INTRODUCTION
THE ECONOMIC ROLE OF THE RATING AGENCY
HISTORY AND DEVELOPMENT
NRSRO
A NEED FOR REGULATORY REFORM?
A JUSTIFICATION FOR REFORM: POTENTIAL DISONANCE BETWEEN REALITY AND PRESUMPTION WITH RESPECT TO RATINGS QUALITY
OTHER POPULAR ARGUMENTS FOR REGULATORY REFORM
MONOPOLY
COST OF REGULATION
POSSIBLE SOLUTIONS
CONCLUSION

1. Introduction

Champions of corporate reform (as well as politicians and bureaucrats hungry for recognition) have trained their spotlight on the collapse of Enron, once the seventh largest firm in the United States,[1] and on the subsequent waves of corporate scandals.[2] How did the vaunted U.S. regulatory system engineered for the purpose of preventing such events appear to have failed so miserably? Indeed, Enron's problems had been apparent for some time before November 28, 2001, yet credit agency review of Enron on that date did not produce a change in the company's investment grade credit rating.[3] Lowering Enron's ratings would have triggered serious financial consequences for the company, which has led some critics to insinuate malfeasance.[4] Others, such as WorldCom (now MCI) and Global Crossing, also enjoyed high ratings prior to their collapse.[5] Lack of transparency, and a false perception of stability and growth-put forward by both company and government actors-is at the center of the maelstrom. If credit agency ratings are intended to be vehicles for enhancing transparency and protecting consumer interest, what does it mean that the agencies failed to signal these socially costly collapses?[6] Nevertheless, it would be hasty to conclude simply because the ratings failed to expose malfeasance occurring within these companies and to telegraph their collapses that: 1) the rating agencies were more at fault than other links in the accountability chain; or 2) that increased oversight of the rating agencies would prevent similar problems in the future. Nonetheless, the fact that the agencies gave the appearance of failing at their gatekeeper function in these instances is intuitively unsettling.[7]

In the wake of these revelations, visions of massive fraud undergirding American's corporate infrastructure quickly spread and, as a result, emotions ran high. In such a context, pragmatism tends to wither, with the effect being a short burst of energetic attention, often manifested as political grandstanding, as opposed to the development of long-term strategy.[8] There indeed has been substantive action, taking various forms, from the passage of the Sarbanes Oxley Act of 2002 ("SOA"), to the campaign waged by the office of New York Attorney General, Eliot Spitzer, which has been front and center in the pursuit of corporate wrongdoing over the past several years. These developments may be effective even if they only represent action for the sake of action. Regulatory reform to preserve confidence in the regulatory system may be a positive even if it is reform only for show but only absent any offsetting negative effects.[9] Moreover, it is not obvious, from a purely pragmatic perspective, what-if any-measures policymakers should pursue, notwithstanding emotive impulses to do something and anything. The specific focus of this paper is on this question: Are regulatory changes vis-à-vis credit rating agencies truly necessary? This paper will seek to parse the cacophony of competing opinions and organize in a coherent way the diversity of factors affecting the decision for or against reform.

Animated by popular concern regarding the rating agencies and perhaps a perceived need to be seen as acting affirmatively to maintain popular confidence in the integrity of the market and in the ability of the government to regulate it, legislators, through the SOA, tasked the SEC with examining the credit rating agencies. The SOA, Section 702, mandated that the SEC "…conduct a study of the role and function of credit rating agencies" and submit a report of its findings.[10] The original text of the act requires no affirmative action beyond the filing of a report 180 days after the effective date of the act.[11] Pursuant to this legislative charge, the SEC began to openly solicit advice from industry actors, government officials and academics. The SEC held hearings on the topic throughout the latter part of 2002 and ultimately produced a report, which provided a roadmap for further deliberations on the issue, but reached few, if any, conclusions about actual action to be taken.[12] Specifically, the report outlines the following areas of inquiry:

1) Information Flow

a. Whether rating agencies should disclose more information about their ratings decisions

b. Whether there be improvements to the extent and quality of disclosure by issuers (including disclosures relating to rating triggers)

2) Potential Conflicts of Interest

a. Whether rating agencies should implement procedures to manage potential conflicts of interest rat arise when issuers pay for ratings

b. Whether rating agencies should prohibit (or severely restrict) direct contracts between rating analyst and subscribers

c. Whether ratings agencies should implement procedures to manage potential conflicts of interest that arise when rating agencies develop ancillary fee-based businesses

3) Alleged Anticompetitive or Unfair Practices

a. The extent to which allegations of anticompetitive or unfair practices by large credit rating agencies have merit and, if so, possible Commission action to address them

4) Reducing Potential Regulatory Barriers to Entry

a. Whether the current regulatory recognition criteria for rating agencies should be clarified

b. Whether timing goals for the evaluation of applications for regulatory recognition should be instituted

c. Whether rating agencies that cover a limited sector in the debt market, or confine their activity to geographic areas, should be recognized for regulatory purposes

d. Whether there are viable alternatives to the recognition of rating agencies in Commission rules and regulation

5) Ongoing Oversight

a. Whether more direct ongoing oversight of rating agencies is warranted and, if so, the appropriate means for doing so (and whether it is advisable to ask Congress for specific legislative oversight authority)

b. Whether rating agencies should incorporate general standards of diligence in performing their rating analysis and with respect to the training and qualification of credit rating analysts.[13]

Doubtless, the forgoing list identifies important areas for consideration.[14] Of concern, however, is that if these points were translated to action, political considerations compounded by emotional zeal might obscure pragmatic cost-benefit assessment of proposed regulation.[15] Therefore, while the points of contemplation appear to target the critical issues, politics and other distortional factors might frustrate a smooth shift from abstract contemplation to useful action. Separately, this agenda, as it is set forth, suffers from some noticeable omissions.

In determining whether regulatory reform of the credit rating agencies is necessary, this paper focuses on whether the rating agencies are fulfilling their obligations, first, to private market investors in their role as private corporations, and, second, to society at large in their role as quasi-governmental organizations. This paper argues that increased regulatory oversight, or at least clarification of some of the relevant regulatory procedures relating to the credit rating agencies, is a necessary component of reformation of the broader corporate landscape. Therefore, the SEC should pursue such policy measures. This argument is qualified, however, by the acknowledgment that, at bottom, the positions that this paper advocates are driven by abstract reasoning, not empiricism. In advance of implementation of any reform measures, further empirical evidence needs to be collected and analyzed to determine the extent to which the market for credit rating is distorted as well as the extent to which such distortion originates with or are exacerbated by the credit agencies or the relevant features of the regulatory framework. This paper will focus only on the threshold issue: providing an analytical framework for considering the extent to which the source of the problem is located with the rating agencies.

The primary intuitions underlying this paper are as follows:

1) Even assuming, for the sake of argument, that the ratings failed to serve their presumed function of telegraphing impending failure in the instance of Enron, Orange County and others, it may be mistaken to allocate to the rating agencies all of the ills that precipitated these situations, an idea summed up pithily by the bromide: "garbage in, garbage out." In other words, if the problem is located earlier along the informational chain, the rating agencies may simply be responsible for perpetuating a problem that does not originate with them.

2) Assuming there is, at least to some extent, an informational problem, there still are concerns about the ongoing and future role of the credit rating agencies. The normative question still stands: should the rating agencies now assume a larger role (or a more limited one)? Or should another organization, such as the government, perhaps under the auspices of the SEC, assume a greater oversight role by erecting a shadow structure? SOA seems to have an aspirational component that says rating agencies may have to take on a larger (or more limited) role in the future.[16]

3) This paper also seeks to offer, as a corrective, the argument that even if empirical data were found to support the intuition that the market for credit ratings is distorted, the ultimate consequence of regulatory rejiggering in the present climate is unlikely to produce significant substantive change, at least with respect to protection of the individual investor. This should provide a reality check against costly and politically motivated window-dressing and motivate the public to assume a gimlet-eyed view of reform measures. The fact that the most significant regulatory maladies (at least those defined as such by magnitude of their impact) are not located at the level of the rating agencies and that political factors have focused attention on other areas (such as on investment banks and accounting firms) suggest that whatever changes are made are unlikely to produce meaningful system-wide improvements because regulators are focused elsewhere.

4) Perhaps most importantly, signs already exist that indicate the market may be a more effective driver of reform than public reordering.

5) Finally, however, despite signs of market driven reform, regulators must not become complacent. Market driven reform may lose steam and, thus, regulators must backup talk with actual vigilance and must generate a genuine appetite for pursuing substantive and pragmatic reform.

2. Paper Summary

To begin, this paper will set the scene, describing the interface between policy and the economy that fueled the development of the rating agencies and their quasi-official status as "Nationally Recognized Statistical Ratings Organizations" (NRSRO). The paper will address the apparent tautology inherent in the designation of NRSRO status, focusing on whether the barrier to entry that results from this designation has regulatory and competitive implications and to what extent these implications affect each other.

The question of whether regulatory reform is needed begs two principal questions:

1) Is there a disconnect between the agencies' expected role and their role in fact? One way of asking the same question is: Are ratings economically valuable given the price? Or, on balance, are they too costly, given the negative effect that they might impose (such as distortions in favor of overvaluing or over-trusting the ratings)?

2) If there is a disconnect or an overvaluation, is the disinformation attributable to the credit agencies themselves or to other actors upon which they rely for information? The second prong is important because in the highly politicized environment surrounding the inquisitions into corporate misfeasance, there exists a danger that those leading the charge and those defending against it may collude and seize upon an easy target, namely the rating agencies, using it as a fig leaf to deflect attention from deeper, thornier problems (a common motivating force for collusion between unlikely partners). Not only might reforms be ineffectual because they are driven by interests with parochial goals, but ineffectual reform might raise the cost imposed on society, since change might give a false sense of improvement.

In an effort to determine whether expected value of the ratings comport with the actual value of the ratings, we must ask: 1) Do consumers understand the relative value of the ratings as merely a component, and perhaps a somewhat unreliable one at that, of the total mix of information? 2) Are customers are sophisticated enough to understand the ratings? 3) Have the pricing mechanisms been distorted by exogenous factors? For example, Moody's Investor Service (Moody's) and Standard & Poor's Rating Service (S&P) have become household names resulting in their services becoming an indispensable part of a securities offer. As such, has the cost of the ratings (misplaced trust in inaccurate ratings in certain instances) exceeded their value?

In an effort to protect their current position in the regulatory fabric, the agencies claim that "ratings are really not that important" and that they simply "add to the mix" of total available information in the market, so that to the extent that inaccuracy exists, the effect on the market is limited. By the same token, the rating agency argument claims that they serve a public good because they do provide some information. Notwithstanding this appearance of doublespeak, agencies punctuate their argument with the claim that ratings are ipso facto of value because people are willing to pay for them, and any added regulation would distort this market signal and inappropriately obscure demand.[17] Nonetheless, if evidence shows that consumers overvalue the product they are purchasing, or that regulatory ordering produces a form of rent extracted by the NRSROs, or shows any other form of abuse of the public trust, then the market is overpricing the ratings and regulatory reform is needed.

Other important questions are: Is there a conflict of interest with respect to the internal structure of the agency or with respect to payment arrangements? Have regulators purposefully or inadvertently created barriers to entry for competitors and do these barriers create excessive regulatory costs? The agencies are often accused of duopolistic control of the sector. This allegation is focused on the NRSROs, which reputedly, because of the obscurity concealing the requirements for access to the NRSRO club as well as other attendant issues, have presented a significant barrier to entry. Is this a fair indictment and, if so, is dismantlement of the duopoly necessary? But is it possible that increased competition alone may exacerbate rather than reduce regulatory concerns? Opening the field to competition may not be the salve needed in this instance and, as such, it is important to ask whether increased public administration is needed and what form it should take.

There is danger in placing too high of a premium on change for the sake of change. That is not to suggest that such changes are totally without value, but the value may be limited to the extent that change serves to reinforce popular confidence in the regulatory system rather than inherent value of the particular change made. The danger, in the instant case, would be to overvalue the substantive element of the change proposed, and conflate the value of "change for the sake of change" with the substantive value of the change itself. These coefficients - the effect of change and the substance of the particular change made - must be disaggregated in order to properly assess whether regulatory reform should be pursued and the extent to which it should be pursued. It is critical that the substantive value of the change is not overstated as efforts conducted pursuant to such an analysis could distract focus from other problem centers, and, as a result, "agency capture"[18] and other distortional forces could drive reform in the wrong direction.

This paper concludes that the sector will likely continue to grow organically and this that natural development may result in the industry self-correcting. The history of the credit rating agency industry evidences grass-roots, market-driven development, as opposed to top-down legislatively prescribed development. Recent industry developments also support this hypothesis. Therefore, although this paper predicts that politics and other distractions may confound meaningful remediation by regulators, particular developments bode well for quiet progress within the credit rating sector.

1. The Credit Market

Contemplation of the role and future of credit rating agencies is a worthy undertaking if only because the credit market has increasingly powerful effects on world economies.[19] Since the 1970s, credit markets have been the driving force in finance and have generated huge trading volumes, massive profits, and a potpourri of fascinating legal and regulatory issues. Although individual investors continue to pour money into equities through mutual fund investments, investors also have increased holdings of bonds.[20] Globalization has provided an additional force driving the rapid expansion of credit markets.

2. Regulatory Role

The importance of the credit rating industry is further compounded by their increasingly expansive quasi-official role. Increasingly, regulation looks to the credit rating as a regulatory measure or trigger for determining the extent to which a diverse set of demands will be imposed on various financial sectors.

3. Effect on Issuers

The rating agencies' activities produce significant effects. "A poor rating can drive up an issuer's borrowing cost or even put it out of business. For example, in February 1991, when S&P and Moody's downgraded a $1.1 billion debt issue by Chrysler, its interest costs jumped by $38 million a year."[21] Credit rating agencies also rate sovereign debt, which has significant implications for international government to government and business to government relations.[22]

4. Market Power of the Rating Agencies

At a time when security markets languished generally, the stock price for Moody's Corp., the parent of Moody's, rose 53% after the company began trading (after being spun off from Dun & Bradstreet in September 2000). Moody's financial performance has been strong, although some analysts have indicated concern that Moody's shares are overvalued. Ratings usage by issuers has increased. In recent years, the credit rating agencies have experienced increasing demand for their rating services. The expansion of the field of securitization has partially fuelled this demand.[23] Also, public regulation has increasingly looked to ratings as a means of increasing regulatory efficiency, increasing public institutional reliance on the rating agencies. For example, the Basel Committee, which sets standards for international banking regulation, has begun setting capital standards by relying on credit rating.[24]

Most reviews of rating agencies begin with such a section, outlining the historical development of the rating agencies.[25] Such an introduction is sensible in view of the fact that the rating agencies developed organically over time rather than in response to a clear regulatory mandate. As such, the question is not simply "has the regulatory system been designed properly?" But rather "how has this function changed over time and what has contributed to these changes?" To the extent that change has occurred as a result of factors exogenous to pragmatic policymaking, such as agency capture or other such distorting forces, these factors must be considered and ideally controlled for in reworking policy.

1. History - Origins

The most important feature of the history of the credit rating agencies is the apparent organic manner in which their role in society has changed and grown. Their significant regulatory function today was not developed in an instant or as a result of top-down pragmatic ordering. Rather they were swept into the regulatory framework through development of the debt markets and the subsequent need for its regulation. By the 1970s, rating agencies had become a prominent feature of the market system and were co-opted by regulators because their market function evinced popular confidence. In other words, they were grandfathered into the regulatory system, not manufactured by the system. This distinction is important insofar as it may have engendered complacency on the part of regulators with respect to implementing monitoring checks on the continued effectiveness of the agencies. From the beginning, regulators were largely unconcerned with the internal operations of the rating agencies, and they have shown a continued reluctance to open the Pandora's box that might result from looking too closely at the inner workings the rating agencies. For regulators, the rating agencies have been the proverbial sausage factories, which, as Bismarck admonished, are better left unexamined.

The credit rating system emerged as a private sector institution in the mid-1800s. Louis Tappan formed the Mercantile Agency, the first mercantile credit agency in 1841.[26] In the years that followed, competition sprang up in the form of R.G. Dun and Company and a company founded by John M. Bradstreet.[27] John Moody applied the rating method to bonds in the early 1900s.[28] Poor's Publishing Company, Standard Statistic's Company, Inc. and Fitch Publishing Company emerged within a short period. [29]

It is significant that the agencies' fee structures historically entailed selling information to investors, rather than charging issuers (today's practice). Some issuers objected to the "intrusion," but they were forced to provide the agencies with information because absence of information would affect their rating. The success of the credit ratings was predicated on their ability to develop and maintain reputational capital. [30] According to some commentators, the development of the NRSRO and regulatory pockets affecting the credit rating industry diminished the importance of reputational capital and with it, the effect of the private market monitoring mechanism. These commentators claims that this transformation produced distortional effects that have allowed the credit agencies to produce inaccurate ratings yet maintain their dominate market position.[31]

2. History - The 1970s

Reform brought by expansion of the debt market, incidents of market failure, and regulatory responses changed and, according to some, compromised the useful role of the credit rating agency.[32] After changes in the 1970s, the credit rating business model shifted. Today, about 95% of the agency's revenue derives from fees to issuers, as opposed to charges to investors for the information provided.[33] Fees are predicated on the size and complexity of the issue. Some have argued that this has produced a conflict of interest, or at least an inappropriate appearance of conflict (discussed in detail below).[34]

This shift in the business model and billing structure may have been driven by the often-coterminous public goods and collective action problems-the "tragedy of the commons." Information is an archetypical public good. Because circulation of information is difficult to restrict once it is released into the marketplace, it becomes cheaper to forbear from paying for the information in the first instance and to wait until it is freely available on the market or available more cheaply. The potential for such free-riding precipitates a collective action problem, where nobody is willing to pay the cost for the information up front because they can get it for free if they wait. The consequence is that nobody pays for the service and the service is not provided, even though aggregate net benefit would accrue from the existence of the service. This traditional collective action problem is compounded further by a decrease in marginal value of the information to the individual investor as bond holdings have become more dispersed. This creates a relatively wider disparity between the value of the information and the cost of the information. Together, the free-riding and collective action problems arguably suppress the profitability of the rating agency and seem to provide a salient explanation for why the rating agencies reorganized their fee-structures in the 1970s.

Although the aforementioned pressures may exist and may have been a principal force in directing change in the rating agency business model in the 1970s, the mere presence of these pressures is not explanatorily dispositive. First, although the collective action problem may have been exacerbated by increasing dispersion of bondholders (which would decrease the marginal utility for each individual holder of seeking information unilaterally), and the free rider problem might have been exacerbated by advent of new technology that aided the efficient and illegitimate distribution of information (e.g. photocopying), to a significant extent these problems existed before the 1970s. Yet credit rating agencies retained this payment structure and survived for years before the 1970s. More important, until today, various organizations-some of which claim not to be rating agencies but derivative entities engaged in distinct business practices, but which do appear to provide services similar, if not identical to the major rating agencies-still base their business model on a membership fee system.[35] Arguably the difference in business models can be attributed to scale, and indeed these new organizations tend to be comparatively smaller than the major rating agencies. On a small scale perhaps information can be more easily controlled and customized, thereby diminishing the free-riding potentiality and attendant costs. It is also possible that larger firms are more exposed to free-riding and information theft because they are better known, more trusted, and their information is more widely dispersed.

A separate and perhaps more compelling explanation for why the large rating agencies have restructured their business model is captured in Partnoy's thesis in "The Siskel and Ebert of Financial Markets?: Two Thumbs Down for the Credit Rating Agencies."[36] His argument is that the large rating agencies are not selling information (or even providing very useful information), but rather are selling a property right in the form of a name brand. The stamp of approval from a major rating agency has become a condition subsequent for any bond issuance. Due to the leverage that emanates from this de facto requirement, the main profit center for the rating agencies has shifted to its contracts with issuers and away from its sales to investors.

Finally, exogenous factors seem to have had a role in the shift from charging investors to charging issuers in the 1970s. One of these factors was the default of Penn Central on $82 million of commercial paper in 1970.[37] Commercial paper had previously not been rated under the assumption that investors could trust any commercial paper issued by a known firm. However, issuers of commercial paper began to solicit ratings to lower their capital costs and revitalize confidence in commercial paper investment.[38] Here is an example where the sudden perception of greater risk drove industry innovation. The risks associated with commercial paper may not have changed, but market perception had, and consequently so had the risk premium on issuance of this financial instrument. As a result, the market niche for charging issuers expanded.

1. Emergence of the NRSRO and Implications for the Sector

The SEC's creation of the NRSRO designation in 1975 had a transformational effect on the industry and especially to those rating agencies that the SEC deemed worthy to carry the vaunted mantle. The term "NRSRO" was originally adopted by the Commission in 1975 solely for the purposes of bolstering Rule 15c3-1. [39] The NRSRO function allowed the SEC to rely on the credit ratings to implement the net capital requirements ("haircut" requirements).[40] The advent of the NRSRO nomenclature represented the veritable knighting of selected rating agencies, transforming them into fixtures of the US financial system's regulatory architecture. As Partnoy indicates, the NRSRO function has been incorporated into hundreds of rules, releases, and regulations. The "resulting web of regulation is so thick that a thorough review would occupy hundreds, perhaps thousands, of pages."[41]

The incorporation was a reflection of substantial public regulatory deference to these private actors. As Amy K. Rhodes notes "in essence the SEC grandfathered the existing rating agencies into a designation without a systematic review of predetermined criteria."[42] Furthermore, Rule 15c3-1 proposed no clear definition of NRSRO, no formal monitoring mechanism, nor any clear method for applying for NRSRO status. The SEC has subsequently used the informal and opaque no-action letter process, which is generally intended as a means for clarification, as the principal mechanism for awarding NRSRO status.[43]

The ill-defined NRSRO has blossomed organically from its obscure origins as an esoteric component of Rule 15c3-1 into a regular feature of regulatory parlance, extending even beyond the restrictive bailiwick of the regulatory agencies to which it was initially confined. For example, Congress uses the undefined term in its definition of a "mortgage rated security" as legislated in the Secondary Mortgage Market Enhancement Act of 1982 as amended, which required that mortgage related security be rated in one of the top two categories by at least one NRSRO.[44]

The NRSRO designation has become a profitable asset for rating agencies. As described, the market for ratings has expanded increasingly in recent years and the NRSRO role has experienced concomitant expansion, providing the anointed credit agencies with access to new business opportunities. Many of these business areas have grown to represent the most profitable areas for the companies concerned.[45] The SEC initially recognized three existing credit rating companies as NRSROs: Moody's, S&P, and Fitch (those not named have disappeared or been folded into these three in the years after 1975).[46]

The incorporation of the private sector into public regulation offers obvious benefits, especially from an efficiency standpoint in that federal regulators need not duplicate work already done, and often done better by the private sector, if demand has supplied sufficient capital to pay for the work.[47] However, drawbacks are equally obvious.[48] If the private sector is not doing what it is supposed to, then duplicative work would be necessary. Regulatory actors' incentives might, unfortunately, be skewed toward presuming that the private sector is doing its job so as to conserve public resources and avoid attracting public scrutiny. This is only likely to change when the public starts to become suspicious over regulatory laxity, which often happens only after a public crisis or scandal. Then, in response to public pressure, relevant regulatory actors may tend to overcompensate and expend excessive resources because the motivating force is public pressure, with "optics" the goal rather than pragmatic policymaking.

2. Effects of NRSRO Recognition - Frank Partnoy's Reputational Capital Model

The SEC did not intend the NRSRO mechanism, upon its inception, to be a device to regulate the industry. Rather the SEC intended originally to create a vehicle for minimizing costs of regulatory oversight. Yet some, such as Partnoy, argue that because the NRSRO designation has inadvertently assumed a gatekeeper function restricting entry, it has turned the designation into a valuable property right. Subsequent to the knighting of a few rating agencies as NRSROs, the market began to value the designation as an independent good (a stamp suggesting a level of quality), and distort the market in favor of the NRSRO designated rating agencies. Strangely, therefore, a tautology developed whereby the NRSRO designation was awarded to those the market favored, but the market soon favored those awarded the NRSRO designation. This tautology was reinforced by express SEC requirements that indicated accession to the NRSRO club required something called "national recognition,"[49] which even exponents admit is defined by terms that are "vague."[50] This development distorted the market check as NRSRO designation developed inherent value.

The NRSRO designation may be at the heart of the problems surrounding the rating agencies, not because, as is often alleged, the NRSRO designation has created a anticompetitive monopoly in the traditional sense (repressing competition and raising barriers to market entry), but rather because the designation has created a market for "regulatory licenses" and weakened the "reputational capital" check.[51]

Partnoy outlines this dynamic in detail.[52] He presents a narrative describing how rating agencies originally developed a market niche and were able to thrive on the basis of their past accrued and, critically, current retention of reputational capital.

Rating agencies were able to accumulate reputational capital during the 1920s because they were able to gather and synthesize valuable information…. Every time an agency assigned a rating, that agency's name, integrity and credibility were subject to inspection and critique by the entire investment community.[53]

Due to changes over time, a dynamic has emerged in which their credibility now derives not from their accretion and retention of high levels of reputational capital, but rather from their NRSRO designation and regulatory role. The tipping point, Partnoy argues, was the SEC's enshrinement of rating agencies through the advent of the NRSRO nomenclature, which served, perversely, to corrode market vigilance with respect to ratings' quality. Partnoy diagrams the economic explanation as follows:

If the applicable regulation imposes costs and a favorable rating eliminates or reduces those costs, then rating agencies will sell regulatory licenses to enable issuers and investors to reduce their costs. Just as, according to the reputational capital view, rating agencies will sell information until the marginal cost of acquiring and transferring information exceeds the marginal benefits from issuer fees, so according to the regulatory license view, rating agencies will sell regulatory licenses until the marginal cost of acquiring and transferring regulatory licenses exceed the marginal benefit from issuer fees."[54]

Once the NRSRO system was established and regulatory licenses became the asset most valuable to the rating agencies, reputation vis-à-vis investors lost its effect as a market check, and reputation vis-à-vis the SEC informed the regulatory body's willingness to persist in validating the companies' vaunted position.[55]

Yet the SEC has not been a vigilant monitor. Partnoy continues his argument by outlining how this dynamic has produced agency capture, which has in turn perpetuated the dynamic. Partnoy contends that "if the approved raters can earn sufficient profits to pay regulators - indirectly through campaign contributions or even directly though bribes - raters may be able to remain in a position of profit and power indefinitely."[56] A further refinement of this argument is the acknowledgement that rating agencies would not necessarily need to engage in such explicit negotiations to preserve the status quo. Change is costly and from a regulator's perspective it is easier to keep things as they are, both economically and politically. Supporting and preserving the status quo allows bureaucrats and politicians to focus limited resources elsewhere as long as the public believes that matters are being effectively handled. Moreover, indirect and unexpected political costs might result from churning the waters. Thus, there is great pressure towards inaction.

In sum, the NRSRO designation has become a powerful label that affects the private market for credit rating agencies even as it simultaneously serves its separate intended public function.[57] The fact that the NRSRO designation boosts rating agency market share reflects a perverse psychology and a logical loop in which regulators and the public alike have deferred to the other to determine the integrity of the rating agencies. Nobody is spending much time looking under the hood. To wit, following the regulatory perspective to its logical conclusion, we arrive at the suggestion that if a rating agency's work-product once reflected care and integrity (upon its achievement of NRSRO status), it will and does continue to reflect these qualities and no further regulatory check is necessary. This logic is flawed in view of the moral hazard that such a presumption invites. In the absence of a regulatory or market check, the agencies are left to operate with impunity.

3. Barriers to Entry, Monopoly and the NRSRO

The number of credit agencies is few. It is generally recognized that three actors, S&P, Moody's and Fitch continue to control the field, with Fitch being the least important of the three. [58] Two new entrants, IBCA and Thomson Bankwatch, have received limited NRSRO recognition. Since the Enron debacle, a collection of articles has surfaced alleging apparent duopolistic sector capture. However, it is important to recognize that the limited number of players, and any monopolistic pressures that have produced this phenotype, may be exogenous to whether regulatory reform is necessary to improve the quality of ratings. However, the limited number of active players presents a problem of optics and a potentially distracting fig-leaf on which regulators may focus at the expense of truly necessary measures.

Consideration of the potential problem of conflating causation and correlation should be preceded by an inquiry as to why the number of players is so limited. Professor Lawrence J. White of the Stern School of Business has several theories regarding why the number of credit rating agencies is so limited despite the fact that bond rating is so profitable.[59] One issue he raises is that companies impose structural limitations on themselves in an effort to avoid conflicts or the perception of conflicts of interest. As such, some potential competitors have been reluctant to break into the sector even though they may be well-suited for it from a business perspective.[60] He speculates that when the SEC grandfathered Moody's, S&P and Fitch in as NRSROs in 1975, it was satisfied with the regime they had set up and was unwilling to expend further resources to create rigorous and fair oversight of subsequent applications to entry. A derivative point, although not one that White discusses, is that these vaunted companies were given an early "boost" by this regulatory decision, which provided them a competitive advantage that compounded over time and has increased entry costs to such a point that the economic attractiveness of attempting to enter the field at this point is severely diminished.

Further, White acknowledges that the economic and regulatory realties of the sector may more readily support a fewer number of large players in the light of the desire for only a few standardized ratings (assuming that the ratings are indeed standardized). Past regulators have articulated similar sentiments supporting this point.[61] White also attempts to justify the difference between fixed-income and equity instrument ratings by suggesting that equity instruments may be considerably more complex (since the extent of gain and loss are important considerations) "and [therefore] investors may well be more open to varied opinions from many sources, none of which command sweeping authority." The weakness of this last point seems to be in its presumption of uniformity of the rating process within each rating agency. That is, White seems to assume that ratings are consistent and standardized and therefore there is no need for a diversity of opinions on a given security. However, if the rating process is highly variable, and if a rating reflects an abundance of subjective factors that make each iteration through the rating process unique even within a single rating agency, the distinction that White draws between the market for rating of equity and rating of debt seems a spurious one. Finally, White does not explore Partnoy's contention that there has been a historical shift from a reputation capital to a regulatory license regime and how that has served to produce market distortion and agency capture.

Partnoy claims that the fact that so few rating agencies exist provides support for his thesis that the agencies are not operating in a world where reputation matters. His claim is not inconsistent with White's recognition that barriers to entry exist, but it explores further the underlying reason for the existence of these barriers (i.e. agency capture). Partnoy's contention is predicated on an argument that such market dominance would be economically impracticable in the face of competition if forces found in a free market (most critically those forces emanating from the presence of reputation capital) were at work. [62] Notwithstanding whether Partnoy's allegation of regulatory capture has explanatory force, regulatory barriers to entry exist, thus explaining to some extent the limited number of players. LACE Financial, for example, has been angling for NRSRO recognition for eight years. As LACE president Barron H. Putnam testified in the recent rounds of the SOA mandated SEC investigation, "LACE Financial's application for NRSRO stature was before the SEC for eight years and then denied for no clear reason."[63]

Barriers on a logistical level are due to regulatory opacity, and reform might begin with an effort to improve transparency.[64] But the existence of this opacity begs the question: is it the cause or effect of barriers to entry and do problems, like Partnoy suggests, result from deeper and more insidious circumstances such as agency capture and an agenda to protect the dominate corporations? NRSRO recognition criteria were most unclear in the early years after the advent of the NRSRO role. Initially, recognition was determined on a case-by-case basis. However, through a subsequent no-action letter process, the Commission staff developed a number of objective criteria for assessing the primary criterion of NRSRO recognition to determine whether the company had "national recognition."[65] Unfortunately, the subsequent clarifications provide little refinement of this amorphous term. As discussed, a strong tautological quality is inherent in the definition of NRSRO. In practice, the NRSRO admission process is defined by an impenetrable set of guidelines with which interested companies must, in theory, familiarize themselves ex ante. This vagueness poses a clear barrier to those who wish to accede to the NRSRO ranks. Fundamentally, to become recognized as an NRSRO, the rating agency must have the sort of industry presence only achievable under present circumstances if one is already an NRSRO.[66] Thus, it is no surprise that those agencies that are designated NRSROs are those that achieved the status from the beginning. The one notable exception of the recent NRSRO assignment of IBC seems a political gesture or symbol of slow change but not an emblem of a major paradigm shift.

4. Is Reform of the NRSRO Accession Process Necessary to Improve the Quality and Oversight of Credit Ratings?

An affirmative answer to this question would suggest that current regulatory opacity is either causative of lax oversight of the credit rating agencies or symptomatic of a deeper problem (like the one Partnoy identifies) so that reform in this area would serve to resurrect the effects of reputational capital on the operation of the industry. However, this intuition is not obviously supported by evidence and therefore should not be assumed a priori. The need for reform will be addressed further below. This section will simply outline the NRSRO approval process as it appears to stand currently.

Again, the critical question is: to what extent are these regulatory barriers causative of derogation of credit rating quality? Although some critics point to these barriers as dispositive evidence of quality control problems, the connection is not inherently obvious and such a leap is reflective of sloppy logic. The most direct problems resulting from NRSRO entry barriers are possibly: 1) from the competitor perspective - competitive equity and economic costs for competition and other rating agencies or 2) from the purchaser perspective - reduced consumer choice and possibility of higher costs due to lack of competition. Whether or not these factors are directly correlated with ratings quality is not obvious and should not be assumed.

Regulators appear ready to tighten (or at least pay lip service to the need for tightening) regulation of the NRSROs even in the absence of unambiguous indication that the NRSRO institution has itself produced regulatory slips. As one SEC commissioner states: "the designation of NRSRO status should be the beginning, and not the end, of the SEC's oversight." Another commissioner acknowledges that "[n]o action letters designating agencies as NRSROs leave it to them to self-police their own activities. But because of the financial impact that withdrawal of NRSRO designation could have on a rating agency, the NRSROs have a strong disincentive to report any such change in circumstances."[67] The solution the commissioners proffer is more monitoring: "To accomplish this the SEC's inspection staff should schedule more frequent examinations of NRSROs"[68] There is a logical problem with this reform proposal. If the agencies are relying on the NRSRO for a regulatory function, can they monitor the rating agencies effectively without being duplicative? Conversely, if such efforts are duplicative, might the regulators not be better off performing the regulatory function independently, rather than deferring to the private market? In other words, are there efficiency gains had by effectively monitoring the regulatory function performed by the private sector, or does this structure pose a logical or structural fallacy?

1. Analytic Framework - Arguments for Regulatory Reform Regulation of The Credit Rating Agency

Justice Stephen Breyer, in Regulation and Its Reform, provides material for constructing a theoretical model against which to assess the value of regulatory reform. His book identifies important factors for consideration when contemplating regulatory change.[69] Regulation is costly in absolute terms and often too little is obtained in return for these large expenditures.[70] Also, there is danger of implementing regulatory procedures that are unwieldy. Furthermore, the procedures might be unfair.[71] Breyer also points out that the regulatory process has been criticized as fundamentally undemocratic and lacking legitimacy due to unpredictable and even random effects (implicating both fairness and efficiency critiques). Given the potentiality of wastefulness, unfairness, and consequential social costs, increased regulation may produce, on balance, more cost than benefit.

From a theoretical perspective, possible arguments in favor of regulation are manifold and are predicated on a diverse set of criteria. Historically, regulatory oversight has been imposed or tightened under a diverse set of circumstances. Breyer provides an inclusive list of reasons for ramping-up regulation. First, a common reason for buttressing regulation is in response to the emergence of monopoly power. On the other hand, monopoly power is not an ipso facto ill. Some industries cannot efficiently support more than one firm. For example, it is progressively cheaper to supply extra units of electricity or telephone service and, as such, it is more efficient to grant monopoly licenses to single companies to provide these services ("natural monopolies"). Economies of scale are sufficiently great that unit costs of service would rise significantly if more than one firm supplied service in a particular area. Rather than have three connecting phone companies laying separate cables where one would do, it may be more efficient to grant one firm a monopoly subject to governmental regulation of its prices and profits.[72]

Breyer justifies the need to control monopoly power through traditional arguments of macro-economic efficiency.[73] Separately, he outlines three additional and derivative reasons for regulating monopoly power, which are: 1) to reduce income transfer; 2) to increase competitive fairness; and 3) to reduce the firm's social and political power.[74] Separate reasons for regulation include rent control and limiting excess (and therefore efficiency degrading) profits.[75] Another reason for regulation derives from a desire to compensate for what Breyer calls "spillovers," i.e. externalities. Negative externalities have justified public regulation to reduce the externalities or pay their cost. Similarly, positive externalities ("as when free public education is argued to have societal benefits far exceeding the amount which students would willingly pay for its provision")[76] often justify public subsidization if private incentive to produce the good or service is insufficient. Breyer's other rationales include: 1) the need to remedy the effects of imperfect information; 2) the inability of the market to meet the supply/demand equilibrium because of the impact of distortional forces on the market; 3) the presence of moral hazard and its resultant effects; and 4) paternalism and the desire to affect private ordering and social policy.[77]

Moving from abstraction to rating agencies specifically, arguments for regulation are equally diverse: from complaints of monopolistic or duopolistic control to complaints of informational inadequacy. Most concerning, however, is that arguments for regulatory tightening have often not been clearly made or justified against a clear analytical model that sets out a precise regulatory goal as well as the specific means by which the proposed regulatory reform is to effect this goal.

1. Does Excessive Pricing and Informational Ambiguity Exist Vis-à-vis Rating Agencies?

In considering whether pursuit of regulatory reform of the rating agencies would be a wise decision, it is important to first consider the most obvious question: is there anything wrong with the quality of the credit rating system as it presently exists? Critics argue that the credit rating oligopoly is pernicious because prices are excessive and the information the credit ratings provide about themselves and about the issuers they rate is qualitatively poor. In response, defenders argue that these concerns are unfounded because rating agency prices are not excessive and rating agency information provision is not clearly insufficient, which would foreclose use of these arguments to justify reform. For example, Professor Schwartz of Duke University Law School, arguing that regulation would not increase efficiency, discounts excessive costs claims and rejects Partnoy's position that the reputational capital mechanism has been retarded by historical developments.[78]

It is not obvious that incidents of "ratings failure"[79] indicate that quality is below expectation and that informational ambiguity is endemic. Assuming arguendo that, 1) the market is completely efficient, 2) Partnoy's model does not hold, and 3) the cost of the ratings is lower than the value of the rating if the value of the rating reflected a 100% accurate representation of reality, then we can assume from a theoretical perspective that the ratings will not be one hundred percent accurate given the implicit price discount. That is, we may assume that the ratings, in practice, both ought to and do provide a less than perfect mechanism for presaging default or company collapse. Accordingly, sometimes default may occur despite a healthy rating, and that this fact militates against using evidence of disconnect between an incident where default occurs despite a healthy rating to support an argument that reform is necessary. Were ratings acknowledged by the market (in view of the discount) to be imperfect as in this hypothetical, it would be unfair to assume that the ratings would always serve to prevent significant loss on the individual level, although they could still serve a somewhat more limited individual function as well as a public function by reducing aggregate loss over time. This is simply a way of saying that purchasers are getting what they are paying for, and that this is not necessarily bad for purchasers or for society.[80] Were the following evidence available, it might vindicate this point and shift it from the world of hypothesis to the world of empirical support for current credit rating agency behavior. If the historical number of "ratings failure" incidents is consistent with probabilistic expectation, this would mean that there would be expected instances of disjunction between rating and reality, assuming probabilistic expectation of default were greater than zero. Assuming that there is no normative requirement that rating agencies predict every failure, rating agencies would be acting in a just manner - selling their services "as advertised" for a fair price.[81]

2. Possible Existing Contributory Factors to Excessive Pricing and Informational Ambiguity

The hypothesis explored in the forgoing paragraph is well and good, but how do we know if purchasers are indeed getting what they are paying for? How could excessive costs be determined? One way to begin this inquiry would be to posit that to the extent that the ratings serve a private function, the cost of ratings should equal the informational value of the ratings to the purchaser. Admittedly, the individual price-value comparison may be only a baseline and therefore inadequate given that ratings serve a macro-economic market function as well. That is, NRSRO status implies that the ratings are to serve a public function as well as a private one to the extent that their value exceeds their private collective cost. Furthermore, it may also be that NRSRO status itself affects and amplifies the rating value and therefore cost beyond the value of the information provided in the rating.

What might be the cause of excessive pricing if it indeed exists? In theory, excessive pricing may be a product of the regulatory license regime outlined in Partnoy's model (purchasers are not buying information, but a property right). Other arguments approach the issue from a different perspective and assert that excessive pricing is a product of the existing duopoly. These arguments are not mutually exclusive, however; duopoly may be a consequence of the regulatory license regime. In any event, if excessive costs result from the duopolistic position of the two, or because the two major rating agencies are selling more than just packaged information (perhaps something akin to rent extraction), or if excessive costs result from any other reason, regulatory adjustment would be necessary since at the very least monopoly rents reduce efficiency and, perhaps worse, the regulatory license regime, if it exists, implies a situation in which market forces are not working efficiently to regulate credit agency work product.

Separately, if excessive costs result neither from the existence of a regulatory license regime nor from monopoly pricing but rather from the purchasers' misapprehension of the value of the rating as a result of imperfect information about the rating agency product, regulation should also respond to this fact. In other words, if as a result of distortional forces, purchasers presume incorrectly that the ratings contain more information than they actually do, it is the responsibility of regulation to respond. (That purchasers may not know what they are paying for has been an oft-mentioned concern[82]).

3. Extrinsic Evidence Supporting a Presumption of Excessive Pricing and Informational Ambiguity

Ratings are too costly if rating agencies are not doing what they purport to do. If there exists a disconnect between regulatory presumption and reality or market presumption and reality regarding the value of ratings, then the pricing mechanism is distorted and it is possible that regulation should be adjusted to compensate. If consumers are paying excessively for a product then it would appear that they believe they are purchasing more than they in fact are. The following tripartite analysis should be used to parse this question: 1) what information is conveyed, 2) how is the quality of that information interpreted by markets and by regulators; and, 3) what is the relative level of sophistication of those interpreting the information.

The credit rating system has limited informational content and application. This fact may be understood by the sophisticated investor, whereas the unsophisticated investor may believe the credit rating to be more valuable than it actually is. A resulting disparity between classes of investors might skew ratings prices. If evidence suggests that that the majority of those who pay attention to ratings are unsophisticated purchasers, we may be more likely to be motivated by paternalistic sentiments than if evidence suggests that the majority of those who use ratings are sophisticated purchasers who use the information only as a reference point in connection with other available information. Carrying this analysis a step further, the following issues arise:

1) What is the Rating Methodology and what Information is Impounded in the Rating?

A common frustration relates to the degree of obfuscation surrounding the agencies' inability to clearly explain the credit rating methodology. Each agency seems to employ a slightly different analytical system, but these differences do not appear substantive according to general information put out by the companies. Further, while each is willing to explain the process and methodology, none is able to define precisely the quality of information examined or the precise manner in which it is processed. The justification commonly put forth for this apparent lack of transparency is that the process is ultimately subjective.[83] Agencies, however, are careful to emphasize the balance between subjective and objective information for fear of betraying incompetence or irrelevance; they argue that their product is objective enough to be of value to the market and the individual, but too subjective to allow the rating agencies to clearly outline the rating process.[84] Kenneth Leh, a professor of business administration at the University of Pittsburgh (and an advisor to Moody's), has concluded that only "75% of the ratings process is based on statistical information and equations and that 25% is subjective."[85] The lesson here is that the rating is not an exact refraction of the all the available information in the market. It is not analogous to the credit spread[86] because there is an interpretative and therefore subjective quality to the rating. Is this difference valuable or does it add, instead, negative distortion and decrease value? On one hand, the subjective quality may add value that cold hard information leaves out. On the other hand, the subjective element may be confusing and ultimately may create more costs than benefits.[87] At the very least, however, it is a source of confusion and promotes informational ambiguity.

What is clear, although not always understood by investors, is that the type of information covered by a rating is limited. The credit rating covers only credit risk[88] and does not include interest rate risk, market risk, or event risk, all of which can also affect the value of the bond. Therefore, even on its face, the credit rating by itself is an insufficient tool for assessing the prudence of an investment.[89] Adding to the confusion, rating agencies have refused to clearly lay out the precise type of information they consider and the manner in which they process this information.[90] An optimistic justification for this vagueness is that every situation demands a unique approach.[91] Ratings, the industry has urged, are important[92] merely to "lower the aggregate costs of borrowing and lending and increase overall market transparency and efficiency for both issuers and investors"[93] but not so important that inaccuracy is fatal to their role because they are simply one factor to be considered as a part of a "total mix" of information. [94] A less optimistic explanation is that the process is not systematized because it is inherently flawed and produces inaccurate and only marginally useful information. Indeed an oft-cited frustration of credit rating agency critics is the agencies' lack of transparency.[95] The admittedly partially subjective analysis undertaken by the rating agencies combined with the potential (although arguably not prolific) use of insider (private and/or proprietary)[96] information for ratings analysis shrouds the process in an impenetrable fog of uncertain practices and irregularity behind which rating agency activities have significant leeway to operate in ways that might be inconsistent with regulatory expectations. Hence, through this self-description, rating agencies are able to disclaim significant responsibility for rating error.[97]

Another source of confusion is whether the ratings anticipate changes in risk profiles or simply respond to information telegraphed through the market. Credit rating, in theory, mitigates problems resulting from information asymmetry, thereby reducing transaction costs. [98] Specifically, ratings reduce individual investors' information costs by reducing the need for independent information gathering which is more costly on a unit-by-unit basis for the individual than for the rating agencies. Consequently, there is more transparency in the market due to the availability of this information and therefore less uncertainly. As such, the market demands a lower risk premium on bond issuances, reducing the cost of capital for issuers.[99] One key presumption underlying this theoretical framework is that the ratings reflect information not already impounded in the market price before the rating.[100] Partnoy rejects this premise, stating, "… Federal Reserve Bank into the movement of sovereign bond yields indicates that yields typically decline several days before the agencies act on a rating, suggesting that the agencies lag behind the market."[101] This evidence, if accurate, may imply that bond ratings are less efficient than alternative market indicators like credit spreads, discussed below. Alternatively, this evidence may also suggest that ratings may act as catalysts of market shifts, rather than vice versa. It is a safe presumption that a company would issue new bonds and subject itself to rating agency scrutiny only if it thought that such scrutiny would yield a good rating, which would result in, among other things, a lower risk premium.[102] Such rating might increase the value of the company's other securities, including bonds, on the secondary market and might also create a presumption of low risk for future issuances by the company. Even in the absence of evidence that the bond ratings themselves affect the market price of the bonds, credit rating agencies may provide added value even if ratings provided no "new information." For example, the rating might serve to package information in a way more accessible to individual or smaller investors. Thus, simplicity of presentation is the value added. Since securities markets, especially bond markets, are still moved by large players that contain in-house analysis departments,[103] ratings may simplify and package the information in a way that helps explain the reasons for market movement, which could be valuable to the smaller investor. In other words, the credit ratings may provide a useful filter for those who are not equipped with the resources or do not have the time to collect and assess all available information in light of market movements. This argument, of course, assumes that small investors seek justification and understanding as opposed to simply responding to market movements and invest accordingly. It also assumes that the small investor understands the informational limits of the credit rating.

The agencies themselves, especially under the current pall of recrimination, have issued significant disclaimers about their service, suggesting a level of guilty defensiveness. S&P asserts that their credit ratings are based principally on public information as well as "other economic, financial and industry information that rating analysts deem relevant and reliable."[104] Further the 2002 S&P report before the SEC states that S&P "does not perform an audit for the rated company or otherwise undertake to verify the information provided by the company; nor does S&P audit or rate the work of the company's auditors or repeat the auditor's accounting review."[105] S&P insists that its rating business is predicated on "the full and fair disclosure regime mandated by the U.S. federal securities laws."[106] Finally, the report reads, "[c]learly the events of the last year have demonstrated the consequence for all market participants, including the credit rating agencies, when companies fail to meet their disclosure obligations, or worse - set out to defraud investors or rating agencies."[107] This victim talk suggests a thinly veiled shell game. By asserting that it can be accepted into the ranks of harmed investors, S&P gives off the appearance of engaging in a defensive tactic to avoid being portrayed as the bad guy.

Contrary to some arguments, the fact that there are two major rating agencies, instead of one, does not seem to have put pressure on the companies to clarify their processes, nor does it suggest that their ratings are more accurate. Some have argued that a second credit rating provides additional information and reduces borrowing costs even if the two ratings are the same.[108] The intuition here, presumably, is that the rating agencies are at least somewhat constrained in their activity if there is a competitor conducting the same rating.[109] In other words, in the absence of competition there is more moral hazard and the incentive for the rating agency to expend fewer resources and engage in a less rigorous analysis of the subject is greater because consumers of ratings can't comparison shop. With two rating agencies, it is more likely that both rating agencies will actually examine the data, lest one does not and produces a rating produced with the proper care; if one takes care and the other does not, there could be an obvious divergence in what is supposed to be an objective process of interpreting the same or similar data. This proposition, however, suggests a darker corollary, which regards the potentiality of explicit or implicit collusion or signaling, whereby one firm may simply copy or respond to the rating produced by the other firm. Both firms have the incentive to uphold the status quo and are unlikely to blow the whistle on the other firm (lest it expose its own poor behavior).[110]

2) Regulatory Presumption of Information Quality

The agencies describe broadly the nature of the information that they incorporate into their analysis. Or more precisely, they describe the nature of information that they might incorporate into their analysis but resist describing under what circumstances what sort of information will be examined. Why is there so much ambiguity surrounding the informational quality and value of the credit rating? Are the agencies engaged in a shell-game, successfully pressing self-characterizations that have lead regulators to place them under an unusually lenient legal regime? Indeed, many have attributed the success of the rating agencies to a few regulatory features and legal interpretations peculiar to the agencies. These features have also been the source of considerable criticism, which critics have marshaled against the agencies. Many argue that in order to liberate the sector and bring the agencies into line with regularity norms, these features should be altered or eliminated:

Critics have argued that the exemption from Regulation Fair Disclosure (Regulation FD) for credit rating agencies represents an improper regulatory subsidy since there is no reason that rating agencies should be given this competitive advantage. Moreover, the exemption introduces significant ambiguity into the market and, consequently, risk for investors.

The Regulation FD exemption provides agencies access to propriety and/or private information without requiring that such information to be simultaneously released to the public as is normally the requirement under Regulation FD.[111] The exemption was granted as a result of lobbying.[112] The exemption reflects the intuition that since the purpose of Regulation FD is to level the playing field between sophisticated and unsophisticated investors,[113] and since the rating agencies use the information they acquire not for their own investment purposes, but rather to pass on to others, it comports with principles of equity to permit rating agencies to be exposed to inside information without requiring the same disclosure required of others privy to the same information. The rating is a public good and all investors have an equal opportunity to benefit from it. Moreover as Glenn Reynolds, CEO and founder of CreditSight, in his testimony before the SEC points out, "the Reg FD exemption is an opportunity for the agencies to take a very activist approach when major credits start to unravel in the market as we have witnessed with disturbing frequency this past year."[114]

It is clear, however, that private information is not always sought and incorporated into the ratings. Barron H. Putnam, president of LACE, stated during his testimony before the SEC:

I always worry about confidential information. It concerns me a great deal. I try and dissuade companies from giving it unless - unless it's very material…. If they feel that somehow this information might significantly affect or be a major part of going into the rating. I just don't want to know it, if, in fact, it's not clearly something that would affect their rating significantly"[115]

If agencies are admittedly skittish about seeking undisclosed information, we cannot assume that information relevant to the rating is necessarily impounded in the rating. Lack of clarity as to whether undisclosed information has affected the credit rating creates noise and enhances the amount of uncertainty in the system. Related to this problem is that it allows rating agencies to further avoid enhancing transparency. It allows them to argue that the public has an interest in allowing them to continue operating in secret because they transmit otherwise publicly unavailable information. But because of this structure, rating agencies are allowed to operate under a veil of secrecy. Indeed, credit rating agencies may have successfully convinced the public that the emperor's clothes are made of the finest cloth.[116]

A major criticism leveled at the agencies and the regulators with respect to Regulation FD is the potential informational confusion it introduces into the market. The criticism derives partly from the general criticism that it promotes lack of clarity vis-à-vis rating agency due diligence and accountability. It appears inconsistent that they then do not necessarily seek and/or use this information that would benefit the marketplace "and support the confidence levels of institutional investors and intermediaries in the rating system."[117] Reynolds offers that "…the hope would be that the agencies would be very demanding in their informational requirements for more volatile issuers." He states further, "[w]e would hope that the active pursuit of this exemption did not just reflect the desire to gain a margin of safety advocated by rating agencies' counsel to keep the agencies and their issuing clients away from any risk of a 'violation.'"[118] Indeed, the agencies seem to be speaking out of two sides of their mouths. Generally, their defensive position is that they are not to be blamed for misrepresenting credit quality in most instances because they depend on public information for their ratings. Yet, as Reynolds asserts, "they should, at a minimum, make it clear when that is the case. Otherwise, the market could assume more is being done."[119]

Reynolds's criticism is not without its own weaknesses. Private or proprietary information can also be fraudulent and access to insider information does not ensure that the agencies will be privy to damning information. Therefore, even if the agencies swore to always seek and incorporate inside information, such a promise may not in fact reduce uncertainty. In other words, the Regulation FD exemption may be too broad and abstract to "confuse the market" in any practical way. It simply suggests that rating agencies are legally allowed access to information provided by the company that others are not. It does not allow the agencies to break and enter, or search and seize, nor does it guarantee that agencies will become omniscient.

It is worth noting that Reynolds's point comes from his perspective as a competitor with the agencies. He alleges that the exemption provides the agencies with a competitive advantage which they should "pay for" by adopting more assiduous rating techniques that would provide them a better nose for sniffing out foul play.[120] Reynolds conclusion is that "[t]he risk of the Reg FD exemption and the manner in which the rating agencies communicate the rating review details sends confusing signals to the market." Again, we must return to the dominant questions to assess the validity of Reynolds's criticism: does the private market demand and assume that the credit rating agencies are conducting more due diligence than they are? Are the markets paying commensurately for the level of due diligence the credit rating agencies are performing? Phrased differently, does the market overprice the value of the ratings given the lack of clarity regarding the inclusion of private information in the generation of ratings?

A further criticism leveled at regulators is their unwillingness to subject agencies to Section 11 liability under the Securities Act of 1933. Amy Lancellota, senior counsel for the Investment Company Institute, argued in her 2002 testimony before the SEC that NRSROs should be legally accountable for their ratings, and that this legal accountability should be achieved by applying Section 11 coverage to credit rating agencies. Lancellota argues that the agencies have enjoyed an "exemption" and would, absent the exemption, be subject to Section 11 liability.[121] Her critique seems to exhibit a somewhat naive reading of Section 11, its intended uses, and its historical application by courts, however. More importantly, perhaps, broadening the purview of Section 11 beyond its historical limits so as to subject rating agencies to Section 11 liability would only subject a narrow swath of ratings to increased liability. Of some interest, however, is the fact that the SEC has not been entirely dismissive of this proposition, suggesting that it may carry more salience that it appears to at first blush.[122]

Historically, Section 11 applies to those who signed the registration statement (the issuer and its chief executive, financial and accounting offers); all the directors at the time of filing (whether or not they singed); the underwriters of the offering; and any expert who consents to such expert's opinion being used in the registration statement, such as an accounting firm that auditioned the company's financial statements (the expert's liability is limited, however, to the information prepared or certified by the expert).[123] This exclusive list has been defined by case law.[124] Presumably, Lancellota and others in her camp, support expanding this last category to include ratings.[125] However, even with this expansion, Section 11 would only apply to a narrow range of ratings if the remedies accorded investors vis-à-vis ratings were consistent with those remedies currently accorded investors vis-à-vis other information sources currently subject to Section 11 liability.

The principal limitation is that Section 11 liability applies only to registration statements, and is subject to limitations developed through case-law. Escott v. BarChris Construction Co. establishes precedent for a civil remedy for purchasers in an offering of registered securities if a "material misrepresentation or omission" in the registration statement can be shown.[126] Critically, this remedy is under the 1933 Act, which essentially covers primary market transactions as opposed to the 1934 Act, which covers secondary market transactions.[127] Therefore, absent more significant restructuring of the securities laws, only ratings provided in the registration statement would be subject to Section 11 liability. Moreover, even ratings produced in this context would be subject to the dictates and qualifications of Section 11 and case law developed thereunder, which further narrows its application.[128]

Yet the rating agencies' counterargument anemically fails to point out these obvious flaws in the proposed regulatory adjustment. Rather than arguing the notable absence of case-law precedent for such an expansion of Section 11 and, more importantly, the legislative trend towards limiting the scope of private action motivated by a desire to reduce opportunistic securities litigation,[129] they take a completely different approach. They argue that they are members of the "media" that are providing their "opinions" as opposed to recommendations and both elements should absolve them of Section 11 liability. Although in the abstract the distinction between their identity as outside commentators and the identity of experts subject to liability is an important one under U.S. securities law, these arguments seem forced.[130] Their claim that they are simply members of the media stretches common-sense notions of what typically falls within the definitional ambit of that industry. Moreover, their claim that they are simply issuing opinions that should not subject them to liability since the market understands that these opinions are not "recommendations," a quality that triggers liability under some areas of securities law, also represents a dubious proposition.[131] On the contrary, the market (and investors) may improperly lump them together with the other financial service providers, inferring incorrectly that a similar level of liability adheres and thus overestimating the extent to which their opinion reduces the risk premium. Furthermore, contrary to their self-assessment, credit rating agencies, in terms of their internal structures, their fee structure, and their purpose, are closer to investment banks than media corporations. Finally, the absence of a "recommendation" element does not absolve those involved in the sale of securities of liability. Actors subject to Section 11 liability, such as company directors, experts, etc., could equally argue that they too are simply providing their opinions as opposed to their recommendations. This rhetorical distinction is not necessarily easy to parse, but given the level of definitional ambiguity between "recommendation" and "opinion-giving," their argument does not seem particularly compelling.

The death knell for the argument for expanding Section 11 liability or other areas of securities law for added protection may be the fact that legislation has trended towards reducing, rather than increasing, liability appurtenant to sales of securities. One of the most prominent recent indications of this effort, The Private Securities Litigation Reform Act (PLSRA) of 1995 was motivated by a desire to decrease opportunistic securities litigation by shareholders.[132]

Finally, from a normative standpoint, although greater exposure to liability would likely incentivize greater care in ratings issuance, it also may have a negative chilling effect. There is the potential for completely chilling the willingness of agencies to issue ratings for certain securities if the profit margins were to become too narrow. The would decrease the amount of available information about these securities and, depending on other factors, possibly result in a net loss of social welfare.

As mentioned in the forgoing section, rating agencies have turned to First Amendment defense against "fraud on the market"[133] or "market conditioning" legal theories. The rating is no more than an opinion or, more simply, a refraction of publicly available information (as well as some private information- maybe). As such, the rating agencies provide information about credit quality similarly to how the media provides information about world events. The latter may have significant effects on the behavior of observers, yet the promulgators of such information are not subject to liability absent incidents of libel or violations based on other legal theories of fraud. The argument that rating agencies are similar to the media, although perhaps facially a stretch, may find justification in a public policy driven intuition that constructive treatment of the rating agencies as media as opposed to something more akin to the an investment bank would better serve the interests of the investment public and regulators.[134]

The rating agencies claim that it is important that they not be inhibited in requesting information and thereafter subjecting that information to vigorous internal analysis and discussion. "In that connection it is critical that the courts afford shield law and journalist privilege protection to rating agencies so that rating agencies are not unduly burdened by third-party discovery and the confidentiality of their deliberative processes are respected."[135] Again, this argument reveals an effort to maintain a façade that keeps the public from discovering that the emperor has no clothes. Also there are traces of the argument that was trotted out by the securities industry against Regulation FD - that it would chill the collection of information. According to observers, evidence has not borne out this claim.[136]

3) Public Information?

The above sections discuss the ambiguity with respect to whether ratings simply reflect available market information or contain something greater (such as private or proprietary information). But until now the presumption in this paper has been that the ratings, at the very least, accurately reflect aggregated market data. However, even this presumption may be hasty as scholars have quibbled over whether ratings accurately impound public information.

For support that ratings fail to fully reflect public information, critics commonly point to instances where a pool of bonds have variant credit spreads but share the same ratings.[137] Exponents of the rating system defend the veracity of their system by arguing that distinctions between reality, market movements, and ratings can be rationalized by recognizing the mistake of looking at ratings as "point estimators rather than probabilistic distributions."[138] They argue that ratings reflect "relative credit worthiness. Not necessarily the absolute, but just the relative creditworthiness."[139] This argument can be bolstered by a recognition of credit rating letters as a rough cut, where each grade may represent a series of underlying values. But is this simply an excuse; is the grading system abstracted to a point of uselessness? If they act as a "relative" measure and if this relative measure is just one component that should be assessed in light of the "total mix of information," do these justifications beg the following questions: 1) If the value of the rating is so diluted, does it have value at all? 2) Does the value justify the cost (i.e. do consumers know what they are paying for in the majority of cases)? 3) Is there negative value to the ratings if consumers are confused over the purpose of the ratings - that they are "recommendations" to buy or sell, or that they reflect more precisely the underlying information?

The test of the extent to which ratings accurately reflect publicly available information should be an analysis of historical correlation between ratings and default rate. Credit rating agencies assert that they have supplied data that shows a statistically significant correlation between ratings and default.[140] S&P has stated, "simply put, AAA-rated companies don't default."[141] Similarly, Rhodes asserts unequivocally that the ratings are proven by empirical evidence to accurately reflect changes in credit risk.[142] Fundamentally, however, the rating process remains opaque and such claims of rating accuracy are difficult to test.

4. Sophistication Disparity Between Investor Groups and the Need for Paternalistic Regulation for Some Groups

Given the express limitation of the credit rating to address only credit risks and the ambiguity surrounding the quality and consistency of information impounded in the credit rating, the question of whether investors understand these nuances is paramount. Therefore, the question of whether rating costs are justified may be irrelevant if investors do not understand what they are buying. The relative sophistication of investors is thus a critical criterion when considering the function of the rating agencies. Sophisticated investors might understand the limitations and qualifications outlined in preceding sections,[143] but unsophisticated investors may not, which is a potential justification for further regulation.[144]

The form of regulation that this distortion would justify would be of the paternalistic sort. Even if there is a willing buyer and seller, we may still feel a level of concern if we believe that some subsection of the buyer group does not realize that it is not acting in its own best interest.[145] This paternalistic impulse can be supported by a fairness argument that our discomfort with the information asymmetry emanates from a normative assumption that because individuals are from time to time badly hurt due to discrepancy between ratings and reality, regardless of the objective predictability of such occurrences, the system needs to incorporate more protection against individual loss. These individuals, this argument says, cannot understand the extent to which this potentiality is costly (on an expected value basis or in real terms for those who are hurt). If individual investors are unwilling to pay for such protection, perhaps government should subsidize an alternative or complementary regulatory structure.[146]

Sophisticated investors seem to understand that ratings have limited value and application. For example, sophisticated investors often work in cooperation with the issuers, which helps them better understand the underlying information informing the rating and supplement that information.[147] Furthermore, according to industry insiders, the ratings play only a limited role in decisions made by sophisticated investors.[148]

What is the profile of the majority of rating consumers? If the sophisticated investors are investigating by interpreting ratings (or ratings alone), then the implication is that unsophisticated individuals are either using them in the total mix, or perhaps more problematically, inappropriately using them as a sole or principal criterion. Unsophisticated purchasers may not understand the limits of the credit rating and overvalue it.

1. Reform for Reform's Sake

Apart from the issues raised in the forgoing discussion, critics have highlighted other areas that provide evidence that supports reform of the credit rating agencies. Much of the above argument is predicated on an assumption of excessive pricing. However, few have argued that prices are in fact excessive. Absent evidence of excessive pricing, there are other reasons why the SEC should pursue reform.

The new political reality may militate in favor of reform irrespective of the precise substantive nature of the reform. In other words, reform may be of value simply because the unveiling of corporate crimes has upset the public and consequently Congress and regulators. Although this rationale may bear a Machiavellian gloss and carries a darker flip side that encourages costly and inefficient production of red-tape (which may encourage, in the long term, negativity towards reform generally), it does not vitiate the importance of factoring this when weighing the value of pursuing reform. The regulators are presently in the spotlight for having been asleep on the job and are under pressure to act to retain public confidence in the market. So, to some extent there is (and perhaps should be) a "show trial" feature to the current reform effort. Indeed the mandate of Section 702 of the SOA is merely to engage in investigative process.[149] The language does not require action beyond the initial inquiry. The SEC itself has not moved past its commitment to "consider." Therefore, the appearance of concern has been emphasized over the need for particular action.[150] This paper has emphasized the potential cost of action for action's sake, which is the possibility of creating a diversion. However, by the same token, action for action's sake may also have a salutary effect if it is pragmatic and measured as opposed to politically driven.

Yet the SOA is not simply an instrumentalist attempt to distract public criticism, but rather places discretion in the hands of regulators who presumably have some personal or professional commitment to systemic improvement. In the long run, transformation of the reform process into Kabuki theater will not resolve problems sufficiently.

2. Conflict of Interest

The credit rating agencies' payment structure implies a facial conflict of interest similar, in some respects, to Wall Street equity analyst and investment bank conflicts. This conflict of interest is also strikingly similar to accounting firms who have been pilloried in recent years because they are paid by the masters they are serving.[151] Indeed, the rating companies are paid by those for whom the credit grade is issued and the bulk of the rating agency revenue derives from this transaction. Some have argued that "the rating agencies' strong interest in preserving their reputation in order to compete effectively with other rating agencies and providers of securities research largely prevent exploitation of the fee structure."[152] Partnoy's argument directly contravenes this position and suggests that reputation would not be an adequate deterrent and the incentive for overrating would remain. Even if Partnoy is correct, are there other mitigating factors? Perhaps because of the NRSRO dynamic, reputation vis-à-vis the consumer is not such a concern to the rating agencies, but it is still a concern vis-à-vis lawmakers; government will start turning the screws if abuses are noticed. The reputational capital system between the rating agencies and the regulators, presumably, is far more elastic, however, than one with the consumers. Nevertheless, the rating agencies simply have no clear incentive to deviate from objective rating because of the payment structure. The rating agencies have responded along these lines:

We do not believe that the fact that the issuer pays a fee to Fitch creates an actual conflict of interest, i.e., a conflict that impairs the objectivity of Fitch's judgment about creditworthiness reflected in Fitch ratings. Rather, for the reasons stated below and based on our experience, it is more appropriately classified as a potential conflict of interest, i.e., something that should be disclosed and managed to assure that it does not become an actual conflict.[153]

Credit rating agencies present further arguments in support of the claim that no rating violation exists, arguing that "we will not forbear to move a rating regardless of the effect it may have on an issuer…." In addition, to further distinguish themselves from the investment bank example, representatives from the rating agencies have emphasized that compensation structures are distinct from those in the investment banking world that showed potential for compromised objectivity.[154]

The regulatory regime is just as interested in perceived conflicts of interest as it is in actual conflicts of interests because conflicts imply structural weakness, lack of transparency and accountability, and a system that consumers do not trust. As such, the credit rating agencies are more interested in the perception of conflicts than actual conflicts to the extent that regulators are concerned with this perception. Therefore, even absent a rational justification for how incentives produce actual conflicts, it is the perception of conflicts that the credit rating agencies must be concerned with. That said, the entire rating agency industry is predicated on this payment structure and it is an understatement to suggest that credit rating agencies would be loath to abandon or rejigger this system.

A more problematic conflicts issue is line of business conflict of interests. For example, conflicts resulting from the expansion of professional services firms into diverse business areas. Another example is when investment banks underwrite clients' stocks and then marshal analysts to publicly comment on the quality of those stocks. Rating agencies have sought to combat the optics problem by insisting that the internal bureaucratic structure of the organizations precludes a similar conflict.[155] They assert that the bureaucratic administrative structure, internal pay system, and strong corporate governance prevent conflicts from developing.[156]

Rating agencies have also claimed that their compensation policies do not incentivize manipulation in the way that compensation programs have in the investment banks' situation (where analysts received express and implicit pressure through a salary keyed to their ability to pump up stocks underwritten by the bank). A representative from Fitch rating states:

We also manage the potential conflict through our compensation philosophy. The revenue Fitch receives from issuers covered by an analyst is not a factor in that analyst's compensation. Instead, an analyst's performance, such as the quality and timeliness of research, and Fitch's overall financial performance determine an analyst's compensation. Similarly, an analyst's performance relative to his or her peers and the overall profitability of Fitch determine an analyst's bonus. The financial performance of analysts' sectors or groups do not factor into their bonuses.[157]

Fitch explains its safety measures further, specifically describing existence of Chinese walls or internal fire walls:

Fitch goes to great efforts to assure that our receipt of fees from issuers does not affect our editorial independence. We have a separate sales and marketing team that works independently of the analysts that cover the issuers. In corporate finance ratings, analysts generally are not involved in fee discussions. Although structured finance analysts may be involved in fee discussions, they are typically senior analysts who understand the need to manage the potential conflict of interest.[158]

These facts hold evidentiary value but are not dispositive. Even absent express pressure that results from linking salaries in various ways, there always exists "implicit" communication and compensation. Pressure to be a "team player" is ever-present within any organization.

Rating agencies are at least paying lip-service to the optics problem and have responded directly to line of business conflict of interest criticism. At the SEC hearing, an industry representative from Moody's asserted when pressed: "I think I can agree that consulting will not be a material part of our business." Whether reality will bear these promises out remains to be seen.[159]

Many have argued that the credit rating industry is controlled by a duopoly.[160] During the 2002 SEC consultations it was noted that "introducing additional NRSROs may also serve to increase the competitiveness of pricing and overall market efficiency."[161] Deconstruction of the monopoly or duopoly, if it exists, is likely to affect pricing, but "accuracy" is a different animal and the correlation to pricing may be nonexistent or even inverse. Further, as noted, Professor Schwartz agues that overpricing is not really a problem with respect to the credit rating agencies. As such, to the extent that the rating agencies market hold is symptomatic of a duopoly, it is still unclear that this dynamic contributes to the problems driving legislative concern under the SOA. As Justice Breyer notes:

"[M]any of those who demand regulation of rents may not be aware of the precise nature of the problem. They may see large producer profits and believe that they are monopoly profits or they may simply see prices rising and without inquiry into the cause, assert political pressure to bring about lower prices through regulation."

Critically, Breyer notes, "mistake, confusion, or political power may cause regulation but cannot justify it."[162]

The potential for new problems if the "duopoly" was dismantled is real. Competition could produce more irregularity rather than less. There is the risk, as the SEC notes, of "unintentionally creating a climate in which issuers of securities will shop around, to try to buy the highest rating."[163] Yet the SEC and political actors do not seem to take this risk seriously. Under the current climate, with only two or three real NRSROs, the opportunity for shopping is limited and the pressure on the companies to attract customers is also limited. Some empirical evidence exists to support the appropriateness of this fear. "There was some academic research done that indicated that the more marginal players and the new entrants tended to rate more highly than the established players.'"[164]

Therefore, market power shields the rating agencies from competitive pressures that might precipitate a race-to-the-bottom type atmosphere in which the bottom line might compel companies to begin engaging in activities in which conflicts are more apparent. If under the current climate there exists little incentive for rating agencies to push these limits, given the relative noncompetitive atmosphere, perhaps destruction of the present structure would bring greater harm than good.

This discussion of duopolies has been focused on social costs. But there are also arguments for the perpetuation of the present regime supportable as an information issue vis-à-vis the consumer. As Commissioner Goldschmid queries:

Why is it really better to have six or seven or eight to look at, than two or three, in terms of rating agencies? When I pick a movie, to use a perfectly imperfect analogy, I'd like to pick one review I trust and having seven or eight may make it more confusing.[165]

This academic discussion notwithstanding, if there is a legislative mandate to undo the monopolistic hold of the rating agencies, policy arguments are less relevant. Some have asserted that the SOA in fact mandates consideration of a monopoly existing within the rating industry separate from and in addition to the principal question at hand in this paper. "Although the subject concerning competition and its effect on the pricing of ratings is not part of the proposed questions of attendees, I believe it is one of the main reasons behind the Sarbanes-Oxley Act of 2002."[166] Held against the legislative text of the SOA, this argument is most likely a spurious one.

Yet whether the monopoly question should or must be considered as an issue in and of itself, the SEC has not been silent on its view with respect to whether undue industry protectionism exists. During the 2002 string of testimonies, a commission on the record states elliptically, the SEC has "not determined" that the national recognition requirement for NRSRO accession creates a substantial barrier to entry (as opposed to stating that the SEC has determined that the national recognition requirement does not…).[167] It has justified this position by suggesting that, "[g]rowth in the businesses of several credit rating agencies not recognized as NRSROs suggests that there may be a growing appetite among market participants for advice about credit quality from all credible sources, and that this makes it possible for new entrants to develop a national following for their credit judgments." Indeed Reynolds, head of CreditSight, proclaims that they have no intention of become an NRSRO. Some agencies have found their market niche by distinguishing themselves from NRSROs.[168]

Ultimately, despite popular opinions to the contrary, it is not obvious that increased competition would be a panacea for regulatory aims. If an immediate goal of a reform effort would be reinstatement of the reputational capital affect (to realign the interests of the agencies and the consumers), encouraging more entrants might exacerbate divergent interests and conflicts. Competition might not further incentivize the emergence of implicit arrangements between issuer and agency, which would work to the detriment of the consumer. There will be more companies competing for the issuers' business, encouraging more aggressive strategies and raising the risk of losing business to competition.

Regulatory tightening would likely mean increased oversight, which would likely necessitate duplicity of rating effort on some level. This would not be costless. As Breyer notes: "Further regulatory oversight might introduce new subjectivities to the rating process that could lead to the misallocation of capital."[169] Moreover, there is a catch-22 feature implicit in the proposed enhanced regulatory structure. To the extent that the regulators have differed from NRSROs in the past to play a quasi-official function, they have done so because the private sector is presumed to rate more efficiently. To the extent that the SEC must monitor the private sector and to the extent that the monitoring requires duplication of analysis, this efficiency savings is depleted. Separately, there exists between the private sector and the regulatory oversight body a misalignment of capability and authority, which would be exacerbated to the extent that regulatory oversight is increased. This misalignment can be understood as follows: the private sector, because of its resources, access to capital, and freedom from abundant public sector strictures (such as bureaucracy), it is likely to have the capacity to arrive at a more accurate rating. Government workers, on the other hand, are comparably less well paid, are perhaps not as expert, and may be under-funded. Consequently, the public sector and private sector rating results may not be entirely consistent. Moreover, the private sector is likely, on balance and absent certain distortional forces, to be more accurate. Nonetheless, the public sector's authority trumps the private sector's authority. The result of this dislocation is likely to be the production of greater inefficiency and market noise.[170] Another reason that regulatory tightening may not be the answer is because agency capture and politically driven reform may have the effect of distracting attention from needed structural readjustment.

Also, focusing on the rating agencies may be losing the forest for the trees. In the case of Enron, Orange County or other such notable moments where credit ratings appear to have been misleading, were there things that the rating agencies could have done differently to prevent such mistakes? Were the rating agencies privy to inside information about the complex structured financing schemes, and yet chose to ignore them because they feared sullying their hands? If the rating agencies were forced to reflect inside information in the credit rating rather than simply having the option to do so through the Regulation FD exemption, information provided by the credit rating could serve to supplement weak accounting, or even incentivize more rigorous accounting if the two industries - accounting and rating - were suddenly placed in a more competitive position relative to each other rather than a mutually reinforcing position.[171] As a practical matter, however, an affirmative obligation to reflect inside information in ratings would pose obvious enforcement and monitoring problems.

Further regulation may also produce a problem of regulatory asymmetry between bonds and stocks. For example, if the government were to assume or at least supplement the rating function, it would mean that equity was left to private market rating while debt was rated publicly. Depending on the distribution of resources between regulation of equity markets versus regulation of the debt markets, government agencies may be inappropriately subsiding debt issuance; that is, adding more muscular regulation may distort and lower debt prices relative to equity, and that may be in conflict with public policy.

Bondholders, in theory, should be able to reduce risk of imperfect ratings through the market. For example, bondholders could contract for further protection via the indenture with the company. Or if there is a greater need for paternalistic policy, state corporate law could mandate that the indenture contains risk reducing provisions to offset the ambiguity of the credit rating.

A cacophony of voices advocate different reform trajectories. In some cases, the old bromide "where you stand is where you sit" seems to hold true in this situation. Putnam, president the beleaguered LACE Financial Corporation, which has tried for eight years to achieve NRSRO status states in his testimony:

I believe the most important point I can make before the SEC Commission is that if a creditable rating company with the proper expertise applies for NRSRO status the SEC should help them through the application process. If the SEC staff feels that changes should be made in the applicants procedures or processes tell them in writing. If the company makes the necessary changes and meets SEC requirements then grant the company NRSRO status. The application process should be accomplished in an expeditious manner.

Those from the academic community, such as Partnoy, advocate massive overhaul - even dismantling of the credit rating cartel. He asserts that ratings are merely reflective of and responsive to public market information. They are therefore inefficient and should be replaced. Partnoy favors credit spread estimation over credit ratings.[172] The credit spread is the market's estimate of the risk of the bond compared to its risk-free counterpart, based both on the probability of default and the expected recovery in the event of default.[173] Partnoy claims: "[t]he fact that credit spreads of bonds in a particular rating categories change over time is strong evidence that the ratings do not contain valuable information." However, this claim is too categorical because ratings are generalized categories that imply a range, not point indicators.

The most obvious problem with abandoning ratings for credit-spread estimation is that the credit risk may cut a wider swath and incorporate information not available in the market. Given noise in the market resulting from imperfect information, the possibility of the existence of private information, and other market inefficiencies, the credit spread may not reflect the true extent of "risk" but rather reflect "available information" or the "understandability" of information to investors, many of whom may not be sophisticated. Market inefficiencies may also result in highly variable interest rates that are subject to exogenous influences. Here also the relative fixity or stability of the rating relative to mercurial market prices may prove of value.

By the same token, credit rating agencies' claim that "ratings have informational value because ratings are highly correlated with actual credit spreads" is a precarious claim.[174] To assess the accuracy of this claim, in theory, analysts must look at a credit spread sans the rating, since the rating itself may have effect on the spread.

But again Partnoy's position is too extreme. After all, the credit rating agencies are still kept in check to some extent by a reputational capital mechanisms running between the public and the rating agencies and more significantly between the rating agencies and the regulators (NRSRO status could always, ostensibly, be revoked by regulators). Moreover, empirical evidence supports that, historically, there has been some level of consistency between ratings and credit spreads. Most of the time, the mechanism works within a margin of error, otherwise we would have had more instances of bankruptcies generally (due to moral hazard generated by the absence of any functional creditor monitoring system) and more bankruptcies in specific circumstances where the companies' credit rating remained inexplicably high. Also, dismantling the regime would be costly. Making adjustments to the current structure would address the public confidence issue more readily. Most importantly, however, the size and influence of the credit rating industry is prodigious, and therefore a solution that entails dismantling of the regime entirely is not a particularly realistic one.

Various organizations, such as Egan Jones, put forth a diverse and pointed set of recommendations, many of which are measured and reasonable.[175] Yet these recommendations seem focused in ways that primarily serve the interest of the recommenders. Ultimately, the implied goal of regulatory reform is to create a regime that shifts the paradigm back to reputational capital market regulation. As discussed in the above section, it is unlikely that the situation is as dichotomized as Partnoy would lead some to believe. Reputational capital undoubtedly still has some effect. Moreover, the notion that opening the NRSRO accession process to greater outside scrutiny to improve the competitive environment may not be the correct tact. First, it is not clear that the NRSRO effect and name branding effect are coterminous. That is, other rating agencies may accede to NRSRO status yet fail to achieve the name branding of S&P and Moody's. The NRSRO designation may have at imputed to them the necessary elements to codify their name brand, but at present nobody is concerned with the pronouncements of S&P and Moody's because they are NRSROs, but rather because they have a name brand, and therefore market power. Moreover, the reason they achieved NRSRO status originally was because they had market power. Focusing on the NRSRO designation and opening up the NRSRO accession process may reflect confusion of correlation for causation and thus be a red herring. This is not to say that new regulation might not be necessary, but should be undertaken for competitive reasons, not for improving credit rating quality.

Also, it is not clear that increasing the number of competitors will directly achieve reinstatement of the reputational affect, which would realign the interest of the agencies and the consumers. In fact, competition might further incentivize the emergence of implicit arrangements between issuer and agency, which would work to the detriment of the consumer. There would be more companies competing for the issuers' business, encouraging more aggressive strategies and raising the risk of losing business to competition.

Finally, the market itself may provide the better mouse-trap. As noted, Egan-Jones Rating Co., rose to prominence and to a competitive position as a result of its aggressive criticism of the rating agencies and its position as a kind of check on the sector.[176] A myriad of smaller organizations have emerged filling a similar market niche. They act in some ways as an intermediary between the investing public and the rating agencies.[177] Correlatively, increased attention on corporate governance has shown a spotlight on the rating agencies, putting them in a more precarious position and requiring that they modify their policies. Assuming they have the resources to marshal, and that the whole rating agency concept as it currently exists is not entirely smoke and mirrors, such a nudge may put enough of a premium on increased diligence that massive structural reform would be unnecessary. Evidence of this happening maybe borne out by the fact that whereas before the rating agencies seemed ready to expand into provision of other financial services, they now appear more reticent. Finally, even stock market skepticism might produce the necessary corrective to reign in rating agency irresponsibility. As noted above, analysts have suggested recently that Moody's shares are overvalued. Only the future will tell the significance of these potential market remedies, which may prove insufficient especially given the stock market's mercurial nature.

The goal of any regulatory reform focused on this sector ought, normatively, to be the reestablishment of market pressure on the operation of the credit rating agency. The best mechanism for the reintroduction of productive market pressure is not clear. What is clear is that, despite marginal value derived from reform for its own sake (i.e. for purposes of optics), blind regulatory tinkering may be more destructive than productive. Ultimately, any adjustments should be responsive to empirical evidence reflecting answers to questions raised in the body of this paper, not emotive reaction. Ideally, pragmatism instead of politics should drive reform. The danger is that to the extent political forces are unleashed to address this issue, they are likely to push for reform initiatives that serve simply to paper over problems or to overcompensate in ways that are destructive. We may, however, take solace in the fact that recently publicized scandals focused new attention on corporate governance reform, which may provide enough of a market corrective that further legislation or other measures would be unnecessary.

To the extent that the concern that regulatory distortions have eclipsed the important market check that reputational capital provides is valid, evidence that new organizations have emerged that pressure the market to place appropriate weight on reputational capital again suggests that the market may produce its own best medicine. For example, Egan-Jones Rating Co. (Egan-Jones), which unlike the large rating companies, sells ratings to investors, downgraded Enron to junk a month before the big agencies did. Egan-Jones has since risen to prominence and to a competitive position as a result of its prescient act.[178] Other similar smaller organizations that have no ambitions of NRSRO status, but rather seek to act as an intermediary between the investing public and the rating agencies, have found themselves filling this market niche.[179]

The fact that the market may provide the necessary fix does not mean that regulators should slip into complacency. Market remedies may fail. Moreover, there are certain areas that are obviously in need of tightening that private market forces will not handle, for example rating agencies should be subject to some liability for their ratings. The First Amendment argument discussed above is a specious one.[180] This problem could be easily remedied without challenging the legal order or incurring significant political costs: one easy way to do it would be to make NRSRO status contingent upon forsaking a claim to a certain level of first amendment rights. In other words, NRSRO recognition would be similar to broker-dealer recognition, and once the organization puts itself out as an NRSRO, certain responsibilities (and liabilities) would inhere.[181]

Critically, the crisis of confidence will have to be addressed if another Enron emerges and rating agencies are held to account more seriously than they in the past have been. Rating agencies may not have been defying a legal or normative mandate up until now, but their assertion of irresponsibly is unsettling. Their claim that they simply serve a narrower purpose than many assume and are simply purveyors of limited information is unlikely to hold up in the long term. Reynolds of CreditSight rightly points out, "[w]e would agree Enron lied to everyone, but the agencies had an insider's advantage that would have allowed them to play a more significant role in the protection of the market."[182] The rating agencies may not have been, compared to the investment banks, the focus of public outcry following Enron, but they were on a watch-list. Notwithstanding whether actual significant and substantive change is necessary, preventive action must be taken to buttress confidence before the next big corporate scandal breaks and public scrutiny focuses on the credit rating agencies. So far, they have successfully remained out of the spotlight. Without a better public relations effort, however, the credit rating agencies and their regulators could find themselves the focus of a witch-hunt when the next corporate scandal breaks.