitized risky/opaque mortgages incur more than 17 percent (in relatively terms) higher loan losses than observably similar portfolio loans. High quality/transparent loans do not show any difference in loan losses. These results complement earlier empirical studies that reported adverse selection on mortgage securitization with respect to probability of default. We also employ a difference-in-difference analysis using a natural experiment to provide evidence that lax securitization standards pre-crisis contributed to higher loan losses.