« Back to Results

Institutional Issues in Mortgage Markets

Paper Session

Sunday, Jan. 5, 2020 10:15 AM - 12:15 PM (PDT)

Manchester Grand Hyatt, Nautical
Hosted By: American Real Estate and Urban Economics Association
  • Chair: Jia He, Nankai University

Securitization and Screening Incentives: Evidence from Mortgage Processing Time

Jung-Eun Kim
,
Federal Reserve Bank of Richmond
Dong Beom Choi
,
Seoul National University

Abstract

"We test if lenders' screening incentives weaken when facing the possibility of eventual loan sales. We adopt a new measure of lending standards, application processing time
for mortgages at the loan level, and use the collapse of the non-agency MBS issuance market as a natural experiment. The market collapse significantly reduced secondary
market liquidity for non-conforming loans, but had little impact on conforming loans. Following the collapse, lenders spent discretely more time screening and processing applications for loans larger than the conforming loan limits than those below the limits. The processing time gap widened more for banks with lower capital, greater involvement in the OTD model, and greater assets."

Dismembered Giants: Bank Divestitures and Local Lending

Yunqi Zhang
,
Nankai University
Yong Kyu Gam
,
Southwestern University of Finance and Economics

Abstract

This paper studies how bank divestitures in M&As affect local credit markets. We find that combined market shares of merging banks in local mortgage markets decline following divestitures in M&A. This decline is less in mortgages to black borrowers and refinance mortgages, which rely more on relationship lending. In contrast, combined market shares of merging banks in local small business lending markets, which heavily rely on relationship lending, do not change. Divestitures incur following negative externalities: mortgage credit availability is deteriorated for minority group borrowers in M&A counties; mortgage interest rates increase more in M&A counties than those in nearby non-M&A counties; mortgages originated in M&A counties are more likely to enter foreclosure than those in adjacent non-M&A counties; and house prices in M&A counties declined more dramatically during the subprime crisis.

FinTech Lending and Mortgage Credit Access

Lauren Lambie-Hanson
,
Federal Reserve Bank of Philadelphia
Julapa Jagtiani
,
Federal Reserve Bank of Philadelphia
Tim Lambie-Hanson
,
Haverford College

Abstract

Following the 2008 Financial Crisis, banks have lost significant mortgage market share to nonbank lenders, including recently to Fintech firms. We compare Fintech and traditional lenders’ mortgage offerings using a dataset of direct mail solicitations and Home Mortgage Disclosure Act data. Fintech lenders cast a wider net when soliciting prospective borrowers than do depository institutions, marketing to consumers with weaker credit scores, to whom banks rarely advertise. Fintech loans tend to be originated in areas where traditional lenders have denied a larger share of loans applications and resident credit scores are lower. Fintech lenders are making inroads in nonmetropolitan areas and places with less lender competition. We use a regression discontinuity design to show that Fintech lenders and other nonbanks respond to Community Reinvestment Act (CRA) incentives by lending more in LMI tracts, even though they are not subject to CRA regulation. Although Fintech firms have made innovations in marketing and the application process, the characteristics of Fintech loans are broadly similar on common observables to those made by other lenders. Unlike in Fintech small business and unsecured consumers loans, Fintech mortgage lenders may not have the same flexibility to utilize alternative data for credit decisions, due to stringent mortgage origination requirements.

Misreporting of Second Liens in Portfolio Mortgages and Privately Securitized Mortgages

Shuang Zhu
,
Kansas State University
Abdullah Yavas
,
University of Wisconsin-Madison

Abstract

Using a unique nationwide mortgage servicing dataset, this paper investigates the underreporting of second liens in both portfolio mortgages and privately securitized mortgages. We document that over 27% of portfolio loans and over 33% of privately securitized loans have second liens, and more than 40% of those second liens are underreported. Loans with misreported second liens have about 90% higher chance of default and incur about 90% higher loan losses in relative terms. Lenders seem to know the existence of the second liens in portfolio loans and intentionally misreport second liens with higher second lien LTV ratios and associated with risky first mortgages. Sold loans ex post have lower misreporting rate than observably similar portfolio loans. The effect is especially strong for loans that are subject to tighter screening such as low documentation loans and loans ex ante perceived harder to sell. In addition, we offer evidence that the lender-MBS servicer affiliation also plays a role in misreporting.
Discussant(s)
Haoyang Liu
,
Federal Reserve Bank of New York
Tong Yu
,
University of Cincinnati
Xiangyu Guo
,
Fudan University
Wei Shi
,
Jinan University
JEL Classifications
  • R2 - Household Analysis
  • G2 - Financial Institutions and Services