Do the credit rating agencies deserve to exist?

• Position: A SYMPOSIUM OF VIEWS

The proverb says it's no use locking the barn door after the horse is gone. But even if the door to the credit rating agencies can't be closed, should these institutions be disenfranchised, as many critics argue?

Do the rating agencies elevate or add to risk? Is the charge credible that these institutions have never been ahead of the curve in predicting the bursting of an economic or financial bubble? Should the U.S. Securities and Exchange Commission and similar international agencies disassociate from the agencies in the evaluation of risk?

Or are effective reforms possible?

Fourteen distinguished experts rate the raters.

[ILLUSTRATION OMITTED]

Yes, but eliminate their conflicts and have them report to the SEC.

MAURICE R. GREENBERG

Chairman and CEO, C. V. Starr and Company

The rating agencies are an integral component of the financial market. Done properly, their evaluations of credit risk are essential to many market participants who lack the resources or skill to make an independent evaluation.

The problem lies in the method whereby rating agencies are paid for their services which has changed adversely over the years. Once rating services sold their ratings to the purchasers of securities and were viewed as independent evaluators of risk. As securities became more complex, investment banks wanted to know in advance what would be the rating of the security before bringing it to the market. The rating agencies became engaged in the structuring of the securities and demanded to be paid for their efforts.

Investment banks soon learned to play one rating agency against another and only pay for the highest two ratings they would receive. The rating agencies, in order to maximize their own income, became co-originators of securities rather than the independent arbiter that was their original role. The rating agencies became reluctant to downgrade securities they helped to create. If institutions want or need the ratings supplied by the rating agencies, then they should pay for them.

The rating agencies must at all times be independent of the investment banks and originators of securities. If they are publicly owned, individual corporate investors should be limited to 5 percent of the outstanding common stock to avoid an appearance of conflict of interest.

In light of the important role they play in the functioning of the securities market, all of the ratings agencies should be subject to a self-regulatory body which would ultimately report to the Securities and Exchange Commission as do the self-regulatory bodies for the securities markets.

[ILLUSTRATION OMITTED]

They should be taken over by a public regulatory agency.

HEINER FLASSBECK

Director, Division on Globalization and Development

Strategies, UNCTAD, Geneva

Credit rating agencies should solve information problems and increase transparency. Indeed, they have played the opposite role and made the market even more opaque. As in all former crises, agencies were too optimistic. This is the systemic problem.

Rating agencies normally respond that their ratings include disclaimers that clarify that they are paid by the companies they rate and that ratings are only opinions and not accurate predictions of the risk of a given instrument. The problem is that rating agencies play an ambiguous role in the current regulatory environment as it renders rating decisions important in establishing what assets can be held by certain types of financial intermediaries.

A fundamental reform of crediting rating agencies and of their role in rating complex financial instruments is an indispensable step towards increasing transparency of the whole financial system. There is no private solution to this matter any more. What is needed is the establishment of a public regulatory agency which takes over the role of credit rating agencies. Thus, just as the Food and Drug Administration has to certify the safety of a given pharmaceutical product, such a non-partisan agency would certify that an AAA asset has indeed minimal probability of default and can be used by risk-averse investors like pension funds.

Given this, the main question today is whether this should be a national or supranational agency? If it is a national agency, should assets rated as AAA in a given country be considered as AAA in other countries? How would such an agency deal with political sensibility linked to rating sovereign bonds? While these are important issues, it is worth noting that three agencies (one in the European Union, one in the United States, and one in Asia) would cover the majority of the world's financial assets and this would be the case even if these agencies were not allowed to supervise the rating of sovereign issuers.

[ILLUSTRATION OMITTED]

Only if they become less opaque.

ROBERT E. LITAN

Vice President for Research and Policy at the Kauffman

Foundation and Senior Fellow, Brookings Institution

There is no question that the credit ratings agencies deserve much of the blame for the subprime lending fiasco and its current, unbelievable aftermath (although, to be fair, there is so much blame to go around). It is not easy, however, to figure out what policymakers should now do with them.

There is mounting support, for example, for removing the legal requirements that ratings be used--by money market funds, insurance companies, pension funds, and the like. Even were this done, however, there will continue to be a demand by both investors and securities issuers for some kind of rating by someone who is trusted. So some entities we call "credit rating agencies" will continue to exist, and as long as the market calls for them, they will "deserve" to exist.

The Securities and Exchange Commission could regulate the agencies more tightly in an effort to enhance their trustworthiness with market participants, but I am skeptical that second-guessers working for the government will be much better than those working at the agencies now. Likewise, I am skeptical that requiring different letter grades for asset-backed versus other securities will make much difference. And I fear that approving more agencies as "nationally recognized" would encourage more ratings shopping (which I think will go on even if the agencies are formally barred from providing consulting services to issuers).

I think the best that can be done to make the ratings process more effective is to require more disclosure by the agencies, not only of their ratings methods, but also the time periods of data used to assess default probabilities. Perhaps a big red star (or...