1 UNITED STATES SECURITIES AND EXCHANGE COMMISSION 2 (REVISED 01-15-03) 3 HEARINGS ON THE CURRENT ROLE AND FUNCTION OF THE 4 CREDIT RATING AGENCIES IN THE OPERATION OF THE 5 SECURITIES MARKETS 6 7 8 PAGES: 1 through 244 9 10 The above-entitled hearing commenced, pursuant to 11 notice, at 9:00 a.m. 12 13 14 15 U.S. Securities and Exchange Commission 16 Washington, D.C. 17 18 19 Friday, November 15, 2002 20 21 22 23 24 Diversified Reporting Services, Inc. 25 (202) 467-9200 1 COMMISSIONERS: 2 Chairman Harvey Pitt 3 Commissioner Roel Campos 4 Commissioner Cynthia Glassman 5 Commissioner Harvey Goldschmid 6 Commissioner Paul Atkins 7 8 OTHERS PRESENT: 9 John McCarthy 10 Michael A. Macchiaroli 11 Larry Harris 12 Annette Nazareth 13 Bob Colby 14 15 PRESENTERS: 16 H. Kent Baker, Ph.D. 17 University Professor of Finance 18 Kogod School of Business 19 American University 20 21 Deborah A. Cunningham 22 Senior Vice President and Senior Portfolio Manager 23 Federated Investors, Inc. 24 25 1 PRESENTERS: (Continued) 2 3 Sean J. Egan 4 President 5 Egan-Jones Ratings Company 6 7 Gay Huey Evans 8 Director, Markets and Exchanges Division 9 The Financial Services Authority 10 11 Frank A. Fernandez 12 Senior Vice President, Chief Economist and 13 Director of Research 14 The Securities Industry Association 15 16 Malcolm S. Macdonald 17 Vice President - Finance and Treasurer 18 Ford Motor Company 19 20 Leo C. O'Neill 21 President 22 Standard & Poor's, a division of the McGraw Hill Companies, 23 Inc. 24 25 1 PRESENTERS: (Continued) 2 3 Glenn L. Reynolds 4 Chief Executive Officer 5 CreditSights, Inc. 6 7 Greg Root 8 Executive Vice President 9 Dominion Bond Rating Service Limited 10 11 Cynthia L. Strauss 12 Director of Taxable Bond Research 13 Fidelity Investments Money Management, Inc. 14 15 16 17 18 19 20 21 22 23 24 25 1 C O N T E N T S 2 PAGE 3 Session I: The Current Role and Function of 4 Credit Rating Agencies 8 5 6 Session II: Information Flow in the Credit 7 Rating Process 71 8 9 Session III: Potential Concerns Regarding the 10 Role of Credit Rating Agencies (Conflicts 11 of Interest/Abusive Practices) 127 12 13 Session IV Potential Barriers to 14 Entry/Regulatory Treatment 184 15 16 17 18 19 20 21 22 23 24 25 1 P R O C E E D I N G S 2 CHAIRMAN PITT: As I think everybody knows, for 3 about a century, credit rating agencies have been providing 4 opinions on the creditworthiness of issues of securities and 5 other financial obligations. 6 These opinions are, obviously, of great importance 7 to investors and other market participants, and they have an 8 enormous influence on the securities markets, all of which 9 has increased significantly, particularly with the increase 10 in the number of issuers in the advent of new and complex 11 financial instruments, such as asset backed securities and 12 credit derivatives. 13 I think the globalization of our financial markets 14 also has served to expand the role of credit ratings to 15 jurisdictions beyond the United States, where the reliance on 16 credit ratings largely was confined for the first half of the 17 Twentieth Century. 18 Today, credit ratings affect securities markets in 19 a number of important and critical ways. 20 During the past 30 years, regulators, such as the 21 Securities and Exchange Commission, have increasingly used 22 credit ratings as a convenient surrogate for the measurement 23 of risk in assessing investments held by regulated entities, 24 because of the quasi public responsibilities in assessing 25 investments held by regulated entities. 1 Because of the quasi public responsibilities of 2 rating agencies, the importance given to ratings by investors 3 and other market participants and the influence of ratings on 4 the securities markets, we announced last March that we would 5 engage in a thorough examination of rating agencies to 6 ascertain facts, conditions, practices, and other matters 7 related to the role of rating agencies in the U.S. securities 8 markets. 9 Since that time, the Sarbanes-Oxley Act has been 10 enacted and it directs us to conduct a study of rating 11 agencies and their impact, as well. 12 The hearing today and the one scheduled for next 13 week represent the culmination of months of effort by the 14 staff and the Divisions of Market Regulation and Investment 15 Management, as well as in the Office of Compliance, 16 Inspections, and Examination. 17 I want to thank the staff for their tremendous 18 efforts that they have expended over the past few months. I 19 would also like to thank the members of our panel for taking 20 the time to appear before us today. 21 With that, I will turn it over to Annette Nazareth, 22 the Director of our Division of Market Regulation. 23 MS. NAZARETH: Thank you, Chairman Pitt. We have 24 assembled this broad section of distinguished and experienced 25 panelists not only to discuss the increasing importance of 1 credit ratings in today's financial markets, but, also, to 2 facilitate a number of initiatives currently underway at the 3 Commission. 4 First, the Commission has been directed by 5 Congress, as Chairman Pitt said, to conduct a study of the 6 role and function of credit rating agencies in the operation 7 of the securities markets. 8 This directive was issued in Section 702 of the 9 recently enacted Sarbanes-Oxley Act of 2002, which also 10 directs the Commission to submit a report on the study to the 11 President and Congress by January 26, 2003. 12 It is our goal and expectation that these will 13 enable us to parse in a more meaningful way the areas of 14 consideration set forth in Section 702. 15 Specifically, these areas include the role of 16 credit rating agencies in the evaluation of issuers of 17 securities; the importance of that role to investors and the 18 functioning of the securities markets; any impediments to the 19 accurate appraisal by credit rating agencies of the financial 20 resources and risks of issuers of securities; any measures 21 which may be required to improve the dissemination of 22 information concerning such resources and risks when credit 23 rating agencies announce credit ratings; any barriers to 24 entry in the business of acting as a credit rating agency and 25 any measures to remove such barriers; and, finally, any 1 conflicts of interest in the operation of credit ratings and 2 measures to prevent such conflicts or ameliorate the 3 consequences of such conflicts. 4 Another goal of these hearings is to assist the 5 Commission in determining how the Commission's regulatory 6 framework should best be applied to credit rating agencies. 7 For over 25 years, the Commission has been using 8 credit ratings from nationally recognized statistical rating 9 organizations, known as NRSROs, in Commission rules and 10 regulations. 11 During this time, the Commission has considered a 12 number of issues related to credit rating agencies, including 13 the need for direct oversight authority over credit rating 14 agencies and the appropriate role of ratings in the federal 15 securities laws. 16 Nonetheless, there remain a number of difficult 17 issues, many of which are reflected in Section 702 of the 18 Sarbanes-Oxley Act, that the Commission would like to review 19 before formally considering possible changes to the rules and 20 regulations regarding rating agencies. 21 Our agenda today reflects these issues. 22 Before I introduce today's participants, I will 23 briefly mention some details concerning our schedule today. 24 This morning there will be two sessions. We will begin our 25 opening session with brief introductory remarks from each of 1 the hearing participants. 2 We will then discuss in some detail the current 3 role and functioning of credit rating agencies. This session 4 will last approximately 90 minutes, and will be followed by a 5 15-minute break. 6 The second session this morning will explore the 7 information flow of the credit rating process; that is, how 8 rating agencies obtain and use information in the rating 9 process and how and what type of information is communicated 10 publicly and privately to issuers of credit ratings. 11 At approximately 12:15, we will break for lunch for 12 one hour and 15 minutes and we will resume our program at 13 1:30 this afternoon. 14 The first afternoon panel, which will last an hour 15 and a half, will focus on concerns regarding the role of 16 credit rating agencies, such as the potential conflicts of 17 interest and abusive practices, and our final panel, which 18 will run from 3:15 to 4:45, will address potential barriers 19 to entry in the rating business and the appropriate degree 20 and manner of regulatory oversight of rating agencies. 21 As you can see, we have a very ambitious agenda to 22 cover in a relatively short time. 23 Now, with that out of the way, I would like to 24 introduce each of the participants and then circle back for 25 any opening remarks that they would like to make. 1 First, I would like to introduce the Commission 2 today. Of course, Chairman Pitt, who we would very much like 3 to thank for calling for these hearings. 4 We also have Commissioner Atkins, Commissioner 5 Goldschmid, Commissioner Glassman, and Commissioner Campos. 6 We also have staff here today, as well, and while 7 we do have one moderator for each session, we fully expect, 8 since there is such a tremendous collective knowledge on the 9 part of the staff on this issue, that there will be a lot of 10 give and take and weighing in by the staff members who will 11 be participating, as well. 12 We have Bob Colby, Larry Harris, and Mike 13 Macchiaroli. We will be joined by John McCarthy, as well. 14 Let me introduce our panelists, as well, and we do 15 very much appreciate people taking the time to be with us 16 today. 17 We have Sean Egan, from Egan-Jones Ratings. We 18 have Malcolm Macdonald, from Ford. We will be joined, I 19 think her flight is delayed, by Deborah Cunningham, from 20 Federated Investors. 21 We have Gay Huey Evans, who very graciously came 22 all the way from London to be with us today. She is with the 23 FSA. We have Frank Fernandez, who is with the SIA and Leo 24 O'Neill, from S&P. 25 We have Greg Root, from Dominion Bond Ratings; 1 Glenn Reynolds, from CreditSights; and H. Kent Baker, from 2 American University. And that's our panel. 3 So I'd like -- oh, I'm sorry. What did I do. 4 Cynthia Strauss, from Fidelity. You know, I got tricked. I 5 said that was going to be a problem. They put two Cynthias 6 next to each other. So I've merged you into one Cynthia. 7 That's very bad, since I have a sister named 8 Cynthia. I will never live that down. 9 So I would like to give each of you an opportunity 10 to make a brief opening statement. Please be mindful that we 11 have a comprehensive and ambitious agenda, so I would ask 12 that, if you could, that you limit your statements to 13 approximately three minutes. 14 Shall we start with you, Sean? 15 MR. EGAN: Thank you. I have passed out and, 16 hopefully, everybody has a letter that we prepared. If you 17 don't have it, feel free. We have additional ones. 18 But it explains our position on the whole rating 19 field. In the first paragraph, the SEC portrays itself as an 20 investor advocate. That's their number one mission, investor 21 advocate. However, in that process, it extends that mission 22 to the rating agencies that it's recognized. Unfortunately, 23 the rating agencies that are currently recognized by the SEC 24 have had some major failures. It includes Enron, the 25 California utilities, Worldcom, Global Crossing, AT&T, AB&B. 1 There's probably a number more that we could list, but these 2 are the big ones. 3 Paragraph three, there is a study that was done by 4 Professor Baker that refers to only 29 percent of investors 5 thought that rating agencies issue timely, accurate ratings. 6 We think there are two problems associated with 7 this. One is that there's a conflict of interest, and I will 8 elaborate on that more this morning. 9 Two is the lack of competition. 10 Let me just, first, touch on the conflict of 11 interest. Moody's, and that's the only one where we can get 12 public information, Moody's obtains 87 percent of their 13 revenue base from issuers. 14 You have a similar problem, and that is the problem 15 with the equity analysts. They are serving the issuers' 16 interests. 17 Now, they can claim that any one issuer is 18 relatively a small portion of their overall revenue base. 19 However, the same argument could be made for the equity 20 analysts, Blodget -- and who is the other big one that is in 21 the news? Grubman, Jack Grubman, the equity analyst. So we 22 don't think that argument that they are a relatively small 23 portion of the revenue base holds much water. 24 The second fact, and that is a lack of competition, 25 is really picked up by the Justice Department's phrase, 1 "partner monopoly." 2 Now, this is not an oligopoly. If you remember 3 from your economics, oligopoly is when you have a couple 4 firms competing for the same revenues. Here, if Moody's 5 gains some revenues, it's not going to take away from S&P's 6 revenues. 7 You need two ratings to have an issue and some 8 people refer to this as a two and a half firm industry. 9 Fitch shows up in less than ten percent of the corporate 10 ratings that we look at. 11 I'm on to page two. In the package, you have a 12 letter and it's rare to get letters like this, but it's a 13 letter from First Albany. Basically, a First Albany broker 14 had clients that had some securities. Those securities were 15 ultimately -- the company was acquired by Allied Signal. 16 Allied Signal asked the rating agencies not to 17 issue a rating, to withdraw their ratings, so that Allied 18 Signal could buy back those securities at a lower price. 19 And the rating agencies said, "I'm sorry. That's 20 what the issuer requested." And they came to us asking for a 21 rating. 22 We thought this was inappropriate, and you have 23 some material. It's very rare to get something in writing 24 with that and it's a couple years old, but it's still there. 25 We have a couple recommendations and they're at the 1 bottom of page two. One is to recognize some non-conflicted 2 rating firms that have succeeded in warning investors. I 3 mean, that's why we're all here is because of these huge 4 losses. 5 Two is to prohibit issuer compensation. If there 6 are problems in the equity analyst field, there are also 7 problems in the ratings field. You're dealing with basically 8 the flip-side of the same coin. 9 Three is prohibit involvement with rated firms and 10 dealers. Moody's Chairman, Clifford Alexander, was on the 11 board of Worldcom up until last year. The biggest bankruptcy 12 in U.S. history and the Chairman of Moody's was on the board. 13 This is crazy. There's no way that should happen. 14 You should also shoot down the board of the National 15 Association of Security Dealers. They're supposed to be 16 independent. 17 Four is remove the exclusion from Regulation FD. 18 Why in the world should these private firms be given an 19 exclusion from Regulation FD, when all their competitors 20 aren't given that? It doesn't make any sense. They're 21 private firms. Why not let them compete on the same business 22 as everybody else. 23 Five is separate ratings from consulting. The same 24 issue with the accounting firms. 25 Six, prohibit the use of rating triggers. One of 1 the problems with Enron is that S&P and Moody's didn't want 2 to downgrade them because these rating triggers would have 3 thrown them into bankruptcy. 4 Get rid of your triggers. Just say forget it, you 5 can't use it in any security documents. 6 Seven is don't hold yourself out as independent if 7 you're not independent. They list in their web site, they 8 describe themselves as independent. They're not. 9 Eight is police monopolistic practices. There are 10 two aspects to it. One is right now in the structured 11 finance area, there are a number of press releases that Fitch 12 has made whereby Moody's is cutting their ratings by up to 13 six notches in the structured finance, which basically makes 14 it very difficult for Fitch to compete. 15 Moody's has no basis for doing that. They haven't 16 done a study. In fact, you could probably argue that in some 17 cases, Moody's ratings should be cut much more, especially 18 right before offering. 19 Nine is prohibit a collar to investors as basically 20 inside information. 21 Ten, if this is too much, get rid of this whole 22 NRSRO designation. If the SEC doesn't feel comfortable in 23 making adjustments to protect investors' interests, and that 24 has to be the guiding light, we think, in this whole process, 25 we agree with the SEC's mission, get rid of it. Get rid of 1 the whole NRSRO designation. The market will sort it out and 2 they will be much better off. 3 We have, afterwards, some additional charts. 4 MS. NAZARETH: Sean, could I cut you a little short 5 now. 6 MR. EGAN: Yes. Thank you. 7 MS. NAZARETH: We have lots of additional time. 8 Thank you. Mr. Macdonald? 9 MR. MACDONALD: Thank you. I'm delighted to be 10 here on behalf of Ford Motor Company. Certainly, we hope to 11 provide our insights into a very important subject to our 12 company and to many other companies in the United States. 13 In our view, the rating agencies provide a very 14 valuable service to the capital markets. At best, they have 15 a very difficult job, with multiple objectives and serving 16 multiple constituencies. 17 Ford has extensive experience with the rating 18 agencies. For many years, we lived in the rarified air of 19 AAA. Unfortunately, that's a thing of the past and, in more 20 recent times, our ratings have varied quite considerably with 21 the economy and the health of the domestic auto industry. 22 The principal issues that we would hope will be 23 raised today include these significant concentration of power 24 in a shrinking number of rating agencies. 25 The potential for conflict with differing forms of 1 communication to differing constituencies which they serve. 2 Finally, the potential lack of transparency in the 3 rating process and the seeming inconsistencies in the rating 4 process over time. 5 Thank you. 6 MS. NAZARETH: Thank you. Gay Huey Evans? 7 MS. EVANS: Thank you. On behalf of the Financial 8 Services Authority, we thank the Commission for inviting us 9 here today. So thanks. 10 Clearly, much can be said about rating agencies, 11 and I know this is to be brief comment, but from the UK, we 12 do believe there is information asymmetry in the securities 13 markets. 14 Issuers do know more about their own 15 creditworthiness than do investors. It may be both feasible 16 and cost-effective for major investors to undertake their own 17 credit analysis, but this is not the case for small 18 investors. So there is a demand for information and to meet 19 this demand, we need a supply of information, reliability, 20 and, firstly comes along the supply of information in the 21 form of rating agencies. 22 So we have demand and supply in the market for 23 information, just as in the markets for any other commodity. 24 Second, investors cannot easily monitor credit 25 rating agencies for the same reasons they cannot easily 1 monitor issuers. 2 Thirdly, the final trick is to find a way of 3 charging for the information provided by credit rating 4 agencies, since, once this information is revealed publicly, 5 it is no longer of value to the credit rating agency. 6 Over time, we have seen, which has been good, a 7 general switch from charging for subscriptions to charging 8 the issuer for a rating. The issuer is prepared to pay 9 because the impact on its cost of funding, access to a 10 broader pool of investors, and by signal of relative quality. 11 So credit rating agencies appear to have solved the 12 equivalent public good problem of how to pay for the what 13 lighthouse. In fact, they spread the costs across investors. 14 Now, the paradox is, is this for the best, in the 15 best of all possible worlds. There has been explosive growth 16 in the use of credit ratings since the 1970s, but at the same 17 time, the informational value of credit ratings appears to 18 have declined. 19 Maybe a lot has been published on the web, but the 20 length and complexity of methodologies is probably not so 21 comprehensible to many, particularly the retail. 22 There also remains a broad correlation between 23 ratings and default experience, but there has also been a 24 lack of predicted power in individual cases that were 25 mentioned earlier. 1 And how much of the broad correlation is due to 2 changes in ratings immediately ahead of defaults? 3 Almost no empirical studies find no added value of 4 credit ratings over and above other indicators of default, 5 and particularly what we watch are credit spreads, despite 6 the access of credit ratings agencies to non-public 7 information. 8 Indeed, some studies find that credit spreads are 9 superior to credit ratings in predicting defaults, although, 10 equally, it must be recognized that credit spreads perform 11 just as badly as credit ratings in the individual cases that 12 we've already mentioned earlier. 13 Now, on the regulatory side, why have credit rating 14 agencies prospered in recent decades? Now, some have 15 attributed this to regulation. 16 Now, this is not the place for a UK regulator, even 17 though I have a non-English accent, to explain the history of 18 this regulation, and I think Chairman Pitt did a little bit 19 of this earlier. 20 But it is substantial and it is noteworthy. 21 However, these regulatory licenses should not be portrayed in 22 black and white. First, investors also impose rating caps 23 for purely market related reasons; for example, to resolve 24 the principal agent problem in monitoring fund managers, 25 insurance companies, et cetera. 1 Second, the use of ratings by the market is much 2 more refined than the use by regulators who tend simply, in 3 Europe, to divide securities into investment grade and sub- 4 investment grade. 5 Third, part of the explanation for the growth in 6 the use of ratings since the '70s must be because of the 7 growth of tradings in securities and, indeed, credit. 8 Nevertheless, we must remain alert to the 9 possibility that the use of credit ratings by regulators may 10 distort the market for ratings information. 11 A new area that has come out, if anybody is 12 watching, as a banking regulator, the new Basel Accord. It 13 is designed to include a reasonably simple standardized 14 approach, but nevertheless, for this term risk sensitive, to 15 including what we call risk buckets, than currently exists 16 today, and credit ratings are to provide a reasonably simple 17 and straightforward means of populating these risk buckets. 18 The use of credit ratings reflects the universal 19 availability and because, as a more risk sensitive approach 20 demands, the ratings are reasonably well correlated with 21 probabilities of default. 22 So, for example, in the standardized approach that 23 the banks have today for putting capital aside and if they 24 used any of the ratings, let's say S&P, sorry, but whereby 25 all corporate exposures are currently included with the 100 1 percent risk bucket, in the Basel II new approach, a AAA to a 2 AA-minus rated corporate can be included at 20 percent, an A- 3 plus to an A-minus at 50 percent, and a BBB-plus to BB-minus 4 at a 100 percent, and then below the BB-minus, at a 150 5 percent. 6 So the ratings are extremely important, at least 7 for the future, if Basel II goes forward, and, similarly, for 8 collateral and guarantees, and it goes on. 9 But the recognition in the Basel Accord II has to 10 be a public processes, and basing off six criteria. 11 The first criteria is objective, objectivity, and 12 the other five are independence, international access and 13 transparency, disclosure of methods, actual default rates, 14 and a likelihood of ratings transitions, sufficient resources 15 and credibility. 16 The least of the eligible rating agencies is made 17 available by national regulators under Basel II. Banks are 18 not allowed to cherry-pick across credit rating agencies and 19 they must use agencies consistently and disclose which ones 20 they use. 21 Banks also have to disclose the amount of exposures 22 they have in each risk rated basket. 23 So Basel II, designed to be more risk sensitive 24 than Basel I, will actually depend on rating agencies more 25 than they do so today. 1 So in conclusion, we can say a whole lot more, but 2 much can be said with regard to the usage of credit ratings 3 for other market purposes, either whether it's rated 4 structures, finance, SPVs and synthetics, and other aspects 5 of rating agencies which we are bound to discuss. 6 But I would like to close on a major problem with 7 ratings, that adjustment is delayed for far too long before 8 the ratings fall to the trigger level, again, the problem of 9 too much reliance being placed on ratings, and I hope that 10 gets covered today. 11 MS. NAZARETH: Thank you, Gay. Frank? 12 MR. FERNANDEZ: Thank you, Annette. I promise to 13 be brief and respect the two minute rule. 14 Since you have my written introductory remarks, I 15 would just like to say my thanks to the staff and the 16 Commission for this opportunity to appear here at these 17 hearings. 18 I applaud you for both the timeliness of this issue 19 and bringing it forward and recognizing the critical role 20 played by the credit rating agencies in providing valuations 21 and risk assessments to financial markets, and for providing 22 this forum to evaluate and assess these functions which are 23 key to securities markets operations, as well as the 24 performance of the credit rating agencies who provide them. 25 Given the breadth and depth of the market expertise 1 around this panel, I'm going to try and confine myself to 2 something I do disproportionately, which is assessing the 3 costs and benefits of supervision and regulation, and looking 4 at the business model that is being applied here. 5 I think it is important to consider some of the 6 broader issues of the provision and the business of 7 information, particularly economic and financial assessments 8 of that business, particularly and specifically looking at 9 them both in the whole and for the segments of both the 10 provision of data analysis and valuation, because I think 11 they are distinct functions with distinct implications. 12 Beyond that, I will address them when we get to 13 specific points. 14 MS. NAZARETH: Thank you. Leo? 15 MR. O'NEILL: Thank you. Well, on behalf of 16 Standard & Poor's, I would like to thank the SEC for the 17 invitation to discuss the role of rating agencies in the 18 United States capital markets. 19 Standard & Poor's has been rating debt securities 20 since 1916 and in view of this long history in the capital 21 markets, we fully recognize the importance that our ratings 22 have played and will continue to play in the assessment of 23 credit risk in this country and abroad. 24 I believe the fundamental reason that Standard & 25 Poor's and other ratings have grown in importance in our 1 capital markets is our long track record of providing 2 independent, objective, and reliable opinions on 3 creditworthiness. 4 Indeed, ratings have found their way into various 5 regulations far back as 1931, when the Comptroller of the 6 Currency sanctioned ratings as a tool for banks to haircut 7 their capital for bonds held in inventory. We are heartened 8 that Basel II, they're seeing the same need. 9 In 1975, when the SEC coined the phrase "nationally 10 recognized statistical rating organization," and in 1976, 11 when the SEC designated Standard & Poor's as an NRSRO, I 12 believe they were simply stating a market reality that 13 Standard & Poor's had earned over 60 years and they were not 14 conferring a newly won status on us. 15 Now, more than 25 years later, the U.S. capital 16 markets, which have experienced enormous growth, continue to 17 be extremely well served by Standard & Poor's ratings. 18 Our ratings are subject to intense market scrutiny 19 every day, but they remain globally accepted benchmarks that 20 are respected for their independence and their integrity. 21 In continuing our important role and extending the 22 benefits of independent, credible rating services, both here 23 and abroad, depends on a framework that continues to preserve 24 the independence of credit rating agencies and recognizes the 25 market, the market is the best judge of a credit rating 1 agency's quality, objectivity, and independence. 2 Standard & Poor's also has a long history of 3 constructive dialogue with the SEC and other government 4 agencies about important issues affecting the capital 5 markets. We are frequently asked our views on key analytical 6 issues and we openly share our rating methodologies and 7 practices with the SEC and other government entities. 8 Today will be no exception. 9 Let me close by stating the obvious. We fully 10 recognize the value that we add to the markets and we 11 understand that it rests on a platform of integrity, 12 objectivity, and independence. 13 We also recognize that in the absence of such 14 qualities, our value in the marketplace diminishes, if not 15 disappears entirely. Hence, all of our processes, all of our 16 standards, all of our methodologies are geared to meeting the 17 objectives of integrity, quality, independence, and the 18 record will show, as it has in the past, that the valued 19 reputation that Standard & Poor's has earned is well-deserved 20 because of that deep commitment and our every intention is to 21 keep it. 22 Thank you. 23 MS. NAZARETH: Thank you. Greg? 24 MR. ROOT: I want to thank the Commission very much 25 for giving us this opportunity to address such an important 1 issue that affects the capital markets. 2 I'd just like to give you a little bit of 3 background about Dominion Bond Rating and summarize what we 4 think is the most critical issue that the rating industry 5 faces at this point in time or that the Commission faces 6 regarding rating agencies. 7 Dominion Bond Rating is a Canadian-based rating 8 agency. We have been in business now for 26 years, so we 9 don't have quite the longevity of Standard & Poor's, but we 10 are one of the early participants in the rating agency 11 business. So we have quite a long experience level in that 12 respect. 13 We are a privately held company. We have no other 14 activity. So we are totally independent in our activities. 15 Just by way of background, the Canadian market, 16 there are no actual formal rules and regulations as there may 17 be here, that exist here in the states in terms of mandating 18 that issuers be rated. 19 So the success of DBRS and our activities are very 20 much driven by market acceptance, and that's the acceptance 21 of the issuers, as well as the investors, as we think that's 22 a very important aspect that drives the rating business 23 overall, not just in Canada, but here as well. 24 DBRS rates and tracks about a thousand different 25 issuers, but because of the nature of the Canadian market, 1 with all the cross-border activity between the U.S. and 2 Canada, a substantial amount of our activity is here in the 3 United States. 4 So, obviously, what affects the rating business 5 here in the U.S. is very much important to Dominion Bond 6 Rating. 7 Overall, having been in the industry for quite some 8 time, I would concur with Leo. I think the rating business 9 has been very effective and has been very supportive of the 10 development of the capital markets overall. 11 I wouldn't say it's been totally flawless. But if 12 you look at the overall track record, I think it stands on 13 its own. 14 The key issue that we see that we think needs to be 15 addressed is with the NRSRO being such an important part of 16 many of the activities in the capital markets, we feel NRSRO 17 should be clearly defined. I think that is the most 18 important issue, is to define what an NRSRO is, have very 19 clear rules and regulations and guidelines for what it takes 20 to become an NRSRO, with an appropriate time frame, and, that 21 way, let the market judge very much who is the NRSRO and who 22 can be and who should not be. 23 So I think if that were to be accomplished, 24 overall, I very much believe that the system works well and I 25 think that today's session is important to continue to 1 discuss it, see if there are areas that could benefit from 2 some adjustment or improvement, but I think we will hopefully 3 find that it's not broken. 4 Thank you very much. 5 MS. NAZARETH: Thank you. Glenn? 6 MR. REYNOLDS: Thank you for inviting me to be part 7 of this very important hearing. 8 I come to this topic from the angle of an 9 independent credit research firm. We are not a rating agency 10 and we are not looking to become a rating agency. 11 Our primary product is market-based research on 12 issuers, industries, and segments of the marketplace. We are 13 not involved with managing assets or with underwriting 14 activity. 15 But like many analysts, I have been spending a fair 16 share of my 20-year career trying to predict the rating 17 agency behavior patterns. It's a significant input into the 18 performance of securities in the market and is a critical 19 portfolio management variable that we all need to watch 20 closely. 21 The last year has been as challenging as any in my 22 career and it's easier to sympathize with the demands placed 23 on the rating agency. That said, they have one of the most 24 important roles in the capital markets. 25 In my prepared written comments, which will be 1 posted to the SEC website, I focused on three areas that I 2 find most pressing and look forward to discussing today. 3 First, the quality of information flows and the 4 transparency of the ratings process has proven very uneven 5 over the past year and the post-Enron response has left many 6 institutional investors alarmed that there is an attempt to 7 over-compensate. 8 Together with some other dislocations in the 9 market, coming off a year of record investment grade 10 issuance, in particular, in 2001, it has inflamed market 11 volatility and created a serious crisis of confidence among 12 bond-holders. 13 If there is anything that is the case, it's that 14 there has never been a time where clarity and specific 15 descriptions and details of the ratings parameters have been 16 more important than they are today and the process has left 17 much to be desired. 18 Investors are intermediaries in the business of 19 taking risks, but they do need to understand fully the rules 20 of engagement. 21 Second, we believe that changes in the market over the 22 past year could bring some new competition to the NRSRO peer 23 group. We don't believe it is appropriate to attempt to burn 24 the NRSRO framework to the ground. We just think it needs to 25 be upgraded and the bar raised. 1 Third, we believe there are some working conflicts 2 of interest that are at least worth exploring, which we will 3 discuss later. 4 I will repeat one point that we made in our written 5 comments, and that is I think it would be worthwhile to more 6 intensively survey the institutional investor base and the 7 processionals in the brokerage and banking sector on the 8 performance of the agencies and the pros and cons of the 9 approaches. 10 We suspect that the power of the agencies and what 11 they wield in the marketplace and the relationships that need 12 to be maintained with the rating agencies may discourage a 13 full public airing and confidentiality could encourage more 14 candor. 15 I look forward to the proceedings. Thank you. 16 MS. NAZARETH: Thank you. Kent? 17 MR. BAKER: Thank you very much. I appreciate the 18 opportunity of being asked to participate in the panel. 19 The role that I perceive for myself is really that 20 on the academic role, because that's my environment. One of 21 the things that I hope to be able to contribute to these 22 hearings involves some empirical research that a colleague 23 Sattar Mansi, and myself have conducted involving credit 24 rating agencies. 25 These publications, one appeared earlier in the 1 year in the Journal of Investing, and the second publication 2 Mr. Egan referred to earlier, at least one of the comments or 3 findings from that particular study will appear in the next 4 issue of the Journal of the Business Finance Accounting, and 5 the title of that paper is "Assessing Credit Rating Agencies 6 by Bond Issuers and Institutional Investors." 7 In those studies, we took, hopefully, a 8 dispassionate view of credit agencies and trying to get input 9 from users of those particular services, specifically looking 10 at the bond issuers, focusing upon industrial issuers, and, 11 also, from institutional investors, primarily looking at the 12 segment relating to mutual fund investors, which, of course, 13 are large purchasers of these bonds. 14 There were four key issues or key research 15 questions that we focused upon in these studies, and I just 16 want to mention those topics, without going into the findings 17 at this particular time, which I can share, where that is 18 appropriate, during these hearings. 19 The first research question focused upon does the 20 number of rating agencies hired by issuers to rate bonds 21 differ from the number that are required by institutional 22 investors of these corporate bonds, given the different 23 perspectives. 24 We'll talk about the importance of having that type 25 of competition within the industry. 1 The second issue involved do the issuers and 2 institutional investors perceive any differences in the 3 accuracy among the various key rating agencies within the 4 U.S., and I'm sure our folks at S&P would like the results of 5 that, since they tend to be more favorable for S&P than the 6 other agencies. But I'd just make that as a quick comment. 7 The last two questions, I think, perhaps are more 8 germane to these hearings. The third question basically 9 focuses upon do issuers and institutional investors believe 10 that agencies maintain corporate bond ratings on a timely 11 basis, and Mr. Egan referred to part of the results that we 12 had, and being able to maintain timely ratings certainly is 13 critical in the marketplace today as opposed to having 14 excessive lags or at least lags that may be viewed as 15 excessive by market participants. 16 The final issue that we investigated involved do 17 issuers and institutional investors believe published ratings 18 of corporate bonds accurately reflect their issuers' 19 creditworthiness, and, of course, that is the key purpose 20 underlying bond ratings in the first place. 21 So we'll try to share some of those results during 22 the course of the hearings today. 23 Thank you very much. 24 MS. NAZARETH: Thank you. Cynthia? 25 MS. STRAUSS: I'm Cynthia Strauss, from Fidelity 1 Investments. I thought in these few minutes I would just 2 give a little background on who we are and how the rating 3 agencies really affect us, as well as a general comment on 4 our views. 5 At Fidelity, we manage about $360 billion of 6 investment grade U.S. dollar fixed income funds, and that's 7 spread about $240 billion in money market funds and about 8 $120 billion in bond funds. 9 We play a fiduciary role for millions of 10 shareholders. Our funds are principally distributed through 11 retail investors and they are in mutual funds, and Fidelity's 12 role is one of a fiduciary. 13 We do our own research for these funds. 14 The rating agencies are incredibly important to us 15 because they have tremendous influence on the markets in 16 which we participate. They are supportive tools, as we've 17 all said, for individual investors in the marketplace, small 18 and large. 19 They are the very structure of the marketplace, 20 just I said, investment grade references their ratings. They 21 are the risk language that we all speak and rely on. 22 They are important, too, because they are embedded, 23 appropriate to the SEC, in many of our regulations, very 24 important to money markets. Again, the cash for many, many 25 individuals in the marketplace. 1 They are also embedded in various state and federal 2 regulations that really we pay a tremendous amount of 3 attention to as fiduciaries. 4 And last, but not least, I think often goes 5 unnoticed is that the rating agencies drive industry 6 standards. They, in essence, are the voice for the 7 marketplace. They may set, for example, the level of 8 liquidity that a corporation should have to be a strong 9 issuer. 10 They actually are at the front line in designing 11 the structure of the ever-growing asset backed market. 12 So with that background, we have tremendous respect 13 for the agencies. We think, in many ways, they do an 14 excellent job with an ever increasing challenge. 15 When you think about it, there are only three. 16 Their business model is a for profit model, and yet they 17 serve a public, a greater public interest, and things really 18 haven't gone too wrong. 19 However, all is not well. I do think I agree it's 20 very timely that we're here. The examples that brought us 21 here were Worldcom and Enron, but as I hope to be able to get 22 into later today, there are many other examples, perhaps more 23 subtle today, of problems that could be facing this model 24 with only few rating agencies in the future. 25 That being the case, I think Fidelity's view is 1 that the NRSROs do need greater oversight to make them more 2 directly accountable to the markets that they serve. 3 Our specific issues are going to be more in the 4 area of the inconsistency and lack of transparency, which you 5 have already heard from a few, as well as their role as 6 driving industry standards, as I already mentioned, and I 7 look forward to getting into those in the examples later on. 8 Thank you. 9 MS. NAZARETH: Thank you. I want to welcome 10 Deborah Cunningham, who has had an arduous trip to get here. 11 We haven't given you time to sort of get settled in. 12 We can offer you a chance to make a statement now 13 or give you a little time to get your bearings and speak up 14 later. 15 MS. CUNNINGHAM: I'm fine. 16 MS. NAZARETH: Great. Wonderful. 17 MS. CUNNINGHAM: I'm Debbie Cunningham. I'm Senior 18 Portfolio Manager and Senior Vice President with Federated 19 Investors. Federators is an asset management firm that is 20 based in Pittsburgh, Pennsylvania. 21 Currently, we have about 200 billion in assets 22 under management. 23 My particular group that I'm in charge of at 24 Federated deals with our stable net value asset taxable 25 funds, and that includes about a $130 billion of those $200 1 billion in assets. 2 It includes our government money market funds and 3 our stable net asset value GIC products. 4 So my intention today is to address the group more 5 as a buy side, particularly with emphasis with regard to 6 money market funds and Rule 2a-7, as a Rule 2a-7 buy side 7 investor. 8 My use of the rating agencies and the NRSROs is 9 basically through the portion of our assets that we invest in 10 what we call our prime money market funds, and those are the 11 ones that can take credit risks. 12 Rule 2a-7, which is part of the Investment Company 13 Act of 1940, is the precedent that we use and that is the 14 NRSROs are required to be utilized within management of funds 15 that are holding themselves out as being 2a-7 funds. 16 Currently, about $70 billion of the assets that we 17 manage are subject to this particular regulation requiring 18 the use of NRSROs. So we've been using them since they've 19 been in business, since we've been in business, and, 20 essentially, when the NRSRO information was added into Rule 21 2a-7, there were seven different rating agencies at that 22 time. 23 We are now down to, through consolidation, through 24 the consolidation process, to three of them. 25 And although I don't really have a problem with the 1 three that exist, our preference is definitely to have more 2 rather than fewer, and this is basically because we like the 3 input that is available from the reputable sources. 4 This would assume that these would be qualified 5 NRSROs and that they would meet some minimum standard that 6 was set in the marketplace. 7 On the other hand, if there were any type of 8 legislative changes that would require us to always follow 9 all NRSROs and always purchase all NRSRO ratings, we would 10 have problems with more than the three or four or five or six 11 that have been in place in the marketplace on a recent 12 historical basis. 13 Currently, Rule 2a-7 is such that it requires 14 what's called a second opinion rule and the ratings 15 information is derived from a second opinion rule. 16 If there is only one rating on the particular 17 issuer that we're looking at, then that rating is the one 18 that counts. 19 If there is more than one rating, then, 20 effectively, you need two, a second opinion, in order for an 21 issuer to be considered first here, topped here. 22 So even if you had seven different rating agencies, 23 really the two that you need, there are two that you would 24 need as confirmation that this issuer would be first tier. 25 This is all utilized within our analytical basis at 1 Federated from an independent credit analysis standpoint of 2 taking inputs directly from the research that's done by the 3 rating agencies and utilizing that in determining minimal 4 credit risk for the securities that we purchase within the 5 portfolios that we use. 6 We would definitely advocate more NRSROs, as I had 7 indicated, and we would absolutely hope that any additional, 8 as well as the current NRSROs would be subject to oversight 9 and accountability, some of which exists today, but a lot of 10 which does not exist today. 11 And we know that the assessment of financial health 12 is not an easy thing to predict going into the future, 13 whether it's short-term or long-term, and our analysts are 14 faced with that on a day-to-day basis. 15 We know there is never going to be a 100 percent 16 certainty in that type of a business, but we feel that with 17 additional oversight, with additional accountability, and 18 with additional competition in that marketplace, definitely 19 the usefulness of these assessments, I think, will be better 20 served and would be better used by the investor base. 21 We also think that if you merge the due diligence 22 process that is typical by analysts to assess analysis or to 23 assess the financial health of a company with what is 24 currently the rating agencies' credit assessment, we, again, 25 think that with more oversight, you could produce better 1 results for investors. 2 Lastly, the last thing I would like to encourage 3 rating agencies to do is really to recognize each others' 4 ratings. There are published sources of information that 5 detail what makes up a AAA rating or what makes up an A1-plus 6 or a P1 or an F1-plus rating across the three NRSROs that 7 exist today, yet, in many instances, the rating agencies will 8 only look to their own internal ratings when viewing 9 information and trying to rate it. 10 So I would promote some sort of an acknowledgment 11 and recognition across the NRSRO base so that we would be 12 able to more fully enjoy the ratings benefits of the three 13 that exist today and hopefully more that exist in the near 14 future. 15 I think that's about it. I can address other 16 things as the questions come forth. 17 MS. NAZARETH: I was going to start with a question 18 that I think Cynthia actually touched on quite directly in 19 her opening statement, which was the relative importance of 20 credit ratings for use for credit analysis. 21 So I thought I would turn to the other two 22 participants who may have some interesting insights into 23 this, which would be Glenn and Deborah. 24 You spoke, Deborah, about how you make use of these 25 credit ratings, partly because they are regulatorily 1 mandated. But could you speak a little bit more about how 2 much they are actually used, even on the buy side, in a 3 sophisticated shop like your own in terms of how you 4 determine what you are going to purchase for various 5 portfolios and how much you do rely on the ratings? 6 MS. CUNNINGHAM: Certainly. We utilize the ratings 7 as an input to our own internal credit assessment. We, in 8 many ways, attempt to duplicate the process that the rating 9 agencies provide for investors that don't have the means of 10 duplicating that internal mechanism themselves, and we 11 essentially utilize, in that process, the rating agencies' 12 information. 13 We utilize other sources of information, as well, 14 from various street firms research, to information from the 15 companies themselves, to basically public information that is 16 available via internet type sources. 17 But the rating agencies' input is absolutely of 18 importance and it's also important, I think, from a 19 regulatory standpoint, that before we can start the process 20 with our own internal analysis, we have to meet the hurdles 21 of having the NRSRO ratings that we're looking for from a 22 regulatory standpoint. 23 MS. NAZARETH: Okay. Glenn? 24 MR. REYNOLDS: If people want to get a sense of how 25 all pervasive the NRSRO designation is, just go to Google and 1 type in those letters and you'll read page one of one 2 thousand pages. So you get the idea, it's everywhere out 3 there. 4 The first line of using the ratings designation is 5 just the explicit parameters that many mutual funds, pension 6 funds, and insurance companies may have put some of its 7 discretionary and in-house and some of its shareholder-driven 8 based on a prospectus which has been delivered to investors. 9 So very often it's hard and fast, but then as you 10 drove down a little bit more, it can very dramatically impact 11 the price behavior of securities, which is a performance 12 issue for these portfolios, which directly affects policy- 13 holders, mom-and-pop retail investors, mutual funds, savings 14 plans, pensions. 15 One of the things in the past that has been a real 16 problem is the minimum rating threshold in some portfolios 17 creates what we used to call a forced seller. 18 Once it drops below a ratings tier, they have to 19 kick the security out. Wall Street traders are in the 20 business of looking to chop trading for those opportunities, 21 because they immediately drop the bid down because they smell 22 a forced seller in the market. 23 That inflames the volatility and heightens the 24 loss. After some difficult experiences with that, people 25 have eased those types of restrictions, where they're even 1 more discretionary in the timing of the sale, but a ratings 2 move goes far beyond just a lower average quality. It can 3 inflame the magnitude of a loss for a given institutional 4 investor. 5 And when we talk about institutional investors, 6 behind every institutional investor is a whole array of 7 retail investors. 8 But then it keeps going from there, as we'll talk 9 about later, ratings triggers in other areas. That security 10 that you may own, whether it be a BB or a BBB rated security, 11 will have follow-on effects from that ratings change that 12 effects the quality of the credit collateral posting 13 requirements on OTC derivative trades, bank line repricing; 14 in some cases, explicit puts, ratings indexed. 15 So it gets very complicated, and we'll spare you 16 all the arcane credit mumbo jumbo, but it goes well beyond 17 the fact that the spread widens. It can have follow-on 18 effects that are concentrated on some investors and some 19 types of portfolios more than others. 20 MS. NAZARETH: Did you have something to say? 21 MR. BAKER: I just had an observation on the uses 22 of those ratings. Part of our study did focus on, from the 23 issuer's perspective, on the number of ratings that they 24 generally used and we found that, from our study, again, 25 relating to the industrial sector, that approximately 20 1 percent of those companies did use a third credit rating 2 agency, and the primary reason that that was done was because 3 there were essential conflicts that existed, that is, a split 4 rating between basically S&P and Moody's. 5 In order to try to clarify that, they went for a 6 third rating, in many respects, to try to provide additional 7 information to the marketplace with that third rating. 8 MS. NAZARETH: Thank you. 9 MR. COLBY: Could I just follow-up? Glenn, I 10 understood you to say that you have to watch the rating 11 because of the impact of the rating itself on the market 12 price. But could I just go back and follow up with Cynthia 13 and Debbie? 14 To what extent are you using the rating because of 15 the clarity of the insights and the analysis of the rating 16 agency itself as opposed to the effect it might have on 17 price? 18 MS. STRAUSS: As Debbie said, as well, we use it in 19 two ways. One is the rating agencies are some of the best, 20 we consider, peer analysts out there, in many cases. So we 21 use their research as an input to our research. 22 But since we do independent research, it's a 23 compliment. It's input. 24 But what really is critical, I think, in some of 25 the issues looking forward is the point that Glenn made, that 1 the direction of the ratings can really impact many things 2 about that company, and I'm sure Mac will have comments about 3 that, as well, and it can really impact that company's access 4 to capital markets, what's going to happen in the 5 marketplace, which is something that we, as market 6 participants, look at. 7 And we'll talk about later that we argue greatly 8 that the rating agencies are trying to figure out how their 9 ratings impact market prices, and we'll get back to that 10 later. It's a very worrisome area. 11 We're looking at market prices, what could happen 12 if the rating goes down, what could happen to that company's 13 fundamental liquidity if they are shut out of the investment 14 grade market. 15 MS. CUNNINGHAM: I echo all the things that Cynthia 16 said. I would just add that we have found, in the past, that 17 the written input that we get on the rating agencies, what is 18 available in written form or what is available in printed 19 form on their web sites, is oftentimes not as useful as the 20 global input that we get from talking to their analysts. 21 It's oftentimes less timely, from a date specific 22 standpoint. So the printed material may be dated, to some 23 degree. It's hard to keep exactly with current times on 24 that, in that regard, yet we are able to, from a verbal 25 standpoint, get very good input, typically, up-to-date input 1 by speaking with their analysts. 2 So I would note that it's not just on a written 3 basis that we utilize their information, but even more 4 important, probably on a speaking basis. 5 MS. NAZARETH: Thank you. Mac? 6 MR. MACDONALD: While I recognize that from an 7 investor perspective, the input, other than the written input 8 from the agencies may be important. 9 It does, I think, give me a great deal of cause for 10 concern that that data, which, in many cases, is data that is 11 being released free of the oversight of Reg FD, is being 12 potentially used to provide information to one individual 13 alone. 14 I have to say up front that the agencies with whom 15 we deal have been absolutely superb in their handling of 16 confidential information. In the 20 or so years that I've 17 had personal experience, we have never come across a 18 situation where confidences have been abused. 19 But having said that, we do find that we receive 20 reports that personnel from the rating agencies have made 21 comments to analysts, investors, and others that they will 22 not make publicly and they will not make to us, and I believe 23 and I am concerned that that can lead to serious distortions 24 in the markets. 25 MS. STRAUSS: Can I just make one comment on that? 1 We may get a chance to -- I know we have FD later on, but I 2 just want to make the linkage, which is very worrisome to 3 Fidelity and I think others in the marketplace. 4 We just mentioned the power of a moving of a rating 5 in impacting market prices, and yet the fact that the rating 6 agencies have access to information that investors can't have 7 creates great concern for us. 8 Our position has always been that we encourage 9 greater disclosure to the marketplace and to investors, 10 because this creates a difficult situation for all parties. 11 MS. NAZARETH: I think this does raise a very 12 interesting issue for the Commission. It is something that 13 has come up in other discussions that we have had with market 14 participants and has been high on Chairman Pitt's list of 15 issues. 16 If the key information that is so important for the 17 buy side is information that, in fact, is not in the public 18 domain, then perhaps there is something with what we are 19 requiring to be in the public domain, and I certainly think 20 that is a significant point that is being made there. 21 MR. O'NEILL: Ms. Nazareth, I feel like -- I'm 22 trying to behave myself here. I can respond to all of these 23 here at one time or another, but I will take your lead on 24 that. 25 On the information that we get, the confidential 1 information we get, we've actually -- this is not something 2 that came out of Reg FD, as you know. Rating agencies, for 3 many, many, many years prior to Reg FD, were the recipients 4 of information that issuers provided to us on the condition 5 that we would not make that public and, also, on the 6 condition, and what we are, is we do not trade securities, we 7 do not buy, hold, sell securities in that regard. 8 That information is there solely for the purpose of 9 providing a credit rating. 10 The disclosure obligation, obviously, is not on the 11 rating agencies. The disclosure obligation is on the 12 corporations that file with the Securities and Exchange 13 Commission. 14 MR. COLBY: Mr. O'Neill, could you address the 15 issue that has been raised about verbal contacts with the 16 analysts at the rating agencies and what sort of information 17 is made available in that context? 18 MR. O'NEILL: Sure. I'll take Deborah and Cynthia, 19 who have direct contact with our analysts and they hear what 20 they hear. 21 I suspect, like I believe, that what we put out on 22 our web site, what we put in our publications is intended to 23 provide a broad audience with the reasons for maintaining our 24 rating or changing our rating or what have you. 25 Certainly, an analyst from outside our organization 1 calls up with a specific question or specific concern, 2 assuming it does not breach, it cannot breach the 3 confidentiality seal that we have put on this information, we 4 will, obviously, respond to those analysts with specifics. 5 I think that's important that we do that and, in 6 fact, we pride ourselves on being available to the analytical 7 community that way. 8 I would encourage our people to continue to do that 9 and I think it would be a bad practice for the rating 10 agencies to put a cloak or a seal over their rating opinions 11 in any way. 12 So, again, we will respect the confidentiality. We 13 will not disclose the confidentiality of that information. 14 MS. NAZARETH: Could I ask the other two rating 15 agency representatives to address the issue, as well? 16 MR. ROOT: First of all, I very much agree with 17 Leo. I think the important thing is transparency and I think 18 if you take a look at the type of information that Dominion 19 Bond Rating publishes and puts out there, we are very clear 20 and open in terms of what is incorporated into our ratings, 21 what the key issues are, and we think that's important. I 22 think that is one of the issues here, is to what extent the 23 rating agencies are communicating, what is behind the 24 ratings, basically. 25 I'm not going to speak for our competitors, but, 1 certainly, if you look at Dominion Bond Rating, this is 2 something that we put very much as a top priority, is making 3 sure that -- the rating itself is one thing. 4 It's what is behind the rating, what you are 5 assuming, because, again, we can sit down and discuss ratings 6 between us and we're going to agree to disagree. 7 So I think what's important for us is that you know 8 what's behind our thinking and you may say that we may be too 9 optimistic about the economy or whatever it is and, 10 therefore, you can adjust our ratings accordingly, but it's 11 important, as long as we are quite open in terms of what 12 we're looking at, then we think that is as important as the 13 rating itself, and that is something that we very much 14 emphasize at Dominion Bond Rating. 15 COMMISSIONER GOLDSCHMID: Just to pin that down a 16 drop. But do you have a confidentiality policy on material 17 information you have been given? 18 MR. ROOT: Absolutely. Again, I'll go back to Leo. 19 Prior to joining Dominion Bond Rating, I was with one of the 20 NRSROs, Thomson BankWatch. I have never come across a 21 situation where having confidential information has been 22 breached, certainly by any of the organizations currently or 23 prior, and I assume Leo will support that at Standard & 24 Poor's and I suspect the other agency. 25 So while it's something that needs to be continued 1 to be looked at and considered, I think if you look at the 2 history, it has never been an issue that has been violated 3 and I think that is very important to communicate in this 4 forum. 5 CHAIRMAN PITT: Mr. Macdonald, do you find that you 6 are asked for the same information from each credit rating 7 agency? 8 MR. MACDONALD: To a great extent, Mr. Chairman, we 9 are. But the nuances and the depth to which the questioning 10 is subjected can be quite different over time. 11 I would give you a current example. The issue of 12 under-funded pensions is of great concern to one rating 13 agency, far less a concern to others. 14 All three have asked about it, but in one case, in 15 great depth, in the other cases, a few questions, an 16 understanding, and move on. 17 MS. NAZARETH: Sean? 18 MR. EGAN: Thank you. Our position at Egan-Jones 19 is that the exclusion of NRSROs from Regulation FD provides 20 the opportunity for abuse, and those opportunities haven't 21 fully been explored 22 In fact, if it is not corrected, we think, over the 23 next couple years, you will see some huge abuses. 24 There is no reason why these private firms should 25 be given information ahead of other participants in the 1 market. If it's relevant, produce a press release to 2 everybody. 3 Gathering information such as on the rating 4 triggers which is done by the two rating firms and then 5 providing that to private clients is inappropriate. It 6 should be sent out to everybody at the same time to even the 7 playing field. 8 CHAIRMAN PITT: There is a general exclusion, 9 though, in FD, if you have a confidentiality agreement, which 10 would seem to me to achieve exactly the same purpose. How 11 would you differentiate that? 12 MR. EGAN: The other market participants aren't 13 going to get that confidentiality. In fact, as a practical 14 matter, what would happen is that if the name S&P or Moody's 15 is on the letterhead, they will be given that confidential 16 information. If it's not, they won't be given it. 17 CHAIRMAN PITT: Maybe I haven't made myself clear. 18 Let's assume you got your wish and there were no specific 19 exemption for NRSROs in the process. There is still an 20 exemption for anybody who receives information based on 21 giving a confidentiality agreement. 22 So what is the difference? 23 MR. EGAN: Our position is exclude that. If it's 24 part of an underwriting, that's fine, but if it's part of an 25 ongoing rating assessment process, why hold these other firms 1 in some superior position? 2 COMMISSIONER GOLDSCHMID: But one thing ought to be 3 clear to everybody. If you've signed a confidentiality 4 agreement or you have such a policy that is understood, if 5 there is then misuse of that information, there are far more 6 serious consequences than FD. That becomes insider trading, 7 with all of that criminal and civil and treble damage 8 penalties potentially available. 9 So this would be a very serious violation, if it 10 occurred. 11 CHAIRMAN PITT: But there is also a shifting of 12 responsibility. It shifts the information away from the 13 originator of the information, the issuer, to the recipient 14 of the information, where it would belong once you've gotten 15 a confidentiality agreement. 16 I don't want to belabor the point, I just don't 17 understand how getting rid of the specific exemption would 18 necessarily change anything if companies complied with the 19 policy of FD to require a confidentiality agreement from any 20 rating agency to whom they gave information. 21 MR. MACDONALD: Following up on the Chairman's 22 point. I would merely say that in terms of if we were 23 required to observe FD with the ratings, the quality of the 24 rating for the investors would undoubtedly go down, because 25 there is no way that we would release confidential product 1 plans for future years and data of that type, which could be 2 used by our competitors. 3 So applying the exemption from FD, I think, 4 improves the rating process. 5 Now, let me change subjects a little bit. One of 6 my concerns about this, shall we call it, informal contact 7 between people from the agencies and individual investors is 8 that the general form of rating comes out of a committee, as 9 I understand it, at most of the agencies. 10 That committee is made up of a group of 11 individuals, clearly, not all of whom share the same opinion 12 on an individual company. 13 To the extent that one individual perhaps has a 14 more positive or a more negative outlook and that individual 15 is talking with an investor, the nuances and phrasing of his 16 or her comments may not necessarily reflect the opinion of 17 the rating committee. 18 For that reason, I would strongly support the data 19 from the rating agencies only be published and, therefore, 20 subject to full oversight of the internal committees. 21 COMMISSIONER GLASSMAN: Can I just ask a follow-up 22 question? I just want to make sure I understand. 23 Are you saying that the exemption from FD does 24 improve the quality of the ratings? 25 MR. MACDONALD: Absolutely. 1 COMMISSIONER GLASSMAN: So for the non-designated, 2 the rating agencies that are not designated NRSROs, who are 3 not exempt from Reg FD, are missing whatever it is that is 4 improving the ratings by this exemption. 5 Is that a correct interpretation? 6 MR. MACDONALD: Basically, I was taking the point 7 that to require us to observe FD with respect to the rating 8 agencies that we choose to use, we do have a choice, gives us 9 the ability to give them better information, which, in turn, 10 benefits investors. 11 CHAIRMAN PITT: The question that I was originally 12 getting at when I posed it to you, which relates to FD, is if 13 you see different rating agencies focusing on different 14 information, do you assume any responsibility to make sure 15 that every rating agency that comes in to talk to you winds 16 up receiving effectively the same information irrespective of 17 how they may use it in their rating? 18 MR. MACDONALD: I think the answer to that is 19 certainly with respect to written information, for example, 20 in the past month, we have met with the three nationally 21 recognized agencies and each one of them received an 22 identical 76 page book of data. 23 Now, the questions that followed from meetings, 24 that were extensive, clearly provided more follow-up and, in 25 fact, subsequent telephone conversations would probably mean 1 that in certain areas, one agency asked more questions and 2 received more information than others. 3 CHAIRMAN PITT: That's what I would assume, simply 4 because different people are going to be focused on different 5 issues. 6 But if you apply what the underlying philosophy of 7 FD is, so the company gives out information that may be 8 significant, it is supposed to provide it to everyone. 9 Now, in the rating agency context, where there is 10 no application of FD, the question still remains if, in the 11 course of doing this, you discover that information that has 12 been ferreted out by questioning may be relevant to a rating 13 or may take on more significance than you would have thought, 14 the question is should there be some obligation to provide 15 the same information to every rating agency that might be 16 interested in covering Ford, for example? 17 MR. O'NEILL: Having done this sort of thing, I 18 think that puts a burden on the issuers to get into the minds 19 of the analysts and try to determine what it is that's truly 20 important to the analysts. 21 I don't think our analysts come in and say to 22 Malcolm, "Malcolm, here's a real big one, you got to get this 23 one right." No. They would ask dozens and dozens and dozens 24 of questions, I'm sure, over periods of days, many times. 25 CHAIRMAN PITT: Thousands and thousands. 1 MR. O'NEILL: It would be wonderful if that could 2 happen. I think, in reality, it's difficult. 3 CHAIRMAN PITT: That is the way FD operates outside 4 that context. If a corporate official says something to an 5 analyst who happens to ask the right question, putting aside 6 the rating agency context, there is an obligation on that 7 official to then consider whether that information is 8 material and should be disclosed to the marketplace as a 9 whole. 10 MR. MACDONALD: I would have -- in fact, it would 11 be a major efficiency if I could see all three rating 12 agencies in the same room on the same day for four hours. 13 I do not believe the agencies necessarily, and I 14 look to Leo as to whether they would support that. 15 MR. O'NEILL: No. No. We think that the value 16 inherent in our ratings is the key value, is the independence 17 of our process from the other rating agencies. So we would 18 definitely not want to participate in a process which 19 creates, in the final analysis, a homogenized singular view 20 to the marketplace. 21 Some comments have been made about rating agencies 22 and transparency and understanding why our ratings are what 23 they are. I think part of the issue is that we all have our 24 own criteria, our own emphasis. Yes, there is a strong 25 correlation between the rating agencies, but there are 1 nuances to our approaches, changes in our approaches, 2 differences in our approaches. 3 It sometimes makes it difficult to understand what 4 the total picture is. You may understand exactly what DBRS 5 does or even Egan Ratings and S&P, but then when you put it 6 all together in one situation, it becomes difficult to 7 understand what is driving the total mix. 8 I think that's appropriate. This is a market that 9 is based upon the independence of opinion and investors' 10 ability to ferret that out and make it ultimately their own 11 decision. 12 MS. STRAUSS: I'd like to just have another voice 13 in this, which is, again, we are here talking about advocacy 14 for the marketplace. 15 Two points. One, Mr. Macdonald, you may feel that 16 the confidential conversations you're allowed to have with 17 the rating agencies makes a better rating process, but we 18 would disagree. From the rating process to the marketplace 19 is really what is important here. 20 Again, I'm not sure that that is always the case. 21 It may feel better that you don't have to give information in 22 a more public forum and, therefore, you think that the 23 outcome of the rating is better, but I don't know that that 24 is the case. 25 I can see many situations where that is not 1 necessarily the case. What makes the rating good is how well 2 the agency analyzes the company, makes an assessment, and 3 clearly articulates that to the marketplace, so the 4 marketplace can make its best assessment of that. 5 I also want to remind you that many investors, such 6 as ourselves and Federated, we're fiduciaries, and the SEC 7 specifically asks us to make independent assessments of these 8 companies. 9 In addition, with just very specifically saying you 10 should not rely on the ratings, with your regulations 11 requiring us to do this, and yet you are also saying the 12 agencies can have access to things that we don't, it puts us 13 in a very -- we think that is where the conflict is. 14 We like the process the rating agencies do. We 15 think they have vigorous discussions that are healthy. But 16 we also, in the marketplace, have responsibilities that are 17 separate from that, which are fiduciary duties, which we take 18 very seriously and we think are appropriately put into Rule 19 2a-7. 20 Different perspective. 21 CHAIRMAN PITT: Mr. Egan, could I just ask you one 22 question? Do you find that there are circumstances where you 23 do not get information you request? 24 MR. EGAN: Yes. In fact, it's rather frustrating 25 that the information is provided to the other NRSROs and it 1 is not disclosed to the market shortly after it is provided 2 to them. 3 So, yes, but we manage. In fact, we tend to think 4 that this whole Regulation FD issue is outweighed by the lack 5 of competition and the conflict of interest. 6 We have been getting ratings right and by measuring 7 of whether or not the other agencies come in to our ratings. 8 So it's very important. We think that Regulation 9 FD should address this issue. Obviously, some other people 10 share our opinion. 11 But it ties into some other things, too, that have 12 -- that are more important. Otherwise, we wouldn't be able 13 to -- we wouldn't be in business. We wouldn't be able to get 14 the ratings early and right. 15 COMMISSIONER CAMPOS: I have a question, Cynthia. 16 We have two perspectives. Mr. Macdonald says he needs 17 confidentiality. You're saying, yes, but from the buy side, 18 I don't get information that the ratings agencies get. 19 How would you square that? What, in your mind, 20 creates the right balance to satisfy the marketplace and 21 satisfy the need of companies to keep things confidential, if 22 you acknowledge that there is such a need? 23 MS. STRAUSS: I guess it's pretty simple. I think 24 we would encourage companies to disclose, if it's that 25 important to a rating, that it should be disclosed publicly. 1 There are finer points to things. 2 COMMISSIONER CAMPOS: What about the future 3 products and if that's going to mess up my competition with 4 GM or something, what would you do about that? 5 MS. STRAUSS: Quite frankly, I personally, and 6 others in our group have been long-time analysts of the auto 7 companies and I don't think the ability for the debt holders 8 -- we're talking about the debt markets here, which are 9 looking at future earnings streams with a particular weight 10 on liquidity, cash flow, sustainability. 11 I don't think you need to know what the latest new 12 design on the auto -- for a new auto is to assess their 13 financial capabilities. 14 Again, I don't know whether all that is released to 15 the ratings agencies, but I haven't felt that we -- with the 16 auto companies, to be fair, I don't think that we feel like 17 we're going without important information. We get, I think, 18 pretty good disclosure. 19 We always want it to be better, but -- so the idea, 20 what is worrisome is that there -- I just don't think it's a 21 good area to be in where there is this confidential 22 information. 23 We don't have it and then we're reliant, we the 24 marketplace, on the rating agencies doing the right thing 25 with it. I just think it creates an uncomfortable imbalance, 1 which can also add volatility. 2 One of the issues I'm very concerned about now is 3 the increased volatility in the debt markets, and there are 4 many factors for that. But one of them is if a rating agency 5 moves its rating in a way that we can't understand, because 6 it may not be transparent and clear, the marketplace right 7 now, one of the things they wonder is why are they doing 8 that, what is it that they know that is not disclosed to us, 9 and I don't think that favors companies. 10 It certainly adds to volatility and I just don't 11 see the benefit for investors or issuers. 12 MR. O'NEILL: I think we try to. We try to go to 13 great, great lengths to explain why we lower our ratings. 14 I think we're in kind of a unique period. I mean, 15 hopefully, it's unique, and the sensitivity about 16 creditworthiness I think is probably at a pretty all time 17 high, maybe going back to the mid '70s or early '70s, when 18 people woke up to credit risk. 19 So, obviously, there is great sensitivity. Credit 20 spreads are very, very wide here. A lot of uncertainty, 21 trust, we've all heard all of these things in the 22 marketplace. 23 There have been periods of time, and not too long 24 ago, when rating agencies lowered ratings and did those sorts 25 of things and the market didn't -- you know, it made 1 absolutely no difference whatsoever in any respect. 2 So our view is stay level, stay consistent, and 3 make sure that we are responsive to the information 4 requirements of the markets, and also stay independent. We 5 can't be adjusting our ratings because other rating agencies 6 do certain things. 7 So that's our mantra. 8 MR. HARRIS: I wanted to follow up on a question 9 for Ms. Strauss, because you all are in the same position, 10 theoretically, as any other stockholder out there. In fact, 11 Reg FD was pretty much aimed at you all, because in the past, 12 companies would share information on road shows with 13 institutional investors rather than with individual 14 investors. 15 So it seems to me that if there is a 16 confidentiality agreement with a credit agency, that that 17 acts as sort of a safety valve, a pressure valve to get 18 information out into the marketplace in more of a rational 19 way. 20 So I was wondering if you all had any opinion about 21 that. 22 MS. STRAUSS: I would just like to -- we did, 23 actually, both Debbie Cunningham and ourselves actually 24 addressed this issue at the time of Reg FD. Reg FD was 25 addressing concerns about companies, I think, in large part, 1 companies giving equity investors information that wasn't 2 broadly available, a concern, a serious concern for all of 3 us. 4 But particularly in this market, we're talking 5 about the debt holders and so we don't particularly have 6 people running to us trying to give us a lot of extra 7 information about their future earnings or try to influence 8 us in that part. 9 MR. HARRIS: But it did in road shows. I mean, 10 that was what FD was aimed at. 11 MS. STRAUSS: I understand that, but I can't -- 12 Fidelity did not feel that they were abusing that, getting 13 tips on that. 14 But let me just get over to the debt markets. One 15 of our concerns at that time, and is always the case, is, 16 again, the position of us. We are fiduciaries and under Rule 17 2a-7, we are specifically required to make an independent 18 assessment of a company's debt metrics, which includes 19 liquidity. 20 Disclosure does not focus on the debt markets. It 21 typically focuses on the equity markets. So we, for years, 22 relied on conversations with companies about their backstop 23 facilities, their liquidity profiles, how they are planning, 24 if the markets throw them -- if they're throwing out of the 25 markets, to manage their short-term debt positions and so 1 forth, and our great concern about Reg FD is that it is not 2 typically disclosed in financial statements because they tend 3 to aim at the stock market, which isn't typically -- maybe a 4 little more recently, but is not typically interested in the 5 liquidity. They are interested in earnings growth. 6 I don't know if that really answers it, but we're 7 not looking -- we don't want non-public information. We're 8 not seeking to get extra information people are not -- our 9 business is taking what information is available and making 10 the best possible decision with it. 11 So that's why I say this whole issue of the rating 12 agencies, potential or theoretical, I mean, I don't know how 13 much they get, just seems to me to create an imbalance. 14 It's not a big thing we worry about, but I think it 15 does add concern in the marketplace when an agency may act in 16 a way that we don't understand. 17 We sit there and one of the first things we wonder 18 is what is it that they know, and I think that that adds 19 unnecessary volatility and uncertainty to the marketplace. 20 That's it. 21 MR. O'NEILL: We would strongly advocate that there 22 be more disclosure, more communication between debt issuers 23 and buy side or sell side analysts. We think there is a 24 great need for it. 25 The debt markets have grown huge in the last 25 1 years and have a huge impact on the future of how retirees 2 are going to finance their future. 3 We need a robust, independent buy side fixed income 4 community. They need to have the right information. They 5 need to have good ratings and good transparency. 6 The debt markets have gotten very complex with 7 securitization and how do you figure all of this out. 8 So clearly there is a need for more open 9 communication between the buy side and the issuer community, 10 and we would advocate that. We encourage that. 11 I don't know what it creates for Malcolm in his 12 work day, but maybe a lot more. 13 MS. NAZARETH: Frank has been very patiently 14 sitting here. He's ready to make a point. 15 MR. FERNANDEZ: First, Cynthia, I'm not 16 unsympathetic. It sounds like you're arguing a lot of points 17 that I argued from our standpoint when it was an equity issue 18 on Reg FD, and I think it would probably be a shock to 19 anybody at the staff of the Commission to find I would agree 20 with you on a Reg FD issue, particularly Commissioner 21 Goldschmid. But I do have to agree with you on this. 22 COMMISSIONER GOLDSCHMID: Goldschmid will note that 23 one basic distinction between the rating agencies and 24 everyone else in terms of analysts is they're not doing it 25 for selfish interest. 1 Everyone else can trade and trade ahead to the 2 public and tip clients, and that is a very serious market 3 problem. 4 MS. NAZARETH: Could I interject, also? I think 5 there has been some confusion on what the exception is for 6 FD, and you know it far better than I do. 7 But it's not limited to NRSROs. It's for people 8 involved in the credit rating business. So whether a firm 9 has that designation or not is not, at least by the words of 10 the exception -- 11 MR. EGAN: The reality in the marketplace, the 12 reality is that companies will only give that confidential 13 information to the currently recognized NRSROs. 14 I think it is quite ironic that Fidelity, one of 15 the biggest investors, is arguing that Regulation FD be 16 adjusted for the rating agencies. What about the middle tier 17 firms or the small firms? 18 There is no reason why a couple of private firms 19 should be given market information ahead of everybody else. 20 It's crazy. I mean, how is the investor being benefitted by 21 that? They're not. 22 MR. FERNANDEZ: I think this is the point -- if I 23 could finish. 24 MS. NAZARETH: Go ahead, Frank. 25 MR. FERNANDEZ: You hit on one of the points here. 1 I think you have to distinguish between the exemption 2 provided for someone who is providing a quasi public function 3 here. 4 You are not trading in the market if you are an 5 NRSRO. You are not acting on this for portfolio adjustments 6 or any other purpose. We hope. 7 But I think that's a point that we should address, 8 because I don't believe that this is a Reg FD issue and I 9 think there are more important issues that need to be 10 considered today than this one. 11 But there is another flaw, I think, in here, is the 12 failure to distinguish, and you alluded to it, between 13 information and analysis. At a certain point, you have to 14 draw the line there in terms of what is disclosed and then 15 the mosaic that may be created by the analysts with that 16 information or conclusions he may draw or his competitive 17 ability to ferret and drive from a piece of information that 18 he is provided. 19 That's different than the disclosure of a piece of 20 material information. So I think that distinction has to be 21 drawn, as well. 22 The second is the inference that is being made here 23 that there is some evidence that there has been misuse of 24 this privilege, this safe harbor, this exemption that has 25 been provided, and I would like to hear what that is, because 1 I've looked at this and we not only look at the application 2 of this for just the NRSROs, but let's remember there's like 3 37 different credit rating agencies, CRAs, globally. 4 Then there is, in addition to the surveys we do of 5 our own risk management functions, the particularly the very 6 sophisticated large institutions and their risk management 7 systems that also have some regulatory obligations in terms 8 of providing oversight. 9 I hear and see nothing either on an anecdotal or a 10 secondhand nature that there has been any abuse of this at 11 all. So I think in conclusion on this generalized point, I 12 hate to agree on a Reg FD issue with you, but I think that 13 this is not the issue that we are treating here. 14 You can argue a lot that there are impediments to 15 accurate appraisal, and if there is a problem, if you have a 16 problem, do what we have had to do. If you think they know 17 something you don't, go to the person who has the obligation 18 to disclose what that material piece of information is and 19 ask the company themselves for an explanation. 20 MS. STRAUSS: We actually have not had a specific 21 concern -- I don't have a specific example with the rating 22 agencies on this. So I was just clarifying the various 23 points here. 24 Also, as you can see by the fact that I don't know 25 what they get that we don't, if they do, they're keeping it 1 very confidentially. We don't get tipped, we don't get -- 2 MR. FERNANDEZ: Quite honestly, on the sell side, 3 when you're doing this, you discreetly or explicitly 4 incorporate all the publicly available credit rating 5 assessments out there as part of a reflection of market 6 sentiment, and that tends to be actually a numerical input 7 into your evaluation process. 8 And it's a two-way flow, in a lot of senses. So 9 when there is a deviation even between your assessment and 10 those of the three rating agencies, you do want to know what 11 the basis of that deviation is. 12 I'm not a lawyer, but if I understand FD correctly, 13 the obligation then for that disclosure is on the part of the 14 issuer and you have a right to certainly go ask the issuer is 15 there something that I am missing here. 16 MS. NAZARETH: Right. Commissioner Goldschmid? 17 COMMISSIONER GOLDSCHMID: How good are these 18 ratings? How good are they over time in terms of correcting? 19 MS. NAZARETH: I think that's a very important 20 question. I think we should start with that. Why don't we 21 take a 15-minute break and come back with that very important 22 question. 23 Thank you. 24 (A brief recess was taken.) 25 MR. HARRIS: Welcome to this session on information 1 flow in the credit rating process. 2 Access to current reliable information is an 3 essential ingredient in making a timely and credible credit 4 assessment. This holds true for credit rating agencies, 5 which need a continuous flow of accurate information to 6 assess the creditworthiness of a specific issuer or issue. 7 It also holds true for many issuers of credit 8 ratings who need explanatory information to understand the 9 rationale upon which a credit rating is based. 10 During the past few years, a rapidly changing 11 credit environment and a number of high profile bankruptcies 12 have highlighted the need for the communication of reliable 13 and timely credit information to credit rating agencies and 14 to the users of credit ratings. 15 To better understand how information flows in the 16 credit rating process to and from credit rating agencies, we 17 will focus this session on issues related to the flow of 18 information in the credit rating process. 19 We will begin the session by discussing a number of 20 issues concerning the flow of information from issuers to 21 rating agencies. We will consider the types of information 22 made available by issuers and deemed significant by rating 23 agencies when issuing credit ratings, the access credit 24 rating analysts have to senior level management of an issuer, 25 the extent to which rating agencies independently verify 1 information obtained from issuers, and steps taken by rating 2 agencies when they are unable to obtain complete or reliable 3 information. 4 Of course, you recognize I am reading from set text 5 here. We have already done much of this and we will adjust 6 ourselves a little bit accordingly. 7 We will then focus on the flow of information from 8 rating agencies to the users of ratings. We will consider 9 the method by which credit agencies disseminate their ratings 10 and notices that ratings are under review to subscribers and 11 the public; whether sufficient information is publicly 12 available regarding credit ratings, the rating process, and 13 the information on which a rating is based; and the extent to 14 which information concerning the rating is available to 15 subscribers that is not available or not publicly as readily 16 or as soon to the public. 17 Two specific areas that we would like to discuss, 18 actually, we have already discussed one of these, are the so- 19 called rating triggers and Regulation FD. I suspect we'll 20 skip much of FD, since we've done a good job of it already. 21 Rating triggers are provisions in financing or 22 operating agreements that require a company to retire its 23 financing or post new collateral with counterparts in the 24 event of creating a credit rating for the firm or if its 25 outstanding obligations decline below a certain level. 1 We would like to discuss the extent to which rating 2 agencies are aware of credit triggers and whether rating 3 agencies are hesitant to issue a downgrade that would 4 activate a credit trigger. 5 We can talk a little further about Regulation FD, 6 but I suspect we've done a pretty good job. 7 There are also, in addition, a few topics that we 8 didn't have a chance to get to from the last session that I 9 would also like to address, and I'm going to start with a 10 very fundamental issue that I think we need to explore 11 further as we consider the flow of information in the credit 12 rating process. 13 What exactly is a credit rating? All of us are 14 familiar with it, but we need to understand exactly what a 15 credit rating is to understand many of the issues that we are 16 discussing today. 17 So let me start by asking Leo, with a blank piece 18 of paper, what is a credit rating? What purpose does it 19 serve? 20 MR. O'NEILL: Actually, I hope I can answer this. 21 COMMISSIONER GOLDSCHMID: And, Leo, I'm going to 22 push you to go on and say how good are they and how good are 23 they over time. 24 MR. O'NEILL: The credit rating is a relative sense 25 of the creditworthiness of an obligor, what is the 1 likelihood, relative likelihood that the obligor will be able 2 to make its debt repayments in a timely fashion over a period 3 of time. 4 A rating is assigned at a point in time. 5 Consequently, if events change or circumstances change, 6 ratings will change accordingly, and that's why we maintain 7 surveillance and that's such an important part of what we do 8 on ratings. 9 In terms of the quality of ratings, how well do 10 they work, clearly, as I think was pointed out and all the 11 statistical evidence will show, there is a very, very strong 12 correlation between the ratings and ultimate default. And 13 I'm proud to say it's a positive correlation. The higher the 14 rating, the less likely that an issuer will default. 15 A rating is not predicted in the sense that I can 16 tell you today which issuers are going to default at some 17 point in the future. 18 What I can tell you is that AAAs, for example, one- 19 half of one percent, roughly, of AAA will, in all likelihood, 20 default sometime over the next 15 years. I can also tell you 21 that if you are rated single B, that likelihood of default 22 rises to about 50 percent over that period of time. 23 Interestingly enough, and I think this supports the 24 earlier discussion, obviously, there is great need for 25 investors to have their own analytical expertise to identify 1 which of those 50 percent don't default and they may disagree 2 about percentages, even. 3 So I think that sometimes people believe because we 4 rate something single A, that that guarantees that this issue 5 is not going to default. No, it doesn't. It just creates a 6 relative likelihood. 7 Finally, the whole quality issue of ratings, I 8 think that you cannot accept the rating just solely on its 9 symbol. That symbol is usually accompanied with something 10 that's called an outlook or credit watch or a rationale that 11 decrease the dependency, the dependency of that rating on 12 certain conditions that might take place in the future. 13 So to just take the number, the letter grade and 14 say that's it, no. You have to look at it in the total 15 context of the opinion that we give. 16 The final comment I will make, the quality of that 17 rating ultimately depends upon the accuracy and the 18 timeliness, and the disclosure that is provided to the rating 19 agencies. 20 I would clearly say that when that disclosure is 21 flawed, as it was in many of the situations we have 22 unfortunately had in the last couple of years, I'm sorry to 23 say that the basic premise of that rating is suspect, because 24 it cannot accurately reflect credit conditions at a point in 25 time without that accurate disclosure. 1 COMMISSIONER GLASSMAN: Can I ask a question on 2 those statistics? What is the time frame that you're talking 3 about in terms of if you have a AAA? 4 MR. O'NEILL: Generally, what we've said, we 5 generally like to take a view that goes out two to three 6 years in terms of what we think we can understand today about 7 current conditions in the future. 8 I think what we've seen in the past several years, 9 and other periods in history, is that the volatility tends to 10 accelerate in periods of economic decline and economic 11 recession. 12 So what happens is the events change. They change 13 quickly. Conditions worsen much more so than what we 14 foresaw, perhaps others foresaw, and hence the ratings tend 15 to change with more volatility during periods of high 16 economic stress. 17 MR. HARRIS: Can you describe more completely the 18 process by which ratings are reissued? 19 MR. O'NEILL: A rating is -- first of all, the 20 process, very simply stated, is we assign two analysts to 21 every rating that we do, a senior analyst or a lead analyst 22 and a junior analyst. 23 We will meet with issuers. We will gather all the 24 information. We will then do all the analysis and assign a 25 rating. We then begin a continuous review process on that 1 rating. We don't just put it on the shelf and bring it out a 2 year later, dust it off, and take a look at it. It's a 3 continuous review process. 4 To the extent, from the earlier discussion on 5 conversations with third party institutional analysts and 6 everything, we control that very, very precisely. 7 The only people that are allowed to speak to the 8 media or to others, including investors, are the analysts who 9 follow the company, and they are clearly instructed that 10 their position is the institutional position of Standard & 11 Poor's. We do not have a Star system. We don't have a 12 Grubman or Blodget or anything like that. 13 This is an institutional process. The rating is 14 assigned by a committee. The committee is responsible for 15 the assignment of that rating. The analysts are responsible 16 for surveilling, monitoring the credit conditions and 17 bringing those changes to the attention of the committee. 18 This -- sorry to go on here, I apologize here, but, 19 obviously, it means a lot to me. 20 The problem that we have is -- and I don't know if 21 it's a problem. It's an operating stance we have. 22 Conditions begin to change, especially in an environment like 23 this. We will put the rating on a credit watch. We think 24 that issuers in some of this environment have a right to 25 address our credit quality concerns, and we have a need to 1 make sure that we fully understand that rating or what those 2 conditions are. 3 So sometimes when you hear, well, S&P didn't rate 4 something to here and to there, actually, we put it on credit 5 watch and sent that very negative signal, in most cases, to 6 the marketplace that this rating is under intense review. 7 So I come back to my earlier point, when looking at 8 the performance of ratings, it's really critically important 9 to understand the context that surrounds and the rationale 10 that surrounds that rating. 11 Now I'll be quiet. 12 MR. HARRIS: Are the procedures at Dominion 13 similar? 14 MR. ROOT: Overall, I'd say definitely. Just a 15 couple other points, again, that Leo made. First of all, 16 again, all our ratings are done on a committee basis. 17 So this way, any one analyst may be biased 18 positively or negatively, which could influence the rating. 19 By doing it on a committee basis, we feel that it avoids that 20 kind of an issue. 21 Another, I think, important aspect is that before 22 we finalize the rating, initial rating or rating change or 23 whatever, we communicate that with the company, so that we're 24 not just blind siding them at all, that they have an 25 opportunity -- and we explain to the company what our 1 concerns are and why and give them a chance to respond, which 2 I think is a very important aspect. 3 The other thing is kind of what Leo is alluding to, 4 is that we have the rating, but in addition to the rating 5 itself, we use trends, so that if there's -- if we try and 6 project, and, again, we don't have the crystal ball, but as 7 we look forward, to the extent we think the next rating 8 change will be downward, we'll have a negative trend. 9 Then to the extent that we have more concerns or 10 there is an event, we'll put it on credit watch, and then the 11 next step. 12 So it's not -- again, there will be circumstances 13 potentially that surprise us and you have to respond quickly, 14 but beyond that, what we try to do is have a communication to 15 the investors and to the issues about where we see things 16 going and have an orderly process, so that rating changes 17 hopefully don't tend to be disruptive, and I think that's 18 very important. 19 Again, the research and the information that we 20 publish with that, I think, is an important aspect of this 21 whole process. 22 MR. HARRIS: What does it mean to have a disruptive 23 rating change? 24 MR. ROOT: I guess I should let the investors speak 25 to that. But I think to the extent that you can at least 1 communicate effectively where your concerns are and people 2 can adjust so that when a rating change happens, they're not 3 surprised, I think the one thing -- noone likes surprises. 4 Rating agencies don't like surprises. 5 Investors don't like surprises, and the issuers 6 don't like surprises. The issuers don't want to come in and 7 meet with you and the next day they see a press release 8 saying you've just downgraded them, which would never happen 9 at DBRS, by the way. 10 So I think, again, to the extent that you can be 11 open and communicative to the marketplace, I think that is 12 what is critical. 13 MR. HARRIS: Just a couple more questions on the 14 issuing of ratings in an attempt to understand exactly what 15 ratings are. 16 I presume that when you reissue a rating, that the 17 analysis is based upon all available information, including, 18 of course, all information that has been revealed since you 19 last did the analysis, is that correct? 20 MR. ROOT: That is correct. 21 MR. HARRIS: And that would be the same for you, 22 too. Which is to say that when an issue is re-rated, any 23 change in the rating must be based on something that was not 24 available at the time you last did it, is that correct? 25 MR. ROOT: Let me, again, give you my view on this. 1 To me, ratings are reviewed every day, in essence. There is 2 a formal review process that would be triggered off of a 3 quarterly earnings statement or a meeting with management or 4 some other event that would maybe trigger a review process so 5 that it's done on a regular basis, but, again, essentially, 6 our view is that we are watching ratings every single day. 7 So to the extent that ratings change, there could 8 be a variety of factors that cause us to change our rating. 9 Again, it could be component specific. It could be an 10 external factor that we think would have an impact on the 11 creditworthiness, positive or negative, on that institution. 12 So there is a whole variety of reasons why a rating 13 may change, but, again, you have to keep in mind this is a 14 collective, educated, and professional opinion. 15 MR. HARRIS: I understand that, but my point is 16 this: if the rating is changed from the last time, if, 17 indeed, the last time it reflected all of the information 18 available to you, the rating change must result from 19 information that was newly revealed in some sense. 20 MR. REYNOLDS: Not necessarily. 21 MR. HARRIS: Or that became more significant in 22 some way. 23 MR. ROOT: Again, I think -- 24 MR. HARRIS: Would that be safe to day? 25 MR. O'NEILL: I think from a classical efficient 1 market theory, that if a market encapsulates all available 2 information and prices it accordingly at a point in time, 3 and, therefore, if, at some point in time, the price changes, 4 there must be new information. 5 I guess a rating is analogous to that model, and so 6 I don't disagree. 7 MR. HARRIS: This is very important. 8 MR. O'NEILL: Now I probably will. 9 MR. HARRIS: Yes, you will. This is a very 10 important point, because as we are trying to understand what 11 ratings are, what you have described, I believe, is 12 inconsistent with the pattern of rating revisions that we 13 see. 14 In particular, we know that rating revisions don't 15 follow random walk. That is to say that a revision is not 16 equally likely to be up as it is down, but rather they tend 17 to have far more continuations. 18 And so the question is what are ratings, you 19 would expect them to follow a random walk if indeed they are 20 fully aggregating information, because all changes then would 21 be based only on things that are fully unanticipated, and 22 unanticipated stuff could be either good or bad. 23 So the question is why are the ratings not 24 reversing, say, 50 percent of the time? I understand the 25 continuation rate when there's a change more on the order of 1 92 or something like that percent of the time. 2 MR. O'NEILL: I'm not sure that there is a cause 3 and effect of what you're saying and what ratings are, but I 4 think the reality is that when we lower a rating, it is in 5 response to a set of conditions that exist and to the extent 6 that those conditions worsen, the rating will be worse. So 7 maybe there is new information in the marketplace. 8 The fact that they tend to continue to go down 9 probably is a reflection of the fact that the situation is 10 getting worse, and then we're adjusting accordingly. 11 I don't think that the role of ratings, though, is 12 to try to guess, when we do a rating and lower it from BB to 13 BB-minus and say, well, I guess it's going to go into 14 default. We're assessing what the current information is in 15 the marketplace at that time. 16 MR. HARRIS: This is a question to everybody, 17 though I can direct it to some, is there a particular problem 18 with ratings that reverse too frequently? 19 MR. REYNOLDS: Can I jump in on that? 20 MR. HARRIS: Sure. 21 MR. REYNOLDS: That has been one of the problems we 22 have had in the market this past year. Because the market 23 has been under pressure, it is easy to justify, but at the 24 same time, there has been a pronounced acceleration in fallen 25 angels, high grade credits moving to speculative grade. 1 To add to your point, it's not just -- 2 MR. HARRIS: I'm confused. Is it a problem that 3 ratings are changing? 4 MR. REYNOLDS: Changing very quickly on the 5 downside in the immediately aftermath of them having printed 6 bond deals. 7 So to your point about new information, it's not 8 just new information. It's new information versus 9 expectations and the problem in the marketplace is it's not 10 clearly laid out exactly what those expectations are. 11 MR. HARRIS: I'm still unclear. Is the problem 12 that they changed or that they didn't change earlier? 13 MR. REYNOLDS: The problem is that they didn't 14 change when the bond deal was priced, leading investors to 15 believe that there's some time horizon where they will get to 16 prove out the variables. 17 So what happens is the rating agencies get a fee. 18 The brokerage houses get a fee. The banks lay off their risk 19 and the one holding the bag at the end of the day is the 20 investor, who has stepped in with a set of assumptions about 21 how the rating agencies rated it in the first place, whether 22 it be a bond deal in October that's downgraded in November or 23 an eight billion dollar debt deal done by two telecom 24 companies in March which are downgraded or watch listed in 25 June. 1 We just need to understand the variables so we can 2 assess accordingly and price risk. 3 So it's not the action itself. It's the sense that 4 you really, on the prior ratings action, didn't have the full 5 picture to understand how the ratings would respond going 6 forward. 7 So in terms of the random walk and all that, it's 8 not a case of people getting new information. It's really 9 not understanding firmly the criteria. 10 MR. HARRIS: Sean? 11 MR. EGAN: What we see in the market is a lot of 12 frustration with some of the actions of the NRSROs. For 13 example, investors just can't swallow that Enron was an 14 investment grade credit four days before bankruptcy or the 15 California utilities were rated A-minus about two weeks 16 before they defaulted, or they have ingenuity, you have a six 17 notch downgrade in the period of two months. 18 It's just -- 19 MR. HARRIS: Sean, how would you explain why the 20 market itself didn't anticipate those effects? 21 MR. EGAN: The market doesn't have the information 22 is the reason why, because the market, historically, has 23 looked to the rating agencies for that information and they 24 haven't fulfilled that role very well. 25 We, fortunately, have stepped in and caught most of 1 these, but I think there is an evolution in the market's 2 perception of what a rating is. 3 I'll give you the story of a person in Colorado. I 4 think they headed up the Colorado state pension fund. They 5 said they couldn't believe that they weren't going to get all 6 their money back on Enron. They said this thing was rated 7 investment grade, how in the world can something that was 8 rated investment grade, how can we get less than ten cents 9 back. 10 Our response is believe it, because the company 11 isn't worth it, and then they got also caught in Worldcom, 12 and then they got caught in Global Crossing. 13 It just is a huge eye opener that some of these 14 ratings aren't as good as they had thought they were. We 15 explained that our business is protecting investors. We're 16 not looking to help companies at all. In fact, we regularly 17 alienate companies, but we have to assign timely, accurate 18 ratings, and we do look to the future, too. 19 We believe that our ratings should reflect what the 20 credit quality is going to be within six months, between six 21 and 12 months out. 22 We also don't buy into this fraud, the company 23 isn't giving us accurate information, therefore, we can't 24 assess credit quality accurately. 25 We don't believe the information that typically 1 comes out from the company. We didn't believe that Bernie 2 Evers had the backing for his $400 million loan from the 3 company. 4 We didn't believe that the capital expenditures 5 that they listed were truly capital expenditures. We treated 6 them as expenses. It's like the fast food business, where 7 you have to put in the capital expenditures to keep up. You 8 have to put in the new lines. You have to upgrade the 9 software. 10 So we made the adjustment. We saw that the 11 coverage was being cut in half. So, basically, we're putting 12 ourselves in the investor's shoes and trying to get to the 13 truth quickly. 14 MR. O'NEILL: I think I really need to -- this is a 15 public record and I think we need to correct the public 16 record. 17 On Enron, on November 9, I believe it was the 18 eighth or ninth, Standard & Poor's put its BBB-minus rating 19 on credit watch with a very strong statement stating that 20 that rating was dependent upon the pending merger agreement 21 with Dynergy and the failure of that agreement would cause us 22 to lower our rating to BB or B. 23 I think Sean would even agree, I think the 24 following day you raised your rating, as a matter of fact, 25 from BB-minus to BB in the middle of November. So you, 1 obviously, were depending upon the disclosure the 2 organization was making. 3 At one point, you had an A-minus when we all had 4 BBB-plus. So you were, obviously, depending upon the level 5 of disclosure and the accuracy of the disclosure that was 6 being made in the marketplace. 7 I'm surprised you're not shocked by what Enron did 8 and Worldcom, to some extent. I would think you would have 9 the same righteous anger that we in the rating agency 10 industry have about this. 11 MR. EGAN: You can look at our Enron record and you 12 can see -- and by the way, it's in the material that was 13 included in the package that I placed -- that everybody 14 should have. 15 You see on June 27, 2001, we cut our rating from 16 BBB -- that was our first negative action, and you see how we 17 cut it to non-investment grade. 18 MR. HARRIS: I'm going to cut you off here. Our 19 purpose today is not to do a forensic investigation of what 20 happened with Enron, except perhaps to allow us to understand 21 more clearly what exactly ratings are. 22 In the case of Enron, it is alleged that there was 23 a lot of hanky-panky by quite a number of people, and, as has 24 been stated here, the ratings can only be good as the 25 information that is given. 1 While ratings will never be perfect, they always 2 could be better, we have some other issues to address. 3 I want to return, though, to the question I asked 4 before, which is this: is there a problem with ratings that 5 reverse? That is to say, a rating that is perhaps -- let's 6 not even talk about new issues, as Glenn spoke of earlier, 7 we have a seasoned issue. It's rated AA. It drops to A and 8 then months or somewhat later it goes back to A. 9 This doesn't happen very often. I'm focusing on 10 this because it is very important to understand whether the 11 ratings are being used for their information content or 12 whether they are being used for some other purpose. 13 So the question is what is the market resistence to 14 ratings that reverse too often? Cynthia? 15 MS. STRAUSS: Thank you. We also observe that that 16 doesn't happen very often, and I think one of the reasons 17 that it doesn't happen is that the agencies are conscious 18 about the statistics they want to follow and have it so they 19 are highly unlikely, given that that is one of the key 20 measures of how well they're doing, they're not going to take 21 their ratings up and down. It actually is quite rare, unless 22 there is a very clear event to justify that. 23 So I don't know how the market would react. When 24 the market gets unhappy is when they don't understand why the 25 ratings are moving, which gets back to this point of clarity 1 of criteria. 2 You made an interesting point, I thought, about the 3 different and disruptive rating changes versus rating 4 changes, and as Leo said, we can be in a volatile time where 5 ratings may, due to the fundamentals, need to change very 6 quickly because the company's profile is changing. 7 But a disruptive rating change is when the ratings 8 are changing very quickly and it's not observable to the 9 market why they are changing so fast at this pace, and we 10 have seen some of that, and I think that does trouble us. 11 An example might be a rating downgrade, some of the 12 telecom names, and the reason that the agency will say we're 13 looking for the company to do certain things and then the 14 company does those things and then the rating goes down. 15 As a market participant, our only question with 16 respect to the agencies is have you changed your criteria, 17 because it seems like you're moving your ratings in a 18 different way and that leaves the market uncertain and, 19 therefore, I think, adds volatility and uncertainty. 20 That's why we say we want clarity of criteria and 21 transparency in living up to that, because it is the rating 22 agencies that have to set their own criteria. 23 That has happened a great deal, and I think we 24 worry sometimes that that makes it very difficult for 25 companies. 1 MR. HARRIS: I would like to go there, but not 2 right now. We're going to stay just a moment more on quality 3 of the ratings in an attempt to understand exactly what this 4 product is. 5 I would like to ask Kent if he could briefly 6 summarize some of the academic evidence, not necessarily your 7 own, on quality of ratings. 8 MR. BAKER: Just as was mentioned by the gentleman 9 from S&P, if you look at the quality of the ratings in 10 relationship to defaults over time, that evidence has been 11 quite strong, very high correlations on what the ratings are 12 and what the eventual defaults have been. 13 There's really a substantial amount of evidence in 14 support of that. 15 MR. HARRIS: What of the relationship between 16 ratings as predictors of default and credit spreads as 17 predictors of default? 18 Again, recognizing that we're not predicting who is 19 going to default, but the probability of defaulting, which is 20 a subtle distinction, but extremely important, if credit 21 spreads do as much or better than ratings, then perhaps 22 ratings are doing something else that we haven't put our 23 finger on yet. 24 MR. BAKER: From my review of the literature, the 25 jury on that particular issue may still be out in the sense 1 that I can show you support for either one of those 2 particular positions, and that's why I'm saying the jury may 3 be still out. 4 There is support in some of the literature that 5 credit ratings do a better job. On the other side, you do 6 have that with the spreads. 7 So I think the jury is out on that as far as my 8 view would be concerned. 9 MR. ROOT: Just one point, and I think it's 10 important because on the issue of spreads and ratings, at 11 Dominion Bond Rating, our approach, our philosophy is to rate 12 to a cycle and, again, it's anybody's own estimate or 13 assessment as to whether a change is cyclical or structural. 14 That's an opinion. 15 So we try to maintain stability in our ratings, if 16 you will. I'm sure many other people on this panel will talk 17 about spreads and spreads will certainly widen and tighten 18 over that time frame. 19 So people will say, well, this thing is trading at 20 500 over, and yet their rating hasn't changed, I think you 21 have to look at this over the entire time frame of the life 22 of the security to determine whether or not the ratings and 23 the spreads -- you know, at some point, they will come 24 together, I believe. 25 Usually, they will come together because the issue 1 will pay off and then the spreads will come back to where 2 they were originally, but there may be cases where they 3 default. 4 So I guess my point there is, again, ratings, I 5 think, are different than day-to-day spreads in the credit 6 markets, and I think that's important because I think there 7 is sometimes too much attention paid to spreads at any one 8 point in time, and, therefore, that's a predictor or that's 9 an indicator whether ratings are accurate or not, and I think 10 that's wrong. 11 MR. HARRIS: Greg, let me paraphrase what you've 12 said and ask Leo whether he is in agreement with my 13 paraphrasing. 14 That is to say that ratings are designed to assess 15 the relative risk of default among the population of firms 16 that are related, whereas credit spreads, to the extent that 17 they are predictive, would be rating an absolute probability 18 of default, and that the rating agency's interest is in 19 assessing the relative risk, is that correct? 20 MR. O'NEILL: I'm not sure I clearly understand, so 21 maybe we can circle back on your point. 22 But I will point out that spreads were at their 23 narrowest, A to ten year treasuries, were at the narrowest 24 days before the stock market peaked at a little over a 100 25 basis points. 1 Today those spreads are about 230 basis points over 2 treasuries. I'm not sure spreads don't follow a certain 3 sense of their own view of what the conditions are, and I 4 think -- and, again, but I think you have to look at any 5 particular security in a spread. 6 MR. HARRIS: I hope, of course, that your point is 7 that the spreads may not be as predictive as might be alleged 8 as opposed to suggesting that perhaps we haven't seen the 9 worst yet. 10 MR. O'NEILL: Also, Larry, I also use single A 11 versus a ten year treasury, so I was doing it with a rated 12 security or a rated universe. It's an interesting 13 discussion. 14 I think our goal is to stay focused on the 15 fundamentals and do what we do best, and the market uses 16 spreads and that's great. 17 MR. HARRIS: Sean, you had wanted to speak, as 18 well. 19 MR. EGAN: Yes. I agree, Leo, that is probably one 20 of the few things that we agree on. 21 MR. O'NEILL: Oh, my god. 22 MR. EGAN: Write it down. It won't happen very 23 often. We think there should be a decoupling of spreads from 24 credit ratings and the information that is derived from it. 25 What I mean is that the way we look at spreads is 1 that is just static information on what the market believes 2 of a particular security. There may be there's a variety of 3 other things, like supply and demand, that affect that. 4 In the package I sent out, you can see some 5 demonstrations of how the spread information doesn't quite 6 capture a lot. About six pages in, you see Worldcom's -- 7 this is not spread, this is prices, but you could just get 8 the reverse of it and get spreads. 9 You see how Worldcom, in fact, traded up from 10 December 2001 to January of 2002, it was trading above a 100, 11 and then it wasn't until February, March, and April that it 12 started dropping. 13 So if you are to take the information back in 14 November of 2001, the conclusion would have been that this is 15 a terrific company and, therefore, don't worry about it. We 16 think that's wrong. You have to look at the underlying 17 fundamentals. 18 Another case where that is demonstrated, where it 19 just sort of falls off very dramatically, is AT&T Canada, 20 about two or three pages afterwards, again, is trading at 100 21 from December of 2000, January of 2001. 22 It wasn't until about February-March and there's a 23 few months, this is printed off of Bloomberg, but it's not 24 until later that it starts dropping dramatically to the point 25 where it goes down to 20. 1 So getting information, driving ratings from this 2 type of pricing information we think is a mistake and if you 3 do it, it will send real aberrations, and I can point to 4 about 20 hedge funds that would love it if a mechanism like 5 that were set up. 6 MR. FERNANDEZ: If I may. If the implication here 7 is that for some reason spreads might be a substitute for 8 ratings, I think that would be a disservice to everyone. 9 Spreads are the reflection of the last trade in the 10 marketplace, and that market may be wrong on any given day 11 about the long-term fundamental value, the probability of 12 default or ultimate recovery value of any security. 13 Indeed, we have just gone through a period where 14 the market collectively was wrong about quite a few issues in 15 that regard. 16 If you are also looking -- and Kent is absolutely 17 right. If you are going to look at the predictive value in 18 terms of probability of default of either spreads or ratings, 19 the jury is still out. 20 It depends upon the time series that you're looking 21 at, the spreads that you're examining, because there isn't 22 one single set of credit spreads. 23 You can look at it with respect to global credits 24 and emerging markets and say that these things predict 15 out 25 of the last five defaults, in many cases. 1 The other thing is that you cannot even address 2 this issue until you recognize that this is the flow of the 3 impact, market impact of both spread changes and rating 4 changes. It's not unidirectional. 5 There is clear feedback and very important 6 threshold effects, and this is where I think you should 7 develop this, because you bring up the issue of ratings 8 triggers, which are becoming growing in their importance, 9 just as the growing importance of proliferation of the user 10 ratings themselves. 11 But the point being made here is that the people 12 who are demanding this are the investors themselves. They 13 want the triggers in there. It's not something that is being 14 driven by the companies or by the rating agencies themselves, 15 because they want a simple, accurate, quick benchmark for 16 them to be responsive to an environment where the nature of 17 risk is changing and becoming more difficult to assess. 18 Risk management practices have to evolve and they 19 are becoming more and more sophisticated, because we are 20 going through a period of very high sustained near record 21 levels of volatility. 22 This has been true for four years now. We have to 23 recognize that we have both cyclical problems and structural 24 problems in terms of the nature of risk and risk management 25 that lead to higher probabilities of default, more rapid 1 shifts in valuation and the perception of those shifts. 2 So to say or to look at any one of these indicators 3 and say, gee, it has more value, these are one of a 100 4 things that any good credit analyst is going to look at. 5 MR. HARRIS: Thanks, Frank, and thanks especially 6 for bringing us to our next topic, the rating triggers. 7 I want to ask Malcolm about the issues that Ford 8 issues, do they have rating triggers in them? 9 MR. MACDONALD: No. 10 MR. HARRIS: And to the extent that you are aware 11 of the market for your issues, do you feel pressure from 12 anybody to put rating triggers into your issues? 13 MR. MACDONALD: We have not come up with any great 14 pressure from underwriters to go for rating triggers, but 15 given where spreads are, I'm staying out of the unsecured 16 market at the moment. 17 MR. HARRIS: And to the best of your knowledge, 18 obviously, it's not in your experience, but you are certainly 19 well informed in this area, why are rating triggers being 20 pushed into these contracts? 21 MR. MACDONALD: The rating triggers, you mentioned 22 that they can call for prepayment. In some cases, a rating 23 trigger will merely recoup on the bond and at that point in 24 time, in effect, what we're doing is we're establishing an 25 automatic mechanism that the investor is being better 1 compensated for presumably a higher risk occurring with a 2 lower rating. 3 MR. HARRIS: And why do you suppose that the 4 covenants are written based on rating triggers as opposed to 5 the alternative, which would be coverage ratios of various 6 types? 7 MR. O'NEILL: I think it's probably because of the 8 lucidity of the identifier. You don't have to get into 9 having to judge what the appropriate earnings are, core 10 earnings, operating earnings, pro forma earnings, that whole 11 discussion. It's clear. 12 But I think we should also point out that rating 13 triggers are only one type of trigger. There's equity price 14 triggers, there's material, there's change triggers. 15 I think the ultimate goal by the investors or 16 whomever it is putting these triggers in is to create, in 17 effect, a preference in time for the repayment of their debt, 18 that they are going to be out early or out first or if it is 19 a complete repayment on it. 20 It could be and I think many more of them are ones 21 that, as you say, recoup on the debt or enter into maybe some 22 kind of accelerated sinking fund or over time whatever. 23 I think an unintended outcome of triggers is one 24 that nobody intends when they do them, and that is to cause a 25 sudden and sharp decline in liquidity of the corporation, 1 which, in effect, creates a situation where the next 2 available step is protection in bankruptcy. 3 MR. FERNANDEZ: That is becoming more likely now. 4 MR. O'NEILL: Which is becoming more likely. 5 MR. HARRIS: Certainly not intended, but certainly 6 not unanticipatable, too. So the next question is directed 7 to Greg, first, then to Leo. 8 What systems or what means do you have of obtaining 9 information about the rating triggers so that you can take 10 that into account, if you do, in forming your ratings? 11 MR. ROOT: To be honest, not a lot of the companies 12 we rate actually have rating triggers, but certainly, in 13 light of the events of the last year, we have a whole series 14 of items that we will cover in the due diligence process with 15 the issuers. 16 So one of the key questions that we do pose to the 17 issuers is are there rating triggers and if so, what are 18 they. 19 So that's just one more piece of information that 20 we put into the puzzle, but, again, we haven't had it at 21 DBRS, where that's been an issue at all. 22 Again, I think disclosing rating triggers is not an 23 issue. Kind of going back almost what Leo said early on, we 24 don't think disclosing rating triggers is a responsibility of 25 the rating agencies, I don't anyway. 1 MR. O'NEILL: I agree with that and I would just 2 say as the rating triggers are really just part of a larger 3 discussion we have about the total liquidity of the 4 corporation, if it is a corporation, and their ability to 5 beat all the liquidity demands on it, triggers could be an 6 important one. 7 When we talk about rating triggers with them, we 8 talk about alternative sources of funding, if, in fact, that 9 trigger is really -- but we will not withhold a rating action 10 because of the existence of a rating trigger. 11 MR. ROOT: I knew that was probably going to come 12 up. At Dominion Bond Rating, that is absolutely the case. 13 MR. HARRIS: You perhaps leave me a little bit 14 confused. If you would, please, rephrase, you will not 15 withhold a rating action based on a trigger. 16 MR. O'NEILL: If there's a rating trigger, let's 17 say single A and we're at A-plus or something, we will go 18 ahead and lower that rating to single A if, in fact, that is 19 the thing we believe. 20 MR. HARRIS: I understand. I'm glad to hear that, 21 of course. 22 The question that I have, though, is this. In 23 deciding what the rating should be, would you take into 24 account the rating trigger? 25 MR. O'NEILL: Yes. If we think that the rating 1 trigger itself is the cause of the potential change in 2 creditworthiness, then yes. We really do. And over the 3 years, we have cautioned issuers about building into these 4 rating triggers in there, because the presumption is that it 5 could create a sudden and sharp decline in liquidity. 6 And I think most of them have been very judicious 7 as a result of that in building those in. 8 MR. HARRIS: Thank you. I want to turn to the buy 9 side and a slight change of direction again, and ask them, 10 we'll start with Debbie, in addition to your own work that 11 you do on credit, you obtain information from the bond rating 12 agencies and I would imagine, given your size, that you are 13 also consulting with independent credit analysts. Is that 14 correct? 15 MS. CUNNINGHAM: That is correct. 16 MR. HARRIS: How do you value the different 17 services you would get from an independent credit analyst 18 from the services that you get from a rating agency? 19 MS. CUNNINGHAM: We would typically value the 20 credit rating agency opinions on a higher plane, a higher 21 echelon than we would the information that would result from 22 Wall Street analysts. 23 MR. HARRIS: And that's based on your experience 24 with the processes or what is it then that you're looking for 25 from the analysts? 1 MS. CUNNINGHAM: It's based on our experiences with 2 the processes, but it's also based on our alignment with 3 effectively trying to create our own internal credit rating 4 agency and knowing what the credit rating agencies do, whose 5 information we are then utilizing as part of our own credit 6 rating process. 7 So we have an internal scale that similarly -- that 8 is a rating scale that we want to adhere to, and I can tell 9 you what a Federated One is versus a Federated Five. 10 I think what we get from the rating agencies and 11 the information on the various types of analysis that we do, 12 looking at liquidity, looking at cash flow, looking at 13 capitalization, looking at earnings and the quality of 14 earnings over time, all that matches up very nicely with the 15 information that we are receiving from the rating agencies. 16 As I said before, sometimes if we feel things are 17 not necessarily as up to date in a written format, we are 18 very much able to have conversations and get those updates 19 from the rating agencies. 20 On the other side of the equation, when we're 21 dealing with Wall Street analysts, we typically have a good 22 information flow of what is already publicly disseminated 23 information and maybe a bit of an opinion based on that 24 public information that may be a little bit different from 25 our own, but there's normally not as much analysis or not as 1 much emphasis placed on the analysis of some of the specific 2 items that we look at through our own credit process. 3 MR. HARRIS: Cynthia, are things similar, do you 4 want to add anything? 5 MS. STRAUSS: I think Debbie said it perfectly, 6 very similar for us. 7 MR. HARRIS: Okay. What information should a 8 credit rating agency disseminate in addition to their 9 ratings? We have heard several people speak to the basis for 10 the ratings and I'm just curious as to how deep should that 11 information go. Glenn? 12 MR. REYNOLDS: I would say there has to be some 13 clarity, at least in the given financial metrics that are 14 required for giving a rating for a given industry. They vary 15 considerably across different industry groups, depending on 16 the volatility of their underlying revenues and cash flows. 17 So you have to be very clear on what the 18 expectations are, what needs to be maintained at a minimum 19 and over what time horizon. 20 It is very difficult, especially in some of the 21 more emerging sectors, such as we have seen in telecom and 22 media and tech in recent years, because they are changing so 23 quickly, to set a time horizon, but that doesn't change the 24 fact that people need to understand what the thought process 25 is and just pure basic objective financial metrics. 1 There probably also has to be some input on what 2 the expectations of the risk profile of the industry is. The 3 telecom, it could be what their minimum expectation is for 4 wireless subscribers, for example, for retail, what the 5 expectation is for store sales, anything that is a key driver 6 of the decision should be laid out there for people to 7 understand, because then it's caveat emptor. 8 You either embrace those ratings given those set of 9 criteria or you don't, or you can have your own set of 10 forecasts around those variables. 11 It allows you to partly predict their behavior, 12 which is part of the game. So one of the things that makes a 13 negative downgrade is when you didn't expect it. 14 At least this would limit the instance of 15 surprises, but the rating agencies are entitled to their 16 views on changing underlying variables, but they are really 17 not entitled to keep the variables to themselves. 18 MR. HARRIS: Would it be fair to paraphrase what 19 you've said as to say that the additional information is 20 valuable because you don't always trust the ratings? 21 MR. REYNOLDS: That's fair. That's fair. Because 22 if I don't understand the ratings, I don't trust them. 23 MR. EGAN: Let me explain the process that we have 24 gone through, and it has been very well received by the 25 market. 1 What we do is when we're looking at a target 2 company, we compare it against other companies within the 3 industry and we come up with benchmarks, if you will, for the 4 different credit ratios that we look at, at, let's say, the 5 AAA level all the way down to the single C level. 6 So that investors can tell what we think is needed 7 at the different gradation for credit quality. 8 Then we provide the investors with historical 9 financial statements, assumptions for projecting financial 10 statements, for both the income statement, balance sheet, and 11 cash flow, and then with those financial statements included. 12 So there is a huge level of transparency. They can 13 see exactly where the company has been, how it stacks up 14 against competitors, how it has developed over time, what we 15 expect will happen. 16 MR. EGAN: Sean, I'm a little confused. Given that 17 that's all in the public domain, why is that a value added 18 service, other than just packaging? 19 MR. EGAN: The assumptions for projecting financial 20 statements are not in the public domain. We put those 21 together. 22 Also, comparison against other companies is 23 something that we drive. You're not going to have all the 24 data points, if you will, for, let's say, a forest products 25 company. You're not going to have a BBB rated, a BB rated, a 1 single B rated or all the way up. 2 MR. HARRIS: And you provide this information to 3 subscribers only or to the public as a whole? 4 MR. EGAN: We provide it to subscribers and we 5 provide the rating to the public as a whole. 6 MR. HARRIS: Now, we've had some discussion already 7 about the possibility that the rating agencies add value not 8 only by virtue of the quality of their analyses, but also 9 perhaps by the high quality of their access to the firms 10 through the exemption, through Reg FD, or perhaps through 11 their market position. 12 It has been suggested that rating agencies are able 13 and need to obtain information that wouldn't otherwise 14 publicly be available. 15 Now, the question is to what extent should that 16 information be shared with the public beyond the rating 17 itself and to whom. 18 We know that some of the rating agencies have early 19 warning services. 20 MR. O'NEILL: Larry, on the one hand, the credit 21 watch is a listing when there's a near-term expectation there 22 will be a change. On the other hand, in language that 23 economists would understand, the outlook is a longer term 24 change in the expectation of change. 25 MR. HARRIS: So the question is what about 1 differential treatment between subscribers and the public as 2 a whole? Should subscribers be permitted to have a peak at 3 information or more information than is available to the 4 public? Should they be allowed to obtain information 5 earlier? 6 MR. O'NEILL: I would ask Malcolm to -- that's the 7 issue, not necessarily ours. 8 MR. MACDONALD: I'm not sure, Leo, that it is in 9 terms of subscribers. Clearly, the point that I had made 10 earlier I will reiterate, that we are comfortable in 11 providing confidential data to the three rating agencies 12 today, because our experience has been that they have been 13 absolutely scrupulous in complying with the confidentiality 14 provisions and because we believe it helps them make more 15 informed judgments as to the credit quality of the company. 16 MR. HARRIS: I understand all about the 17 confidentiality and we've been there and I don't believe that 18 we need to go there again. I have little doubt that there 19 has been very little of confidential nature revealed. 20 The more interesting question, though, is whether 21 opinions that are based on confidential information are 22 being revealed in an unfair way. 23 So let me ask the question in a different way. Who 24 here would object if the SEC said that given the unique 25 position of the rating agencies and their exemption from Reg 1 FD, that they must display all information that they give to 2 anybody to everybody and at the same time? 3 We've got some hands moving here. Great. 4 MR. FERNANDEZ: I think it brings us back. The 5 reason I would object to this is I think we are starting with 6 a flawed understanding of the business model that is involved 7 here. 8 First of all, I don't think anybody has made the 9 case for me that to provide a timely, a prescient credit 10 valuation, or any kind of risk assessment, that you require 11 access to inside information or selective disclosure. I 12 don't think that is the case and I don't think that is what a 13 good analyst of any sort depends upon. 14 MR. HARRIS: Surely you wouldn't suggest, though, 15 that if such information existed, that the analysts couldn't 16 do better with it. 17 MR. FERNANDEZ: A trading advantage that dissipates 18 very quickly. I don't think that's what we're talking about 19 here in terms of a rating agency that has to publish and 20 stand by a rating that they can't change minute by minute. 21 If they were to disclose to provide somebody with a trading 22 advantage, then we've got a whole different issue here. 23 We're talking about insider trading, and I don't think that's 24 even the suggestion that we're dealing with. 25 MR. HARRIS: Then if there is no advantage here, 1 then why wouldn't we just have it displayed to everybody? 2 What is it that the subscribers are paying for? 3 MR. FERNANDEZ: The subscribers are paying for 4 something that has value added. To say that information 5 should be freely available is fine, but you can't assume that 6 it's costless to prepare. Now, taking information that you 7 have access to, performing analysis, coming to a valuation is 8 hardly a cheap process. 9 To do it on an ad hoc basis or for some segment, a 10 sector or industry, fine, you can do that for something under 11 eight figures. But to do it in a comprehensive way, in a 12 studious way, in a readily accessible, comprehensible way, 13 and in a timely fashion, you have to do this across, how 14 many, 6,400 and some odd equity securities, tens of thousands 15 of fixed income securities, not to mention derivative 16 structures that come off of these two type of things. 17 This is a very heavy thing. This is -- 18 MR. HARRIS: The point is well taken. 19 MR. FERNANDEZ: No, it's not, because I think we've 20 got a problem. The SEC started with a premise in the last 21 leadership that if we simply provided this as a free public 22 good, we would empower investors out there and that they 23 would be able to use raw information. 24 Raw information is not particularly useful for an 25 individual investor. The analysis -- 1 MR. HARRIS: Let me clarify just a bit. First of 2 all, I want to make it very clear that I'm not here to 3 present any opinions and, in fact, the purpose of holding 4 these hearings is that the Commission approaches this issue 5 with an open slate. 6 So my job here is only to bring out issues. If I 7 have to be provocative to get people to speak, that's where 8 I'm coming from. 9 So I do want to point out, though, that I think 10 everybody here recognizes that it is very costly to do these 11 analyses and that they have to be paid for. At issue is not 12 whether they're costly. What is at issue is to the extent 13 that rating agencies have special privileges that other folks 14 don't have, is it appropriate that they can trade on that 15 special privilege. 16 MR. FERNANDEZ: They're not trading. 17 MR. HARRIS: By trading -- 18 MR. FERNANDEZ: By distribution of the services, 19 perhaps you could say that, but what they're doing here is 20 providing a quasi public good in terms of a rating. 21 To the extent that they proffer from that, from the 22 distribution of the ratings, clearly then this is a business 23 they are engaged in. 24 MR. HARRIS: So you're argument then is that indeed 25 there may be some of that, but the public good that they 1 provide without being paid directly more than compensates 2 for -- 3 MR. FERNANDEZ: Going back to what you use in the 4 market data, yes. They are generating positive 5 externalities. Who do these accrue to? Hopefully to the 6 individual investor more so than any institutional investor 7 out there. That is our hope. 8 MR. HARRIS: I notice that this issue is resonating 9 with quite a few people. I had Deborah first, and then we'll 10 come to Leo and I suspect Cynthia also wants to speak. 11 MS. CUNNINGHAM: Despite the fact that I'm never 12 really excited about the fact that the rating agency bills 13 are probably one of my highest budgetary items within my 14 group, nonetheless, as an investor, I feel comforted to some 15 degree that I am paying for that service, because right now, 16 the information that is disseminated from the rating agencies 17 is paid for by the issuer and there are always questions 18 about conflicts of interest and whether or not that payment 19 from the issuer to the rating agency represents something 20 more than just a payment for information received, that is 21 then being publicly disseminated in an informational form. 22 I guess I'm offset to some degree, by the fact that 23 I am paying them in a different fashion for the subscription 24 of those services and the dissemination of that information 25 to my analysts. 1 MR. HARRIS: And just so everybody should be aware, 2 exactly what is it that you receive when you pay for the 3 service that you wouldn't otherwise obtain if you didn't pay 4 for it, presumably for different channels? 5 MS. CUNNINGHAM: There are definitely different 6 levels of service per agency and different levels of service 7 across the various agencies. 8 But by and large, what you are paying for is the 9 information that is key in understanding why their ratings 10 are as they are. 11 MR. HARRIS: Leo, Cynthia, then Malcolm. 12 MR. O'NEILL: The access to so-called confidential 13 information. Can I put it into context here? I think there 14 is a view, implied or otherwise, that there is this one 15 single fact that drives the credit rating that there is one 16 single fact that drives the credit rating in a situation and 17 that is an inside secret. Not true. 18 S&P over the years has maintained many, many, many 19 times that the disclosure that corporations make to the 20 Securities and Exchange Commission is sufficient upon which 21 to make a rating determination and if it wasn't, we would 22 lobby hard and fast that it be sufficient. 23 The information that we get, in most cases, is 24 corporations, projections, call it what you will, their view 25 of their future cash requirements, how their capital spending 1 is going to go, what their cash flow is going to be to 2 support that, what their access to the debt market might be 3 in the future. 4 I think the situation that Malcolm highlighted 5 earlier about product plans, in many cases, those plans are 6 used to supplement the number of what their capital 7 expenditures might be and that sort of thing. 8 It's pretty arcane, but it is supplementary in 9 nature. It's important, but it's supplementary. 10 COMMISSIONER GLASSMAN: Does it change your 11 ratings? 12 MR. O'NEILL: The information that is disclosed to 13 us, yes. It could very well change our rating, very much so. 14 MR. HARRIS: Cynthia? 15 MS. STRAUSS: The importance of our being able to 16 speak to the agencies is high, but it's basically because we 17 want to understand their analysis and, again, how they are 18 thinking about the ratings, their criteria, because they have 19 such incredible power on the marketplace in which we operate. 20 So it's really pretty simple, but it's really 21 critical to us. 22 MR. HARRIS: Malcolm? 23 MR. MACDONALD: Two points. I think, one, in terms 24 of data made available, clearly, one of the things that we 25 make available to the agencies is a fair amount of forward- 1 looking data, which, by definition, is not made available 2 without financial statements. 3 Therefore, I would concur totally with Leo that 4 that information has to be particularly useful in evaluating 5 a forward-looking view of an issuer. That is one point. 6 In response to your direct question, I would 7 probably favor total release of all information in writing, 8 because, quite frankly, I would like to hear some of the 9 things that are said about Ford Motor Company as an issuer 10 that are not necessarily included in the formal rating 11 notice. 12 MS. STRAUSS: If I could just say, the clarifying 13 point for me, and I think Debbie also mentioned it, by the 14 fact that we pay for it, the rating agencies have tremendous 15 influence on the markets. It is huge, and we don't have a 16 market voice back to them that is commensurate with that. 17 So it's really important that we try to understand 18 their analytical process and if you just, at least to date, 19 look at what they have in writing, that is not sufficient to 20 understand that. 21 So that imbalance is somewhat helped by our ability 22 to talk with them and say why is your analysis so, and we get 23 that only because we're paying fees. I know that we may be 24 very big, because we get access to the agencies because of 25 the magnitude of the fees we pay and that gives us, in other 1 words, our countervailing force against the fees and the 2 attention that would be put on issuers or underwriters on 3 Wall Street. 4 So it kind of gives us a stake in the game. I 5 don't think it's evil, I don't think it's awful, but there is 6 a natural imbalance here and that has influence on good 7 people. 8 MR. O'NEILL: The balance we're trying to strike, 9 Cynthia, and, obviously, we haven't done a good job of 10 communicating that and it certainly appears that we're every 11 bit as open and as transparent to our process as you and 12 Deborah would like us to be, and we'll go to work on it. 13 But I think what the balance is we're trying to 14 strike here is issuers pay us such a significant part of the 15 fee. Investors pay us just a small, small part of the total 16 value proposition, if you will. 17 So we're trying to balance that, to some extent. 18 What can I tell you, I think it's -- in relationship to the 19 cost that goes into the analysis that we provide investors, 20 if we did not have a model that generated fees from issuers, 21 I don't think we could continue to provide it into the 22 marketplace. 23 The cost to the investing public would be so, so 24 great, that the economies of scale would certainly be lost. 25 MR. HARRIS: We are approaching lunch, but I want 1 to turn direction yet again, but this time away from the 2 credit rating agencies and to the SEC. 3 We have spoken about the importance of financial 4 disclosure to the rating agencies in the formation of their 5 opinions. It is extraordinarily important to essentially 6 everybody who, in one way or another, forms an opinion about 7 values in the markets. 8 The question I would like to offer, and this may be 9 a unique opportunity, it's not unique, you can always write 10 us, is what additional disclosure should there be obligated 11 that would improve the process? 12 What more information would, say, Greg, what do you 13 see in Canada that you would like to see done here or what 14 would be your silver bullets for disclosure that you think 15 would particularly benefit your clients? 16 MR. ROOT: It's probably best to ask the clients. 17 MR. HARRIS: I will. 18 MR. ROOT: I'll be honest with you. I mean, I may 19 have a little different view, kind of going back to what 20 Malcolm and Leo were saying a little bit. 21 I think in this day and age, disclosure is very 22 good. I started in the rating business back in the early 23 1970s and you do have to go back that long, and the amount of 24 public information out there was extremely limited, and to 25 really do a credible job in any type of analysis, whether it 1 be credit, equity, you name it, it was very difficult. 2 If you look at Canada and you look at the United 3 States today, I don't think that is the case for publicly 4 traded companies. We rate companies and entities that are 5 not public and that haven't had access to information, it's 6 critical. 7 That's a whole different animal, I think, than 8 we're doing here. 9 So I believe that you can do a very credible job of 10 evaluating a company, from whatever perspective, with public 11 information. That said, obviously, to the extent that you 12 have access to management -- access to management is very 13 important, because one of the things I think is important 14 here, and it's going to the issue of transparency, we do 15 everything we possibly can, we make a strong effort to be as 16 transparent as possible, because I know that's very 17 important. 18 But one of the things that people say, well, we 19 want to know exactly what the parameters are for a rating 20 change. Therefore, they want three ratios that, if it's a 21 ratio, a rating is going to change. 22 Well, a rating is a much more comprehensive 23 analysis than that. What people want is something very 24 simplistic that anybody can just say it it's there, boom, 25 something is going to happen. And that's not the rating 1 business. 2 So I think being transparent, conveying your views 3 and what assessments and assumptions you are making is very 4 important, but it's not such a mechanical issue, because if 5 it was that mechanical, we wouldn't have a staff of analysts 6 with the tremendous amount of training to go out and 7 understand the industries they're following and the companies 8 they're following and put that together in a comprehensive 9 service, which is the rating, the analysis and the company. 10 We do a lot of industry analysis. We do a lot of 11 industry study. So I guess my long-winded answer is I'm not 12 sure that public disclosure -- there's a lot being talked 13 about here. I'm not sure it's the critical issue in this 14 whole conversation, quite honestly, because it's never been 15 abused, as far as I'm aware, first of all, and, secondly, I 16 think the value of that over the course of the rating has 17 been overblown. 18 I think it's being overblown, to an extent, because 19 I don't think it is as important today as maybe the attention 20 that it's being given. I'm not trying to be negative, but I 21 think that's our personal view on that. 22 COMMISSIONER GLASSMAN: I just have a quick 23 question to follow up on that, and this is to the rating 24 agencies. 25 How much of the rating is based on objective and 1 measurable factors as opposed to the subjective issues, like 2 quality of management? 3 MR. O'NEILL: I wish I could give you a definitive 4 ratio. I would have to say that the publicly available 5 information, by and large, accounts for a vast, vast, vast 6 percentage of the rating assignment. 7 I think one of the roles that non-public 8 information plays is it allows the issuer to, if you will, 9 portray or posture their view of their future, what their 10 plans are and how are they going to do that, and I think most 11 rating analysts, I think virtually all of them accept that in 12 the spirit which is intended, but also surrounded with a 13 great deal of skepticism. 14 We're talking about the future here. But I think 15 it's important that it's an input. It's a very important 16 input, because it tells us what the expectation the issuer 17 has for its future and where they're going, and it gives them 18 the opportunity to disclose to us something that may be 19 disclosed at a later date would cause uncertainty, a so- 20 called surprise factor, cause some tremor in our industry. 21 COMMISSIONER GLASSMAN: What about, even more 22 subjective, your view of them, management, and your belief 23 that they can actually do what they're saying? How important 24 is that? 25 MR. O'NEILL: I think the -- I don't -- well, 1 Malcolm, with all due respect, I don't think -- and we 2 understand what is happening here. I think it is 3 management's longevity, management's credibility, the 4 dialogue. It's a continuum. It's the dialogue that we have 5 with that corporation that goes back, in some cases, 30 years 6 and the continuity of management, continuity of thinking, of 7 planning, strategy, financing. 8 That's all that really counts, and that, by the 9 way, and I hope we get a chance to talk about it, that really 10 is the essence of governance. 11 MR. HARRIS: Commissioner Campos? 12 COMMISSIONER CAMPOS: I have a quick question. 13 It's going back a little bit to the prior discussion. 14 Let me ask it this way. I understand the business 15 model. I understand you draw fees from issuers. I 16 understand you draw fees from subscribers. 17 Assuming the fees could be adjusted, what is wrong 18 with making publicly available the data that you provide 19 subscribers? What would be the negative of that? 20 MR. O'NEILL: Well, we do make available all of the 21 public data that we use in our rating process and all the 22 rationales for it. I think the issue is should we disclose 23 the confidential information that the issuers provide to us. 24 I suspect that if that was our policy -- 25 COMMISSIONER CAMPOS: I assume there is something 1 different between what the public and what subscribers are 2 getting. I haven't studied lately, because I haven't been a 3 subscriber, but I assume there is something in the subscriber 4 materials that is different than what is publicly available. 5 MR. O'NEILL: What we make publicly available over 6 our web site is all our ratings and, in many of the most 7 important issuers, the rationale behind the rating. It's 8 maybe three, four paragraphs long, of why we rated it the way 9 we did and what the key factors are. 10 What we do also provide to subscribers is a much 11 more in depth report that supports the summary rationale. 12 COMMISSIONER CAMPOS: What's wrong with having that 13 publicly disclosed? 14 MR. O'NEILL: Making it available gratuitously? 15 COMMISSIONER CAMPOS: I said if you adjusted the 16 fees and had issuers pay for it, what is wrong with doing 17 that? 18 MR. O'NEILL: I don't think there's probably 19 anything wrong with that. I think all the third party 20 providers of credit information might have some difficulty if 21 all of us began disseminating this for free. 22 MR. ROOT: From Dominion Bond Rating, I hope I'm 23 not varying from my owner here, but I don't see anything 24 wrong with it. 25 I think your point is if we could offset in lesser 1 revenue by some -- and, again, our fee structure is very low 2 and anybody can have access to it. We are not precluding 3 anybody from accessing information. I think that 4 is important. So we don't want to say if you want to 5 subscribe, you can't. 6 But as far as I'm concerned, I don't think there is 7 anything wrong with that, to the extent that we can be 8 compensated and so we can maintain the quality of our 9 efforts, which is critical to this whole discussion, by the 10 way. I see no reason why the information, especially in this 11 day and age. 12 In the old days, with having to mail everything, 13 was very difficult. Everything is on the web. It's very 14 easy to access. So I personally would have no problem with 15 that at all. 16 MR. O'NEILL: Frank wants to make a comment, but 17 let me just clearly point out. The major users of this 18 information are major investment institutions and the number 19 of them is relatively small. 20 This information is not being sold to individual 21 investors, per se. The bond market is very much of an 22 institutional marketplace. So I guess it would make all this 23 information available to individual investors, and then we 24 would have to figure out how we're going to come to grips 25 with that. 1 MR. FERNANDEZ: The point I was trying to make was 2 the one I was getting to before. That would be fine, if you 3 then provide for that. If it is to be a public subsidy, 4 great. But the problem that I confront and I see out there 5 is that that our government is very good at creating 6 expectations of delivery and then not funding the mandate 7 behind it. 8 COMMISSIONER CAMPOS: I didn't say the government 9 would take over. I made it very clear in the hypothetical 10 here that you could adjust fees. 11 MR. FERNANDEZ: As long as somebody pays for it, 12 it's a value added product and somebody has to come up with 13 the money to deliver it. I don't think we've seen an 14 economic model where anybody has suggested this replaces the 15 marketplace in this regard. 16 Indeed, if there were greater people willing to pay 17 for this, then it would be provided in that manner. 18 COMMISSIONER CAMPOS: Why not go the other way? 19 Why have the issuers pay for it? Why not the investors? 20 Then you get rid of the conflict of interest. 21 MR. FERNANDEZ: That's one of the issues that you 22 can consider. The second thing I want to -- 23 MR. HARRIS: We will shortly be talking about 24 conflict of interest in the next session. I note, though, 25 that even with the investors, there is a conflict of 1 interest. The holder has an interest in maintaining the 2 current credit rating. 3 MS. STRAUSS: You asked for some specific -- 4 MR. HARRIS: The silver bullet. 5 MS. STRAUSS: What could the SEC do in terms of 6 disclosure, is that not correct? 7 MR. HARRIS: Yes. That was the silver bullet 8 question. 9 MS. STRAUSS: Oh, sorry. 10 MR. HARRIS: So what is your silver bullet? 11 MS. STRAUSS: I have a couple -- well, no. I wish 12 there was one. 13 MR. HARRIS: I'll give you a canon. 14 MS. STRAUSS: The two we faced is -- one is 15 requiring companies to disclose the full details of contracts 16 which are critical to their short-term liquidity. A great 17 example is bank facilities. That is now provided on a 18 discretionary basis. Some provide them, some don't. 19 That is something that, again, we talk about Reg 20 FD, it doesn't sound as bad, but it's one of the areas that 21 we've had trouble with where people won't disclose it and 22 it's critical for us. 23 MR. HARRIS: Cynthia, the ones who don't disclose 24 presumably are penalized in the market with higher credits. 25 MS. STRAUSS: No, that's not true. They have a 1 good rating. 2 MR. O'NEILL: They do disclose it to us. 3 MS. STRAUSS: Right. 4 MR. HARRIS: I see. 5 MS. STRAUSS: So that's discretionary in terms of 6 whether the company -- and, again, that puts -- we don't have 7 that information and we don't always get the details of that 8 from the rating agency, for good reason, because -- 9 MR. HARRIS: And as a consequence, don't you avoid 10 those issues? 11 MS. STRAUSS: No. Not necessarily, because our job 12 is to look at -- that's a critical input to that, but, again 13 -- and there will be other contracts. I think more of this 14 -- Leo also mentioned that ratings triggers, in many cases, 15 disclosure, that might come in through that as well, what 16 could be a short-term support or drag on liquidity, again, a 17 critical thing for debt holders. So that might be one thing. 18 Another thing is an area that is very esoteric and 19 it is hard to get into, which is asset backed securities. 20 Currently, under Rule 3a-7, under the Investment 21 Company Act, that gives any asset backed securities a blanket 22 exemption from all provisions of the 1940 Act. 23 So there is not good disclosure on the details of 24 asset backed securities. This is a huge market. It is 25 growing in complexities. Rating agencies are very important 1 in that and they stand in -- yet another topic that I don't 2 think we should go into now is the rating agencies are 3 perceived as standing in in many, many areas in that, because 4 disclosure is so weak. 5 I'm not sure really they are undertaking, not that 6 they are wrong, all the due diligence that one -- if one 7 would look at it and say, well, if it's not disclosed, look 8 at the complexity. You only have the rating agencies, is 9 this a reasonable balance. 10 So that's an area to look into. 11 MR. HARRIS: Well, I thank everybody for their 12 candor. This was a very enjoyable session for me and I have 13 learned a lot. We will break for lunch now, and we will 14 restart again at 1:30. 15 (Whereupon, at 12:24 p.m., a luncheon recess was 16 taken.) 17 AFTERNOON SESSION 18 MS. NAZARETH: This session, which will be 19 moderated by John McCarthy, from the Office of Compliance, 20 Inspections, and Examinations, is on potential concerns 21 regarding the role of rating agencies. 22 We will talk about things such as conflicts of 23 interest and potential abusive practices. 24 John? 25 MR. McCARTHY: Good afternoon. During this 1 session, we intend to explore the potential concerns 2 regarding the role of credit rating agencies, particularly 3 the potential for conflicts of interest in the credit rating 4 business and the potential for credit rating agencies to 5 abuse their power in the marketplace. 6 On a number of occasions in the past, the 7 Commission has studied the extent to which conflicts of 8 interest in the credit rating business may influence the 9 credit rating process. 10 In 1994, the Commission solicited comment on the 11 practice of NRSROs, charging issuers for ratings and the 12 appropriateness for an NRSRO to charge an issuer fees based 13 on the size of the transaction. 14 As a general matter, commenters at that time did 15 not oppose NRSROs charging issuers for ratings. Nonetheless, 16 the Commission again solicited comment on this issue in 1997, 17 after expressing concern that the rating agencies may be 18 tempted to give a more favorable rating to a large issue 19 because of the large fee and to encourage the issuer to 20 submit future large issues to the agency. 21 Presently, Commission staff is in the process of 22 examining the role and the function of rating agencies in the 23 securities markets and has focused a lot of their attention 24 on the various potential conflicts of interest. 25 Conflicts of interest in the equity research 1 process are now known and they have been the subject of much 2 recent media attention, such as the pressure on equity 3 research analysts to maintain positive ratings on issuers in 4 order to gain lucrative investment banking fees for their 5 firms, and, also, to get into various nursery schools, 6 apparently. 7 One goal of this session is to discuss the extent 8 to which conflicts, such as those that have been alleged in 9 the equity research process, represent themselves in the 10 credit rating process, and, obviously, I'm aware of the many 11 differences. 12 Further, we should identify and discuss any 13 potential conflicts that may be unique to the credit rating 14 process. 15 Accordingly, we will press your opinion today as to 16 whether potential conflicts of interest in the credit rating 17 business influence the credit rating process. I will outline 18 the conflicts that I have identified and then hopefully we 19 can discuss those in this forum. 20 The first one is whether the fact that a credit 21 rating agency is paid by an issuer calls into question the 22 objectivity or quality of the ratings. If so, what about the 23 agency's practice of basing fees on the size of the debt 24 issuance? Does this practice increase the potential conflict 25 to the extent the credit rating agencies could become 1 dependent on a few large issuers for a disproportionate share 2 of their revenue? 3 Secondly, whether conflicts of interest are created 4 when a credit rating agency provides consulting or other 5 advisory services to issuers it rates. And, thirdly, whether 6 there are any other conflicts of interest in the credit 7 rating business that influence the rating process. 8 I would like to note there are a couple of the 9 rating agencies present today that do not charge issuers for 10 ratings and thus avoid certain influences of issuers. We'll 11 be interested to know each participant's views on whether a 12 similar conflict of interest exists for those subscriber- 13 based firms who could be subject to the influence of 14 subscribers, particularly if the firm is experiencing 15 economic pressures. 16 The NRSROs also maintain substantial power in the 17 marketplace. An NRSRO can directly impact an issuer's cost 18 of issuing debt and most issuers seek ratings from at least 19 one NRSRO. 20 Accordingly, commenters have stated that the 21 potential exists for the NRSROs to engage in abusive 22 practices, such as improperly pressuring others who are to 23 pay for ratings in which they don't maintain a relationship. 24 In this regard, we will be interested to hear 25 participants' comments on questions such as whether rating 1 agencies rate issuers that have not requested a rating and 2 whether they do this in a means to generate business or 3 additional revenues, whether these -- and they are termed 4 unsolicited ratings -- are less reliable than solicited 5 ratings, and whether the use of unsolicited ratings should be 6 restricted or regulated to a lesser or greater extent. 7 Secondly, whether rating agencies engage in any 8 practices that are coercive or anti-competitive. 9 To the extent you believe that conflicts of 10 interest or abusive practices exist in the credit ratings 11 business, we welcome your comments on ways to prevent such 12 conflicts or ameliorate their consequences. 13 So with that said, why don't we start by addressing 14 the practice of issuer paid ratings. Obviously, it has been 15 touched upon in the morning session, but we will focus on it 16 now. 17 Let me just tee up the question. Does the practice 18 of issuers paying for ratings create a conflict of interest, 19 and if so, what effect does this have on ratings? Leo, since 20 you're the only NRSRO present. 21 MR. O'NEILL: I'm shocked that you would ask me to 22 start. No. I mean, obviously, what else would you expect me 23 to say? No. 24 The fact that we're paid by an issuer, I don't 25 think, creates a conflict of interest. To the extent people 1 believe it does, I think that conflict is extremely well 2 managed with a rating process. 3 All of our ratings are assigned by a committee. No 4 analyst has any financial stake in the outcome of his or her 5 rating recommendation to that committee. 6 No one issuer accounts for any significant -- I 7 mean, less than one-half of one percent of some very, very 8 small amount, in terms of its rating fee. 9 We all recognize, and perhaps this is the most 10 important point, we know that the whole credibility of this 11 process really depends upon our objectivity in the rating 12 process. 13 The bond market, I guess all markets are very, very 14 strict disciplinarians. And to the extent that those markets 15 are very, very knowledgeable and very learned about ratings, 16 and to the extent that those markets see practices seeped 17 into it that are questionable by rating agencies in terms of 18 ratings inflation or whatever, they would discipline. They 19 would discipline that organization to an extent that no 20 organization I think could survive over the longer term with 21 that sort of opinion hanging over its head. 22 So there's strong internal controls, exceptionally 23 strong incentives, and I guess the third one, as I say, we 24 have been charging issuers since 1968, 34 years, I guess that 25 is, as long as I have been with S&P, and I don't think there 1 is any smoking gun that states acceptance of those fees by 2 Standard & Poor's from those issuers has created any bias and 3 it's in favor of issuers. 4 I guess if there was, we wouldn't have lowered the 5 ratings on how many thousands and thousands of issuers over 6 that period of time. 7 COMMISSIONER GLASSMAN: On what basis do you 8 evaluate the performance of the analysts? 9 MR. O'NEILL: On the basis of their recommendations 10 to the rating committee, their insights. We have peer 11 reviews of the analysts. We have reviews by their lead 12 supervisors and we have -- it's a 360 process, how well they 13 not only perform in their participation and recommendations 14 on their own companies, but all our analysts also participate 15 in rating committees where other analysts are presenting. 16 So that's basically the system. We won't get into 17 the very specifics, but that's essentially it. 18 COMMISSIONER GLASSMAN: Thanks. 19 COMMISSIONER GOLDSCHMID: How about ratings when 20 you don't have fees? Where you're rating an issuer and 21 they're not paying fees? Any differences there? 22 MR. O'NEILL: No. As a practical matter, let's put 23 it this way. The only place where we may do an unsolicited 24 rating is in the United States public securities markets, 25 because we do believe, as I said earlier this morning, there 1 is sufficient information provided through the SEC disclosure 2 process that allows us to make a learned judgment as to the 3 creditworthiness. 4 As a practical matter, though, I must say that 5 virtually no issuer that sells a meaningful, $50 million or 6 more debt issues doesn't come to S&P and request a rating. 7 I would be very uncomfortable, by the way, 8 assigning an unsolicited rating in a market that doesn't have 9 the disclosure requirements that we have in this country. 10 MR. EGAN: We at Egan-Jones take a little bit 11 different view. We think how people are paid, how 12 organizations are paid are critical in how they behave. 13 If you have a five-year-old child and if that child 14 misbehaves, you somehow punish them or restrict some 15 compensation. They don't have dessert if they hit their 16 brother or sister. 17 The same way with organizations. If they are 18 getting 85 or 87 percent of their compensation from the 19 issuers, in the case of Moody's, if it's getting calls from 20 the ex-Secretary of Treasury, the New York Stock Exchange 21 Chairman, the heads of the major investment banks and 22 commercial banks not to change a rating, it would be crazy 23 not to think that it doesn't influence things. 24 There is a basic conflict of interest. We don't 25 have the luxury that Standard & Poor's and Moody's has. 1 The reason why nothing has changed, I mean, look at 2 the rating history on Enron and on Worldcom and Global 3 Crossing. The Moody's earnings did change it all. In fact, 4 they didn't go down at all. They went up 40 percent. They 5 went up 40 percent because there is no competition in this 6 industry. Zero. It's a monopoly. That's why they're able 7 to get away with it. 8 We think it's a huge problem and it needs to be 9 addressed. 10 As far as the percentage of each individual rating 11 being relatively small for the overall revenue base, the same 12 argument could be made for Jack Grubman and Henry Blodget of 13 Salomon Smith Barney or CitiGroup. Those are relatively 14 small percentages, but it resulted, their actions resulted in 15 -- they were pressured to maintain higher ratings than 16 normal. 17 So we think there is a big problem in this area 18 that needs to be addressed. 19 MR. ROOT: Can I comment on this, please? Because 20 I actually have been on both sides of this, prior to joining 21 Dominion Bond Rating, when I ran one of the NRSROs, which was 22 Thomson BankWatch. 23 Thomson BankWatch, for the first ten years of its 24 existence, did not charge the issuer, it got all its revenues 25 from the subscriber, from the investor. 1 Then when Thomson acquired BankWatch from an 2 investment banking firm, we changed that strategy and we 3 ended up charging the issuer. 4 I would have to disagree very much, Sean, on that, 5 and I concur totally with Leo. In fact, Dominion Bond 6 Rating, also, I should say, does charge the companies we 7 rate. From my 25 years of experience in the rating 8 business and being on both sides of this, I absolutely see no 9 evidence of any conflict of interest that has arisen or 10 exists in this day and age, and for the many reasons that Leo 11 cited. 12 So as I say, I have seen this on both sides and I 13 just don't think it's a real issue or has been an issue of 14 abuse, if you will. 15 MR. BAKER: If I can, just for a moment, give you 16 some data from our survey, focusing upon issuers. We did 17 focus upon the area regarding whether or not there were 18 conflicts of interest or perceptions of conflicts of interest 19 as far as the incentive systems would go, and I just want to 20 share a piece of data with you about these potential 21 conflicts of interest as rating agencies, and these are data 22 from issuing companies, as a matter of fact, a 112 issuing 23 companies, on their perception. 24 Slightly over 97 percent said that the rating 25 agencies have an overriding incentive to maintain a 1 reputation for high quality, accurate ratings. They sort of 2 live and die by the quality of the ratings and the 3 perceptions in the marketplace of the value that is added. 4 Slightly less than three percent indicated that the 5 current structure for paying rating agencies encourages 6 agencies to assign high bond ratings to satisfy customers. 7 So there was a small part of the three percent that 8 the perception was that it could have some influence, but the 9 vast majority, over 97 percent, believed that there were not 10 conflicts of interest, at least on this particular issue. 11 MR. EGAN: What about the investor side? Want to 12 address the percentages? Over 28 -- 29 percent thought that 13 rating agencies issued timely, accurate ratings. So that 14 would be, what, one percent of the investors. 15 MR. BAKER: That's true. 16 MR. EGAN: You have to look at it from the investor 17 side, not the issuers. 18 MR. REYNOLDS: I would say we have to look at it 19 from both sides. 20 MR. McCARTHY: Malcolm, could you give us the 21 issuer perspective on this issue? 22 MR. MACDONALD: I'd be happy to. Certainly, in all 23 my experience with the rating agencies, which is substantial, 24 I have never come across nor even heard anything that would 25 remotely suggest a conflict of interest. 1 In saying that, I am speaking from a position where 2 I have been actively negotiating fees with one of the 3 national recognized agencies, at a time when the rating has 4 been under review. 5 At that point in time, the rating agency people 6 involved recused themselves from consideration of a rating. 7 So I am totally comfortable that there is no 8 conflict of interest. 9 With respect to Sean's comment, my view is on most 10 polls, because 29 percent of the people say something, it 11 doesn't mean that 71 percent of them say the opposite thing. 12 It may be that some percentage says the opposite thing and 13 some percentage has no view at all. 14 MR. McCARTHY: I guess the question is do you feel 15 like you are getting what you're paying for? 16 MR. MACDONALD: The labor is worthy of its hire. 17 MR. McCARTHY: So yes, I guess. Deborah? 18 MS. CUNNINGHAM: The answer I'm going to give is 19 really not an answer. I'm just going to give you some 20 comments. It's both from an investor standpoint, which is 21 what I am, but I'm also an issuer. My funds that are 22 available for purchase in the institutional money market 23 arena are, by and large, a large group of them are actually 24 rated. 25 So I go through two groups of rating agency people. 1 On the issuer side, those that are rating are funds. So I 2 know what their credit analysis process is and I'm very 3 comfortable that it is not a process where I am influencing 4 them in a way that that would be considered a conflict of 5 interest. 6 That actually gives me some benefit, I think, in 7 actually -- as an investor, it gives me a little bit more 8 clarity. 9 I think if I were using the rating agency ratings 10 as the end of my analysis, I would have a bit more concerns. 11 But in the process that we use it, which is from an input 12 standpoint, one of many different inputs into the analysis 13 process, we are comfortable with the fact that we do pay for 14 a portion of it, although that payment is small in regard to 15 the payment that the issuers make. 16 We feel that that is compensation that needs to be 17 adequately received by the rating agencies and because of 18 that receipt by the rating agencies, guarantees us or gives 19 us a little bit stronger conviction than what we receive from 20 them as analysis is, in fact, very good. 21 I also think that evidence that the process is not 22 necessarily tainted on a biased basis can be seen from the 23 asset backed marketplace, and, again, we talked about this a 24 little bit earlier, but the asset backed market and the 25 rating process that occurs there is a little bit different 1 than what you see on the corporate side, and, on the asset 2 backed side, it's a very iterative process. 3 The sponsor of the program is essentially looking 4 to achieve a rating. It's not a corporation saying this is 5 what I have, tell me what I get from a rating process 6 standpoint. It's going in and saying what do I need to do to 7 achieve a AAA rating or what do I need to do, how do I need 8 to structure this transaction to achieve a AA rating. 9 And because it is an iterative process, I think 10 that almost a 100 percent guarantees that that removes a lot 11 of that conflict of interest. 12 It's a process. You know what the end result needs 13 to be. It's a question of taking the input as to how you get 14 to that end result. 15 MR. EGAN: I would follow up to Malcolm's point. I 16 refer people, and I can give a copy of this later, this is a 17 study that is done June 18, 2001, by Professor Baker, and 18 looking at Table 9, it indicates that 71 percent of the 19 respondents, this is the issuer respondents, say that, no, 20 this is in regards to the timeliness of the ratings, 71 21 percent believe that they are not timely. 22 So it addresses the flip-side of that question. 23 Not only do 29 percent think that they are timely. The bulk 24 of them think that they are not timely. 25 Also, obviously, this panel is not representing a 1 huge portion of the market. I made that comment in my 2 letter. It's not representing the medium and smaller 3 issuers. Forget about the individuals. They should be here, 4 too, because individuals do use credit ratings. Some of the 5 issues are targeted to individuals. There is noone 6 representing individuals here. 7 They need timely, accurate ratings, and if you ask 8 them, after Worldcom, after Global Crossing, after Enron, if 9 there is something wrong, they will all respond that, yes, 10 there is something wrong with it. 11 Elliott Spitzer, I think, is resonating because he 12 believes there is a conflict of interest. Why is it any 13 different in this field? 14 MR. FERNANDEZ: I would just ask what timeliness 15 has to do with the potential conflict. I'm curious. I don't 16 see the connection. 17 MR. EGAN: The problem is that they are not issuing 18 timely, accurate ratings, and there is a vested interest on 19 the part of the issuers to maintain as high a rating as 20 possible. 21 The environment that we are in now is that ratings 22 were delayed in being cut. With all these debacles, the 23 issue is when were the problems reflected in the rating and 24 whether or not there was a delay. As a matter of fact, there 25 was a delay in all of them. 1 MR. McCARTHY: Thank you. Would disclosure of 2 payments or more specificity about the payments be a 3 worthwhile disclosure? 4 MR. EGAN: Clean it up. Clean up this industry. 5 It's a mess. You have problems on the equity side because of 6 misaligned interests. Why not take a good look at this 7 industry. If you are going to recognize rating firms, 8 certainly, you'll recognize those that have a conflict of 9 interest. 10 Set it up so that you restrict the payment from the 11 issuers. Make it with the investors. 12 You know, we can survive on it. Why can't other 13 firms survive on it? Also, S&P and Moody's got started that 14 way. That's how they established a reputation. 15 MR. McCARTHY: Glenn, did you have something to 16 say? 17 MR. REYNOLDS: Yes. I've never been known as 18 necessarily a major fan of the rating agencies, but I think 19 on this one there is not an inherent conflict of interest. 20 Using the five year old-parent metaphor, I think we 21 have to reverse on who is the parent and who is the five year 22 old here, because Moody's is the one who is doing the 23 spanking and S&P. 24 They don't necessarily have an inherent conflict of 25 interest in making a ratings move, because we'll lose sight 1 of the fact that as companies get downgraded, they still have 2 financing needs. 3 So they are going to get paid as a single A, they 4 are going to get paid as a BBB, they're going to get paid as 5 a BB. They are going to get paid one way or the other. So 6 let's get straight who the parent is here. 7 One other thing, too, does the market want the 8 bonds rated? Absolutely. Do investors want to pay for it? 9 I haven't met too many investors who enjoy writing checks. 10 So for the issuers and a system to work, just look 11 at it this way and this is what the market has decided. The 12 spread on the issue without the ratings is going to be 13 significantly higher than the all in costs of the spread with 14 the ratings plus the fee. 15 Therefore, the market is saying that this is an 16 efficient way to get bonds rated, and we are totally 17 supportive of that. 18 We have some asterisks around the timing of rating 19 decisions and the fee stream, but the actual idea that the 20 issuer creates a conflict of interest, someone has to pay 21 someone and if you are paid by investor, you could just as 22 easily skew your research to not run down a company that he 23 has in his portfolio because he is paying you. 24 At some point, there has to be some trust in the 25 system, but it has to be backed up with good execution. 1 COMMISSIONER GLASSMAN: I have a question for Leo. 2 Why did S&P change its model? If I understand correctly, 3 before '68, you were paid by the investors and you are paid 4 by the issuers. Why did you change? 5 MR. O'NEILL: Actually, I said this morning, we 6 started in 1916 and it as a part of our subscription products 7 doing these credit ratings. 8 In 1968, if we go back in time, if you recall what 9 was happening at that point with the great society programs 10 and the Johnson Administration and the war in Vietnam heating 11 up, inflation took off, the economy went into a crunch, 12 credit defaults began to rise dramatically. 13 S&P took a look around and said, well, my god, we 14 have a certain amount of resources dedicated to this. They 15 are going to be woefully inadequate for this new credit 16 environment that has emerged in this country. 17 So we embarked upon a research program. We went 18 out, we met with investment bankers, we met with issuers, we 19 met with investors, and we said to them, look, we don't think 20 that we can afford to support the business that we think the 21 markets need in terms of rating without tapping into issuers. 22 The response we got back, and I guess we still get 23 back to say, is that's okay, but you better make sure you 24 don't manage to that business model through your ratings 25 process. Keep the conflicts down. Isolate the analysts from 1 the financial aspects of the business. 2 Much like a publication keeps its advertising 3 separate from its editorials. So that's what we did and 4 that's when began charging the issuers. 5 And I might add, just as a last statement, we began 6 in 1968. We started charging for commercial paper in 1969, I 7 guess it was. We began charging corporate issuers, I think, 8 in 1972. 9 All of this started taking place and was bought 10 into by the marketplace prior to the SEC designating S&P as 11 an NRSRO. 12 The model was in place. The NRSRO status did not 13 create the model. I guess now the last thing I'm going to 14 say is we understand all this. We're going to maintain this 15 status and we are just going to make sure that we do nothing 16 in our business that impacts on our ability to maintain that 17 status. 18 It's there, it's reality, and we manage 19 accordingly. 20 MR. McCARTHY: Cynthia first, then Malcolm. 21 MS. STRAUSS: I did have a point I think Sean 22 touched on, which is the timeliness issue. We can come back 23 to it later. It is an issue for us on the timeliness. I 24 think that could be the perception and could be the reason 25 why there is this issue on -- there is a problem with 1 timeliness with some of the agencies. 2 I don't know whether -- we haven't perceived it as 3 the conflict of why the fee is being paid, in part, because 4 we are right now watching the rating agencies downgrade many 5 companies massively and really be very hard on them. 6 So we don't see the fee element making them too 7 much in a company's pocket. I don't think we've seen that. 8 We do, though, see a need for a balancing factor of 9 holding the agencies accountable for their criteria and 10 transparency, which gives us a voice to the open market. 11 We'll come back to that later. 12 But that is why, Sean, we don't see that as the 13 linkage of the timeliness and the fee so much, but we do see 14 that there is an issue there. 15 MR. EGAN: What we see is that for the large, 16 important issuers in each industry, that they are given every 17 benefit of the doubt. That has changed within probably the 18 past four months or so since the Enron debacle, but if you 19 look at the ratings of these major -- the telecom, Worldcom, 20 for example, Global Crossing, both huge issuers. 21 There's a lot of stickiness and it begs the 22 question of whether or not the investor is being protected, 23 whether the issuer's interests are being addressed. 24 I know that, Malcolm, you don't think you have a 25 rating that's high enough right now and that's typical. But 1 the medium and smaller investors' interests need to be 2 represented, and they're getting hurt badly. 3 MS. STRAUSS: I do have a specific example, or a 4 few of them, where we have felt that the timeliness and 5 transparency was not good and we have come to the rating 6 agencies, where they've changed the criteria that we felt was 7 inappropriate. 8 Our voices don't always get heard and boom markets, 9 where -- and I've often had people say to me, "Well, we're 10 not hearing that from any other investor," well, the 11 investors don't often have voices, and we're not hearing that 12 from the company or Wall Street. 13 I mean, maybe there is a conflict that they are 14 paying more of the fees and not. I had a specific example 15 with one agency, where I came in the late '90s and many of 16 the agencies have reduced their requirements for backstop 17 liquidity for commercial paper issuers. 18 It was done in a boom market. Nobody was worried 19 about it, but a good, disciplined debt investor was -- we 20 were all worried about it, and the agency said to me, "Well, 21 noone else is interested." 22 And because they don't really have to listen to us, 23 they didn't, and I asked that same agency just this very year 24 how do they feel now about having reduced the backstop 25 liquidity requirements the way they did several years ago and 1 they said they regretted it. 2 Now, they have the right to change their criteria. 3 This agency did appropriately signal the change in the 4 criteria to the marketplace, but there was no -- they were 5 feeling more pressure clearly from Wall Street and the 6 companies to why should we do it, the markets are great, and 7 making them look a little more at the here and now in their 8 costs. 9 So maybe that's an example of where there's not an 10 equal balance for the other voices in this. 11 MS. NAZARETH: I just wanted to make one point to 12 Sean, also. Obviously, credit ratings are used for fixed 13 income products and that market is overwhelmingly 14 institutional. But that is somewhat of a misnomer. There is 15 a lot of retail investment in fixed income products, but it's 16 through the kinds of representatives who are here at the 17 table. 18 So it certainly was not our intention to not 19 represent the interests of individual investors here. We 20 just think that we have the folks here who, in fact, are the 21 ones making investments for them. 22 MR. EGAN: Understood. Could I make one more 23 point? 24 MS. NAZARETH: I think, Sean, we have to move on. 25 MR. EGAN: Okay. 1 MR. McCARTHY: Malcolm, you wanted to make a point. 2 MR. MACDONALD: At this point in time, I would like 3 to make two, and I will try to make them quickly. 4 Firstly, just for the record, I can only speak for 5 one institution, but I want to say very clearly that the 6 concept of our being able to delay a rating is totally 7 reprehensible and something I have never come across. 8 As to the idea that the agencies have suddenly 9 become proactive, I would ask that the facts speak for 10 themselves. 11 Between 1979 and 1981, my company was downgraded 12 from AAA to BBB in 26 months. That was 23 years ago. 13 Thank you. 14 MR. FERNANDEZ: I, too, have to take exception to 15 some of the things Sean said there. First of all, I don't 16 think that the small and medium sized investor or issuer are 17 not represented here. 18 I think the staff will confirm that endless number 19 of comment letters we find continue to warn about the dangers 20 of some actions being taken that are disproportionately borne 21 by small and medium sized businesses in this country. 22 Particularly, I draw attention to some of the 23 efforts to improve the situation with respect to information 24 flow that are actually reducing research available on the 25 small and medium sized stock listings. Hardwick Simmons 1 brought this point forward and I believe he is actually 2 correct here. 3 It's also limiting the ability of small and medium 4 sized issuers to approach capital markets. These firms 5 produce more in terms of jobs, profits, taxes, and 6 contribution to the general gross national product than firms 7 of comparable dollar for dollar compared to larger firms. 8 So these are also our customers. On the equity 9 side it is more so. The majority of the investors and the 10 volume that we handle now are retail and not institutional. 11 So that's the first thing. 12 The second misconception here is that the 13 appearance of a conflict of interest implies misconduct or 14 even the lesser standard of inefficiency. I think that is 15 totally wrong. These are very efficient firms that we're 16 talking about. Nor does the appearance of a duopoly or 17 oligopoly or even a monopoly indicate that there is some 18 inefficiencies prevailing. 19 All one has to do is look at a natural monopoly in 20 DTCC and see some of the highest efficiency ratings around. 21 So, again, it's more about there are certain types of 22 business that naturally lend themselves to certain types of 23 market structure, with higher barriers to entry, high cost of 24 product. It lends itself to an oligopolistic structure. 25 The other thing is if you're looking for the reason 1 why there is an issue of timeliness here, you don't have to 2 look any further than this room. 3 I would like to read a quote here. "SEC filings 4 contain a wealth of under-utilized business, competitive and 5 financial information and the market for this information is 6 much broader and deeper than the old line data vendors 7 realize. The key to unlocking the hidden value in this data 8 is to find ways to extract and deliver it in real time to the 9 large number of people who can use this information to make 10 business and investment decisions." Harvey Pitt, May of 11 2002. 12 Now, this has been one of my soap boxes to get up 13 on here. This could easily improve the timeliness if, 14 indeed, Congress had mandated that you conduct a study, which 15 we haven't really even addressed two of the major points, 16 impediments to the accurate appraisal of credit rating 17 agencies and barriers to entry, yet in this discussion. 18 If Congress, who mandates that you do this, would 19 simply provide you with the funds that have already been 20 promised, so that you could updated antiquated computer 21 systems, move into the 21st Century before it's over. 22 In terms of processing data that is already being 23 supplied to you, that would improve the timeliness of the 24 input that these gentlemen use and hence the timeliness of 25 your output. 1 If I'm wrong on this, Mr. O'Neill, please correct 2 me. 3 MR. O'NEILL: You are absolutely correct. 4 COMMISSIONER GOLDSCHMID: Let me ask, and I'm 5 learning and finding this very helpful. Whatever the 6 conflicts out there, there are two basic constraints I have 7 heard so far this afternoon. 8 One is what I would call a market power constraint, 9 that the rating agencies actually have lots of power compared 10 to the issuers. That I understand and it seems right to me. 11 But then there is the question we are going to be 12 taking up, which is what happens if we remove some of the 13 entry barriers, and we'll have to think about how those two 14 are going to interface. 15 A second thing people have been counting on, Leo, 16 you fairly raised this and there's vast economical literature 17 about reputational gains. 18 But one of the things that we know, too, from 19 recent events is that the reputational story doesn't work 20 perfectly. 21 It would have been viewed most avidly about the 22 accounting profession. So one thing you'll want to think 23 about is how well the constraints will work, at least on some 24 of these conflicts, as we keep working our way through. 25 MR. O'NEILL: Commissioner, if I may. I believe 1 you can fool some of the people some of the time. You scam 2 the game, if you will, for a while, but ultimately the 3 pattern emerges and then when that emerges, then I think the 4 market really makes it happen. 5 I think what happened to Arthur Andersen had a 6 visible impact within our organization. I mean, I didn't 7 think we had to get out there. We don't go out there and 8 preach every day about objectivity and all of that. 9 But when we saw what happened at Arthur Andersen, I 10 can assure you people were saying it won't happen here. 11 COMMISSIONER GOLDSCHMID: That is good to hear, and 12 it is a powerful statement. I mean, I don't want to 13 underplay reputation. I think it counts. It's not perfect. 14 MR. O'NEILL: That's correct. 15 MS. NAZARETH: I'm not sure we'll have time to get 16 to all of the issues, but one of the issues, analogizing to 17 the accounting profession, isn't just the reputational issue, 18 but involves a variety of the services provided by the rating 19 agencies. 20 Do we start running into the same issues and 21 questions that arose in that industry, as well? 22 MR. McCARTHY: This is a good time to take on 23 conflict number two. Conflict number two is the practice of 24 NRSROs and, I guess, other rating agencies marketing 25 additional products to issuers such as rating assessment 1 services and other consulting services. 2 So let me frame it by does the marketing or 3 provision of these products pose any potential conflicts? 4 For instance, the NRSROs that provide pre-rated assessment or 5 valuation services, and that's basically where they will 6 analyze a hypothetical fact scenario, such as a hypothetical 7 merger or acquisition, and provide how they would, I guess, 8 rate it if it were to be a real life situation. 9 The fact that they provide this rating assessment 10 service, does it become more difficult for them to deviate 11 from the hypothetical rating that they gave if the issuer 12 actually follows through with the situation that they paid 13 for them to rate? 14 Leo, of course. 15 MR. O'NEILL: Obviously, Standard & Poor's has 16 built a firewall around its rating activities. We have, for 17 many, many years, built a firewall around our rating 18 activities and all the other businesses that Standard & 19 Poor's engages in, including its equity research and 20 whatever. 21 That is because we do get into the rating business. 22 We get confidential information, and so we don't share that 23 with anyone. 24 So there is not a conflict there where we have our 25 rating analysts doing things of another nature that I think 1 is apparent in the accounting industry, where auditors are 2 perhaps -- or the consulting business got so great that it 3 had an impact. 4 As far as rating valuation services, rating 5 assessments are done. For many years, going back in time, 6 corporations would come in to us and ask us questions like we 7 intend to do a stock buyback; if we do it, what will happen 8 to our rating. 9 Then we would provide them with, well, if you do 10 it, then we're going to change the rating or keep it the same 11 or do whatever. 12 We think it is an inherent part of the rating 13 process itself. It is part of the rating process. 14 Obviously, if there is a change in the 15 creditworthiness of the organization between the time that we 16 get the theoretical answer and the time that the action is 17 taken, we will change that rating, even though the action 18 taken might be right in line with what they told us. 19 We think that that is the value of the service, is 20 that it gives the issuer the feedback. We stay in a 21 surveillance mode during the process and we rate the entity 22 as we see fit at that time. 23 So there is absolutely nothing in our best interest 24 to do anything other than that. 25 MR. McCARTHY: Malcolm, has Ford ever bought these 1 services? 2 MR. MACDONALD: Yes. Yes, we have. I have to 3 agree with Leo that, for certain items that are under 4 consideration, our board of directors certainly wants to have 5 a strong idea of what are the rating considerations 6 associated with the action under review. 7 Therefore, I agree it's an inherent part of the 8 process. The only place it breaks down, in my opinion, and I 9 know of no way to solve it, is if we go in with something 10 that perhaps could be considered extreme and ask what if and 11 get an answer, it would be terrible, your rating would go to 12 single B, and we don't do it. 13 Have we, in fact, somehow sent the message about 14 our thinking to the agency? And we've known that going in 15 and we accept that that is something we can do nothing about 16 and we must live with. 17 So I am very comfortable in saying this is an in 18 integral part of the overall rating process. 19 MR. McCARTHY: Say you pay for the service and they 20 say it's not going to affect your rating at all and then a 21 week later you go through whatever the hypothetical was and 22 they say things have changed in this week and now it actually 23 will affect your rating negatively. I can't imagine that you 24 would say thank you very much. 25 MR. MACDONALD: You are quite right. I would be in 1 Leo's office within four hours, I suspect. 2 MR. O'NEILL: And I'd say I'm on the other side of 3 the firewall. 4 MR. MACDONALD: John, the only answer to that is 5 I've never seen that. 6 MR. McCARTHY: It's fair to say there is clearly a 7 potential conflict, one that, given certain facts, may put 8 pressure on. 9 MR. O'NEILL: There would not -- again, Malcolm 10 might be upset about it, but the fact of the matter is we're 11 making the current judgment at that time. 12 You can set up a theoretical here, a hypothetical 13 that says in this very, very, very short period of time, 14 something fundamentally changes. I think it would have to be 15 a very visible and cataclysmic sort of event in that sort of 16 framework that clearly was outside the theoretical action 17 they might take. 18 MS. STRAUSS: John, there is another situation, as 19 I understand S&P's model. There is another rating agency 20 that I know -- it came to our attention that they were 21 soliciting this advisory service that was for an extra fee. 22 It was clearly staying outside of the standard rating process 23 that you're talking about as an advisory. 24 And we were actually quite concerned about that, as 25 a user of the ratings, because we did see that getting into a 1 slippery slope. In this situation, the analysts were 2 involved. 3 So it was not the model you're describing, Leo. I 4 wanted to add that when agencies get into other businesses, 5 as well, where they are selling services, I can't say I would 6 say it's a problem yet, but we do worry about if a rating 7 agency has a group that is selling quantitative models and 8 then their analysts use that quantitative model, doesn't that 9 up the -- isn't that saying, well, then, we all have to use 10 that quantitative model. 11 I mean, there is some trickiness when you get in 12 other business. But your model we know goes on all the time 13 and we feel perfectly comfortable about it, because the 14 advice is part of the ongoing relationship and probably 15 critical to it. 16 But there are other models out there that are 17 worrisome, from our point of view. 18 MR. McCARTHY: Would you like to know which ratings 19 are part of these other services? Would that be helpful for 20 you? 21 MS. STRAUSS: The issue is if an agency is getting 22 into these other businesses, they need to make that very 23 clear and very public and give descriptions of that to, 24 again, the otherwise users of their basic traditional 25 services, which I once again remind you that we are obliged 1 to use because they have so much influence on the market and 2 our regulations. 3 MR. McCARTHY: It sounds like the specificity of 4 the information hasn't been that clear. It sounds like you 5 only vaguely know about it. 6 MS. STRAUSS: I don't think that there is a lot of 7 this going on, that I'm aware. I only was aware of another 8 rating agency that was embarking on this and their analysts, 9 there was no firewall. 10 But at some point, firewalls -- it raises another 11 complication that has to be managed and I think you're 12 definitely in dangerous waters when you're looking at it as 13 an incremental revenue source, which, again, is not part of 14 the model described here. 15 MR. McCARTHY: Glenn, in your written submission, 16 you mentioned this issue. Would you like to comment? 17 MR. REYNOLDS: First of all, a lot of these 18 services will be sold to clients they rate. You start to get 19 into a difficult area. Regardless of the firewalls, my 20 experience with firewalls, being on Wall Street for 16 years, 21 is they work pretty well until it gets really hot, and you 22 have to be very careful about that. 23 If you're in one office pitching a million dollar 24 risk advisory service and in the next office you're saying 25 it's time for our annual ratings review, it's really tough to 1 reconcile, even if it's all above board and a 100 percent. 2 There is going to be that rated entity, who is a potential 3 customer, and the high value added service is going to have 4 to be -- it's going to be pulling the collar a little bit, 5 you know, how should I proceed here, what should be my 6 approach. 7 So it's the same debate, the kind of debate that 8 has been raging on forever, and that one was never resolved 9 clearly and it still goes on even after the fact. 10 There is no way to win or lose this argument, but 11 it is an appearance problem. If you consider that you have 12 mutual funds that are rated, you have insurance companies 13 with claims pending, you have banks that are being rated, 14 these are the national users of the risk products and when 15 you talk about the M&A pro forma product, then you're getting 16 very close to Wall Street firms, the investment banks, and 17 it's pretty hard to stay out of the gray zone whenever you 18 walk into an investment banking department. 19 So there is going to be an inherent skepticism in 20 the marketplace about this. 21 MR. McCARTHY: Anybody have any other thoughts on 22 this issue? I'll move on to the next conflict that I dreamed 23 up last night. 24 This is, which we touched on earlier, the practice 25 whereby a rating agency initiates a ratings process without 1 first being requested by the issuer. A lot of people term 2 this area to be unsolicited ratings. 3 Some commenters have alleged that this practice of 4 conducting what are, in effect, unsolicited ratings places 5 pressure on the issuers to pay for the ratings out of the 6 belief that if they don't, then the rating will be negatively 7 impacted. 8 So let me just open that up to anyone who has 9 thoughts on the unsolicited rating issue. Leo, I won't start 10 with you, unless you want to. 11 MR. O'NEILL: No, no. Just very simple. We don't 12 do unsolicited ratings. I will qualify that. We reserve the 13 right here in the United States to do a rating on an issuer, 14 an important issuer, because, again, we feel we have the 15 information that is available to do it. 16 But by and large, if we're not asked to rate, I 17 mean, more than by and large, by and very large, we will not 18 rate unless we are asked to rate. 19 MR. ROOT: At Dominion Bond Rating, generally, 20 again, similar to Standard & Poor's, generally speaking, we 21 do not do unsolicited ratings, but, again, for the same 22 reasons that S&P does here in the states, there are certain 23 circumstances where we actually would. 24 But I think this whole issue of unsolicited ratings 25 gives a negative connotation and I think it's important to 1 talk about the process one goes through in doing unsolicited, 2 because I don't think unsolicited ratings necessarily are a 3 bad thing. 4 Obviously, if they are abused, that's one thing, 5 but I think the important element -- is a couple things if 6 you're going to engage in unsolicited ratings. 7 One, to the extent -- and we do have a couple of 8 the companies we rate. We designate which ratings are 9 unsolicited. So first of all, at least the marketplace knows 10 which ratings are unsolicited, which are not. 11 Secondly, even in the case of unsolicited ratings, 12 we go through the exact same process as rating situations. 13 So before doing an unsolicited rating, we will communicate 14 with management. 15 We will talk with them, we will go through whatever 16 available -- we make every effort to approach that in a 17 similar fashion as if they were a rated institution. 18 Then to the extent we actually publish that rating, 19 again, we will communicate that with management so that this 20 is not something that they would be surprised about. 21 I think the other important thing you have to look 22 at is are the unsolicited ratings really -- does one think 23 that they are being punitive because they are unsolicited. 24 And I can tell you, again, going to the fact that, 25 at least in Dominion Bond Rating, our success and our 1 acceptance in the market is only as good as our credibility. 2 To the extent that you go about doing unsolicited 3 ratings as a means to generate business, it is going to come 4 back and haunt you very, very quickly. 5 So, again, I think unsolicited ratings can play an 6 important element in the marketplace and I think they can 7 certainly help the investors, which is, again, I think, the 8 function of why we're here very much. 9 So I think if properly disclosed and handled 10 appropriately, there is absolutely nothing wrong with 11 unsolicited ratings. 12 Again, I can go back to the days when BankWatch was 13 started, and that's how we got started. If you want to call 14 it unsolicited, the companies didn't come to us, but we did 15 the ratings for the purposes of the benefit of the investors 16 and we communicated that. 17 So as I say, I don't think unsolicited is this 18 dirty word that sometimes it gets viewed in the marketplace. 19 MR. McCARTHY: Deborah? 20 MS. CUNNINGHAM: First of all, I would say we don't 21 see very many unsolicited ratings. They are few and far 22 between. When we do, it is typically in conjunction with 23 solicited ratings from another NRSRO. So we oftentimes use a 24 comparing contrast, sort of what does the solicited rating 25 tell me versus the unsolicited rating, and, by and large, 1 it's really typically not much differential between them. 2 So I guess I agree that I don't think it's a 3 terrible thing, at least not from an investor perspective. 4 Maybe from an issuer perspective, but not from an investor 5 perspective. 6 We see more private ratings than we do unsolicited 7 ratings, and, again, I don't have any problems with those 8 either, but there's really very few of the unsolicited. 9 I would note that we, at points in time, if we note 10 that there is an NRSRO missing from a transaction, an issue, 11 we will ask the one that is missing if they, in fact, have 12 looked at the transaction and turned it down, and they will 13 tell us -- they won't tell us that without asking the 14 question, but if we ask the question, they will answer that 15 for us, and we find that very helpful. 16 MR. McCARTHY: Are the unsolicited ratings 17 disclosed somewhere? How do you know it was unsolicited? 18 MS. CUNNINGHAM: Typically, the issuer will tell 19 you that. The issuer will let you know if it is unsolicited, 20 yes. I've never seen it actually in the form of here is the 21 rating and it's unsolicited. 22 MR. ROOT: I would just mention, again, for DBRS, 23 we do actually put the designation "public information 24 rating" when it is unsolicited. So we do make sure that if 25 anybody is looking at the rating, they will be able to 1 differentiate between a solicited and unsolicited. 2 MR. EGAN: I'd like to put this on the table. 3 There is an article from May 2, 1996, from the Wall Street 4 Journal and it describes how Moody's issued unsolicited 5 ratings in the municipal area, whereby apparently the 6 municipality did not like either the charge or the initial 7 rating that Moody's gave, and so it didn't decide to hire 8 Moody's. 9 Moody's went ahead and issued a punishment rating 10 for the issuer. The issuers sued and they were unsuccessful 11 and the reason why they were unsuccessful is Moody's argued 12 that the ratings were their opinion and they have the right 13 to exercise their freedom of speech. 14 This ties into the whole consolidated aspect of the 15 ratings field. If you had a vibrant market with a number of 16 participants, they couldn't get away with things like this. 17 Many of the issuers in this case have very little 18 recourse. They were forced -- according to this article, it 19 cost one issuer close to a million dollars additional 20 interest expense because they couldn't do anything. 21 It would have been much cheaper for that issuer to 22 pay the Moody's fee. 23 So I'm not quite sure what the answer is, but I 24 think it does tie back in to the consolidated nature. 25 We think that, as I mentioned earlier, that no 1 ratings should be paid by the issuer. So do with that what 2 you want, but I can give this reference to anybody who wants 3 it afterwards. 4 MR. McCARTHY: Any other thoughts on this issue, 5 this conflict? 6 MR. BAKER: Just one brief comment. We also 7 investigated the issue of unsolicited ratings and found that 8 there was a relatively small percentage that unsolicited 9 ratings occurred, but there was an overwhelming view, at 10 least from the issuers, that there would be a clear 11 demarcation if there is an unsolicited rating, that it be 12 identified. 13 We then continued by asking the perceptions of the 14 accuracy of the rating that was unsolicited compared to those 15 that may have been ones they paid for. 16 Not surprisingly, none of the respondents felt that 17 the unsolicited ratings were more accurate, because they 18 have, of course, less information in order to try to come up 19 with the judgments regarding that rating. 20 I think the key point here is the importance -- at 21 least as perceived by issuers, we didn't look at it from the 22 investor side, of having that clear demarcation as mentioned 23 by the gentleman to my left. 24 MR. O'NEILL: But I think Deborah probably saw no 25 substantial difference between unsolicited and solicited 1 ratings. I think I heard you say that, Deborah, which, if 2 you did say that and that is your view, what that speaks to 3 really is the importance of public information within the 4 rating process and it really is strong evidence that we in 5 this country are fortunate enough to have a disclosure that 6 allows us to make credit judgments, buy/hold stock 7 recommendations, that sort of thing, based upon the public 8 disclosure. 9 So we don't do them, as I say, and if we do them 10 abroad, we indicate so. But I think rating judgments can be 11 made on the basis of the public disclosure. 12 MR. McCARTHY: Our data actually supports that view 13 in the sense that about 58 percent indicated they believed 14 the unsolicited ratings were just as accurate as the 15 solicited ratings, which indicates that public information is 16 available in order to develop a rating. 17 COMMISSIONER GLASSMAN: I may be wrong in this, but 18 I thought there was some research that shows that the 19 unsolicited ratings tend to be lower than the solicited. Is 20 that not accurate? 21 MR. O'NEILL: I honestly don't know, Commissioner. 22 I'll have to defer on it. I don't believe they are. I 23 actually have not heard that before. I don't assume they 24 are. Again, we don't do them here in the United States. 25 MR. REYNOLDS: I believe that was the case with 1 asset backed and some of the accusations leveled against 2 Moody's several years back. That was the asset backed 3 sector. 4 MR. O'NEILL: I think on the asset backed sector, 5 there is no widespread public disclosure. I think at the 6 beginning of the transaction, the transaction structure. So 7 you don't have a database of knowledge on it. 8 So if a rating agency should rate that on an 9 unsolicited basis, which is something we would never do, the 10 tendency would be, I guess, to -- because the information 11 risk factor, you don't have the information, you might rate 12 it lower. 13 It's not an experience that we have. 14 MR. HARRIS: Even if it weren't discounted, even if 15 they were assigned on the basis of all available information, 16 there is a significant selection problem in that the folks 17 who are least likely to ask for a rating are the ones who are 18 least likely to want to see what the rating would be. 19 MR. McCARTHY: The next conflict, which was touched 20 on this morning, it's not so much a conflict that affects 21 maybe the quality of the ratings, but in talking to smaller 22 institutions, et cetera, there is this perception that the 23 institutions that may pay more from a subscriber perspective 24 have preferential access or get information quicker. 25 I'm wondering if that's a topic from the 1 institutions. Are your expectations in paying subscriber 2 fees such that you feel that you have preferential access or 3 expect preferential treatment vis-a-vis the information 4 getting out to the general public? 5 MS. STRAUSS: I think we already went over this 6 this morning, that the agency said that they actually release 7 more to subscribers in detail in writing to begin with than 8 they do to the public. So that's already out there. 9 We do, which was also discussed this morning, have 10 access to be able to speak to an analyst and try to 11 understand their analytical thinking. So we do have that 12 access. 13 Is that something that the broader market is 14 greatly missing? I don't know. I think particularly those 15 conversations, those are -- because we talked already about 16 maybe the written piece would be useful, but the 17 conversations we have are typically kind of an analyst-to- 18 analyst, my number of share whacks, how come you're doing 19 this way, and about their criteria, we thought your criteria 20 was X, how does this foot with that and how does that help us 21 think in the future about how you're going to meet your 22 ratings out on this issuer. 23 I would say that's about 90 percent of what the 24 verbal conversation is about. So that's more of an analyst- 25 to-analyst conversation. 1 We get our ratings electronically, Bloomberg. 2 They're pretty quickly released now. In the past, we may 3 have gotten a little quicker release data, but I don't think 4 that's the case today. 5 MS. CUNNINGHAM: I would tend to agree. I think 6 the discussions we have with the rating agencies is where we 7 place most of our value and that's what we feel that we're 8 subscribing for, that's what we're paying for. 9 Whether that's -- I think what we're getting there 10 is, as was said before, analyst-to-analyst discussion. The 11 general public would not be asking the questions that the 12 analysts would be asking from an investment firm just because 13 it's not general -- it's not typically information that they 14 would have the ability to disseminate and try to interpret. 15 So that type of conversation is where we derive 16 most of the value from the subscription fees. 17 I don't think it at all is a timing issue, though, 18 because the public dissemination of the rating and any rating 19 changes occurs just as quickly to anybody that wants to pull 20 that information as it does to those that subscribe. 21 It is only in the value of the information that you 22 get with the analysts afterwards, questioning about that that 23 we feel that our subscription services are being utilized. 24 MR. O'NEILL: And we also make very, very certain, 25 when we have these conversations, that the information we 1 impart to these institutions, and we don't limit that, by the 2 way. I mean, get dozens of calls from non-subscribers, as 3 well, I might add, the media and sometimes others. 4 We don't want to check to see if they're a 5 subscriber, but the control we have is that the information 6 that we provide is consistent in that conversation with what 7 we provide in our printed publications or over the web. 8 We're not imparting a different set of information. 9 Maybe more detailed and what have you, but it has to be 10 consistent. 11 MR. HARRIS: On occasion, the Commission has to 12 discriminate between the providers of investment information 13 and the providers of investment advice. I am wondering 14 whether our panelists can provide us with some advice on the 15 subscriber services that we've been talking about now. Are 16 they clearly providers of investment information or are there 17 occasions where they run into what we might call investment 18 advice? 19 Let me ask the users of those services. Cynthia? 20 MS. STRAUSS: I'm thinking on this, because I'm 21 trying to understand the distinction between information and 22 advice. 23 MR. McCARTHY: I wouldn't have asked if we weren't 24 confused ourselves. 25 MS. STRAUSS: Because I do think that what is -- in 1 this front, the rating is a risk assessment. I know Moody's, 2 in particular, says this is just our view, but in essence, 3 the way it's used in the marketplace, it is used as advice 4 because it is a forward looking risk assessment and, 5 therefore, people use it to say, oh, advice on how they might 6 make their investments, but they don't specifically tell us 7 how to invest. 8 So it really is a gray zone. 9 MR. HARRIS: To what extent are your communications 10 personalized for you? 11 MS. STRAUSS: We do not -- I mean, it is very 12 clear, in our mind, that we do not get investment advice, 13 absolutely not, from them, nor do we look to them. I think 14 we think our investments would be much poorer if we did. 15 Some of it is timing and some of it is, quite, 16 frankly, we're in a different business. They are in a 17 ratings business, as so described. We are in the investing 18 business. 19 So we spend a lot of time thinking about the 20 markets and how our securities will perform and what our 21 shareholders are looking for. 22 MS. CUNNINGHAM: I think this goes back to one of 23 the discussions we were having this morning, too. Advice 24 needs to take into account not just ratings. That's part of 25 the input for advice. But it also needs to take into account 1 what the market conditions are that are prevailing, what the 2 spreads are that are in the marketplace. 3 You may have a BBB issuer that has a very wide 4 spread that, on an advice base from an investment advisor 5 standpoint, looks relatively attractive in the marketplace 6 and, on an advice basis, may be recommended for ownership of 7 that particular security. 8 On the other token, you may have a AAA rated issuer 9 that clearly is much higher in quality, but much lower in 10 spread and, on a relatively basis, versus other AA or other 11 AAA types of investments, does not equate value on a relative 12 scale. 13 So I think you've got to look at the relative scale 14 value in taking not only the rating of the issuer, but, also, 15 the spread and the market conditions that are prevailing at 16 the time. 17 MR. ROOT: Let me add. One of the things that we 18 make very clear, as a rating agency, is that people do not 19 make any kind of recommendation, like this is a good buy or 20 this looks like a good opportunity. 21 You could say that the rating itself does that, 22 maybe, but we make it very clear that we're not to suggest 23 that one should buy or sell or make any kind of investment 24 decision. That is something we avoid completely. 25 MR. FERNANDEZ: To try and clear this up, I think, 1 the best and still definitive source for this is Graham and 2 Dodd from 1934, where they divided into three areas. There 3 is information, which is purely descriptive. 4 In this context, it's the filings at the SEC, 5 whether they be twos, threes, fours, 144s, Qs or Ks. 6 Then you go to the next level, which is analysis, 7 and this is the advice element of it here, where you are 8 actually putting together associative values of largely 9 objective indicators and making some sort of an assessment 10 based on your best judgment. So there is an interrupt of 11 some subjective evaluation that is provided. 12 Then a third level is valuation and this is purely 13 one that is a normative judgment here. In the securities 14 analysis side, this is where you provide the buy, sell, hold 15 type of thing. 16 But to me, I think the distinction is going to be 17 easily drawn and even easily drawn in credit analysis as to 18 whether it's securities analysis or risk analysis of any 19 sort. 20 MR. McCARTHY: Great. Thanks. The last thing I 21 have is rating triggers, which was well talked about this 22 morning. 23 So that's it, unless anyone has something they're 24 dying to say. 25 Why don't we take 15 minutes and then come back. 1 Thanks. 2 (A brief recess was taken.) 3 MS. NAZARETH: What I would like to do for this 4 last session is to promise you that we will either finish 5 when we're finished as opposed to talking til 4:45, or, at 6 the very least, we will not end later than 4:45. But I think 7 there's a chance that we might be able to finish early and I 8 think we can do that without embarrassment. 9 Before I introduce the section on the potential 10 barriers to entry, Mike Macchiaroli pointed out that there is 11 one area that is an area of consideration for the report 12 about which we could use a little bit more input from this 13 group before we move on to the last topic. 14 The topic is, and we did touch on it a little bit, 15 the impediments to the accurate appraisal by credit rating 16 agencies of issuers of securities. That really goes to the 17 question of what information is not public that would be 18 helpful to be public, and I think Cynthia talked about some 19 of those things, like loan documentation and the like that 20 are understanding about ratings triggers and things of that 21 nature, that perhaps there is not a requirement to be public. 22 The other issue is, basically, is there some fundamental 23 information that you find there is resistance on the part of 24 issuers to giving you this information. 25 I think, again, we are being asked to analyze this. 1 I know, Frank, you said at the break that you had some 2 comments to make on this, and then I will call on those who 3 do credit ratings to tell us their experiences. 4 MR. FERNANDEZ: My concern is the same as Mike's 5 and I think it grows out of the same source, and he spends a 6 good deal of time watching what our risk management committee 7 does and our risk management practices. 8 I think they are quite distinct from the approaches 9 to risk management that are done by the credit rating 10 agencies and I'm arrogant enough to think that ours are 11 superior. 12 I think, one, they recognize, first that the nature 13 of risk is dynamic and we are entering a period of very rapid 14 structural and cyclical change. That does influence the type 15 of parameters that lead to payment interruptions and the 16 frequency of those interruptions. 17 So the type of analysis one needs to do needs to be 18 a reflection of where you are at a particular point in time. 19 The second thing, I think, the most critical 20 impediment, in my mind, is the one I have already identified. 21 It is when we survey research units in how the cost 22 of producing information and analysis, we find 30 to 35 23 percent of the overall expenditures are spent on preparing 24 standard statistical information, and it is done by every 25 firm over and over and over again. 1 Oftentimes, sophisticated senior analysts are 2 simply taking numbers down from one source, reformatting them 3 and putting them into another. It's a huge economic 4 efficiency loss. 5 That could be remedied by improving the 6 organization processing and display of basic financial 7 information filings. 8 The last impediment, I think, that exists that I 9 think is significant, there's lots of them, but the 10 significant one here is -- and then I get to be critical of 11 S&P -- is the mix between evaluation and prospective views 12 here. 13 I think they seem to concentrate more on the 14 current period and use -- I believe Mr. Fisher, in a recent 15 speech, brought this point up. Rather than looking out 16 prospectively and the forecasting element of this, and it's 17 almost like an 80/20 mix, it's often been described as that. 18 I think more weight has to be given to prospective 19 payments problems. I think scenario analysis has to be used 20 more exhaustively. 21 One thing I have learned in the last few years is 22 that anything you can conceive as a worst case is easily 23 realizable and it is much more prudent to plan for those 24 worst case estimates and be proven wrong than to neglect them 25 and try and shoot for the middle of the road outcome. 1 So these are not small things, because it changes 2 the way you process information and the type of analysis that 3 you apply. 4 Leo? 5 MR. O'NEILL: Sure. The whole thrust of what we do 6 is to try to get it right and try to get it accurate. I 7 think we, obviously, input into our process a view of the 8 future. Obviously, you have to do that. The bonds are going 9 to be paid in the future. 10 I think what the issue is really is that the future 11 these days becomes -- I think the word that has been used 12 many times today, much more volatile, much more difficult to 13 predict, certainly much less reliant than perhaps in more 14 stable economic times. 15 So I think we try to do, Frank, what you are 16 suggesting what we do and, obviously, as you might expect, I 17 think we do it right. But I think we would all agree that 18 judging the future in these times is extremely difficult and 19 you do so -- if you try to do it precisely, you do get into 20 difficulty. 21 On a somewhat light note, but accurate not, I think 22 this whole idea of analysts in dealing with the statistics 23 and -- I think there's a cultural thing that I have seen 24 that's been around for about 35 years. 25 The analysts like to do their own numbers, for some 1 reason, and we really have this wonderful database called 2 Compustat and we can provide these numbers any way the 3 analysts want to see them and we stand ready to do that. 4 Yet, we still, I'm sure, we do at S&P have analysts 5 that still are cranking the calculator. The only thing they 6 don't do is flip it anymore. 7 I don't know how to get to that one. But I do 8 think the street pays an enormous amount of money on research 9 and that perhaps is a discussion for another day on 10 independent equity research and the amounts that are going 11 there. 12 But I think it's a real issue that needs to be 13 addressed. Again, I don't know anyone else -- 14 MR. ROOT: Let me just add, from DBRS, a couple 15 things. Again, I tend to agree with Leo. We try to make our 16 ratings forward looking, prospective, but at the same time, 17 maybe just in your issue there, Frank, you've got to build in 18 what you think is a reasonable scenario, because ratings are 19 just as wrong if they're too low as if they're too high and I 20 think that's important here, because I know the focus today 21 is on ratings that, after the fact, maybe were too high, but 22 you want to be accurate. That's the key. 23 One can sit back and take a very pessimistic view 24 of the future, rate everything very low, and then when a 25 couple credits go bad, you raise up and say, boy, I got that 1 right. But what about all the others that you got wrong? 2 Noone talks about that. 3 So I think the key is to try not to get too extreme 4 on your prospective views, because I think that can be very 5 dangerous, but, again, communicate what you are 6 incorporating, what your scenario is for your ratings. 7 This way, when I convey that, communicate that to 8 whomever, if you're going to think that auto sales, to use an 9 example, sorry, Malcolm, are going to be -- we think they're 10 going to be 16 million or 16 and a half, and you think 11 they're going to be 14, then you should maybe adjust 12 accordingly. 13 Another thing that we do in our written analysis is 14 we do stress testing. What if scenarios. Not in a 15 predictive nature, in the sense that we're suggesting that 16 these things will happen, but as a means to show that if 17 certain environments or issues do happen, what are the 18 potential implications, assuming all other things being 19 equal. 20 So, again, I think the key is kind of what Leo 21 says, the key is to try and be correct. Again, we use our 22 judgment, professional judgment and opinion, and that's why 23 transparency is key and, again, trying to be accurate on both 24 sides is as important in this whole issue as just trying to 25 guess the next problem. 1 That's my view. 2 MR. FERNANDEZ: Still not to forget, though, that 3 the great impediment to this recently, and will be again, 4 will be the garbage in garbage out problem. Until we get the 5 corporate governance issues and the accounting issues solved, 6 you're not going to improve the systems at all. 7 MR. REYNOLDS: Just in terms of frustration on 8 getting information flows. We very often have to depend on 9 the rating agencies. A couple areas where we just know the 10 disclosure is not adequate, it is garbage in and garbage out, 11 also. 12 But the typical credit analyst is different than 13 the typical equity analyst. They tend to be less optimistic, 14 saying we smell flowers and look around for a casket. 15 But the main point is we need to get the downside 16 scenarios covered. Loan documentation was already addressed. 17 A lot of companies do not provide that information and 18 structural risks have been real makers and breakers in this 19 market, and we need to know if the agencies have looked at 20 it. 21 On the other side of the Reg FD wall, they can at 22 least say they've looked at it. So we can be aware, so we 23 don't have to guess. We don't have to necessarily provide 24 that disclosure. 25 On counter party credit, which has blown up the 1 better part of the power sector this year, there the 2 disclosure is wholly inadequate. So if they have looked at 3 the collateral postings and the liquidity flows within that 4 business, we just need to know they have done that. 5 I think it's possible to have faith in a third 6 party to have done the work, but too often they don't 7 highlight that that work has been done, because they hold the 8 dialogue with the issuer, hold it confidential, and, very 9 often, and we had this discussion directly with some senior 10 people at Moody's, they said we won't even tell you the 11 questions we've asked, which doesn't necessarily help when 12 we're trying to make assessments. 13 Another area, and this would probably be the last 14 one I would highlight, is asbestos. It is going to be the 15 crippler of many cooperates right now and there is a 16 tremendous amount of disclosure in terms of litigation with 17 insurers, numbers of claims, the spread of the claims, two 18 different legal venues. 19 We don't have that information and maybe the 20 agencies can at least take a look at it, because until the 21 disclosure issues are addressed, it's going to continue to 22 lurk as a confidence breaker in the market right now. 23 MR. O'NEILL: I think, again, our objective is to 24 disclose all of the rationale, background data, whatever it 25 is that supports the rating. 1 I want to make sure that we are clear, and we have 2 to be careful. We are not the disclosure agency for issuers. 3 I mean, if there is valuable information that should be in 4 the marketplace, I think they have the full first obligation. 5 Our obligation follows close behind, and that's to 6 make sure the market understands what it is we're rating, why 7 we did it, and when we have more work to do in that, hey, 8 give me a call, we'll set up a meeting, and we'll go to work 9 on it and see if we can get this information to you. 10 MS. NAZARETH: Yes, Sean. 11 MR. EGAN: I think there should be more disclosure 12 in the area of future pension fund liabilities. It's not 13 very good. I think I'm sitting next to the right person 14 here. 15 Also, future -- 16 MR. MACDONALD: I hope to retire soon. You're 17 quite right. 18 MR. EGAN: I set you up for that. 19 MR. MACDONALD: A few questions on the SEC and your 20 retirement package would be welcome, maybe. Just a couple 21 routine questions. 22 MR. EGAN: Also, the liabilities for retiree health 23 care costs, that hasn't been disclosed properly. We don't 24 like just one number, but give us the underlying assumptions, 25 what kind of rates of return are being assumed in the pension 1 fund, how quickly they're growing, so that people can do 2 their own independent analysis. 3 Also, asbestos liabilities. There are some 4 companies that have significant asbestos liabilities. They 5 are not disclosing them because they, for whatever reason, 6 they think that they can win the cases. 7 So we think this is a problem. 8 Lastly, in the asset backed market, there should be 9 some kind of centralized data gathering source that is 10 available to all investors in the market and make it much, 11 much easier. 12 You have a whole set of guidelines in the corporate 13 area for filing 10-Ks, 10-Qs, 8-Ks, whatever, and you don't 14 have the similar regulatory structure on the asset backed 15 side. 16 We saw some huge downgrades, I think it's probably 17 a 12 notch downgrade for a major issuer within the past two 18 weeks. If we had that information available on a regular 19 basis, we would have much deeper participation, you would 20 have much more efficient markets. 21 It's a huge hole, from our perspective, in the 22 financial markets. 23 MS. NAZARETH: Thank you. I think we will switch 24 now to our last topic, which is potential barriers to entry 25 and regulatory treatment. 1 Just to tee it up a bit. As you know, and it has 2 been stated several times here today, this NRSRO concept that 3 the Commission uses has found its way into not only the 4 federal securities laws, but a number of other regulations, 5 as well. 6 To assess whether a rating agency may be considered 7 an NRSRO for purposes of the Commission's rules, the 8 Commission staff considers a number of criteria. 9 The single most important criterion is that the 10 rating agency is nationally recognized, which means that the 11 rating organization is widely accepted in the United States 12 as an issuer of credible and reliable ratings by the 13 predominant users of securities ratings. 14 Plus, the designation is intended largely to 15 reflect the views of the marketplace as to the credibility of 16 the ratings rather than to represent a seal of approval of a 17 federal regulatory agency. 18 As you know, the staff also considers and reviews 19 the operational capability and reliability of each rating 20 organization, and I won't go into all those criteria right 21 now. 22 Although the number of firms designated as NRSROs 23 increased from three in 1975 to seven in 1991, as we said 24 earlier today, we are back down now to three by virtue of 25 consolidation. 1 Several observers have commented that the national 2 recognition requirement creates a barrier to entry for new 3 credit rating agencies. Generally, this argument is based on 4 the premise that users of securities ratings have a 5 regulatory incentive to use ratings issued by NRSROs rather 6 than those that don't have the designation, so that new 7 entrants don't find it difficult to achieve national 8 recognition without having the designation in the first 9 place. 10 Because of the quasi public responsibilities of 11 rating agencies, the importance given to ratings by investors 12 and other market participants and the influence of ratings on 13 the securities markets, we will also focus this afternoon on 14 whether rating agencies should be subject to greater 15 regulatory oversight or whether other adjustments should be 16 made to the Commission's regulatory approach to rating 17 agencies. 18 Since the Commission began using credit ratings in 19 its rules and regulations, it has considered a number of 20 issues regarding oversight of credit rating agencies. 21 Ten years ago, the Commission seriously considered 22 whether it should seek from Congress more direct oversight 23 authority of credit rating agencies, given the increased role 24 in the financial and regulatory systems. The Commission, at 25 that time, did not reach a consensus on the need for 1 regulation. 2 In 1994, the Commission did, however, issue a 3 concept release soliciting public comment on the appropriate 4 role of ratings in the federal securities laws and the need 5 to establish formal procedures for designating and monitoring 6 the activities of the NRSROs. 7 The Commission received 25 comment letters in 8 response to that release, which generally supported continued 9 use of the concept, but recommended that the Commission adopt 10 a formalized process for approving NRSROs. 11 Commenters generally opposed additional regulatory 12 oversight of NRSROs at that time. 13 In 1997, the Commission published a rule proposal 14 that would have adopted a definition of the term NRSRO, that, 15 among other things, set forth the criteria a rating 16 organization would have to satisfy to be acknowledged as an 17 NRSRO. 18 Generally, under the proposed amendments, the 19 Commission would consider the same criteria currently used by 20 the staff in the no-action letter process. 21 To a large extent, the proposal was designed to 22 bring greater transparency to the existing process and to 23 provide for a formal appeal process. 24 The Commission, as of yet, has not acted on that 25 proposal. 1 So with this background, I hope our discussion with 2 regard to regulatory oversight will shed light on two key 3 issues. First, whether rating agencies should be subject to 4 additional regulatory oversight, and, if so, the appropriate 5 form of such oversight. Secondly, if participants believe 6 that the NRSRO framework has merit and should be retained, 7 whether the Commission should adopt its 1997 proposed rule 8 amendments or if it should adopt additional rules to fine 9 tune that framework. 10 But I thought where we would begin is a topic that 11 has been alluded to several times earlier today, which really 12 goes to the barriers to entry in the business of acting as a 13 credit rating agency. 14 Some here have said that there are some natural 15 monopolies. I think it would be interesting at some point in 16 this discussion to impose upon Gay to tell us what is it like 17 in the world out there that doesn't have this designation and 18 to learn, because clearly somewhere else in the world, people 19 have figured this out. 20 Obviously, we're hoping to get some light shed on 21 this. 22 Frank, you seemed to have some views earlier about 23 the sort of the economic model here. Do you think there are 24 barriers to entry? Do you think there are some natural 25 monopolies? 1 MR. FERNANDEZ: Yes. I think there are some very 2 high barriers to entry with respect to the NRSRO designation. 3 Some of them I believe are natural, some I believe are 4 necessary. 5 For example, the NR aspect of it. I think if you 6 didn't have that, the market would have to invent it, because 7 by the very nature of having a bond rating, there has to be 8 generalized acceptance of it in the mind of the investors. 9 I think you see that by the consolidation that has 10 occurred and by the fact that only a small percentage use 11 Fitch. We find that investors and issuers themselves seek 12 out the two largest, simply because of the name recognition. 13 So that constitutes a barrier of entry, because it 14 takes years, it takes substantial financial resources to 15 persist and grow large enough and to have broad enough 16 overage over that period of time to obtain national 17 recognition. 18 So NR does impose that, but it may be a necessary 19 barrier. That's why I say there is a certain natural degree, 20 because, if you're going to use a benchmark or a shorthand 21 definition that you can apply that is easily recognizable, it 22 has to be nationally recognized. 23 COMMISSIONER GLASSMAN: Is it possible to be 24 nationally recognized, but not have a broad national scope? 25 MR. FERNANDEZ: You mean regional? 1 COMMISSIONER GLASSMAN: Either regional or in the 2 industry sector. 3 MR. FERNANDEZ: That does exist and those people 4 are used and they provide a useful service and are a natural 5 provider, and some of them have extremely high standards. 6 But that's not the same thing as providing a bond 7 rating system. 8 MS. CUNNINGHAM: Those are all the ones that have 9 been consolidated into Fitch, too. 10 MR. FERNANDEZ: Yes. 11 MS. NAZARETH: We actually designated NRSROs in a 12 particular industry sector, for instance, with Thomson. 13 Greg, you can tell us. 14 MR. ROOT: I've had the luxury of having this NRSRO 15 experience probably as much as Mike has. I've become an 16 authority on this. 17 But let me say a couple things, because I think it 18 is certainly very important. I think the barriers to entry 19 need to be high as long as there is going to be an NRSRO 20 designation and ratings are going to play the role they do. 21 They need to be high. Not insurmountable, but high. 22 So I think that's appropriate, however you want to 23 define that. 24 Is it virtually impossible or whatever to become an 25 NRSRO? No. Again, I look at my previous employer, Thomson 1 BankWatch, and, again, going to the issue of national 2 recognition, as well. I mean, we were a specialized agency. 3 Actually, at that time, we didn't even -- we rated 4 companies. We didn't rate specific issues initially. But I 5 think the fact that we, over a period of time, provided what 6 is viewed as a credible, valuable service in the marketplace 7 and had all the other attributes, which I think are 8 absolutely crucial to becoming an SRO, we were able to gain 9 that recognition, as were some of the other firms. 10 So I think the issue is not so much are the 11 barriers to entry high. I think the issue of national 12 recognition is one that needs to be looked at for the 13 reasons, because I don't believe you have to do everything to 14 be viewed as a credible and invaluable provider of ratings 15 within the context of what NRSRO is meant to do here in the 16 United States. 17 But, again, as I alluded to in my opening comments, 18 I think the important thing, which you mentioned in your 19 statement, and going back to 1997, the key is to define 20 this, so at least people know what the ground rules are and 21 what it takes to get there. 22 Now, you don't want a situation where the barriers 23 are so low, where you have an influx of participants that can 24 destroy the whole concept of what NRSRO is meant to do, and I 25 think that is very effectively in the marketplace. 1 So we really -- again, I went through this actually 2 twice, believe it or not, and now with Dominion Bond Rating, 3 we're, obviously, very involved and interested in this whole 4 process. 5 So I think if you want to say what's broken, I 6 think what's broken is the fact -- I hope and I think that 7 the people in SEC would agree that let's define this thing 8 and have a clear road map for how you get there, and I think 9 everybody would be better off. 10 So I think the comments that were made, I think, by 11 most constituents back in 1997 should be very seriously 12 looked at and, in my view, obviously, adopted, and I think it 13 will be for the betterment of this entire process. 14 COMMISSIONER CAMPOS: Could I jump in here just a 15 little bit? Would it be wrong, if anything, if our 16 definition at the SEC were to permit, under the nationally 17 recognized standard, and maybe some could argue it already 18 does, a specialized narrower scope, much as your prior 19 employers were in, and if we ended up finding a way to 20 encourage entry, albeit maybe in a specialized way? 21 Maybe we are in a new era in terms of fast moving 22 companies that are reaching prominence. Capital can be 23 raised quicky if somebody saw a business plan, an 24 opportunity. 25 There is no reason, it seems to me, that capital 1 couldn't be put together to create another rating agency. 2 What's wrong with that picture if that were to occur? 3 MR. O'NEILL: Speaking as one who -- we have the 4 designation, I don't think there is really anything wrong 5 with that, quite frankly. I think the key is to have -- and 6 Greg said it quite articulately -- to have an open and 7 transparent process with clearly defined criteria as to what 8 it takes, and I think the criteria really should begin with 9 market recognition, that your ratings are used, that the 10 investors and people at this table use those ratings, like 11 those opinions, and they use them in their own investment- 12 making decisions. 13 And the other part of it is that the ratings are 14 derived independently and objectively and that process is 15 transparent, as well. 16 So I think, from Standard & Poor's standpoint, 17 that's more of a critical need than it is to elect barriers 18 that prevent other entrants into the marketplace. 19 We think other entrants should be in this market, 20 absolutely, and we just want to make sure that the process is 21 open and it's transparent. 22 Eventually, you will hear, I think, if you haven't 23 already, that one of the concerns is this rating shopping, 24 that agencies get in and they give high ratings and other 25 rating agencies who may be calling it like it is don't get 1 the business, that sort of thing. 2 That might happen short-term, that could happen, 3 but I think ultimately the market sorts it out and the market 4 will recognize it and the market will do its thing for that 5 rating agency. 6 Of course, that may not be very helpful for the 7 Securities and Exchange Commission, who designated that 8 NRSRO, but -- 9 MS. NAZARETH: Also, it's a problem from a 10 regulatory perspective if it's being used for regulatory 11 purposes to sort of take a shot at it. 12 COMMISSIONER GOLDSCHMID: Let me follow up on this 13 nationally recognized designation. The second part of what 14 you were saying, Leo, in terms of quality and independent, 15 that kind of ground rule I understand. 16 But why do you need nationally recognized as 17 opposed to regionally recognized or not nationally 18 recognized? When I do a startup, I put together a large 19 organization, the highest quality. I don't have a customer 20 out there, but I'm going to get them, or at least I hope so, 21 and the market will work, presumably, if I've got quality and 22 people think it's going to be reputable, presumably people 23 will come to me. 24 If they don't, the market has said something, too. 25 Why wouldn't that work perfectly well? 1 MR. O'NEILL: I think the two words national 2 recognition were first used without a great deal of thought. 3 I think noone ever contemplated that would be a subject of 4 discussion 25 years later at the Securities and Exchange 5 Commission. 6 I don't adhere to national recognition. There are 7 some great institutions out there with some great research 8 and I think they have market recognition. 9 I think the larger issue, though, is the one 10 Annette was mentioning, is how you put these ratings into the 11 regulation or whatever and then it's is the quality 12 maintained. It creates -- I think eventually it sorts out, 13 as I said, but it doesn't -- I don't think it's very helpful 14 in a short term situation. 15 MR. ROOT: If I could just add. I think, again, 16 because of the importance of the NRSRO designation and the 17 implications of the ratings, I think having a track record is 18 very important because nothing to say that a startup company 19 couldn't get there, but I do think because of the importance 20 and the influence of ratings in this context, I think it's 21 very, very important to have an established track record over 22 an extended period of time, in addition to having all the 23 other -- 24 MS. NAZARETH: If I could just go back for a 25 second. Obviously, were we not using this designation for 1 certain regulatory purposes, such as letting broker-dealers 2 take lower haircuts on their capital because of positions 3 that are investment grade, if we could throw it out for 4 regulatory purposes, then, obviously, you could just let the 5 marketplace decide. 6 So let the best rating agency win. Whoever gets 7 the most business, obviously, that is showing that people 8 have a high degree of faith in that rating agency, which is 9 why they're going there. 10 But we've got this reverse situation. Rightly or 11 wrongly, many years ago, the Commission started using this 12 designation as a proxy for the fact that the marketplace 13 found certain of these rating agencies to be credible for 14 purposes of letting, for instance, broker-dealers enjoy 15 certain capital benefits, if some of the securities in their 16 portfolios had certain ratings that were not shopped for, but 17 obtained from nationally recognized agencies. 18 Unfortunately, what happened with that was it 19 became such a handy proxy, it was used not only for a variety 20 of other purposes at this agency, but I'm sure, as Gay said 21 earlier, it's not NRSRO per se, but the concept of the 22 importance of credible ratings, and they have a different 23 proxy of what credible is, I guess, in the new Basel accord. 24 It's being used for any variety of purposes worldwide. 25 So I think our dilemma is how do we balance it. 1 Do we sort of throw out the NRSRO designation and 2 try to find another proxy for this? If there was, I know 3 some economists have talked about, well, why don't you use 4 credit spreads, just let the marketplace. 5 Well, that doesn't work for new issues. 6 MS. STRAUSS: I think there is something of a core, 7 simple concept here: that credit research is not easy. It 8 is not something that someone can assign themselves as a good 9 credit analyst. It is very difficult and the only way you 10 have credibility is if you're able to mete it out with 11 discipline and some success over time. I mean, that's the 12 way it goes everywhere. 13 So, unfortunately for this situation, someone has 14 to have a proven track record and you can't let them self- 15 assess that they're good at this. I think the best way to 16 lower the barriers, as many of us have said, is to possibly 17 make it a more specialized area, because that also parallels 18 how complicated research is today, that deeper specialization 19 can ever bring better insights. 20 Again, we have some great ones from those that 21 started specializing in financial institutions. There's a 22 reason for that. There are big issuers and they're 23 complicated. They are some of the tougher things to analyze, 24 and that's why IBCA and Thomson got their start there. 25 So I think perhaps encouraging, in many of our 1 specialized areas, perhaps even asset backed or real estate, 2 there's lots of areas, I think that could help what somebody 3 saw and have a credible -- 4 I wasn't around for the national designation, when 5 that came to fruition, but I think what it means is it's 6 broad. It has recognition, because it's good. 7 COMMISSIONER GOLDSCHMID: I take it you would look 8 at their credentials, too, though. In other words, we do use 9 credentials in all kinds of areas, whether it's a member of 10 the bar or wherever else. 11 You might look at their credentials and have 12 specialized areas and let them grow. 13 MS. STRAUSS: No. I would say, unfortunately, good 14 credit analysis is, like any analysis, there is no credential 15 for it. I mean, look at long term capital, look at other 16 situations. 17 Having great degrees in intelligence doesn't mean 18 you have the discipline, the understanding of it, and a lot 19 of this just takes discipline in doing it day in and day out 20 and not losing your heads with the markets overly influencing 21 you. 22 As all markets are more volatile, I see that's a 23 big challenge for even our very well respected existing 24 agencies. It's tough. 25 But if you go in there, you show you understand the 1 basics and you have the disciplines and are able to attract, 2 as many are starting to do here, good talent, which means 3 good talent attracts good talent, I want a big research job, 4 that's the way you do it. 5 COMMISSIONER GLASSMAN: Don't we have a chicken and 6 egg problem here? Isn't it difficult to grow, no matter how 7 good you are, if you don't have the designation? And, if you 8 can't have the designation without a track record, then how 9 do we ever get past where we are? 10 MR. EGAN: I have a rating firm that we started, I 11 guess the base firm was started about ten years ago, we 12 started issuing ratings in December of 1995. 13 We commissioned a survey of all the major users of 14 credit ratings and the result of it -- we asked the question 15 which other firm besides the current NRSROs do you recognize 16 as issuing timely, credible ratings. 17 Our name came up more than four times more than the 18 next firm. So my point is that it's possible to establish 19 yourself in the market, to grow the firm, to get the critical 20 mass, to have a sustainable operation. 21 We have no problems, even when my partner refers to 22 the NRSROs, no room, standing room only, we have no problems 23 with the current rules as they are stated in 1997 or 1999. 24 Our problem is the way that we are applied. When 25 we got into the process, and some of the criticisms are fair, 1 but we feel as though the investors are intelligent. The 2 regulators should respect the regular users of these ratings 3 as to whether or not a firm is nationally recognized. 4 Don't need to get into how many analysts we have 5 and whether we're more efficient than Standard & Poor's. We 6 have to be. Otherwise, we won't survive. 7 So we have established in the corporate area a very 8 strong position to the point where many institutions will not 9 invest in securities until they hear our ratings, and they'll 10 throw out the ratings of what they consider to be the 11 conflicted firm. 12 So I think it's possible, in the corporate area, to 13 establish a major national presence and it is just the issue 14 of how big we have to be. 15 We continue to grow, but we would like the rules to 16 be applied in such a way that some of the non-conflicted 17 firms can get into the NRSRO designation. 18 MS. NAZARETH: Greg, did you want to address the 19 issue? 20 MR. ROOT: I just had a point here. I don't think 21 you actually have to do anything to establish credibility in 22 the credit markets. Basically, the way to get in, as well, 23 is, in addition to -- is provide valuable research. 24 Basically, my view is the marketplace works. If 25 you provide a valuable service and you're effective in your 1 execution and delivery, then the market will reward that, if 2 you will. 3 So you don't necessarily have to have rating 4 authority, if you will, to establish and build credibility in 5 a segment of the market, whether it be fully across the board 6 or regionalized or certain industries or others. 7 So, again, I'll go back to my predecessor, the 8 BankWatch days, and, again, we were recognized as having very 9 strong analysts who understood banking and that was the 10 entry. 11 It wasn't the rating that got us into the business. 12 It was having good analysts who understood a very important 13 segment of the market in a group. 14 MS. NAZARETH: It was our experience with Thomson, 15 I think, that Thomson had become very large and credible 16 through the research side and, in fact, obviously, with the 17 soft dollar business, there's plenty of money out there to be 18 used for good research that people want and that that is how 19 they grew and gained the national recognition. 20 Maybe, Gay, you could talk about the European 21 experience. Obviously, you're not burdened with this 22 designation, but maybe you can tell us how it works there. 23 MS. EVANS: There is quite little to say, I must 24 admit. I don't think the UK regulation actually creates any 25 special regimes for any rating agencies. 1 So we have no specific barriers to entry other than 2 the normal commercial ones that we've talked about. 3 Now, whether that's right or wrong, that's sort of 4 something I think I mentioned earlier, Basel is going to 5 change that, if Basel II goes through, but I'm not saying 6 that we don't have a regulatory regime going forward. 7 But what we do have now clearly is that it's a low 8 key -- it's very hard to say this, because it's low key use 9 of ratings and regulation. 10 What they have done is a group of agencies, what we 11 call and what are recognized for that purpose, and the 12 criteria for inclusion into that is what they call market 13 recognition. 14 It's just pure market recognition. It's based on a 15 survey, not done by the regulators, of which agencies are 16 used by banks for these purposes. 17 Now, we don't have the same issues, because the 18 rating culture was not big in Europe and the UK until 19 recently. It has been growing recently with the enhancement 20 of more U.S. institutions in the UK, but in the past, it's 21 not been one of those things that one has wanted to -- they 22 haven't used, and some of the foreign countries and European 23 countries still don't use rating agencies at all. 24 So they do a survey and right now the UK recognizes 25 probably more than ten agencies and that is more than any 1 other Basel banking committee. 2 There are probably at least about a 150 agencies in 3 total around the world and they vary in size and scope. 4 There is one that was brought to my attention, 5 which is, in comparison to your service, they have about a 6 160 analysts and they rate less than 800,000 companies or 7 just a little over 750,000 companies. 8 But they do it purely on an analysis of the 9 financial data and that is well regarded and it is 10 effectively, what I've been told, extremely transparent in 11 terms of what are the criteria you use. 12 So people can say this is the criteria we like, 13 this is what we need, and if we want to add anything else to 14 it, we can do so. 15 So we in the UK have not done much to date at all, 16 because there is no where in our regulation that mandates 17 that we should do anything. 18 Now, where we do do a little bit, I think, is 19 listing rules. It just goes back to the conversation from 20 this morning about FD. 21 In the UK, the rules surrounding selective 22 disclosure do not allow companies to give any kind of 23 information to the rating agencies. 24 Now, I must admit, we, in our listing review which 25 is underway, are under an immense amount of pressure from 1 many institutions to change that with regard to Reg FD. So 2 being here today and listening to all this has been very 3 valuable. 4 Also, there are a massive amount of directives 5 going through Europe right now with regard to financial 6 regulation, one being a market abuse directive, and that is 7 considering right now who and what receives this kind of 8 privileged, selective information. 9 There has been a lot of discussion on whether 10 rating agencies deserve that or not. So it's coming all the 11 way around from the UK to Europe, but I can say, at this 12 point in time, our regulation doesn't create any special 13 regimes. So what's done is very, very minimal. 14 COMMISSIONER CAMPOS: I'd like to pose a question. 15 Would there be anything wrong if, in our interpretation of 16 nationally recognized, that we took into account the 17 individual members of the credit agency. 18 So for example, if people who had been in the 19 business for a while were to make up a new company and they 20 individually had years of experience doing credit ratings and 21 having that expertise, could we give that particular firm 22 the designation if these people had done work on a national 23 basis and not worry if it was a startup, if it had otherwise 24 capital to do business? 25 MS. NAZARETH: Maybe it would be helpful if I go 1 through, because I skipped over it rather quickly, what the 2 criteria is that we do look at. 3 COMMISSIONER CAMPOS: You're not going to let them 4 answer my question. 5 MS. NAZARETH: It actually is something that is 6 taken into account, but it's only one of the factors. The 7 size and experience and training of the staff is an important 8 factor, although, obviously it's something that requires 9 judgment, which is part of the problem. It puts us in a 10 position of, to some extent, anointing somebody with this 11 designation. 12 The best thing I heard from Gay is that they 13 figured out a way to be out of the business entirely, which 14 certainly would be the preferred approach. 15 The approach that we have is intended to be sort of 16 a proxy for the marketplace making the determination, but it 17 inextricably puts us in the middle. 18 I think the factor that you raise, I mean, others 19 can opine as to whether they think it's the correct factor, 20 certainly is a factor in what we consider today, but we don't 21 give weight as in, well, you've got so many experienced 22 people that you don't have to have a history in this entity. 23 Maybe someone like Deborah could say how you would view 24 something like that as the buy side. 25 MS. CUNNINGHAM: I would rather go back to the 1 specific industry or concentrated one more so than that one. 2 That one I haven't really thought about. It inherently makes 3 a little bit of sense to me, but I would like to think about 4 it a little bit more before I give you an answer. 5 From the specific regional or specific industry or 6 sector side, going through that particular mechanism, I think 7 the danger lies there in what happens when that entity that 8 is specified or is particular to one sector decides to go 9 forward. 10 And I think we've actually seen this, to some 11 degree, when Keefe was analyzing banks. That grew into 12 broker-dealers. That grew into ultimately finance companies, 13 and I'm not sure whether those all link together of one sort 14 or not, but the original recognition from an NRSRO standpoint 15 for Keefe was through the banking world. 16 Effectively, when Mercury Finance defaulted, it was 17 considered a first-tier entity. One of the NRSROs that 18 gave it a first-tier designation was Keefe, and the question 19 was, was that because of their experience in that finance 20 company sector and as an NRSRO designee? Were they thought 21 about as being a broad-based designee versus an S&P and a 22 Moody's? Or were they thought of as being just simply a 23 banking designee and somebody didn't realize the disconnect 24 between those two? 25 I think that is the danger that you run into when 1 you're talking about specified firms. You have to have a way 2 then of disseminating to the marketplace what the specifics 3 are of each of these. 4 This is an NRSRO for just the Midwest region, this 5 is an NRSRO just for municipalities, this is an NRSRO just 6 for electric utilities. 7 There's a whole host of other questions and 8 problems in disseminating that information, I think, that 9 comes about when you go that route. 10 I don't think it's a bad route. I think it poses 11 other problems as to how you disseminate that information 12 properly, so that people aren't using it in a way that is not 13 appropriate. 14 MR. ROOT: I'm not going to defend the Mercury. 15 Other than that, we were not alone. It was one situation, 16 but we were not alone. 17 MS. CUNNINGHAM: And, actually, we were kind of 18 surprised how it -- it wasn't an issuer that we owned or 19 followed and when we kind of reviewed the facts, we were kind 20 of surprised about it. 21 MR. ROOT: And, again, going to, I think the common 22 -- in general, we're not flawless, but I think the important 23 thing, again, is -- you know, we had this discussion at 24 Thomson with the SEC. 25 To me, the issue is being recognized as a rating 1 agency. Let me use the argument I used at the time when we 2 were first designated. 3 We were rating financial institutions and that's 4 what we were given clear designation for. 5 Then all of a sudden, the dot-com world grows up. 6 Now, no offense, Leo, but S&P had no history of following 7 dot-coms. Why should they be allowed to follow dot-coms with 8 potentially inexperienced people - - I'm not saying you 9 would, but -- and where we may have gone out and hired far 10 more experienced people. 11 So to me, being in that world, we're coming to the 12 regulatory aspect or whatever, but I think the key, again, 13 even if you get the NRSRO, is you're not going to get very 14 far if you don't have credibility. 15 So I think the key is if you want to try and expand 16 outside of what you are currently doing, the key is that 17 you're doing it the right way, with the professional people. 18 And to the extent you do it the right way, the 19 market will accept that. And if you do it the wrong way, 20 people like yourself, with user ratings, will see through 21 that. 22 So I think that's the important thing. You start 23 with the company as an agency. I don't think the SEC would 24 want to go into each segment and say, okay, you guys do a 25 good job in this sector and not so good here and that type of 1 thing. 2 And, again, obviously, as a company in this whole 3 process, obviously, our motivation is to do everything right 4 if we're going to be successful. 5 So, again, we may have made some mistakes, but the 6 point is that you approach a new area in the exact same way 7 that you do at your existing business. 8 MR. O'NEILL: I think the only thing I would 9 disagree with you on, Greg, is that I think the presumption 10 would be, because of our long track record, that we would go 11 out and bring the resources to bear on the problem. 12 MR. ROOT: So would we. That's what I'm saying. 13 That was my point, but I know you would, and we will do the 14 same, so why should we be any different in that respect. 15 MR. O'NEILL: The proof, though, as you point out 16 is in the performance. You know, it's interesting, I would 17 just make one little observation. 18 When was Mercury? Was it 1991 was it or 1989? 19 MS. CUNNINGHAM: It was 1997. 20 MR. O'NEILL: But there you go, five years or 12 21 years later, the people who are in the market are still 22 commenting on and, in some respects, I guess, criticizing the 23 role of rating agencies and what we do. 24 And I think that's what I said earlier, the market 25 really does enact a lot of discipline on how rating agencies 1 behave. This market has a vested interest in how we do it, 2 and I recognize the national recognition problem the SEC has, 3 but I do believe the market is pretty tough on those. 4 MS. STRAUSS: I have a response and then maybe a 5 suggestion. On the response on whether you can just take a 6 group of people that are experienced and say they are 7 independently experienced, and, therefore, as a group, they 8 can do ratings well and mete out some kind of a rating 9 system, I just don't think that works. 10 A group of smart people in different environments 11 perform differently when you're asking them to do a task, and 12 I still think that there needs to be a corporate culture, a 13 set of standards. 14 Perhaps you say, just like S&P, we can trust them 15 because they have a corporate environment that they have been 16 able to show that they can hire the right people to follow a 17 new sector, just a group of smart people will not give you 18 what you need. They need a track record together. 19 The suggestion I might make is one where, I think 20 Debbie made some good points, as well, but how do we make 21 sure rating agencies continue -- perhaps they start narrow, 22 but how do they continue strong. 23 It's a suggestion that there really is, as part of 24 the NRSRO process, a formal element which requires that there 25 is public assessment and feedback on a regular basis about 1 those ratings. 2 That could help you in something where, say, 3 someone is starting narrower. That's why, for you all, we're 4 market practitioners, there's a lot of people out there that 5 could come back and say these ratings don't look as good. 6 Here are some examples: You have a chance for the 7 marketplace to say maybe they're a little out of their league 8 and they could get that feedback. Similarly, for the bigger 9 rating agencies, and, again, you say there isn't a market 10 dynamic. There are only three. There isn't a nice way for 11 us, other than perhaps trying to make a silence, and while, 12 in the right markets, we'll be listened to, to perhaps, in a 13 constructive way, say you're getting a real -- maybe this 14 doesn't work as well. 15 These are the issues that you've heard many of us 16 say. For example, right now, I think you'd hear back on 17 consistency and using a criteria that we know about. 18 There are things going on in the marketplace that 19 you're not going to get to know, but it's our everyday 20 business and if there was part of this systematic feedback, I 21 think it would make the system strong. 22 It could allow you to do more flexible things. I 23 know we were one of many who made comments in your 1997 24 process, and we talked a lot about that. 25 A good point, I mean, in there is in other quasi 1 regulatory areas, such as radio stations and so forth, 2 there's a process to get systematic feedback from the public 3 on how is someone doing, and that's not done here. I think 4 it really could make the current process stronger and give 5 you a better way to be more creative in the future. 6 COMMISSIONER CAMPOS: I think the basic question 7 here, and I'm jumping the gun here and stealing your thunder, 8 is do we need the government to be in this business at all? 9 What I'm hearing is, in fact, there are barriers 10 and maybe they're not insurmountable, but they're pretty 11 close to it. Indeed, you have a company and you can't put 12 experienced, talented people together and be able to do this 13 work. Then you have effectively a de facto monopoly that is 14 going to continue, it seems to me, forever. 15 That's the side comment. The real issue is, in 16 light of what Gay said seems to work in Europe, and maybe 17 this is just a broad issue here, why wouldn't a system like 18 that work here? 19 If we somehow were willing to do the effort to take 20 away all our regulatory references to NRSROs, a lot of work, 21 but let's say we were to do so, just a big hypothetical, what 22 would be the downside? 23 Would the marketplace be able to take over and 24 handle good ratings? 25 MS. STRAUSS: Let me just quickly jump in. With 1 full respect for the European markets, the U.S. has a much 2 more, a much deeper, a much longer history in their debt 3 markets, where many more investors are going directly into 4 those markets. I would start with the money market funds, 5 where there is a regulation where your ratings are very much 6 embedded in that. 7 That is a humongous market that does not exist in 8 Europe and so it's not a handwave to say let's look at 9 Europe. I think Europe is more, as Gay suggested, trying to, 10 as they are building their capital markets and opening them 11 up, trying to look at where to add our regulations. 12 So, just an aside on that. 13 MS. CUNNINGHAM: I would think, too, that what we 14 have in Europe is essentially migrating from what has been a 15 "I know this company, I recognize the name, I know the people 16 that run it because they happen to be on the school board of 17 my kid's -- of the educational system that we're in," 18 basically a knowledge base of the regional investors in the 19 marketplace has been what that market has been based on. 20 To a large degree in Germany, to a very large 21 degree in Spain, lesser degree in the UK, but other 22 countries, as well. I think as that market opens up and 23 becomes more developed and realizing the small problems that 24 have occurred in that market, that have basically been bailed 25 out by the banks, which have ultimately been bailed out by 1 the countries themselves, have not taken place in the U.S. 2 and is not going to continue as the markets grow and develop 3 and become larger and stronger and bigger issuers there. 4 So there has to be some sort of a tie-in to high 5 credit quality and just individual assessment and knowledge 6 basis on who the issuer is is no longer going to be the 7 precedence. 8 COMMISSIONER CAMPOS: Bailouts in the U.S.? 9 MS. CUNNINGHAM: They have occurred. 10 MR. REYNOLDS: The European front, though, I used 11 to run, for several years, their credit research department 12 at Deutsche Bank, and they are rife with problems in terms of 13 inefficiencies and information flows. 14 You have the slew of unrated bonds out of the 15 Milschtein that are just starting to come home to roost right 16 now in the downturn. 17 So what's going on in Europe right now is really, 18 three years into it, a change in the credit cycle and a 19 change in how companies borrow. Disintermediation has really 20 just kicked in. 21 COMMISSIONER CAMPOS: I'm sorry, but knocking 22 Europe really isn't getting to the question. 23 MR. REYNOLDS: I'm not knocking Europe. I'm just 24 saying the structure. 25 MS. STRAUSS: It's not a good parallel. 1 COMMISSIONER CAMPOS: That's fine, but what if the 2 marketplace took over and allowed credit rating agencies to 3 exist on a free-market basis. 4 MR. REYNOLDS: I don't necessarily disagree with 5 the point. Take the NRSRO designation away and when the 6 smoke clears, Moody's and S&P will dominate if you leave it 7 there. Moody's and S&P will dominate. 8 I totally agree with that. I'm just saying that to 9 go totally away from a model where you have some type of 10 oversight, I think the government still needs to stay in this 11 business, at least from a regulatory and oversight 12 standpoint, to assure quality, because in this country in 13 particular, the dispersion of holdings of debt instruments 14 throughout the system from the institutional level to the 15 retail level requires some additional oversight, as we have 16 seen in dramatic fashion over the past year. 17 I think Europe is just starting to see that now, 18 because so many bonds have been put in the market over the 19 past three years since the advent of the Euro, they're in the 20 credit cycle now where they're starting to feel the pain and 21 they're going to sit back and say, okay, how do we make sure 22 there is a better mechanism in place to assure there is some 23 independent body watching the quality of these securities. 24 MS. NAZARETH: Perhaps I should clarify, because I 25 think Roel's question sort of raises this issue. Clearly, I 1 think everybody thinks there are some barriers to entry for 2 credit rating agencies, but are the barriers to entry because 3 of this NRSRO designation or are they separate and apart from 4 what the Commission has done or what the designation has 5 done? 6 MR. FERNANDEZ: I'd like to answer that and then to 7 come back to Harvey's question, which never really got 8 answered in the first place, which is similar. 9 To me, I kind of do a reverse engineering here. If 10 that were the case, why don't we have them now. What is it 11 that is an effective barrier? 12 We do have various credit rating agencies and I 13 think you'll agree that are of good quality that could step 14 up and become an NRSRO. I don't see anybody clamoring for 15 that designation at the moment. 16 Maybe I'm wrong. Why do we need it then? Because 17 I think there are already certain standards one has to meet 18 to meet that. 19 Having put that forward there, please, let me get 20 through this, I think the first thing you have to have, it's 21 been identified, Cynthia identified some of them, Greg 22 identified some of them, you have to have a track record, you 23 have to have a good team, you have to have the management, 24 you have to have all these things in place, and you have to 25 be prepared to lose money for an extended period of time. 1 It takes a while to build this kind of expertise, a 2 while to build the name recognition, the brand recognition. 3 So you have to be adequately in capital and funded through 4 this period. 5 So let's say you do all those things. The business 6 model fails you to take the next step, in my opinion, because 7 how do you get there, well, you develop expertise and the 8 higher value products, selling your research to institutional 9 investors who are willing to pay for it. 10 But then you're going to be asked to broadly 11 disseminate this as a public good, where you're not going to 12 be adequately compensated because who you are providing it to 13 is a totally different audience. Now you're talking about 14 providing this to a broad spectrum of individual investors 15 who are not disposed to pay for it, who don't see an adequate 16 need for anything more than two ratings. 17 Because this is what they have -- the market has 18 already decided. They would, I think, find it confusing to 19 have a panoply of ratings. If you go back historically, when 20 there were more, again, people developed preferences for who 21 has the greatest integrity and the greatest standard or brand 22 image out there. 23 So when you have someone who has reached the 24 critical mass, who has paid for the track record, who has 25 reached this threshold, I think that they are not willing to 1 give up the higher value added product to produce something. 2 This is the same problem we have with independent 3 equity research here. Sure, maybe you want to step up, but 4 institutional investors might not find that S&P tear sheets 5 provide the same level of analysis that they get from some of 6 the more sophisticated boutiques who could provide 7 independent equity research. 8 So it's almost that old saying, is that the more 9 broadly disseminated something is, the less value it has. So 10 that's part of the answer. And, again, how many do you need? 11 If you need comfort levels in terms of rating, if the job is 12 well done, and I think it is, how many do you need? 13 Are we going to be better served with six as 14 opposed to three? I don't know. Tell me in the economic 15 literature, what is the critical point in an oligopoly? I 16 mean, you do get better efficiency, but I don't see it. 17 I don't see that you gain much by going too far 18 out. 19 COMMISSIONER CAMPOS: So are you saying government 20 should maintain two or three and that should be our 21 objective? 22 MR. FERNANDEZ: No. I don't know if that's the 23 answer either. I'm just trying to answer your question of 24 why haven't they naturally -- what would it take to 25 stimulate -- 1 COMMISSIONER CAMPOS: There's a lot of regulation. 2 MR. FERNANDEZ: Right. That's part of it. There 3 are also natural barriers that the markets and the investors 4 themselves impose, and I think part of it has to do with the 5 way that we value information and analysis, which is very 6 bizarre and not market oriented. 7 COMMISSIONER GOLDSCHMID: Frank, but if it's a 8 natural barrier, and I think there are some natural barriers 9 here, the answer is if you get rid of it all, the natural 10 barriers will remain and that's how it works. You've got to 11 come further than that to say what we're getting here is 12 something more efficient, broader, of higher quality, the 13 kind of looking to market that's giving us more. 14 Now, that may be the case. 15 Assuming we're going to keep the barriers for the 16 second and assuming they work to some degree, which appear to 17 be the case, what are we losing by not having a more dynamic, 18 more wide open system? 19 If I gave Malcolm what you've got here and said no 20 one can sell automobiles in the United States unless they put 21 100,000 on the road, then it would be wonderful for the auto 22 company, but not so good for the consumer. 23 Now, you say this market is different than autos. 24 Tell me why, how it's working and what we're gaining and what 25 we lose. 1 MR. FERNANDEZ: It may not be that different than 2 autos, to be quite honest. 3 MR. O'NEILL: Commissioner, I think it is 4 different. The key difference is what we're dealing with 5 here is intellectual property and there are no barriers to 6 entry to intellectual property. 7 There are people -- we say that, well, we created 8 this oligopoly or these three rating agencies, well, I submit 9 that, again, both of us doing it for 85 years or something, 10 and all that, has its own merits. 11 But there are hundreds of fixed income -- thousands 12 of fixed income analysts in the marketplace who are looking 13 at our ratings, assessing their own views of 14 creditworthiness, questioning, criticizing, Sean's role, 15 others. 16 I don't think we have an oligopoly of intellectual 17 opinion here. I think we've got a multitude of intellectual 18 opinion. 19 I think the issue and the problem are that what 20 we've done has found its way into regulation, and that is 21 really it. 22 MS. NAZARETH: That's the problem. The question 23 is, could we wean ourselves off this designation? 24 Whatever happens, from an economic perspective, 25 happens. The marketplace sort of votes with their feet in 1 their pocketbooks, but is there some substitute for this? 2 Obviously, what we're concerned about is because of 3 the way it's used in the regulations. You don't want issuers 4 to have the ability to just shop for a rating to get the 5 benefits that each of these many, many disparate regulations 6 provides. 7 Is there a substitute for that? Is there some 8 other way that you could have that assurance? Again, some 9 other market based substitute? 10 MS. STRAUSS: Just one thing. I think, also, in 11 the regulations, I mean, these ratings are in regulations, 12 whenever the research is done, I think there should be 13 situations, which is a little tougher, to talk about whether 14 those regulations have protected people, meaning that the 15 regulations where ratings -- 16 Again, the ratings have a pretty good track record. 17 They are in regulations for good reason. In many places, 18 they have protected investors through those regulations. 19 So if you dilute it by saying we will have any 20 rating agency with a group of six smart people, to the 21 extreme, then you could have that flow back through the 22 regulations and, in essence, not have those you're trying to 23 protect through those regulations left unprotected. 24 So, again, some of the regulations I'm familiar 25 with that relate to us, I think a lot of them are good 1 regulations. 2 So the question shouldn't be just how to let any 3 rating agency get into those regulations. It's how to find 4 something -- and I'm not sure, in the debt markets, that it's 5 so easy to find a market-based regulatory factor. 6 Don't forget, debt securities, by their nature, are 7 different. They have very little upside and tremendous 8 downside. On the shorter end, it's typically seen for 9 safety. 10 Except for the high yield market, which is really 11 quite small, most investors are looking for relative safety 12 here and so it's looking at this downside that I find rather 13 difficult. You need somebody looking at it and helping 14 people through looking at the downside of this. 15 So it's an odd thing, just as a general backdrop to 16 thinking about this. 17 COMMISSIONER GOLDSCHMID: Are you hearing the end 18 users and the issuers saying this system is working fairly 19 well, even when the imperfections or barriers are drawn? 20 MS. NAZARETH: Did you say that, Malcolm? 21 MR. MACDONALD: No, but I will happily comment. I 22 do think the system is working fairly well. I think Cynthia 23 put her finger on it. 24 The requirement for a rating agency from the 25 issuer's perspective is competence, rigorous analysis, and 1 ultimately credibility in the marketplace. 2 Now, quite frankly, I find it interesting. If lots 3 of people had wanted to jump in to the rating game and get 4 the designation, I'm amazed that five companies ended up 5 commingling into Fitch. 6 I would have argued that somebody would have come 7 in and said here's my entry, but we didn't see that. 8 So I would be happier if I had six, because I would 9 not say I want to go shopping, but I would like the idea, if 10 I found one incompetent, I could get rid of them. 11 At the moment, where there are three, that puts -- 12 that makes it very difficult to do. 13 MR. HARRIS: The government has written ratings 14 into its regulations and, indeed, its statutes to address 15 various agency problems. This is the reason there are laws 16 in the first place. 17 There were alternatives available to the drafters 18 at the time that probably weren't thought of, it really seems 19 ridiculous to think of it in the following way. 20 They could have tried to draw an entire list of 21 what they were talking about. These securities are risky and 22 these aren't, and they needed to discriminate between the 23 different types of securities for a variety of purposes. 24 Clearly, they decided not to do that, because such 25 a list wouldn't be enduring, it wouldn't be stable, and 1 various other problems. 2 They also could have hired people to maintain such 3 a list, but they also decided not to do that, and instead 4 looked out and saw that there were rating agencies out there 5 who provided such lists because there was a private demand 6 for them. They decided to free ride or piggyback off of 7 those rating agencies, and, in doing so, established rating 8 agencies in the law through the NRSRO designation. 9 We've been talking about how many we ought to 10 establish, but I would like to pose, as an alternative, the 11 following question. 12 Clearly, the NRSROs have benefitted from this 13 establishment. That is the whole purpose of this discussion 14 here. And once established, given that the government, by 15 intention or by accident, has conferred a fair amount of 16 benefit to these entities, the question is what 17 responsibility do those entities owe the government or 18 otherwise? What responsibilities does the government have 19 over people for whom they have given special privilege? 20 Normally when we give people privilege, we have a 21 regulatory oversight. So the question here, no matter how 22 many we have, is what regulatory oversight should the SEC 23 exercise over the NRSROs? 24 And the issue that we face, as the SEC, is not 25 whether we presently have a problem, I think we're all agreed 1 that the three main entities seem to be doing a reasonably 2 good job, with some of us dissenting, but, by and large, that 3 seems to be the perception. The issue is not whether they 4 are presently doing a good job, but whether they will always 5 be doing a good job, because it will be on our watch that a 6 failure will occur. 7 So what is necessary to ensure that there will be 8 no such failures in the future? To what extent should the 9 government be able to reach in and look over the shoulders of 10 the people at S&P or Fitch or Moody's or anybody else and 11 ensure that having given them privilege, that we are not 12 creating a situation that will be a lurking time bomb. 13 MR. EGAN: I have a comment. If your oversight 14 gets into the underlying ratings that are issued, I think 15 it's a big mistake. 16 Let me explain our experience. We are not always 17 bearish, but we have been bearish for probably the last two 18 years, and we have been taken to task for a lot of our 19 ratings. 20 Worldcom, Enron, Ford, our clients thought that our 21 ratings lacked credibility because they are significantly 22 below the major rating agencies. 23 When Ford was cut three notches by Standard & 24 Poor's and then put on negative watch, they started believing 25 that we knew what we were talking about. 1 Likewise, with some of these other major calls. So 2 don't get into the business of trying to second guess these 3 things, as you are going to be pressured by the issuers. 4 MR. HARRIS: My guess is that there probably won't 5 be any dissent anywhere on this table for that position. But 6 what of the things like document retention? Should we be 7 allowed to inspect these firms to ensure that the 8 compensation contracts of the analysts are further unlinked 9 to the results that they produce? 10 Remembering, again, that what's not necessary now 11 may someday be necessary, and it's our job to ensure that 12 nothing blows up on our watch. 13 COMMISSIONER CAMPOS: Let's be more specific. I 14 think the major NRSROs are essentially taking the position 15 that they are voluntarily part of the investment advisory 16 situation and that they are reserving, at least, the right 17 not to provide documents based on inspections, and that there 18 has been resistence to our compliance audits. 19 So that should be addressed. 20 MR. HARRIS: In particular, should any firm wanting 21 to be an NRSRO waive the First Amendment defense and subject 22 themselves to oversight as a quasi or directly an investment 23 advisor as opposed to being treated as a publisher of 24 information, much like a newspaper. 25 Should that be the condition of the NRSRO 1 designation presented in exchange for the privilege of the 2 franchise? 3 MR. O'NEILL: I'm glad you got back to the 4 privilege of the franchise. I think that the privilege was 5 granted by the marketplace long before the SEC granted its 6 privilege, in all due respect. 7 I think they exist today because of the market 8 recognition, the national recognition, and not because of the 9 NRSRO status. As I've said, I think, several times, these 10 ratings are looked at every day. They are challenged every 11 day. They are assessed every day, and I think that is what 12 really creates the power behind this rating methodology. 13 And if we were not an NRSRO, for some reason, I 14 don't think Standard & Poor's would just -- I mean, we would 15 go right on doing what we're doing and we would do it just as 16 aggressively as we have in the past, quite frankly. 17 I don't think that we should be asked to waive our 18 rights under the First Amendment. I think there are ways 19 that the SEC can monitor the progress of NRSROs, taking a 20 look at some of the things, codes of ethics, insider trading 21 policies, factors that bear on the economic and political 22 independence. 23 If you want to look at the compensation, we don't 24 have contracts with our analysts, but the compensation 25 arrangements of our analysts to ensure that there is no 1 direct tie, we would be happy to share that with the 2 Commission. 3 Track record I think is really critically 4 important, a visible track record, default studies, 5 transition studies, market research to the investment 6 community, what do you think, how do you think they're doing, 7 and, quite frankly, we do that market research every couple 8 of years as to how our ratings are being used by investors 9 and issuers, and I've heard some things today and we did some 10 market research today and we're going to go work on that when 11 I get back. 12 But I think, quite frankly, the point that the SEC 13 should really be careful, and I guess you've said it and I 14 agree, I don't think you really want to participate in the 15 rating process through regulation and I think that would be a 16 tremendously, tremendously negative thing to happen in our 17 capital markets. 18 MS. NAZARETH: Let me ask this question more 19 explicitly. As you know, currently, each of the NRSROs is 20 registered as an investment advisor. Certainly, we've heard 21 from each of the rating agencies in the past, that the 22 regulatory construct really doesn't fit your business. On 23 the other hand, it does bring you within our jurisdiction. 24 We certainly take the position that you have to retain books 25 and records and we have examination authority and the like. 1 I guess what you were talking about here was the 2 kinds of things that you think that the Commission should be 3 looking at or you certainly would find appropriate, whether 4 that's done in a separate regime or whatever is another 5 issue. 6 You said that you did conduct some surveys. Was 7 that a one time survey? I thought it sort of tied in a bit 8 with what Cynthia had said about getting feedback from the 9 marketplace, staying up with whether rating agencies are sort 10 of doing their job. 11 It shouldn't be the case, clearly, that you give 12 someone this designation and then never look at it again. 13 MS. EVANS: And I think we're lucky, fortunate. As 14 I said before, we don't have anything about rating agencies 15 in any of our regulations. 16 So that sort of saves us from all of this. But I 17 think in terms of when we do say, okay, how do you go forward 18 for that little bit of regulation that's included in terms of 19 designating whether you're investment grade or not investment 20 grade, those are done by some of the trade associations. So 21 there's a little bit of difference. 22 So it's not like the BBA being done by banks. We 23 would work with them and say, okay, could you do that, let's 24 talk through it afterwards. 25 So that is really how it has been done in the past, 1 but we do keep dialogue and have technical discussions with 2 rating agencies, but that's about it. 3 I actually thought Commissioner's Campos' question 4 of could you have a market based solution for all this is 5 really quite interesting. You're right. Europe is different 6 and actually now we're getting into more cross-border 7 business, the debt business is growing, and may rival, 8 obviously, the U.S. at some point in time if you looked at 9 national debt, et cetera. 10 But I think one of the issues in the Basel approach 11 is that some of these European countries do not use rating 12 agencies, as we still know, they use internal ratings. Now, 13 there are criteria on what is an internal rating, and then 14 here come Cynthia and Debbie about how much they do in 15 research. Yes, it's valuable to have a rating agency's input 16 on other aspects, because that's another check and balance, 17 but you've got to question -- and I'm not saying we could 18 deal without rating agencies, but why would that now ever 19 work. 20 Why not be another solution? You have rating 21 agencies that provide it, you change your rules, but I think 22 the whole thing, to put it on the table and have it answered, 23 why can't a market-based approach work, is actually quite 24 interesting and I don't think I actually heard a real answer 25 to that. 1 MS. STRAUSS: Market-based, I don't know what you 2 mean by market-based, a lot of that term is broad-based. I 3 think one thing I thought you meant market-based by the fact 4 that each of us do our own independent research. We have 5 some excellent regulation on that front. 6 I mentioned, again, Rule 2a-7, which incorporates 7 the ratings, but also very clearly says that we must do 8 independent assessments. It's very powerful and it works 9 really very well, because it holds us accountable, as we 10 should be held. 11 It also allows a smaller shop. It helps set a 12 standard, which is well respected, which are the rating 13 agencies, and it could help a smaller shop that can do some 14 additional research, but maybe not as much. 15 It's a nice mix. So it's a combination of both, 16 which I think, having lived under it for years, I think it 17 works very well because it holds us accountable. 18 So if market based means that we're supposed to do 19 our own research, I think that's great. 20 So often, though, and particularly in today's new 21 world, market based means look at market prices as a proxy 22 for credit analysis. 23 That's really dangerous, and I think most people 24 here know enough to know that. I cannot tell you how 25 frequent it is, and including from our own rating agencies, 1 many of which have been thinking about the big three, all 2 have been under pressure to have your ratings be more market 3 predictive. 4 It is very much en vogue, it is very powerful here. 5 There are academic papers written about it, but that's really 6 dangerous territory, because market prices are more short 7 term driven and influenced by many other things. 8 MS. NAZARETH: Because we're talking about credit 9 ratings in the debt markets, notwithstanding all of our 10 regulation, the debt markets still remain somewhat the dark 11 corner of our own securities markets. They do not have 12 anywhere near the level of transparency that we have in the 13 equity markets or the ability to, for instance, influence 14 credit spreads through a small number of trades in order to 15 achieve benefits under our existing regulations could be 16 relatively high. 17 So I agree with you that the market based solution 18 could have some consequences if it's not controlled in some 19 way. 20 MS. CUNNINGHAM: I think that existing regulations 21 is the key phrase there, because I agree with what Leo is 22 saying, that if S&P and Moody's and even Fitch, to some 23 degree, did not have NRSRO ratings, the marketplace today 24 would still interpret and utilize them in a way that is very 25 similar to what our current pattern is. 1 But because they are built into the regulations, 2 there needs to be -- it's already in Rule 2a-7, it's in other 3 parts of statutory, municipality criterion. All of that 4 leads to the need then for some sort of oversight, because 5 it's already built in as part of the mechanism. 6 I think you've got two paths you can take. Take it 7 out and have no oversight, in which case then you go back to 8 market based without any type of underlying kind of cushion 9 that the rating agencies currently provide or that they can 10 continue to provide, but without the NRSRO designation, or 11 keep the NRSRO designation, but provide some mechanism of 12 oversight. 13 Oversight can be taken from many different avenues 14 of oversight that the SEC currently does, and sort of the way 15 that an inspection regime develops can evolve. 16 MR. HARRIS: There is a third alternative that 17 maybe we should explore and ask about, and that is to write 18 into our regulations and statutes what we mean by a rating 19 and then tell everybody who has to comply with the statutes 20 that they have to get their ratings and if the ratings aren't 21 of sufficient quality, then we will expose them to liability 22 for failing to do what's proper. 23 So, in this way, we don't establish a national 24 NRSRO designation, but instead put the liability upon the 25 corporate directors and the fund managers and so forth to 1 obtain quality ratings. 2 Does anybody see the pros and cons of that 3 proposal? 4 MR. EGAN: I think it will substantially increase 5 the cost to the rating firms and it will just erect another 6 entry barrier, and that is the liability for making a wrong 7 call. 8 MR. HARRIS: The liability would not be on the 9 rating firms, but rather on the people who use the ratings to 10 comply with the various laws. 11 How do you feel about that, Cynthia? 12 MS. STRAUSS: We already feel that, I don't know 13 what the form of it would be, but we already feel that we 14 have pretty serious liability anyhow. 15 MR. HARRIS: You have serious responsibility. 16 MS. STRAUSS: Right. We have responsibility, 17 again, already under Rule 2a-7, it's specifically in the 18 regulation. I think it is very powerful. 19 Then we have it through what we feel is our 20 fiduciary duty on our other funds. 21 So it's intriguing. We would, as Debbie already 22 said, probably use similar rating agencies that we use now 23 and if new ones come forth, we would assess them certainly 24 analytically. 25 So I don't know, when you say the word liability, 1 of course, what does that mean. We would use them in our 2 process. 3 Are you saying we would be held accountable to use 4 whatever rating agencies we see appropriate or we find 5 appropriate? 6 MR. HARRIS: That's correct. The law, for whatever 7 purposes, distinguishes between certain types of debt and 8 some more appropriate for some purposes, according to 9 whatever the wisdom of the folks who wrote this stuff. 10 And then, the question is, how do we classify 11 various securities, and one way is we say whatever the NRSROs 12 say. 13 The other way is to say this is what we mean by the 14 classification, you've got to do it yourself, you should rely 15 on one or two opinions, you solicit the opinion, but if you 16 don't do it right, you're going to be exposed to liability 17 through civil action or through the enforcement efforts of 18 the Commission. 19 MR. MACDONALD: Larry, forgive me, but I think 20 we're going off on a wild goose chase, because let me tell 21 you, ratings are not a six sigma item. You cannot have a 100 22 percent accuracy. 23 If you could have a 100 percent accuracy, I think 24 you would have something, but you can't. If you take a small 25 company that is being rated and a meteor falls on them, I'd 1 love to hold Leo accountable for it, but I don't really think 2 it's reasonable. 3 MR. O'NEILL: It's a risk I'll take. 4 MR. MACDONALD: But you can wipe a company out from 5 an event that is totally and completely unpredictable. 6 MR. HARRIS: Again, the liability is not on the 7 rating agency, but it's on the people who are applying the 8 ratings, and, of course, the issuer would make a responsible 9 choice. 10 The reason I'm pressing this is because the 11 Commission has in front of it this problem, which is how 12 would you get a market based solution to this problem. 13 That is, the debt, I think, is the marketplace 14 solution and if it's a problem, we need to know about it so 15 that we can rest more confidently with the status quo. 16 COMMISSIONER GLASSMAN: Wouldn't it further 17 entrench the existing brand names, because they would be the 18 safest choice? 19 MR. HARRIS: I certainly suspect so. 20 COMMISSIONER GLASSMAN: So it would be a further 21 barrier to entry, because people would go to the ones that 22 they already know. 23 MR. HARRIS: That's probably the case. 24 MR. FERNANDEZ: Not necessarily. We have an actual 25 example of this, and I'm going to turn it to Gay here, 1 because I think we've both been dealing with this nightmare 2 for how long now? 3 MS. EVANS: Too many years. 4 MR. FERNANDEZ: We're actually trying to assign 5 products into risk buckets, and this is the Basel II, and 6 we're trying to do this to avoid the use of the rating 7 agencies as the be all and end all and come up with a market 8 solution. 9 I have only been dealing with it for three years. 10 I don't know how long you have been working on it. 11 But it's possible, yes, we can do this. It's not 12 as easy as anybody ever thought when we headed down this 13 road. 14 The whole idea, I think, correct me, was to try to 15 find a way to harmonize the cross-border standards and to 16 come up with something that was a little bit easier to deal 17 with on international issues. 18 You can relate your own stories on this one. 19 MS. EVANS: Well, I think it's a little bit off 20 track, but clearly it's the concept of risk buckets and if 21 you use rating agencies to make a simplified format, that 22 would be easy. 23 Then it goes back, then it's, oh, you can have 24 internal risk buckets and internal analysis and that allows 25 you risk buckets. But you have to break all that down and 1 that has been taking an immense amount of time. 2 So when you get into this concept of risk and 3 opening up to the wide variety, you just have to make sure 4 you use the right credit rating in the long run or use your 5 own internal analysis to back it up. 6 We can go on forever, because that's a whole -- 7 MR. HARRIS: And it will go on forever. 8 MS. EVANS: Years. 9 MR. O'NEILL: From our perspective, I just don't 10 think it's workable. You will hear one of the other rating 11 agencies, at some point, make the point that a bond has 12 really only two things, it either pays off or it doesn't and 13 it's binary that way. 14 But a rating talks about the relative likelihood 15 and it ranks it out, and I guess what would happen is all 16 those bonds that defaulted would give a cause of action or 17 something against the purchasers of those bonds, I guess, and 18 I don't think that should be a liability or risk the 19 purchasers of bonds, albeit for the benefit of their ultimate 20 customers, should take on. 21 MR. EGAN: From our perspective, it would be a big 22 positive, because what would happen is that the institutions 23 would have to make the assessment who has the credible 24 ratings and who doesn't have the conflict of interest, and it 25 would be great for us. It would be terrific. 1 MS. NAZARETH: Cynthia? 2 MS. STRAUSS: Sometimes if you put the burden, you 3 get this secondary behavior if you put greater liability on 4 us, and, then, again, I think I would fear that our behavior 5 would be that we would just then, with that additional 6 liability, why take any risk. 7 We would always stick to the same traditional ones 8 and then we would go on doing what we always do, which is, as 9 we do now, we look at everything in the marketplace. 10 So I don't know that that really fundamentally 11 opens it up and makes a more welcoming marketplace for it. 12 MS. NAZARETH: The NRSRO process, if it were 13 retained, and, obviously, I think we've got a lot of open 14 questions here, if it were retained, is there any reason to 15 argue against the whole process being more transparent. 16 I think you are aware of what the Commission had 17 proposed several years ago. Does anybody have any reason to 18 object to go to transparency there? 19 MR. O'NEILL: No. We would endorse a lot more 20 transparency around how the designation is made, clearly. 21 However, again, I would caution that if that transparency, 22 under the guise of that transparency, becomes a direct 23 oversight regulation of the opinions that are coming out of 24 the rating agencies, then we'd have -- 25 MR. ROOT: I totally agree. Again, I think the 1 oversight should be that once you set up the standards for 2 becoming an NRSRO, that you've got to continue to meet those, 3 and that, to me, is the regulation, and if you no longer meet 4 those, then you should lose that designation, the same way 5 you should never have gotten it in the first place if you 6 didn't have those standards. 7 But as long as the process becomes transparent, I 8 think -- I don't think it's broken, I guess, personally. 9 There's been a lot of talk here, but overall, I don't think 10 the whole concept of the activity of ratings in the 11 marketplace is broken. I think it's very efficient. 12 MS. STRAUSS: Again, I don't want to -- but I do 13 think a nice way, Larry, as you mentioned, of looking ahead 14 and what are the issues ahead is to get some more regular 15 market feedback about what we're feeling in the trenches 16 today. 17 Again, I started off my comments today, I don't 18 think it's broken, but I do think that we are gathered here 19 because it's not perfect, which is fine, but how do we get 20 some mechanisms to keep it strong. 21 MS. NAZARETH: Does anyone else have any last 22 words? 23 COMMISSIONER CAMPOS: I have a question. Back to 24 Larry's discussion a minute ago. 25 There's a certain intellectual elegance to what he 1 was proposing, which is that companies buy their own ratings. 2 I, for one, am not convinced that just because there's two or 3 three overwhelming companies, that it would always be the 4 case that they would always have the business. 5 Indeed, Mr. Egan is eager to try to convince 6 companies that he can provide something better. I guess 7 there used to be an old saying that no office manager ever 8 lost his or her job when they selected IBM to do their office 9 xerox. Well, times have changed and times do change. We 10 have an ongoing economy and market forces and things move 11 very quickly these days. 12 So I would be interested to know more about why 13 that wouldn't work. I realize we got off into, well, the 14 topic of there's too much liability, but I don't think that's 15 really the issue. The liability regime can be dealt with and 16 tweaked and so on. But issuers buying and being responsible 17 for their own credit ratings, why not? 18 So maybe you can send us back notes on that. I, 19 for one, would be interested in it. 20 MR. EGAN: Just one final point. We are not paid 21 by any companies and we don't want to. Hopefully, we can 22 continue that business model, it's been very successful, of 23 stripping out the conflict of interest. So it's just being 24 paid. We're only paid by the investors, mainly institutional 25 investors. 1 MS. NAZARETH: Does anyone else have a final point? 2 Well, thank you very much. We've finished a full three 3 minutes early. 4 Thank you very much for your help. 5 (Whereupon, at 4:42 p.m., the hearing was 6 concluded.) 7 * * * * * 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25