The three large U.S.-based credit rating agencies – Moody’s, Standard & Poor’s, and Fitch – provided excessively optimistic ratings of subprime residential mortgage-backed securities (RMBS) in the middle years of this decade actions that played a central role in the financial debacle of the past two years. The strong political sentiment for heightened regulation of the rating agencies – as expressed in legislative proposals by the Obama Administration in July 2009, specific provisions in the financial regulatory reform legislation (H.R. 4173) that was passed by the House of Representatives in December, and recent regulations that have been promulgated by the Securities and Exchange Commission (SEC) – is understandable, given this context and history. The hope, of course, is to forestall future such debacles.
The advocates of such regulation want to grab the rating agencies by the lapels, shake them, and shout “Do a better job!” But while the urge for expanded regulation is well intentioned, its results are potentially quite harmful. Expanded regulation of the rating agencies is likely to:
As a result, ironically, the incumbent credit rating agencies will be even more central to the bond markets, but are unlikely to produce better ratings.
There is a better policy route, which starts with an understanding of the basic purpose of the rating agencies: to provide information (in the form of judgments, or “ratings”) about the creditworthiness of bonds and their issuers. If the information is accurate, it helps bond investors – primarily financial institutions, such as banks, insurance companies, pension funds, mutual funds, etc. – make better investment decisions. It also helps the more creditworthy bond issuers stand out from the less creditworthy. If the information is inaccurate, of course, it does the opposite. As an example of the latter, the major agencies had “investment grade” ratings on Lehman Brothers’ debt on the morning that it filed for bankruptcy. Luckily the large incumbent rating agencies are not – and never have been – the sole sources of creditworthiness information. Many large institutions do their own research; there are also smaller advisory firms; and most large securities firms employ “fixed income analysts” who provide information and recommendations to their firms’ clients.
The next step along this better policy route is the recognition that the centrality of only the three major rating agencies for the bond information process is a major part of the problem. This central role of the agencies has been mandated by more than 70 years of “safety-and-soundness” financial regulation of banks and other financial institutions, including insurance companies, pension funds, money market mutual funds, and securities firms. In essence, the regulators rely on the ratings to determine the safety of institutional bond portfolios. For example, bank regulators currently forbid (and have done so since 1936) banks from holding “speculative” (i.e., “junk”) bonds, as determined by the rating agencies’ ratings. This kind of regulatory reliance on ratings has imbued these third-party judgments about the creditworthiness of bonds with the force of law!
This problem was compounded when the SEC created the category of “nationally recognized statistical rating organization” (NRSRO) in 1975 and in doing so created a major barrier to entry into the rating business. As of year-end 2000 there were only three NRSROs to whom bond issuers could obtain their all-important ratings: Moody’s, Standard & Poor’s, and Fitch. (Because of subsequent prodding by the Congress, and then the specific barrier-reduction provisions of the Credit Rating Agency Reform Act of 2006, there are now ten NRSROs. But, because of the inertia of incumbency, the three large rating agencies continue to dominate the business.)
When this (literal) handful of rating firms stumbled badly in their excessively optimistic ratings of the subprime RMBS, the consequences were disastrous because of their regulation-induced centrality.
A better policy prescription would increase competition in the provision of bond information by eliminating regulatory reliance on ratings altogether. Since the bond markets are primarily institutional markets (and not retail securities markets, where retail customers are likely to need more help from regulators), market forces with respect to the provision of information about bonds can be expected to function well, rendering the detailed regulation that has been proposed (and partly embodied already in SEC regulations) unnecessary. Indeed, if regulatory reliance on ratings were eliminated, the entire NRSRO superstructure could be dismantled, and the NRSRO category could be eliminated, which would bring many new sources of information into the market and in so doing also increase the quality of information.
The regulatory requirements that prudentially regulated financial institutions must maintain appropriately safe bond portfolios should remain in force. But the burden should be placed directly on the regulated institutions to demonstrate and justify to their regulators that their bond portfolios are safe and appropriate – either by doing the research themselves, or by relying on third-party advisors. Since financial institutions could then call upon a wider array of sources of advice on the safety of their bond portfolios, the bond information market would be opened to innovation and entry in ways that have not been possible since the 1930s.
The “Issuer Pays” Business Model of the Major Credit Rating Agencies
The politically popular proposals for expanding the regulation of the credit rating agencies (as well as the SEC’s recent regulations) are devoted primarily to efforts to increase the transparency of ratings and to address issues of conflicts of interest. The latter arise largely from the major rating agencies’ business model of relying on payments from the bond issuers (an “issuer pays” business model) in return for rating their bonds.
Again, the underlying urge to “do something” in the wake of the mistakes of the major credit rating agencies during the middle years of the decade of the 2000s is understandable. Further, the “issuer pays” business model of those rating agencies presents obvious potential conflict-of-interest problems that appear to be crying out for correction. But the major credit rating agencies switched to the “issuer pays” model in the early 1970s (they previously sold their ratings directly to investors – an “investor pays” business model); yet the serious problems only arose three decades later. The agencies’ concerns for their long-run reputations and the transparency and multiplicity of issuers prior to the current decade all served to keep the potential conflict-of-interest problems in check during those three intervening decades.
In the decade of the 2000s, however, this reputation-based integrity eroded. The profit margins on RMBS instruments were substantially larger than those on ordinary debt issuances, and the issuers of RMBS were far fewer than the thousands of issuers of “plain vanilla” corporate and municipal bonds. This made the threat by a RMBS issuer to take its business elsewhere unless a rating agency provided favorable ratings far more potent. Also, the RMBS instruments were far more complex and opaque than “plain vanilla” corporate and municipal debt, so mistakes and errors (unintentional, or otherwise) were less likely to be noticed quickly by others. And the major credit rating agencies, like so many other participants in the RMBS process, came to believe that housing prices would always increase, so that even subprime mortgages – and the debt securities that were structured from those mortgages – would never be a problem. The result? A tight, protected oligopoly became careless and complacent.
In many ways, it was “The Perfect Storm.”
Even so, this storm would not have had such devastating consequences if financial regulators had not propelled the three major agencies into the center of the bond markets, where regulated financial institutions were forced to heed the judgments of just those three.
The Dangers of Expanded Regulation of the Rating Agencies
The dangers of expanded regulation of the rating agencies are substantial. They require the SEC to delve ever deeper into the processes and procedures and methodologies of credit judgments. In so doing, such expanded regulation is likely to rigidify the industry along the lines of whatever specific implementing regulations the SEC devises. It is also likely to increase the costs of being a credit rating agency. Expanded regulation will discourage entry and impede innovation in new ways of gathering and assessing information, in new methodologies, in new technologies, and in new models – including new business models. Even requirements for greater transparency, such as more information about the rating agencies’ methodologies, rating histories, and track records, could have adverse consequences if they force the revelation of proprietary information about the modeling and thereby discourage firms from developing new models.
Further, expanded regulation may well fail to achieve the goal of improving ratings. One common complaint about the large agencies is that they are slow to adjust their ratings in response to new information. This criticism surfaced strongly in the wake of the Enron bankruptcy in November 2001, with the revelation that the major rating agencies had maintained “investment grade” ratings on Enron’s debt until five days before that company’s bankruptcy filing. More recently, as mentioned above, the major agencies had “investment grade” ratings on Lehman Brothers’ debt on the morning that it filed for bankruptcy. But this sluggishness appears to be a business culture phenomenon for the incumbent rating agencies that long precedes the emergence of the “issuer pays” business model.
As for the disastrous over-optimism about the RMBS in this decade, the rating agencies were far from alone in “drinking the Kool-Aid” that housing prices could only increase and that even subprime mortgages consequently would not have problems. The kinds of regulations that have been proposed (as well as those already implemented) would not necessarily curb such herd behavior. The incumbent rating agencies are quite aware of the damage to their reputations that has occurred and have announced measures – including increased transparency and enhanced efforts to address potential conflicts – to repair that damage.
The harm to innovation from restrictive regulation is illustrated by the experience in another field: telecommunications regulation and the development of cellphone technology in the U.S. Although cellphones could have been introduced in the late 1960s, restrictive regulation held them back until the early 1980s. Cellphone usage didn’t really flourish until the mid 1990s, when a less restrictive regulatory regime took hold.
The Way Forward
The rating agencies’ promises to reform their ways are easy to make and could fall by the wayside after political attention shifts to other issues. Consequently, enforcement mechanisms are necessary. The rating agencies’ concerns about their long-run reputations provide one potential mechanism. But that mechanism proved too weak in the near past, so something stronger is needed. Expanded regulation of the rating agencies (to address the transparency and conflict of interest issues) is certainly another potential route – but the dangers, as outlined above, are substantial.
Expanded competition among current and potential providers of information about the creditworthiness of bonds and bond issuers is a third – and preferable – route. New competition could come from the smaller bond advisory firms or from advisory firms in other parts of the securities business (e.g., in December 2009 Morningstar, Inc., which is known primarily for its assessments of mutual funds, announced that it would begin rating some companies’ bonds). Competition could also come from some of the fixed income analysts at large securities firms who might (in a less regulated environment) decide to establish their own advisory companies, or from new entrants that no one has ever heard of before. Since the bond markets are primarily institutional markets, the bond managers of the financial institutions in these markets can be expected to have the ability to choose reliable advisors. Expanded competition would be enabled by the elimination of regulatory reliance on ratings, and enhanced by a reduction in (or, ideally, an absence of) regulation of the bond information advisory/rating process.
This withdrawal of regulatory reliance on ratings must be accompanied by an enhanced approach by prudential regulators of banks and other financial institutions in how they enforce requirements that their regulated financial institutions maintain appropriately safe bond portfolios. In essence, the regulators must place the burden for safe bonds directly on the financial institutions, thereby replacing the regulators’ current delegation (or, equivalently, outsourcing) of the safety decision to a handful of third-party rating agencies. The financial institutions could do the research themselves, or enlist the help of an advisory firm, which could be one of the incumbent rating agencies or a new competitor. The prudential regulators would have to maintain surveillance of the advisory process; but the primary focus would be on the safety of the bonds themselves.
The SEC has taken some recent steps in the direction of this third route by eliminating some regulatory references to ratings; but no other financial regulatory agency has followed the SEC’s lead.1 The SEC has simultaneously expanded its regulation of the rating agencies. The financial regulatory reform legislation (H.R. 4173) that was passed by the House of Representatives in December would eliminate legislative references to ratings and instruct financial regulators to eliminate reliance on ratings in their regulations; but it would also greatly expand the regulation of the rating agencies.
In essence, public policy currently appears to be two-minded about the credit rating agencies: The wisdom of eliminating regulatory reliance on ratings has gained some recognition; but the political pressures to heighten the regulation of the rating agencies are clearly formidable.
There is a better policy route than relying on the incumbent credit rating agencies to police themselves, or on the politically popular route of expanded regulation of the rating agencies. This better alternative would entail:
The institutional participants in the bond markets - with appropriate oversight by financial regulators - could then more readily make use of a wider set of providers of information. As a consequence, the bond information market would be opened to new ideas and new entry in a way that has not been possible for over 70 years.
1. However, in late 2009 there were two small steps in a favorable direction: In October the Federal Reserve announced that it would be more selective with respect to which ratings it would accept in connection with the collateral provided by borrowers under the Fed’s “Term Asset-Backed Securities Lending Facility” (TALF) and would also conduct its own risk assessments of proposed collateral; and in November the National Association of Insurance Commissioners (NAIC) announced that it had asked the Pacific Investment Management Company (PIMCO) – which is not a NRSRO – to provide a separate risk assessment of residential mortgage-backed securities that were held by insurance companies that are regulated by the 50 state insurance regulators.
Lawrence J. White
Lawrence J. White is Arthur E. Imperatore Professor of Economics at New York University’s Stern School of Business and Deputy Chair of the Economics Department at Stern. During 1986-1989 he served as a Board Member for the Federal Home Loan Bank Board, and during 1982-1983 he served as Director of the Economic Policy Office, Antitrust Division, U.S. Department of Justice.
The views expressed in this paper are those of the author and do not necessarily reflect the positions of the Roosevelt Institute, its officers, or its directors.