Transparency and the (E)valuation of Asset-Backed Securities

In 2011, the commission appointed by Congress to investigate the causes of the financial crisis concluded that “a combination of excessive borrowing, risky investments, and lack of transparency put the financial system on a collision course with crisis” (The Financial Crisis Inquiry Report, 2011, p. xix).  In particular, the opacity of asset-backed securities (ABS) greatly inhibited the ability of investors and regulators to fully understand the risks held by institutions that owned these products.  As part of the post-financial crisis effort to reform the securitization process, the Dodd-Frank Act directed the SEC to “require issuers of asset-backed securities, at a minimum, to disclose asset-level or loan-level data, if such data are necessary for investors to independently perform due diligence” (Section 942[b]). In response, the SEC developed asset-level disclosure requirements in Regulation AB II (hereafter, Reg AB II), which became effective on November 23, 2016 (SEC, 2014).

Public disclosures of asset-level data by ABS issuers represent a significant increase in transparency for certain classes of ABS.  For example, prior to these requirements, issuers provided investors in auto loan and lease ABS (which are typically collateralized by pools of 20,000 to 100,000 auto loans or leases) with only pool-level summary statistics of individual underwriting criteria such as borrower FICO scores and loan-to-value ratios. While useful, these pool-level disclosures suppress important features of the underlying assets, such as the degree of risk layering (i.e., the degree to which individual loans have multiple underwriting red flags, such as low credit score, high loan-to-value, no income documentation, etc.) (Ryan, 2018). Under Reg AB II’s asset-level disclosure requirements, issuers provide investors with underwriting criteria, such as the FICO score and loan-to-value ratio, for each asset in the pool.

Our paper uses the implementation of the asset-level disclosure requirements in Reg AB II to examine the impact of asset-level transparency on the (e)valuation of ABS by investors and credit rating agencies (CRAs). These asset-level disclosure requirements represent an ideal setting to address our research question for two primary reasons. First, these requirements constitute the first and still most significant post-crisis expansion of public information about the assets underlying ABS. Second, Reg AB II’s asset-level disclosure requirements only substantially increase disclosures for Auto ABS, but not for ABS backed by equipment loans, credit cards, or commercial mortgages.[1] This allows us to partition ABS securities into a treatment group (Auto ABS) and a control group (equipment, credit card, and CMBS), which improves our ability to identify the effects of the increased transparency experienced by the treatment group.

While it might seem obvious that increasing transparency would improve evaluation of these securities by investors and CRAs, there are reasons why these improvements may not materialize. Due to investors’ information-processing constraints, the complexity of securitizations, and the highly disaggregated nature of asset-level disclosures, Richardson et al. (2011, p. 482) conclude that “it is not clear how investors or regulators can use this voluminous information.” Similarly, in its comment letter on the SEC’s 2010 proposed rule that led to Reg AB II, which is representative of the letters from other ABS issuers, Bank of America (2010, p. 11) states that “while some investors may suspect that the information would be helpful, the lack of any historic reliance on some of this data suggests that it may be per se immaterial.” In addition, prior academic research suggests that public disclosure can crowd out the production of private information and that greater public disclosure may reduce price efficiency by shifting investors’ attention from fundamental analysis to the determination of other investors’ beliefs (Goldstein and Yang, 2017; Banerjee et al., 2018).  Finally, prior to Reg AB II, the big three CRAs (S&P, Moody’s, and Fitch) already had access to non-public information that was more granular than information publicly available to investors, and none of the CRAs substantially changed their rating methodologies for these products around the implementation of the asset-level disclosure requirements in Reg AB II.

We first analyze the extent to which asset-level disclosures improved the ability of investors to evaluate ABS risk. We find that initial yield spreads became more predictive of future delinquency for the treatment group (relative to the control group) after Reg AB II, suggesting that investors were better able to assess ABS risk.  Further tests suggest that this improvement was due to investors’ ability to analyze certain risks, such as the degree of risk-layering within each tranche, which only would have been apparent from asset-level information.

Similar to yield spreads, we also find that credit ratings became more predictive of future delinquencies, and that investors placed greater reliance on credit ratings as a result of the asset-level disclosures.  Because CRA’s did not change their rating methodologies or reliance on asset-level information around Reg AB II, the improved credit rating quality suggests that public asset-level disclosures either discipline the agencies’ evaluation of ABS or increase their use of improved ABS price-related information.

We find that our documented effects of asset-level disclosures occur sharply around the compliance date of the asset-level disclosure requirements in Reg AB II. Further, we find that our primary results obtain only in ABS deals with above-median risk layering in the underlying assets, for which pool-level summary statistics of credit risk metrics (e.g., pool-level average borrower FICO scores) are less informative, and with above-median complexity in the tranching of credit risk.

In addition to contributing to academic research on the effects of transparency in financial markets and the banking system, our findings yield several policy implications. First, Reg AB II’s asset-level disclosure requirements improve market participants’ (e)valuation of ABS, likely improving price efficiency in ABS markets. Second, these improvements are attributable in part to more effective due diligence by ABS investors and in part to higher quality credit ratings of ABS. Third, contrary to a stated primary objective of the SEC for Reg AB II’s asset-level disclosure requirements, these disclosures do not reduce investors’ reliance on credit ratings.

ENDNOTE

[1] Reg AB II’s asset-level disclosure requirements only apply to the following types of publicly offered (i.e., SEC-registered) ABS: Auto ABS, CMBS, residential mortgage backed securities (“RMBS”), and resecuritizations of these ABS or debt securities (CDO).  We do not include RMBS or CDOs in our treatment sample, because there were an insufficient number of securitizations during our sample period (non-agency ABS deals with issuances date from June 2013 – May 2018).  While CMBS are subject to the asset-level requirements, they are included in our control sample because asset-level information for these deals has historically been provided in the prospectus (i.e., there was no change in asset-level transparency for CMBS as a result of Reg AB II).

REFERENCES

Banerjee, S., J. Davis, N. Gondhi. 2018. When transparency improves, must prices reflect fundamentals better? Review of Financial Studies 31 (6): 2377-2414.

Bank of America. 2010. Comment letter on asset-backed securities proposed rule (Release Nos. 33-9117, 34-61858; RIN 3235-AK37; File No. S7-08-10), August 2. Available at https://www.sec.gov/comments/s7-08-10/s70810-108.pdf.

Financial Crisis Inquiry Commission. 2011. The financial crisis inquiry report: Final report of the national commission on the causes of the financial and economic crisis in the United States. New York, NY: Public Affairs.

Goldstein, I., L. Yang, 2017. Information disclosure in financial markets. Annual Review of Financial Economics 9: 101-125.

Richardson, M., J. Ronen, M. Subrahmanyam, 2011. Securitization reform. Chapter 16 in Regulating Wall Street: The Dodd-Frank Act and the New Architecture of Global Finance, eds. Acharya, V. V., T. F. Cooley, M. P. Richardson, and I. Walter, John Wiley & Sons: Hoboken, NJ.

Ryan, S. 2018. Recent research on banks’ financial reporting and financial stability. Annual Review of Financial Economics, forthcoming.

Securities and Exchange Commission. 2014. Asset-backed securities disclosure and registration, Final Rule. 17 CFR Parts 229, 230, 232, et al. Federal Register 79 (185). Available at https://www.gpo.gov/fdsys/pkg/FR-2014-09-24/pdf/2014-21375.pdf.

This post comes to us from professors Jed J. Neilson at Pennsylvania State University’s Smeal College of Business, Stephen G. Ryan at New York University’s Stern School of Business, K. Philip Wang at the University of Florida’s Fisher School of Accounting, and Biqin Xie at Pennsylvania State University’s Smeal College of Business. It is based on their recent article, “Asset-Level Transparency and the (E)Valuation of Asset-Backed Securities,” available here.