Henry Kaufman Fellow in Business History, Center for Financial Policy
This post is the first in a series drawing on Professor Sicilia’s ongoing research on the evolution and reform of credit rating agencies.
As pundits, politicians, and scholars dissect the great financial crisis of 2007, several varieties of culprits have emerged – not least, the leading credit rating agencies (CRAs). Moody’s Investor Services, Standard & Poor’s Rating Services, and (a distant third) Fitch Ratings are responsible for rating between 95 and 98 percent of the nation’s municipal and corporate bonds, including those pesky “structured financial instruments” that played a central role in the recent financial meltdown. Why, an angry public wants to know, did the big three CRAs – awash in profits from the post-1970s surge in securitization – give high ratings to what too often turned out to be “toxic assets”? For its part, Congress became focused on CRAs a few years before the 2007 crisis, for their dubious role in the collapse of Enron.
But the roots of the problem run deeper – back to in the mid-1970s, if not earlier.
In 1975, the SEC, as part of its revision of rule 15c3-1 (the so-called “haircut” requirements for broker-dealers), designated a new category of CRA called the Nationally Recognized Statistical Ratings Organization, although the criteria for achieving NRSRO status were not defined for many years. In retrospect, it was a momentous step – but one that drew little notice in the financial community. Nor did the SEC – then focused on clamping down on Nixon-style “slush fund” corporate payments and abolishing fixed commission rates in the securities industry – make more than passing reference to the change.
Investors, jittery because of the recent collapse of the Penn Central and other financial turmoil, were willing to pay well for reliable investment information. The CRAs, especially the newly sanctioned NRSROs, recognized the premium this placed on their ratings and switched from a user-fee business model to an issuer-fee model. That is, they began charging the issuers of bonds for their rating services rather than potential or existing bond investors. Over time, the CRAs would take a more active role in actually structuring the investments that they would then turn around and rate for their clients.
Meanwhile, more and more laws were structured around CRA ratings – a practice that dated back to the 1930s, when, for instance, banks were required to mark to market bonds rated BBB or lower and were forbidden from purchasing “speculative” (below BBB) securities. A more recent example is the 1989 regulation allowing pension funds to invest in asset-backed securities rated A or higher. Federal laws, state constitutions, and municipal laws across the land use CRA-issued ratings to limit or encourage particular kinds of investment.
Therefore, a risky combination of principal-agent conflicts and of the outsourcing of regulatory functions to private entities lay dormant until Enron-style asset shell games and the explosion of derivatives and other arcane financial instruments made accurate and timely rating much more difficult. At least that’s what many regulator and CRA representatives suggest. History tells a different story.
CRAs and the Financial Collapse of New York City
Let’s return to 1975, the year the SEC created NRSROs and the CRAs began switching to issuer-pay. The biggest financial crisis that year was New York City’s plunge toward bankruptcy. The budgets of many major U.S. cities were straining as pensions and social welfare programs expanded aggressively while industrial employment was eroding and higher-income residents moved to surrounding suburbs. Yet in the face of decade-long rising tide of dire financial news, Moody’s and S&P raised the city’s ratings, then held firm. It was not until October 1974 that Moody’s downgraded its New York City bond rating from A to B (marginally speculative) and then to Caa (very speculative). By that time, fixed income investors had effectively accomplished what the harsh downgrade would have accomplished in a more stable bond market. That is to say, in effect, Moody’s move was moot. (The City’s last resort, the federal government, initially drew a rebuke from President Ford, but ultimately came through with a bailout loan.)
In a follow-up investigation of New York City’s financial calamity of the early 1970s, the Securities and Exchange Commission identified a number of culpable parties, from the “deceptive practices” of Mayor “Abe” Beame and Comptroller Harrison J. Goldin; to leading underwriters that issued optimistic investment pronouncements while quietly selling off their positions; to attorneys for the key players.
As it would in the wake of the 2007 financial crisis, the SEC deemed the big three credit rating agencies major contributors to largest financial calamity of the period. The regulators concluded that the CRAs had failed to perform their function, to the detriment of thousands of investors, of providing objective, timely, and accurate ratings. This failure occurred for three reasons. First, the CRAs based their ratings on “largely unverified” information that had been provided by the institution whose securities they were rating. Indeed, the SEC found the CRAs to be unrepentant on this score. “The agencies expressly disclaimed any responsibility for the accuracy of the information upon which they acted,” investigators noted. Second, the SEC concluded that even if data verification had been justified initially, the CRAs should have investigated once serious questions were raised. But the CRAs “apparently [did not] recognize a responsibility to make diligent inquiry even in the face of adverse facts which came to their attention.” Finally, the SEC concluded that “the agencies failed to make timely adjustments to their assigned ratings.”
Echoes with Enron are striking. As Malcolm Salter of HBS has documented in his comprehensive 2008 study of Enron’s collapse, Innovation Corrupted, the CRAs were accused of sole reliance on client-provided data; failure to investigate in the face of persistent and profound questions about the veracity of that data; and slowness to act in downgrading ratings.
In the wake of New York City’s debacle, a writer for the New York Times noted that “One of the major casualties of that debacle appears to be the rating agencies’ creditability. Why didn’t the agencies predict what was going to happen? Is the widespread complaint.”
But the same writer reported that demand for ratings was never stronger, and salaries at Moody’s had been rising “sharply.”
The regulatory and business model changes in the CRA industry in the mid-1970s – and how they played out in the financial collapse of New York City in 1975 – point to some key policy recommendations. One would be to roll back the 1975 issuer-pay business model change, with its obvious potential for conflict of interest. Rather, the original user-fee model better aligned CRA and investor (rather than borrower) incentives.
In addition, as long as scores of laws and regulations (public) depend on CRA ratings (private), the public interest remains a risk in some domains. This is not to say that the government should go into the credit rating business whenever public debt is involved. But as long as CRAs claim no responsibility for verifying the accuracy of borrower-supplied information on which they base ratings, regulators need to ensure that independent auditors perform their work diligently and in ways that support the ratings function. If the auditing function fails, so will the ratings function.
The role of CRAs in New York City’s 1975 tale of woe also suggests that CRAs are biased toward extraordinary forbearance when the stakes are especially high … whether sharp ratings downgrades would trigger the financial collapse of the nation’s financial capital, of the world’s largest energy trading company, or – as more recent events suggest – of the U.S. government’s ability to borrow. In my next entry on Credit Ratings Agencies, I will explore more fully this particular dimension of the too-big-to-fail dilemma.