Today’s top left editorial feature in the Wall Street Journal led into today’s Commentary very nicely. The editorial alleges that the financial system needs something more than mere cheap money to stimulate credit market recovery. Precisely, and that’s why myself, Sean Mathis, and Julia Whitehead were asked by Congressman Ackerman to draft legislation that can provide a first step toward fixing some of the fundamental structural flaws in the system that caused and now perpetuate the crisis.
Unfortunately, the legislation was introduced Monday amid the side-show antics of the latest Treasury proposal and got a bit lost in the newsday. As today’s Journal noted, it shouldn’t have. Whether the problem is Fannie and Freddie or short sellers – the topics du jour – one big reason investors are skittish is that they have no reliable accurate means by which to evaluate firms and investment without financial reporting adequate for a world of structured finance nor bond ratings.
While financial reporting, generally, is a private sector transparency issue, ratings – specifically ratings produced by Nationally Recognized Statistical Ratings Organizations, NRSROs – difficulties were caused primarily by the value placed on them by the SEC and their widespread use in laws and regulations worldwide.
The government can fix how they use ratings right now, by simply putting out requirements for government (nay, NRSRO) ratings and allowing firms that can produce to those specification to meet the need for ratings suitable in terms of accuracy and reliability, to meet the purpose of restrictions on investment-grade securities. If existing ratings agencies find those restrictions too onerous, they can still produce ratings, just not those used for regulations in laws worldwide. I have a feeling that the major ratings agencies will gladly meet the government-led demand.
Following is a brief background description on the bill. The bill, itself, along with a press release issued at its introduction, can be found at this link from Congressman Ackerman’s web site: http://www.house.gov/list/press/ny05_ackerman/PR_071408.html
Summary: The proposed bill directs the SEC to clarify and amend the limitations on the types of investments that are eligible for NRSRO ratings, most specifically with respect to asset-backed securities, and set out a process by which new innovative financial products can eventually, after exhibiting a reasonable record of historical performance, achieve similar status.
1. The passage of this bill will regulate the meaning of credit ratings that the government uses for its own regulations and legislation. It is important to note that this bill does not add regulation; it simply revises certain aspects of an existing regulatory framework to reflect the lessons of recent events.
2. The bill will help stabilize markets by restoring confidence in ratings on asset-backed securities that have performed reliably over the years, adding much needed liquidity to the economy.
3. The bill will foster financial innovation, by allowing instruments to become NRSRO-eligible after they exhibit some degree of appropriate performance predictability.
4. The bill will also allow investors to isolate, within their own portfolios, the investments that fall outside of the new NRSRO-eligibility definition from the rest of their asset-backed instruments.
Why this bill: The subprime/credit bubble could never have grown so big so fast without access to easy money raised through securitization. That securitization, in turn, was fueled by the unbridled readiness of credit rating agencies to award investment grade labels on deeply flawed structured finance instruments. Not surprisingly, we are now seeing a substantial number of proposals to change rating agency behavior, lest we find ourselves in this same predicament in the future.
While current calls for enhanced rating agency transparency, disclosure and self-policing are laudable, they do not really address the issue of inflated ratings for products whose performance in some cases cannot be predicted with any reasonable degree of accuracy. The source of the problem is that the regulatory construct which initially established the rating agency Nationally Recognized Statistical Rating Organization (NRSRO) framework, has been expanded and stretched by myriad statutes, regulations and private investment policies that have also assigned importance to NRSRO ratings, giving NRSRO agencies a power far beyond what was ever envisioned.
The real issue, therefore, is that many securities sold to banks, public pension funds, and other regulated entities must have NRSRO ratings – even if the NRSRO ratings are not accurate. Effectively, then, the government has awarded substantial regulatory authority to ratings agencies, since they make the decisions that determine which securities are safe enough for regulated intermediaries. Since so many financial market participants are required to rely on NRSRO ratings the standard market penalty for being wrong, the loss of business, does not work – but neither is there a government oversight mechanism to deal with the massive scale of ratings errors that has been demonstrated most recently in structured finance.
Thus, it is time for regulators to impose definitions and limits on the NRSRO framework so that it can serve the purposes for which it was originally intended: protecting investors; financial institutions; and the economy from untoward risk and keeping public pension funds and banks from participating in speculative bubbles like the one we are presently witnessing.
The root of the problem: In 1975, the concept of an NRSRO was created to identify large, national, established rating agencies whose ratings could be used to help the SEC set capital requirements for broker-dealers.
Fundamental to the establishment of this scheme was the SEC’s belief that AAA meant safe and liquid, an understandable assumption since, at the time, ratings were primarily used for corporate, municipal and government securities and those carrying the AAA label generally exhibited such characteristics.
Two parallel trends caused that construct to break down. First, during the three decades since the establishment of the NRSRO system, innumerable domestic and global financial regulations, investment mandates, and statutory requirements piggybacked on the SEC’s use of NRSROs to differentiate securities that were safe and liquid from those that were not. Soon after the SEC used NRSRO rating for broker-dealer capital requirements, ERISA laid special importance on NRSRO ratings for public pension fund investments, a practice followed shortly thereafter by most public pension plans nationwide. Then, bank regulators began relying on ratings to determine safe and sound investments, and plan to expand the use of ratings internationally in the Basel II round of banking supervision.
Today, NRSROs have become the de facto gatekeepers to the investment grade universe – they determine what securities are eligible for investment by an enormous segment of the investor universe – and quasi-arbiters of how much capital regulated financial institutions may need to hold against their investments and other assets. But until the events of the last year, the full scope of their influence was not well understood and, as a result, there was little appreciation of the devastation that could be caused by a lapse in the reliability of their ratings on the scale that has recently occurred.
Second, asset-backed securities emerged as a significant investment category. The simply structured, homogenously collateralized, and strongly underwritten asset-backed securities of the 1980s and 1990s were easily absorbed into the NRSRO ratings scheme and, indeed, most of these securities performed as their ratings would have suggested. As we all now know, the asset-backed securities of the last few years were something very different – characterized by exceedingly complex structures, new and exotic collateral, and untested models and assumptions, which presented a recipe for highly unreliable ratings.
The confluence of these trends was that fiduciaries and financial institutions that were supposed to remain safe and sound in the public interest poured money into instruments whose performance would prove disastrous. Hence, the impact of bad credit ratings is vast, extending from pension funds whose ability to meet obligations is deeply impaired, to municipalities who struggle for cash to meet expenses, to money market funds which proved to be less than a safe haven for investors’ short term funds, to bank balance sheets which have been dramatically weakened by losses and an inability to liquefy assets.
What the bill does: It is important to note that regulatory distinctions between tested financial products and those too new to be evaluated with any certainty have been made before. Even in the most recent NRSRO legislation, the Credit Rating Agency Reform Act of 2006, the Reg AB definition of eligible asset-backed securities purposely omits synthetic securities that were substantially different from traditional asset-backed structures. Unfortunately for our markets, ratings on synthetic securities appear to have been treated as NRSRO for all intents and purposes nonetheless; worse, non-synthetic products proved to perform in troubling and unpredicted ways as well.
Recognizing that the use of NRSRO ratings is deeply embedded at all levels of the international financial system, discontinuing their regulatory use is not practical. However, if ratings are to be used for regulatory purposes, they must meet the test of reliability, which means that the instruments to which they are applied must also demonstrate adequate predictability. Clearly, then, new financial products need to demonstrate a history of performance before they can be deemed acceptable for NRSRO ratings. The proposed bill directs the SEC to clarify the limitations on the types of investments that are eligible for NRSRO ratings, most specifically with respect to asset-backed securities, and set out a process by which new innovative financial products can eventually, after exhibiting a reasonable record of historical performance, achieve similar status.
It should be noted that nothing in this bill prevents rating agencies from privately assigning ratings to non-NRSRO approved financial products, and those ratings can be awarded on whatever basis and whatever scale the ratings agencies deem prudent and the market finds acceptable. But those ratings should and will not in any way be confused with NRSRO ratings, which are the only ratings assigned meaning and importance in the laws and regulations of the United States of America.
Implications and ramifications of the bill’s passage: First and foremost, the passage of this bill will regulate the meaning of credit ratings that the government uses for its own regulations and legislation. The bill does so by preventing untested asset-backed securities from being confused with stable, predictable, mainstream investments. If this bill had been in place five years ago, the subprime bubble would likely never have occurred, since NRSRO-constrained investors would not have been able to buy the subprime securities that have caused the credit crisis.
Second, the bill will stabilize asset-backed securities markets by providing a meaningful benchmark for the vast majority of asset-backed securities that have reliable performance. A substantial reason that credit markets are currently faltering is that lenders can no longer securitize loans as they did previously. But standard mortgages, credit cards, auto loans, and student loans were not the problem. Those markets are shut largely because the meaning of AAA and other bond ratings applied to structured finance securities have been so perverted that investors thinking of buying those securities have little idea of the risks they are assuming at a given rating grade. When investors regain confidence in the underlying bond ratings, markets for consumer and commercial finance will again re-open, public pension funds can invest with confidence, and citizens can once again be secure that their homes and their pension funds will retain the value they are relying on.
Third, this bill provides a stable path for financial innovation, by allowing instruments to become NRSRO-eligible after they exhibit some degree of appropriate performance predictability. Providing that stable path for innovation, however, will also require an acknowledgement that financial markets have strayed from that path in recent years.
Fourth, the bill will also allow investors to isolate investments within their own portfolios that fall outside of the new NRSRO-eligibility definition from the rest of the asset-backed securities market which can then operate on a stable foundation of meaningful and reliable ratings. Clearly, there may be a need to allow investors to hold what will now be considered “ineligible” securities already in existence until losses can be properly accounted for and/or the securities can be disposed on in an orderly fashion. Neither such provisions, nor the clarifications introduced by this bill, can change the fundamental complexion of the ineligible securities; they can only acknowledge that such instruments were never what they were represented to be in the first place.
Footnote: (1) The Credit Rating Agency Reform Act of 2006 requires NRSRO applicants to register for defined categories of issuances and the definition of asset-backed securities referenced in CFR 17, section 224, para 1101(c), plainly excludes synthetic CDOs and ABS. However, while 1101(c) clearly demonstrates a regulatory sensitivity to the risks of newer asset-backed classes, it failed to prevent the current crisis for two reasons: first, synthetic asset-backeds which were awarded ratings by credit rating agencies who were registered as NRSROs appear to have been treated, for all intents and purposes, as NRSRO securities by regulators and investors notwithstanding 1101(c)’s limitations. Second, even if the 1101(c) definition had prevailed, it was developed before it became clear that innovations within asset-backed collateral asset classes, such as mortgages, could cause just as much unpredictability as entirely new types of asset-backed securities, such as synthetics. Hence, new iterations of subprime and alt-a mortgages which had no performance history on which ratings could be reliably issued would have been able to creep in under the umbrella established by 1101(c).