« Back to Results


Paper Session

Sunday, Jan. 3, 2021 10:00 AM - 12:00 PM (EST)

Hosted By: American Real Estate and Urban Economics Association
  • Chair: Andra Ghent, University of North Carolina-Chapel Hill

Tranching is Also Catering

Jefferson Duarte
Rice University
Justin Balthrop
University of Kansas
Douglas McManus
Freddie Mac


The Financial Crisis literature (e.g. Caballero and Krishnamurthy (2009)) argues that demand for safe assets caused the pre-crisis growth in securitization, but empirical evidence for this theory is lacking. We provide the first evidence that securitization is 'catering'. We show that a 1% increase in excess demand for long-term U.S. Treasury bonds increases next month's issuance of long-term Collateralized Mortgage Obligations (CMOs) with low prepayment risk by 1.2%. This substitution effect does not exist for CMOs with characteristics that are very different from long-term government bonds, such as short duration or high prepayment sensitivity.

Securitization of Assets with Payment Delay Risk: A Financial Innovation in the Real Estate Market

Chao Ma
Xiamen University
Hao Zhang
Country Garden Asset Management Company
Hongbiao Zhao
Shanghai University of Finance and Economics


We study a new type of securitization, mortgage-receivable-backed securities (MRBSs) issued by real estate developers. Unlike traditional mortgage-backed securities (MBSs), the major risk of underlying assets of MRBSs is payment delay instead of default and prepayment. Different from default risk, delay risk refers to the uncertainty on how long it takes to obtain payments instead of whether the payments can be successfully obtained. Using unique loan-level data, we estimate proportional hazard models and detect factors that affect the delay risk of underlying assets of MRBSs, including bank characteristics, property-loan-household characteristics, local market conditions, and macroeconomic conditions. Especially, we find that the effects of house prices and LTVs on MRBS risk are the opposite of those on traditional MBS risk. Based on the estimates, we simulate cash flows of an underlying-asset pool and analyze the shortfall risk of the corresponding security tranches. We find that the securitization process imposes a natural adverse selection on the underlying assets. Our analyses provide a benchmark for conducting appropriate security designs based on the composition of the underlying asset pool, increase the transparency for investors on the risk pattern of MRBSs, and provide implications for pricing and regulation.

Price Discovery Limits in the Credit Default Swap Market in the Financial Crisis

Susan Wachter
University of Pennsylvania
Andrey Pavlov
Simon Frazier University
Eduardo Schwartz
Simon Fraser University


We derive the credit default swap (CDS) premium a financial institution requires to assume the default risk of fixed income instruments and the maximum premium a CDS buyer is willing to offer. These premiums are functions of the institution’s capital and current risk exposure. In most cases, an institution requires an increasing premium to assume additional risk. However, we show that an under-capitalized institution that already has substantial default risk exposure would engage in risk-shifting and assume more risk at lower CDS premiums. Consistent with this, prior to the 2008 financial crisis credit default swap issuance increased substantially, as did the volume of the underlying mortgage-backed securities, but the data suggests that required CDS premiums remained constant or declined.

Bubbles, Fragility and Optionality in Mortgage Securitization

Robert Van Order
George Washington University
Rose Neng Lai
University of Macau


The defining characteristic of financial panics is fragility in the form of discontinuous changes in investor or depositor asset demand. We derive a model of fragility for portfolios of assets put into securitization deals. We show, in the context of adverse selection, that fragility is caused by convexity of the payoffs of the assets due to optionality in the assets. The optionality can come from the assets in the deal (like mortgages secured by properties), or from structuring by itself (like Collateralized Debt Obligations (CDOs)). Not accounting for optionality can cause misspecification and underestimation of tail risks. Some types of assets will not produce the nonlinearity that causes fragility, and some types of stress test models will miss fragility when it exists, causing underestimation of insolvency risk and capital needs.
Sean Flynn
Tulane University
Haoyang Liu
Federal Reserve Bank of New York
Nancy Wallace
University of California-Berkeley
Ally Quan Zhang
Lancaster University
JEL Classifications
  • G1 - Asset Markets and Pricing