Introduction In recent months, there have been several ten-year retrospectives about the Great

Size: px

Start display at page:

Download "Introduction In recent months, there have been several ten-year retrospectives about the Great"

Sophie Little
5 years ago
Views:

1 Research Proposal SFAS 140/ASC 860 Disclosures and MBS Securitization Credit Ratings Downgrades and Defaults Arthur Wharton, Ph.D. Virginia State University Introduction In recent months, there have been several ten-year retrospectives about the Great Recession/Financial Crisis/Mortgage Crisis. In these retrospective views blame for the crisis has been attributed to the greed of market participants: mortgage lenders, investment bankers and the toxic mortgage securities they were creating and selling to the marketplace. Some steps have been taken in the interim to better safeguard the market. One such underappreciated step is the inclusion of a requirement in the Dodd-Frank legislation (SEC. 942) that requires issuers of securitizations to make supplemental and periodic information disclosures to the marketplace. One not-widely known fact about the securitization market is that most issuers are able to file Certifications of Termination of Registration under SEC Rules 12g-4 or Suspension of Duty to File Reports under Rule 12h-3. With both of these rules if the affected securities were held by less than 300 persons the securities were able to avoid filing ongoing periodic reports about the performance of the underlying collateral (assets) such as mortgages. Because most securitization issues were bought by institutional investors, nearly all securitization issues were able to use Rule 12h-3 to get around ongoing SEC reporting. This resulted in credit rating agencies being the only entities that provided any periodic information on securitization issues. This reporting was based on statistical models rather than widely available fundamental accounting information that occurs with most other sectors of the fixed income market. This information void was brought into clear view during the Financial Crisis when many market participants had no idea what was 1

2 actually happening with their securitization investments. Investment decisions were based on initial credit ratings of the issues. Credit rating agencies were accused of being slow to downgrade credit ratings of securitization issues whose underlying asset pools that had seriously deteriorated and significantly compromised ability to make timely interest and/or principal payments to investors. Market participants relied heavily on credit rating agencies to supply timely and accurate information about changes in the credit worthiness of securitization issues. The actual credit worthiness of a given securitization issue is tied directly to the payments made by the obligors of the underlying assets such as mortgages, auto loans, credit cards etc. The performance of the underlying assets in aggregate, which will directly impact the ability of the securitization issue to make timely interest and principal payments to investors, is reported in filings to the Securities and Exchange Commission (SEC). The information that is disclosed to the marketplace in these filings is required by the Financial Accounting Standards Board (FASB) in Statement of Financial Accounting Standard 140 (SFAS 140/ASC 860). To date, SFAS 140/ASC 860 disclosures have not figured prominently in academic research on securitizations. The proposed research project will contribute to the understanding of the value relevance of publicly available accounting information required by SFAS 140 for securitizations. Purpose The purpose of this research project is to examine the association between the credit ratings downgrades and/or defaults of mortgage securitization issues and the SFAS 140/ASC 860 disclosures that provide information on the performance of the underlying assets of a 2

3 securitization issue. SFAS 140/ASC 860 requires a securitizations issuer to provide disclosures about the performance of the assets that underlie that securitization issue. The SFAS 140/ASC 806 required disclosures are publicly available and provided in monthly SEC filings. To date, the extant literature has not examined securitization credit ratings downgrades and defaults within the context of SFAS 140/ASC 860 disclosures. Most analysis of securitizations is statistical in nature and is usually focused at the asset-class level. The proposed research will contribute to the understanding of the ability of accounting information to predict securitization downgrades and/or defaults at the individual security level. This contribution will help direct securitization investor s attention to publicly available sources of credit information about securitization issues. This should lessen the dependence of investors on the securitization credit information provided by credit rating agencies. It is anticipated that the results of the research project should demonstrate a statistically significant association between SFAS 140/ASC 860 disclosures and securitization credit rating downgrade/defaults. In addition, it should also shed light on the extent to which these disclosures can predict eventual credit rating downgrade and defaults. This will add a fundamental analysis approach to the existing primarily statistical model based analysis of securitizations. Proposed Next Steps 1) Data collection (hand data collection) 2) Preliminary Analysis 3) Analysis and development of draft report 4) Final report Proposed Research Design This research project will compare the SFAS 140/ASC 806 disclosures of a control sample of MBS that have not had credit ratings downgraded, to a MBS sample that have had 3

4 credit ratings downgraded. An event study methodology will be employed to examine the behavior of individual tranches of MBS time series disclosures data between 1 and 6 months or more (depending on data findings) before being placed on the On Watch list for a possible rating action, typically a downgrade. Proposed Model: Downgrade Status = β0 + β1 CreditLosses + β2 Delinquencies + β3 WAL + β4yldslope + β5princpl + β6fxflt + β7maturity + β8 Rating + β9 Enhance + ε The dependent variable, downgrade status, will be a dummy variable, 1= downgraded, 0 = not downgraded. The main independent variables are italicized in bold fonts and remaining independent variables are control variables. Preliminary Literature Review Ammer Clinton (2004) examines the impact of credit ratings on the pricing of assetbacked securities (ABS) and finds that negative returns and widening spreads are associated with ratings downgrades. These effects are stronger for ABS than what has been reported in prior research on corporate or sovereign bond ratings. Also, ABS investors rely more heavily on credit rating agencies as a source for news about changes in credit risk of ABS. Violi (2007) examines credit ratings migration (changes in the credit rating from one level to another ie. AAA to AA etc.) in the structured finance (SF)market and then compares among various sectors and products to those of traditional corporate finance (CF) bond market. The study finds that SF ratings are more stable in the short to medium term and that mobility can be a reasonable proxy for credit risk. 4

5 Gutter and Raupach (2010) analyze the impact of credit rating downgrades on credit portfolio risk using an estimated transition matrix. They find that investors that use transition matrices that do not take into account previous downgrades tend to underestimate the momentum sensitive Value-at-Risk for a given bond portfolio. This result suggests that bond portfolio management would do well to use downgrade-sensitive default probabilities in determining relative portfolio risks. Saretto (2005) examines how to predict corporate bond defaults using financial information and proposes a model to forecast default. It then evaluates how the measure is related to Fama French factors at both the firm and aggregate economy level. It finds that its model outperforms other models [Altman (1968), Ohlson (1980), Zmijeski (1984) and Shumway (2001)] in correctly classifying firms into (1) default and (2) non-default categories. Wharton (2013) examines the effect of SFAS 140 paragraph 17 disclosures on the launch spreads of mortgage-backed-securities (MBS) brought to market between 1999 and It finds a statistically significant association between the SFAS 140 disclosures and MBS launch spreads. It concludes that the SFAS 140 disclosures provide investors with value relevant information to assess credit risk as measured by yield spread. Data Sources SFAS 140/ASC 860 disclosure data will be hand collected from SEC Edgar filings. MBS On Watch dates and downgrade data will be hand collected from Moody s Investors Service. References Ammer, J. & Clinton, N.. (2004). Good News Is No News? The Impact of Credit Rating Changes on the Pricing of Asset-Backed Securities. Federal Reserve System International Finance Discussion Papers, n

6 Boot, A., Milbourn, T., & Schmeits, A.. (2006). Credit Ratings as Coordinating Mechanisms. The Review of Financial Studies,Spring v. 19, n. 1, pp Cantor, R., & Mann, C.. (2003). Measuring the Performance of Corporate Bond Ratings. Moody s Investors Service Special Comment, April. Chakravarty, S., & Chiyachantana, C.. (2009). On the Informativeness of Credit Watch Placements. Working Paper. Purdue University. Cho, D., Kim, H., & Shn, J.. (2008). The Effect of Seniority and Security Covenants on Bond Price Reactions to Credit News. Working Paper. State University of New York at Buffalo. Güttler, A., & Raupach, P.. (2010). The Impact of Downward Rating Momentum. Journal of Financial Services Research, 37(1), Saretto, A.. (2005). Predicting and Pricing the Probability of Default. Working Paper. University of California at Los Angeles. Violi, R.. (2008). Credit Ratings Transition in Structured Finance. Journal of Financial Transformation, n. 22. Wharton, A.. (2013). The Effect of Accounting Disclosures on the Mortgage-Backed-Securities Market: SFAS 140 and Launch Spreads. Working Paper. Towson University. 6

Mechanics and Benefits of Securitization

Mechanics and Benefits of Securitization Executive Summary Securitization is not a new concept. In its most basic form, securitization dates back to the late 18th century. The first modern residential