Cliff Risk and the Credit Crisis
Joseph R. Mason
Louisiana State University - Ourso School of Business; University of Pennsylvania - Wharton Financial Institutions Center
November 10, 2008
Many recent financial institution failures have taken investors and the public by surprise. In recent years, investors and regulators have witnessed steady - even high-flying - firm performance that seems to suddenly fall off a cliff. Investigation of the sudden cliff-like failures we have witnessed in the credit crisis reveals continuous processes stemming from shortcomings in firm rights and responsibilities in securitizations that can explain recent firm failures and help better understand firm performance and guide regulation in the new world of "off-balance sheet" finance. The ability to skew earnings is a powerful temptation to increase securitization period after period to maintain growth in accounting (non-cash) earnings. One way to increase securitization from period to period is to "sell" the loans for accounting purposes without really selling them in any true economic sense. In recent years, "representations and warranties" replaced holding the riskiest bottom securities as the chief credit enhancement mechanism for investors, subsidizing pool performance so that no asset- or mortgage-backed security investor ever experienced losses. Regulators, misunderstanding (or refusing to see) the economics of securitization, misguidedly claim that incentives of sellers and investors, as well as borrowers, can be aligned by having sellers retain some risk in their securitizations. The reality is that sellers have always retained risk, and that retained risk is precisely what has created the seeming cliff.
Number of Pages in PDF File: 14
Keywords: Securitization, RMBS, Servicing, Gain-on-sale Accounting, Credit Crisis
JEL Classification: G28, G32, E44, E58
Date posted: November 8, 2008 ; Last revised: November 12, 2008
© 2015 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo4 in 0.250 seconds