1 1 UNITED STATES SECURITIES AND EXCHANGE COMMISSION 2 3 4 ROUNDTABLE TO EXAMINE 5 OVERSIGHT OF CREDIT RATING AGENCIES 6 7 Wednesday, April 15, 2009 8 9 Securities and Exchange Commission 10 100 F Street, N.E. 11 Washington, D.C. 20549 12 13 14 15 16 17 18 19 20 21 22 23 24 Diversified Reporting Services, Inc. 25 (202)467-9200 2 1 C O N T E N T S 2 Page 3 Welcoming Remarks 3 4 Panel One: Current NRSRO Perspectives 6 5 Panel Two: Competition Issues 65 6 Panel Three: Users' Perspectives 114 7 Panel Four: Approaches to Improve Credit Rating 8 Agency Oversight 171 9 Concluding Remarks 230 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 3 1 P R O C E E D I N G S 2 10:07 a.m. 3 WELCOME BY SEC CHAIRMAN MARY L. SCHAPIRO 4 CHAIRMAN SCHAPIRO: Good morning everyone. Welcome 5 to the Securities and Exchange Commission's roundtable to 6 examine the oversight of the credit rating agencies. The 7 Commission is truly grateful that so many have agreed to 8 participate in today's meeting. We look forward to your 9 comments, insights, and recommendations. 10 As you all know, the SEC is the investor's 11 advocate, and since I arrived here just about 10 weeks ago, I 12 have focused singularly on how the SEC can best serve the 13 needs of investors. And clearly, the role of credit rating 14 agencies must be an area for our intense review as we think 15 about how to promote investor protection and market integrity 16 and restore confidence in our financial system. 17 In late 2006, the Credit Rating Agency Reform Act 18 gave the Commission the exclusive authority over rating 19 agency registration and qualifications. In the less than 20 three years since, the Commission has undertaken no fewer 21 than five rulemakings. Clearly the Commission has been very 22 active in its efforts to fully implement the authority 23 granted by Congress, all with an eye to providing the 24 investing public with more confidence in the rating system. 25 These rules, which are still relatively new, relate 4 1 to a myriad of topics ranging from registration and 2 recordkeeping to disclosure and managing conflicts of 3 interest. The last of these rules, some of which were 4 adopted this past December and some of which are still 5 pending, were significantly influenced by the findings of the 6 SEC's extensive 10 month examination of three major credit 7 rating agencies. 8 But as much as we have done, there is still more to 9 do. The status quo isn't good enough. Rating agency 10 performance in the area of mortgage-backed securities backed 11 by residential subprime loans and the collateralized debt 12 obligations linked to such securities has shaken investor 13 confidence to its core. 14 Our purpose today is to ask some very basic 15 questions. Should the Commission consider additional rules 16 to better align the raters' interest with those who rely on 17 the ratings, that is, principally, investors? State another 18 way, does one form of rating agency business model represent 19 a better way of managing conflicts of interest than another? 20 Is there a way to realign incentives so that rating agencies 21 view investors as the ultimate customer? 22 Do users of ratings, whether they are issuers or 23 investors, have all the information they need to make the 24 most informed decisions? For example, is there more 25 information about performance, expertise with regard to 5 1 certain types of securities products, or fees that would be 2 meaningful in restoring investor confidence or would provide 3 investors with the tools to discern the value of the rating? 4 Should we borrow a page from the research analyst conflicts 5 of interest settlement of several years ago and require a 6 mechanism that provides for the issuance of multiple ratings 7 for every security, including one generated independently? 8 Are there additional behaviors, for example, 9 concerning the way that agencies bid for work, that should be 10 examined or modified? Would increased competition in the 11 rating agency space benefit investors and how would we 12 achieve that? 13 Are some securities products so inherently 14 complicated or risky that ratings are at best meaningless, or 15 even worse, misleading? In other words, is everything really 16 ratable? 17 Should investors reexamine the way they look at 18 ratings to ensure that ratings represent the beginning of due 19 diligence and not the end? Should the government end its 20 reliance on ratings or limit that reliance? 21 We are so fortunate today to have with us a number 22 of panelists who will help us find answers to these and other 23 questions. The first panel will help us understand what the 24 rating agencies themselves are doing to address the problems. 25 The second will discuss competition issues both domestically 6 1 and from a global perspective. The third panel will provide 2 the views of different types of users, and the fourth will 3 primarily focus on identifying possible new approaches to 4 oversight. 5 I will now turn the meeting over to Dan Gallagher, 6 Deputy Director of the Division of Trading and Markets, who 7 will introduce and moderate our first panel. Dan? 8 PANEL ONE: CURRENT NRSRO PERSPECTIVES 9 MR. GALLAGHER: Thank you, Chairman Schapiro. I 10 want to thank you and the Commission for your leadership in 11 this area. On behalf of the Division of Trading and Markets, 12 as you know, the staff has been very active since the passage 13 of the Credit Rating Agency Reform Act of 2006. It always 14 makes our life a lot easier when we get such great leadership 15 from the Commission, so thank you very much. 16 So I will be the moderator today. My instructions 17 from the Commission are to provide a format that allows free 18 flow of information between you and the Commission. That 19 said, you are in the SEC, and so we have some rules. Let me 20 tell you what they are. 21 I'm going introduce each of the panelists here in a 22 minute. After that, you will have four minutes, in the same 23 order that I introduce you, to provide your prepared remarks. 24 When you are done with your prepared remarks, I will start 25 the Q&A session. I will start off with one question, but I 7 1 assume shortly thereafter the Commissioners will have plenty 2 of questions that we will take into account and pass on to 3 the panelists. 4 The panel is scheduled to go until 11:25, so it 5 sounds like a lot of time, but it's not. We all know it will 6 fly by, so I ask that you respect the other panelists and the 7 Commissioners' questions and not take up too much time. I 8 will have here a yellow card that I will flash to you during 9 your prepared remarks when you have one minute left, and then 10 after that when you are just taking up too much time, so 11 please be mindful of that. 12 So let's get moving here with the introductions. 13 First on the panel is Mr. Daniel Curry, the President of 14 DBRS, Inc. Sitting next to him is Mr. Robert Dobilas, CEO 15 and President of Realpoint LLC. Next to Robert is Sean Egan, 16 President of Egan-Jones Ratings Company, and then Mr. Stephen 17 W. Joynt, CEO of Fitch Group, Inc. Then Raymond W. McDaniel, 18 Jr., Chairman and CEO of Moody's Corporation, and finally, 19 Mr. Deven Sharma, President of Standard & Poor's. 20 So now I will turn it over to you, Mr. Curry, if 21 you could begin. Again, four minutes for prepared comments. 22 I will show you the yellow card when you have one minute 23 left. 24 Thank you very much. 25 MR. CURRY: Thank you. Good morning Chairman 8 1 Schapiro and members of the Commission. We appreciate the 2 opportunity to participate in this discussion regarding 3 oversight of credit rating agencies. 4 A great deal has happened in this area over the 5 last two years. Not only did the Commission create a new 6 regulatory regime for rating agencies, it also adopted a 7 second set of rules that took effect just last week. It 8 already has a third set of rules out for comment. In view of 9 all this activity, DBRS believes that it is important to take 10 stock of what has been accomplished to see where we are now 11 and to identify what if anything is left to do. 12 In answering the question what went wrong with the 13 industry, we think it is important to distinguish between 14 ratings in the corporate area, on the one hand, and ratings 15 in the structured finance area on the other. There is no 16 evidence of a systemic problem in the corporate ratings 17 market. On the contrary, with very few exceptions, these 18 ratings have performed well over time. 19 Unfortunately, this has not been the case with 20 structured finance ratings. While the underlying causes of 21 the problems in this area will be studied for years, one 22 thing is clear: market concentration in the credit rating 23 industry exacerbated the turmoil in the structured finance 24 market because changes in the assumptions underlying the 25 rating models of the largest agencies lead to rapid and 9 1 dramatic downgrades over a short period of time. We believe 2 that increasing competition in the credit rating agency 3 industry will minimize the risk of this type of problem in 4 the future. 5 The need to enhance the quality and transparency of 6 credit ratings has not gone unnoticed by the rating agencies. 7 Over the past few years, the NRSROs have worked with the 8 Commission and other standards setting bodies like IOSCO to 9 identify and implement a host of reforms. At DBRS, we have 10 adopted a range of measures to eliminate or manage conflicts 11 of interest. These include the adoption of a business code 12 of conduct in accordance with the IOSCO code. 13 As for transparency, DBRS makes its ratings 14 available to the public free of charge, and our public 15 website also discloses the firm's ratings policies and 16 methodologies, as well as extensive information about how our 17 ratings have performed over time. We will also be making 18 additional information about our ratings history available to 19 allow investors to compare our ratings to those of our 20 competitors. 21 Because the NRSRO rules are so new, DBRS believes 22 that the best course is to monitor their effectiveness before 23 pursuing regulatory measures. In particular, we urge the 24 Commission to examine the impact of these rules in three 25 interrelated areas: conflicts of interest, competition, and 10 1 transparency. We believe that conflicts of interest exist in 2 all business models. We also believe that there is no 3 substitute for rating agencies maintaining high ethical 4 standards and being transparent about their methodologies and 5 results. For this reason, we endorse the Commission's 6 current approach of eliminating conflicts wherever possible 7 and disclosing and managing the rest. 8 However, it would be a mistake for the Commission 9 to start hiring NRSROs. In addition to creating a moral 10 hazard, such an approach would present significant practical 11 problems. Depending on the selection methodology used, the 12 Commission could wind up picking unqualified agencies, 13 perpetuate the existing market concentration, and could 14 discourage new investment in the industry. 15 On the question of competition, DBRS believes that 16 an open, competitive rating agency is critical to the 17 stability and long term efficiency of the capital markets. 18 Despite its best intention, regulation can hurt competition 19 if it is not evenly applied or if it is so costly that it 20 drives smaller firms from the market. In this regard, DBRS 21 urges the Commission to apply its rules uniformly to both 22 issuer and subscriber pay models. Meaningful comparisons of 23 ratings quality cannot be made unless investors have access 24 to ratings performance histories of all agencies. 25 MR. GALLAGHER: Thank you Mr. Curry. Mr. Dobilas? 11 1 MR. DOBILAS: Chairman Schapiro and Commissioners, 2 thank you for the opportunity to participate in the rating 3 agency roundtable. 4 Realpoint is the most recent company to be 5 designated as an NRSRO. We were authorized in June 2008 for 6 asset-backed securities. Within that area, our specialty has 7 been commercial mortgage-backed securities, or CMBS. 8 Please allow me to begin by commending the 9 Commission on its implementation of the Credit Rating Agency 10 Reform Act of 2006. During our application process, the 11 Division of Trading and Markets was readily available for 12 meetings and calls. We also found the ongoing proposals to 13 amend and improve the regulation of NRSROs to be 14 user-friendly. 15 The CMBS market has become an essential part of the 16 overall lending environment for commercial real estate. The 17 CMBS market now stands at approximately $1 trillion. Market 18 growth peaked in 2007 with approximately $230 billion in 19 securitized financings and then fell to almost zero in 2008. 20 In 2009, there has yet to be a significant transaction 21 brought to market. We share the hope that the 22 administration's TALF program will prove to be a catalyst for 23 a revival of CMBS and the rest of the securitization markets. 24 At Realpoint, we provide subscription-based ratings 25 to over 200 CMBS investors and fund managers. Our 12 1 traditional focus has been the ongoing performance of the 2 underlying real estate collateral. Our principle work 3 product is in the form of monthly rating reports on CMBS 4 transactions, credit events, and industry trends. We about 5 50 employees focused on providing our clients with the 6 insight and transparency they need to make sound investment 7 decisions. 8 Because we are conducting thorough monthly 9 analysis, we typically have very stable ratings and are ahead 10 of any major downgrades by issuer-paid NRSROs by 6 to 12 11 months. For example, Realpoint downgraded 537 CMBS tranches 12 over the past six months. The same CMBS tranches were 13 subsequently downgraded by one or more other NRSROs that rate 14 CMBS. 15 These recent downgrades reflect rising commercial 16 mortgage delinquency rates at a time where there are an 17 estimated $154 billion of securitized commercial mortgages 18 coming due between now and 2012. In order to effectuate 19 successful re-financings, investors will insist that CMBS 20 carry credit ratings that are timely and accurate from new 21 issue throughout the life cycle of the transaction. 22 In our view, the largest impediment to the 23 integrity of the current process for structured finance is 24 rating shopping. This means that an issuer or arranger of 25 credit starts the process of selecting its issuer-paid rating 13 1 agencies by providing advanced information to certain NRSROs. 2 These NRSROs then provide preliminary feedback regarding 3 subordination levels and the supposed tranches for the 4 securities backed by the pool of mortgages. This process 5 allows the issuer or arranger to select the NRSROs that 6 provide the most favorable preliminary feedback to be the 7 rating agencies for the new issued transaction. Only these 8 NRSROs are provided with the remainder of the information 9 necessary to develop the final ratings on tranches. 10 As can be seen, rating shopping fosters lack of 11 independence, accountability, and transparency. The 12 Commission has already begun to address this problem. The 13 Commission had proposed requiring issuers to simultaneously 14 disclose to all NRSROs the information that the issuer 15 provides to its solicited NRSROs and also providing 16 unsolicited subscriber-based NRSROs to deliver pre-sale 17 reports to their subscribers. Realpoint strongly supports 18 these initiatives. 19 In addition to these two proposals, the Commission 20 may also want to consider requiring disclosure of the 21 preliminary levels provided by issuer-paid NRSROs that 22 participated in the bidding process. This disclosure would 23 make this highly controversial process more transparent and 24 accessible to investors and regulators. In conjunction, 25 these proposals will increase competition, as well as 14 1 accountability and transparency in the credit rating process. 2 For surveillance purposes, Realpoint suggests that 3 the Commission likewise require that any material information 4 provided by a trustee or servicer to an issuer-hired NRSRO be 5 simultaneously disclosed to the other NRSROs. Providing the 6 unsolicited NRSRO with all the same property level 7 information, such as appraisals and rent rolls, that is given 8 to the issuer-paid NRSROs will level the playing field and 9 force NRSROs to compete for business on the basis of 10 providing timely and accurate ratings. 11 Implementation of these proposals in tandem with 12 the TALF program will result in greater transparency and 13 informational disclosures in the credit rating process, which 14 will improve the credit rating process for the benefit of 15 capital markets and facilitate the return of securitization 16 as a viable means to liquidity investment risk 17 diversification in the CMBS markets. 18 Thank you for your time. 19 MR. GALLAGHER: Thank you Mr. Dobilas. Mr. Egan? 20 MR. EGAN: Thank you. We are fortunate to have a 21 host of industry representatives and numerous academic and 22 other experts to provide affirmative replies to the question 23 the SEC has presented: what went wrong and how to improve the 24 credit rating agency oversight. 25 This morning, however, I want to begin by stressing 15 1 some things that were not the case. First, the much 2 publicized failures of the industry's largest companies to 3 provide timely and accurate ratings are not a new 4 development. Second, these problems are not limited to 5 structured finance. And third, the industry problems are not 6 the result of a lack of competition. 7 Enron brought much attention to the rating industry 8 when that company failed in 2002 despite the fact that S&P 9 and Moody's had its debt at investment grade as late as four 10 days before the bankruptcy filing. The fact is, however, 11 that the work product of the major rating agencies was 12 equally dismal in numerous other instances, including Orange 13 County, California, Pacific Gas and Electric, Worldcom, 14 Delphi, General Motors, Ford, Collins & Aikman, MBIA, Ambac, 15 IndyMac, Countrywide, and many others. 16 Well before the current crisis, as noted, for 17 example, by professor Jonathan Macey of Yale Law School in 18 Congressional testimony leading up to the enactment of the 19 2006 reform legislation, there was a plethora of academic 20 studies showing that credit rating changes lag the market. 21 Second, today's problems were not just in 22 structured finance, but also in the unsecured bonds and other 23 plain vanilla debt offerings of many corporate entities, 24 including Fannie Mae, Freddie Mac, Countrywide, and many 25 others, Bear Stearns, Lehman Brothers, AIG, Washington 16 1 Mutual, the Reserve Fund, et cetera. Deliminating this issue 2 to structured finance would likewise ignore the more recent 3 situation confronting this so-called monoline or 4 bond-insurers such as Ambac, MBIA, ACA, and FGIC, which 5 carried triple A ratings up through and even during the time 6 when state insurance officials were actively pursuing 7 multi-billion dollar restructurings of these companies. 8 Third and most important to note is that 9 competition per se is not the answer. In fact, Chairman 10 Arthur Levitt was known to admonish that additional 11 competitors would produce rating inflation, and in fact, the 12 growth of Fitch as a viable competitor to S&P and Moody's has 13 actual produced less, rather than more accurate ratings. 14 Paine Webber said in a 1997 report that it was the 15 emergence of Fitch and Duff & Phelps in the early 1990s that 16 gave issuers the opportunity to play S&P and Moody's against 17 each other. They would shop each deal at both S&P and 18 Moody's, choose the agency that gave them the best execution, 19 namely the highest level credit quality, and then use that as 20 a swing rating with either Fitch or Duff & Phelps. 21 Fortune magazine recently identified Egan-Jones as 22 the leading firm to warn about the credit crisis and the Fed 23 has identified Egan-Jones as leading, by far, the big credit 24 rating agencies. For example, at a Congressional testimony 25 in 2003, I stated that Fannie Mae and Freddie Mac did not 17 1 merit triple A ratings, which Moody's, S&P, and Fitch quoted 2 them. At that same time, we issued a rating call to the same 3 effect with respect to MBIA, which our competitors were still 4 rating at triple A five years later. 5 How is it that the major rating agencies, which 6 have approximately 400 employees for every analyst at Egan 7 Jones, have been consistently wrong across a broad spectrum 8 of debt offerings over such an extended period of time? I 9 would like to say that we have more sophisticated computer 10 models or that our people are just more talented, and I hope 11 that some of that is true. 12 However, the real answer is that Egan-Jones is paid 13 by investors to be in the business of issuing timely, 14 accurate credit ratings, whereas S&P, Moody's, and Fitch have 15 gravitated to the business of being paid by the issuers of 16 securities to facilitate the sale of those securities. 17 Investors want credible ratings. Issuers, on the 18 other hand, want the highest rating possible since that 19 reduces their funding cost. Under the issuer-paid model, a 20 rating agency which does not come in with the highest rating 21 will before long be an unemployed ratings firm. If simple 22 and all the explanations and excuses cannot refute the market 23 evidence -- 24 MR. GALLAGHER: Thank you Mr. Egan, we are going to 25 have to move on to Mr. Joynt. 18 1 MR. JOYNT: Thank you. I have four minutes, so I 2 will keep it crisp. 3 Fitch has thoroughly reviewed our ratings policies, 4 procedures, and operations, and aligned our organization 5 around best practices as outlines by IOSCO, CESR, the SEC, 6 and other regulatory bodies we dialogue with. We have a 7 senior management team committed to continuous reassessment 8 of ourselves to make sure we are acting in fact 9 appropriately, not just in theory. 10 We are open to additional measures that 11 constructively support a better analytical product and 12 greater transparency. But if you ask what steps we are 13 taking in light of the current credit crisis, our main focus 14 is on vigorously reviewing our analysis, research, and 15 changing ratings to reflect the current risk profile of 16 securities we rate. In many cases, that is generating 17 significant numbers of downgrades of structured securities, 18 RMBS, CMBS, and CDOs. 19 We are releasing our newest analysis transparently 20 and publicly, and we are communicating directly with 21 investors the latest information and analysis we have. These 22 are sometimes difficult meetings as investors are unhappy 23 about their own decision making and the part that ratings 24 played in their decisions, but we must reassess the risk 25 today and move forward. 19 1 Fitch Ratings traditionally has assessed the 2 probability of first dollar of loss on a relative scale. 3 Many subprime and Alt-A mortgage securities are now 4 anticipated to have losses, so that ratings are the lowest 5 possible in the C category and not particularly useful to 6 investors. This broad, high-risk categorization does not 7 help distinguish any of the widely divergent recovery values 8 possible with these securities. For this reason, we have 9 realigned our analysis to identify the principle recovery 10 possibilities for these securities. 11 Our goal is to help market participants, banks, 12 insurance companies, or their accountants or regulators 13 better assess the loss potential or aide price discovery on 14 these complicated instruments. We have been continuously 15 offering help to the Fed and other regulators as they attempt 16 to rebuild credit market flows through TALF and TARP 17 programs. 18 We have also been reexamining the structure of our 19 financial institutions ratings. Issuer default and bond 20 ratings have long reflected expected government support, 21 while our individual ratings, B or CD, reflect company 22 performance. The BIS has recently suggested reworking 23 guidelines to use ratings that don't incorporate government 24 support to provide more focus, transparency, and clarity 25 about the bank's own credit ratings. We need to consider 20 1 changing our rating definitions and/or systems if 2 appropriate. 3 I would like to comment on the use of ratings in 4 regulation. Ratings have been used constructively in many 5 places in regulation. You will hear from some industry 6 participants that ratings are an important common benchmark. 7 In many cases, if you eliminate the use of NRSRO ratings and 8 regulation, company and industry participants will develop 9 their own guidelines and use credit ratings anyways. I 10 believe they will often default to the brand name rating 11 agencies, which is not a positive if one of your objectives 12 is increasing competition and thereby a better work product. 13 If you are serious about eliminating ratings and 14 regulation, I suggest you transition to elimination over an 15 intermediate term timeframe with careful consideration of 16 each regulation rather than wholesale elimination. 17 MR. GALLAGHER: Thank you very much Mr. Joynt. Mr. 18 McDaniel? 19 MR. McDANIEL: Good morning and thank you for the 20 opportunity to contribute Moody's views to the discussion 21 under way regarding the credit rating industry. My comments 22 today focus on events that have transpired over the past 23 several years in the financial markets, particularly 24 structured finance, what Moody's has done in response to 25 those events, and additional measures that we believe will be 21 1 helpful in restoring confidence. 2 Events of the past two years have exposed 3 vulnerabilities in multiple areas of the global financial 4 sector. Weaknesses include exceptional leverage, loss of 5 liquidity in periods of stress, rapid changes of asset 6 valuations and capital needs, insufficient risk management 7 practices, interlinked market participants, and limited 8 transparency. 9 With respect to Moody's, many have asked us why 10 mortgage-backed securities have demonstrated such poor 11 performance under stress and been subject to such widespread 12 downgrades. While there are numerous contributing factors, I 13 believe it was the interaction of three that was most 14 consequential. 15 First, magnitude and speed of home price 16 depreciation, second, the nationwide character of the 17 downturn in housing, which undermined credit protection 18 historically provided by geographic diversification in 19 mortgages, and third, the suddenness of transition to 20 restrictive lending in the housing market. We were certainly 21 not alone in our expectations, but we should be at the 22 leading edge for predictive opinions about future credit 23 risks, and we have learned important lessons from these 24 rapidly changing market conditions. 25 Moody's is well aware of the loss of confidence in 22 1 credit ratings over the past 18 months. To address this, we 2 have implemented and continue to implement a range of 3 measures to enhance the quality, independence, and 4 transparency of our ratings. These measures are listed in 5 our publication about strengthening analytic quality and 6 transparency, and fall into four principal areas. 7 First, strengthening our rating methodologies, for 8 example, by creating feasibility reviews for new structured 9 products. Next, enhancing consistency of cross-rating 10 groups, for example, by incorporating more rigorous and 11 systematic macro-risk scenarios. Third, bolstering measures 12 to avoid conflicts of interest, for example, by strengthening 13 the separation of our credit professionals from commercial 14 activities at the firm. And finally, improving transparency 15 of ratings and the rating process, for example, by publishing 16 volatility scores and loss sensitivity analysis on new 17 structured securities. 18 In addition, we have adapted our policy systems and 19 organization to implement the recently adopted NRSRO rules. 20 Nonetheless, we can always do more, and we are continuing to 21 communicate with market participants and public sector 22 authorities to better understand various concerns and 23 recommendations. For example, measures that discourage the 24 use of credit ratings as measures of liquidity or volatility 25 would be highly beneficial and help to restore market 23 1 discipline. 2 We also believe that one of the more significant 3 steps to be taken is to make more underlying information 4 about structured securities available to investors. In our 5 view, this would yield at least three important benefits. 6 First, giving investors access to more information will 7 reduce the risk of over-reliance on credit ratings. Second, 8 making more information available through offering documents 9 is likely to improve the quality of that information. And 10 third, making more information publicly available will 11 broaden the range of opinions and analysis, including from 12 those credit rating agencies not to select -- not to rate a 13 securitization. 14 In conclusion, we believe that all market 15 participants now need to consider how to improve the U.S. 16 mortgage origination and securitization process, and for our 17 part, we have made changes that range from revised analysis 18 to greater process transparency, to more independent rating 19 system management and governance. 20 Thank you. 21 MR. GALLAGHER: Thank you Mr. McDaniel. Mr. 22 Sharma? 23 MR. SHARMA: Good morning. Thank you for the 24 opportunity to participate in today's roundtable. 25 Standard & Poor's, we can assure you, is doing is 24 1 its part to restore confidence in ratings and have enacted a 2 number of important measures to enhance our analytics, 3 transparency, and governance to prevent conflicts of 4 interest. These steps, along with continuing effort to 5 educate the markets about ratings should go a long way in 6 promoting accountability and restoring confidence. 7 We believe private and regulatory action should be 8 guided by a core set of principles to better serve investors 9 and the markets. The first is transparency. Transparency 10 extends to both the way a rating firm's policies and 11 procedures operate and to the analysis underlying particular 12 ratings. Transparency and market scrutiny serve to promote 13 investor confidence and ratings quality. The second 14 principle is accountability. Accountability provides comfort 15 to investors that the ratings process has integrity and that 16 potential conflicts of interest are effectively managed. 17 Currently accountability exists first in the form 18 of regulatory oversight, including the ability of the SEC to 19 sanction and fine NRSROs, second in market scrutiny, third 20 through internal controls, and fourth, in certain 21 circumstances, private litigation. It is important that any 22 accountability framework takes into consideration the nature 23 of ratings and the assessment of credit risk. Specifically, 24 some portion of trillions of dollars of debt we rate will 25 inevitably default, including some highly rated debt. 25 1 Indeed, that is the essence of ratings. 2 The third principle is critical, that the rating 3 firms engage and give -- and listen to all users, and in our 4 case, especially investors. We have been very active in this 5 area. Our efforts have resulted in a number of constructive 6 measures including, for example, adoption of criteria that 7 explicitly addresses rating stability, an important factor 8 for many investors, particularly given recent market events. 9 As noted, we believe regulation has played and 10 should continue to play an important role in furthering these 11 principles. Regulation is most beneficial when geared 12 towards specific goals such as transparency and increased 13 quality, and when striving for end to end solutions. That 14 is, an appropriate regulatory framework for ratings should go 15 on not just rating firms but all entities that can play a 16 role in impacting the quality and use of ratings. 17 International regulatory consistency is also 18 critical given the global nature of the capital markets. The 19 G20's recent comments and the need for international 20 consistency and the modeled code of conduct published by 21 IOSCO are constructive. 22 Lastly, a word about different rating firms' 23 business models. Each business model, whether it be the 24 issuer fee model or the subscriber fee model or the utility 25 model has its own strengths and weaknesses with respect to 26 1 what market participants are looking for from ratings, 2 including first, transparency, second, quality, third, 3 on-going market scrutiny, and fourth, broad market coverage. 4 Each of these models presents the potential for conflicts of 5 interest which can be addressed by new regulation promoting 6 transparency in rating models, methodologies, and 7 performance. 8 We believe the key in this area is not to mandate 9 one business model over the others, but rather for regulators 10 to provide oversight of all rating business models and allow 11 market participants to chose the firms that best serve the 12 particular needs. The market is large and diverse, with 13 different investors having distinct needs, and investor 14 choice and rating firm competition can foster analytic 15 quality when joined with appropriate regulatory oversight and 16 accountability. 17 To conclude, we are committed to restoring 18 confidence in our ratings. Standard & Poor's has taken and 19 will continue to take steps to enhance our transparency, 20 prevention of conflicts, analytics, and accountability. 21 Perhaps most importantly, we are working to improve our 22 responsiveness to the needs of the investors. We will 23 continue to seek ways to bring greater voice -- investor 24 voice and accountability into the ratings process. 25 Thank you again for taking the opportunity to 27 1 participate in this roundtable. 2 MR. GALLAGHER: Thank you Mr. Sharma and thank you 3 to all the panelists for keeping your remarks generally 4 within the time limit, so that is terrific. 5 We will now move on to the Q&A section of the 6 panel. At any time, I would ask the Commission to indicate 7 if they have any questions for us -- the panelists too, 8 although I will read out here the first question, and we will 9 proceed I think in the order that you gave your formal 10 comments. But to the Commission, please feel free to 11 indicate that you have a question at any time and we will 12 jump in. 13 The first question I have for the panel is, here 14 they are, you have the Commission sitting right here. If you 15 could ask them for one thing, or conversely if you could ask 16 them to not do one thing, what would that one thing be? 17 MR. CURRY: As a small firm, the one thing that I 18 would ask is that you look at the overall expanse of the 19 total regulatory regime and the effect that that has on 20 smaller firms and our ability to participate in the market. 21 MR. DOBILAS: Well I would definitely endorse that 22 statement. But most, I think, dear to the securitization 23 market is don't over-regulate, and keep in mind data 24 protection. The last thing we would want to see is -- a lot 25 of the securitization information is private data, and 28 1 opening up all that data to the public or data vendors could 2 really undermine the general principles of the securitization 3 market. 4 MR. EGAN: I'm going to have one sentence, but it 5 is ridiculous, but it is going to have three things in it, 6 okay? 7 MR. GALLAGHER: Fair enough. 8 MR. EGAN: You can deal with it, believe me. 9 One is that the SEC should not withdraw from the 10 rating agency review. They are out there, they are needed, 11 they are critical. 12 Two is that if you restrict freedom of speech, you 13 better be darn careful because you are going to cause 14 problems for the smaller rating firms that have done a good 15 job. We are threatened with suit on a weekly basis by big 16 issuers that don't like the ratings that we have out there. 17 And three is that the government is not capable of 18 assigning ratings. We have proof of that with Chairman Cox 19 unfortunately three days before Bear Stearns' bankruptcy 20 standing up on CNBC saying the broker/dealers have plenty of 21 capital. 22 MR. JOYNT: I think there is real value in the 23 NRSRO system and the fact that there is increased oversight 24 and responsibility being brought to the system, I think, is 25 constructive and healthy, so I would hate to see a radical 29 1 change in that under -- in a time of duress and stress, 2 although I accept the fact that change over time, 3 thoughtfully, especially with changes in individual 4 regulations, I think that is constructive. 5 MR. McDANIEL: As I mentioned in my opening 6 remarks, Moody's believes that making sufficient information 7 on both assets and transactions structures in the 8 securitization market available to institutional investors, 9 to the people who are going to be holding these instruments, 10 is absolutely critical to a restoration of health in the 11 securitization market. It also will allow rating agencies 12 that are not selected to rate securities, to offer their 13 opinions, and will avoid the problem that we foresee with the 14 current proposal, which is that the rating shopping which has 15 plagued the industry would be transformed into information 16 shopping. 17 If only NRSROs have the information available, 18 there will be a flight to the bottom for which NRSRO requires 19 the least amount of information, which is then made available 20 to the others. I think it is much better if that is made 21 available to the investing public as part of the offering 22 document process. 23 MR. SHARMA: We would suggest the SEC should 24 continue to foster competition and accountability, and part 25 of that means sort of letting different business models 30 1 thrive and let the market dictate and choose which business 2 models work for them the best. In terms of what not to do, 3 we do need more regulation and rules, and I think that will 4 bring a lot of confidence. But we also have to watch for 5 unintended consequences, and in that context, letting 6 analytical independence remain is a very important request 7 from our side not to do. 8 MR. GALLAGHER: Chairman Schapiro? 9 CHAIRMAN SCHAPIRO: Thank you. I actually have 10 several questions, but let me just start with one so my 11 colleagues have plenty of opportunity as well. 12 Clearly the government and not just the SEC by any 13 means has referenced ratings throughout our regulatory 14 framework, and I have to believe that has helped NRSRO's 15 business grow and flourish over the last number of years. 16 But is it the right thing to do? Should the SEC and other 17 agencies of the government continue to mandate in law and 18 regulation reliance on ratings? 19 MR. GALLAGHER: Why don't we just go in the same 20 order. Mr. Curry? 21 MR. CURRY: I would say that to have less reliance 22 on ratings is probably good. I just would suggest that it is 23 important to do that gradually, because I think you have to 24 allow market participants time to adjust to ratings being 25 withdrawn. 31 1 MR. DOBILAS: You know, I'm in favor of having 2 reference to ratings but not limiting those references to 3 three rating agencies and opening up to the NRSROs in 4 general. I think the market needs a minimum standard or 5 quality to at least reference. I don't think you can leave 6 it in the hands of fund managers to grow these themselves. 7 Broker/dealers tried it with broker/dealer research. 8 I think the rating agencies do offer, when properly 9 regulated and if the proper transparency is in place -- offer 10 a minimum level of analysis that the public should come to 11 depend on. Thank you. 12 MR. EGAN: It makes perfect sense that the 13 government and other agencies rely on ratings. The 14 underlying problem, however, is under the current industry 15 structure, the majority of the firms are not paid to protect 16 investors, they are paid to issue the highest rating possible 17 to facilitate issuance of securities. So solve the 18 underlying problem and then you don't have to worry about any 19 reliance because the firm that doesn't issue the timely, 20 accurate ratings will be punished. 21 Under the current system, somebody says "Let the 22 market work out." Well the market is the issuers. They are 23 90 percent of the compensation in this area. The issuers 24 want the highest rating. If the SEC is in the business of 25 protecting investors, then they have to make sure that their 32 1 agents are also protecting investors. We don't have that 2 case right now. 3 MR. JOYNT: I think I would separate your question 4 because I don't believe that the rating agencies have only 5 made money because they are NRSROs. I believe there is real 6 value in the common benchmark that ratings represent, and 7 even if you were to take them out of NRSRO regulation, I 8 believe many investors would use ratings as a common 9 benchmark in any case. So for that part of your question, I 10 don't see the linkage. 11 I see the value, therefore, in having NRSROs and 12 ratings used by regulators and others as benchmarks, so I 13 think they are constructive. They shouldn't be the only 14 benchmark. I'm sure it's not the only thing bank regulators 15 rely on. They have their own benchmarking systems and 16 scoring, insurance company regulators, and others as well. 17 But many non-regulated entities use ratings also as 18 benchmarks, nothing to do with NRSRO designation. 19 MR. McDANIEL: Moody's has been consistent in 20 supporting the reduction or elimination of the use in ratings 21 in regulation. I think that it is most important for us to 22 serve a market-based need. Our aim should be to produce the 23 highest possible quality ratings from a market-user 24 perspective as opposed to for regulatory purposes. And the 25 use of ratings in regulation I think ultimately confuses the 33 1 aim of the rating agencies in terms of the constituencies 2 they should be trying to serve. It should be principally the 3 investing public, and the use of ratings, especially the 4 widespread use, risks altering that perspective. 5 MR. SHARMA: Chairman, we also do not advocate 6 ratings being used in the regulation. And in Europe, we 7 built a business when there was no regulation requiring 8 ratings. As you examine this question, we do suggest that 9 the questioner will also look at what the role and usefulness 10 is of common benchmarks that investors use in the decision 11 making process. But as long as the benchmarks remain in the 12 regulation, ratings remains as a benchmark, we do encourage 13 and suggest perhaps other benchmarks should also be included 14 to avoid a singular focus on ratings. 15 MR. GALLAGHER: Commissioner Walter. 16 MS. WALTER: Thank you Dan. Back in the early 17 1980s, we made a decision not to require rating agencies, the 18 NRSROs, to give consent to appear in registration statement 19 and be subject to the liabilities of other experts are 20 subject to under the 1933 act. And in doing that, we reached 21 that balance that because we were concerned about the 22 practical problems that people had raised, and particularly 23 the possible interference with the rating process and the 24 difficulty in obtaining consent, and we were satisfied that 25 the liability structure under the anti-fraud provisions would 34 1 suffice. 2 It seems to me that as a result of that and other 3 things, we really have helped to foster the use of ratings in 4 disclosure documents, but I wonder whether you think it is 5 time to perhaps not change the answer, but at least reexamine 6 that decision in light of recent events. 7 MR. CURRY: I think any change that would subject 8 rating agencies to additional liability for opinions runs the 9 risk of removing ratings from public use and it becomes more 10 of a private contract between parties. And speaking as a 11 small rating agency, I think that makes it very difficult for 12 firms of our scale to operate in that kind of a business 13 model. I think for the larger agencies that could be managed 14 through legal agreements and ratings shared privately. I 15 also wonder if ratings are not freely available to the public 16 how that would effect secondary market trading, particularly 17 for structure financed securities. 18 MS. WALTER: Can I follow up with a question about 19 that? I guess my question is given the way the state of play 20 is today, I think the marketplace and the public marketplace, 21 it would be my guess -- and I'm not an expert, so take this 22 as a personal guess -- would demand that ratings be publicly 23 available, and so how does that fit in to the scenario? 24 Because I think we are in a very different place today than 25 we were in the early 80s when some of these decisions were 35 1 made. 2 MR. CURRY: It is an interesting question. I just 3 haven't given it a lot of thought. It makes me question the 4 viability of that business model if the ratings have to be 5 made publicly available, but anyone who disagrees with that 6 opinion can file a lawsuit against the agency. Even if we 7 ultimately prevail in those court cases, particularly for 8 smaller firms, I think just the burden of managing the legal 9 process would be pretty dramatic. 10 MR. DOBILAS: Yes, I would agree. I think that is 11 very important. The subscription-based model is a little 12 different than the new issue model. You know, having our 13 ratings going public would definitely go against our business 14 model. 15 But I think if you take a look at the process and 16 you break it up into two pieces where you have new issue and 17 you have ongoing surveillance and you really identify what 18 went wrong in those processes, I think you will see that, as 19 Sean indicated, the rating shopping component of that plays a 20 large role in that. With regards to surveillance, we 21 wouldn't be in existence today if the rating agencies did a 22 good job in the surveillance market, the secondary market. 23 Subscription-based agencies like Realpoint thrive 24 at customer satisfaction, and clients demand a certain level 25 of service, and they are willing to pay for that service in 36 1 the secondary market side. I think on the primary market 2 side, the new issue side, they are not willing to pay for 3 that service. I think the subscription-based model will be 4 very hard to implement in those markets. I think the 5 issuer-paid model is more likely to be prevalent in that 6 market. 7 And I think we have to look at the process and fix 8 it, and I think in making that process of preliminary ratings 9 transparent to the investors and let them make their decision 10 on which rating agencies they would have on that deal, I 11 think the investors need to speak up now more than ever and 12 weigh in. So, thank you. 13 MR. EGAN: My view is that you have to be careful 14 of the unintended consequences. If you increase the 15 potential liability for the non -- for all rating firms, the 16 likely result will be an increase in cost, but the bigger 17 firms will probably get an indemnification from the issuer, 18 so they don't have to worry about it, or they will increase 19 their rating fees from the issuers. It doesn't really 20 matter, the same result is the case. And so you restrict the 21 available information in the market, so I think you want to 22 be very careful about that. 23 Secondly, as far as demanding information from 24 rating firms be publicly available, the further you push 25 towards a complete availability of all the information for 37 1 all rating firms, the more you imperil the existence of the 2 investor-paid rating firms, so you want to be careful about 3 that too. 4 MR. JOYNT: I don't recall the pros and cons of the 5 decision that was made in 1980, the specific question you had 6 about prospectus requirements, although at the time I seem to 7 remember thinking about it. So I can't be specific on that 8 particular question. 9 In general, I think I agree that the competitive 10 aspects of making a change towards liability may be more 11 impactful on smaller firms than the larger firms, so one 12 should be cautious about that. I feel differently about the 13 disclosure of ratings, where I think you are going to have a 14 system that recognizes NRSROs, then probably they should be 15 treated all the same, including the disclosure of ratings. 16 And I'm not as convinced that that has as much business 17 implications for investor-paid rating agencies versus 18 issuer-paid rating agencies. 19 MR. McDANIEL: Just, I guess, as a reminder to all 20 of us, we do currently have liability, certainly for actions 21 that would constitute securities fraud or issuing opinions 22 that are not truly our own or reckless actions. So the 23 question is whether there should be additional accountability 24 in the form of liability, and I think it is important to 25 separate the concepts of accountability from liability 38 1 because it may be very desirable to enhance accountability 2 and yet we may find it very undesirable or risk the 3 unintended consequences of a different standard of liability. 4 Ratings are forward-looking statements. I know 5 there is interest in describing them as something other than 6 that, but that is what they are. They are predictions about 7 an uncertain future, and even if they are done to a very, 8 very high quality standard, they are going to be unpopular. 9 It is the nature of the business. 10 And just as a final question in the area of 11 unintended consequences, I think we would at least have to 12 ask whether a different liability standard would encourage 13 more conservative rating actions or more liberal rating 14 actions. Certainly our litigation history has primarily been 15 on the side of issuers suing Moody's for having ratings that 16 are assigned that they believe are too low. So I would put 17 that forward as at least a question concerning the liability 18 status. 19 MR. SHARMA: We look at your question from the lens 20 of accountability also, and there are two aspects to it. 21 One, we are held accountable by SEC, which has the right to 22 fine and put penalties. There are other measure to be taken 23 to further make that more stringent and we think that should 24 be considered. On the private right of action, we do have a 25 number of lawsuits that we are at the moment defending, so 39 1 investors do have the option to bring private lawsuits 2 against us at this stage. 3 In terms of disclosure, again, that is another 4 element of accountability, which sort of allows a comparison 5 of performance across all the rating firms, irrespective of 6 the business model, and to hold us accountable to what we 7 have said we will do in the marketplace. 8 MR. GALLAGHER: Commissioner Casey, questions? 9 MS. CASEY: Thank you very much. I know there has 10 been a great deal of focus on addressing the conflicts in the 11 investor paid model. And I'm curious as to what conflicts 12 are particular to the subscriber-based model that the 13 Commission should be concerned with, if it all. 14 And then secondarily, if you could speak a little 15 bit about -- actually, Sean, I wanted to hear a little bit 16 from you -- you talked about the limitations of government in 17 being able to assign ratings. And then you also note in your 18 statement the fact that you think it would be a benefit to 19 addressing some of the conflicts in the issuer-paid model 20 enhanced disclosure around who pays for those, and I would 21 like to hear your thoughts on that. 22 So more generally on the different conflicts that 23 exist in the two different models and the issues that the 24 Commission should be concerned with in addition to what we 25 have already done, and then specifically, Sean, if you could 40 1 talk about your comments. 2 MR. GALLAGHER: Why don't we start with Sean, mix 3 it up here a little bit. Mr. Curry has been on the hot seat, 4 so we will shift over a little bit, if you don't mind, Sean. 5 MR. EGAN: No problem. I'm glad you brought that 6 up because it is in the written testimony of various people. 7 Some people say that there are conflicts on the issuer-paid 8 and conflicts on the investor-paid. That is a red herring, 9 and it doesn't hold up to scrutiny. 10 The reality is that if you are an investor-paid 11 rating firm, you don't know what your clients' positions are. 12 You don't know if they are long certain issues, you don't 13 know if they are short. We don't get -- we don't ask for 14 that information and we don't get it. And even if some 15 investors gave it to us, it doesn't help us for the 500 other 16 subscribers that are out there, so it does us no good. And 17 even in the odd case that we knew exactly what their current 18 positions are and what they will be in another week, which we 19 will never get anyway, we don't know if they are -- it 20 doesn't help us because we are paid to give them information 21 on where the credit quality is likely to be. 22 So if all our clients were bullish on MBIA, we 23 would be a terrible service to them by supporting that 24 bullishness. The better result is to say "No, there are 25 problems with this issuer" -- and I'm not trying to pick on 41 1 MBIA -- "There is problems with this issuer, that issuer, for 2 this reason, that reason, and the other reason." Then the 3 investors can take their time, find out if our analysis is 4 accurate, and make adjustments. 5 However, if we were paid by the issuers, it is very 6 clear what the issuers want. They want the highest rating 7 possible so it can facilitate the issuance. So it is 8 misleading, it is disruptive, it is not productive to say 9 there are conflicts on both sides. The reality is that there 10 are not if you are doing your job properly. 11 Another item for investor-paid ratings firms is -- 12 what was it -- the distribution of information. We did not 13 support the carve out for firm Regulation FD for the rating 14 firms. The thought is that why in the world should a rating 15 firm be put on a pedestal vis-a-vis all the other 16 participants in the market. So it is like the head of equity 17 research at Goldman Sachs going to the SEC and saying "Give 18 us the information for first. They are for the issuers. 19 Give us the information first because we will come up with a 20 really good report." Hopefully the SEC would say "Forget it, 21 it is not fair to everybody else," and that is the right 22 answer. 23 And unfortunately, my partner Bruce Jones -- and 24 co-founder, he was with Moody's for a number of years -- he 25 insisted and other analysts insisted on getting advanced 42 1 information from the issuers and they would go ahead and 2 punish the issuer if they didn't get that advanced 3 information. That should be prohibited. It should be out 4 there to everybody in the market at the same time, not only 5 on corporate matters, corporate issuers, but all the other 6 issuers in the market. 7 In the structured finance -- fortunately, Annette 8 Nazareth supported my position in this -- in structured 9 finance, there should be a 10K, 10Q type system. Get the 10 information out there so that people can do their own work. 11 And also, prevent -- for a firm to do a good job in 12 structured finance on the residential mortgage side, you have 13 to spend about a million and a half dollars and data just to 14 get the ratings out the door. It's ridiculous. It should be 15 just like a 10K and 10Q system so you have the information so 16 people can make the judgments themselves. 17 MR. GALLAGHER: I move on to Mr. Joynt. 18 MR. JOYNT: So I guess I don't agree with Mr. Egan 19 here. He seems to posit that the investor-pay model has 20 absolutely no conflicts and the other rating agencies paid by 21 issuers are riddled with conflicts. 22 MR. EGAN: No. 23 MR. JOYNT: I don't feel that way. Fitch, we spend 24 the vast majority of our time meeting with investors. We 25 have many subscribers to our research. We send teams of 43 1 analysts out to talk about the companies we rate and the 2 securities we rate. We hold seminars, and we have as much 3 and probably far more investor contact that Mr. Egan has with 4 his investors. 5 So it is true we are paid by issuers so we need to 6 manage any potential conflicts that are there. We feel like 7 we have done that with great integrity. And so I see the 8 potential for a conflict that needs to be managed and 9 probably needs to be managed in both places. I don't have a 10 lot of comments about the investor part of that. 11 I do think that the point that he has made about 12 getting information out into the market, and including on 13 securitizations, and it was made earlier as well, is an 14 important area for research. So as pointed our earlier, it 15 needs to be done carefully and respectfully for issuers that 16 have private information. But it seems to me there ought to 17 be a way, and it has been pointed out in many places, that a 18 lot of the information is available to many parties in the 19 market, at least the critical information on portfolio, 20 collateral, and deal structure. So it seems to me that that 21 information could be made more openly available and then many 22 people could use that for analysis. 23 MR. GALLAGHER: Why don't we go to Mr. Dobilas next 24 on that? 25 MR. DOBILAS: I think subscriber versus issuer-paid 44 1 both have conflicts. I think both have merits as well. I 2 think it is the process that needs to be looked at. 3 I think for subscriber paid, a lot of conflicts are 4 mitigated by the fact that if we don't do a good job, we are 5 just going to lose subscribers. We have cancellation clauses 6 in our contracts, so we are forced to do a better job every 7 year and enhance the process. 8 On the issuer-paid side, we recently are 9 participating in looking at a new issue deal in the market. 10 Investors have come to us and said "We want to see the 11 analysis you put out on the surveillance side in the new 12 issue market," and we believe that providing openness to our 13 models and transparency in the process, we could make that 14 transition to an issuer-paid new issue ratings model from a 15 subscription-based surveillance model. So that is why I say 16 if you look at the process, the new issue process needs to be 17 more transparent. It needs to be opened up to investors. 18 As far as data and complexity, I mean structured 19 finance is a very time-sensitive analytical process. When we 20 are looking at a bunch of commercial mortgages, we have to 21 get rent rolls, we have to get financial statements, we have 22 to dive into the details. It is very different from the 23 corporate world, it is very different from the residential 24 model-driven world as well. Data is very sensitive, and why 25 I say just don't open up the doors and let this data out to 45 1 everybody out there is, you know, I'm sure people borrowing 2 money in the condo markets do not want to see their rent 3 roles being laid out to the public so that competition can 4 see what other people are paying in those buildings. 5 So some semblance of control needs to be 6 implemented, and I think that is what the NRSROs are. We 7 could be a gatekeeper sort of ensuring that private data 8 remains private and analyzed properly and put out there for 9 investors to look at. And investors have the right to that 10 data too. What worries me is when data companies get hold of 11 this data, what can that do to the securitization market? 12 And I think that needs to be safeguarded. 13 But I think there are conflicts on both sides. It 14 is the process that needs to be addressed. Thank you. 15 MR. GALLAGHER: Mr. McDaniel, do you want to join 16 in? 17 MR. McDANIEL: Sure. There is some information in 18 our written submission that I just briefly highlight here in 19 looking at the subscriber pay versus issuer pay models. And 20 that is that because of our long history and the fact that in 21 the first 50 years of our history we did have an investor 22 pays or subscriber pays model, we were able to compare the 23 ratings for the 1920 to 1970 period with our ratings from the 24 1971 through 2008 period, which is the issuer pays era of our 25 history. And the ratings performance was superior, and in 46 1 many cases substantially superior during the issuer pays era. 2 This is a comparison of corporate ratings. We also 3 included data on structured finance, but the comparison 4 cannot be made back to the 1920 to 1970 period because it 5 didn't exist. And the research that went into this from our 6 credit policy group requires some in-depth discussion and 7 understanding, and I know we don't have time to offer that 8 here in this meeting. But I would encourage a review of the 9 statistical measures that we were able to include in our 10 written submission as well as the underlying methodology, 11 which can be tested by others as far as supporting our 12 conclusions on the before and after comparison. 13 Two quick comments. One, I do want to remind all 14 of us that we do have the substantial public good coming from 15 the issuer pays model, which is that the ratings are made 16 available to the investing public free of charge. That is, I 17 think, a significant benefit. And if we are looking at what 18 we might test or track over time, I think it is probably as 19 important to look a customer concentration levels as opposed 20 to the nature of the customer themselves. Any customer who 21 represents a substantial portion of revenue for a firm is a 22 customer who may have undue influence, so it is another area 23 that I think is worth looking at. 24 MR. GALLAGHER: We have a few more questions, but 25 this is such a critical issue I think Mr. Sharma and Mr. 47 1 Curry probably want to be heard before we move on to the next 2 question, so Mr. Curry? 3 MR. CURRY: Thank you. I think conflicts exist 4 with both models. Issuers obviously have a desire to have 5 higher ratings in order to lower their interest rates. 6 Investors have the opposite interest. They would prefer 7 lower initial ratings to get wider spreads. They also would 8 prefer to avoid rating changes, which in many cases can force 9 them to sell at an inopportune time. So I think the question 10 that is more important is how do you manage this, and I think 11 that the key is really transparency in the process, in the 12 methodologies that are used, and the results. 13 But I think another critical factor that does not 14 exist right now in my view is investors need a much greater 15 voice, particularly in structured finance, when these 16 securities are being structured where they can have a real 17 impact on who rates the transaction eventually. 18 MR. GALLAGHER: Mr. Sharma? 19 MR. SHARMA: We do also think -- we have to 20 consider the fact that both issuers and investors have 21 different objectives in this. And the notion of conflict 22 comes from, when anybody pays any money, there is an 23 opportunity to influence and create a conflict. So it 24 doesn't matter who pays. When that happens, there is an 25 opportunity and potential for conflict to occur. And the 48 1 greater the customer concentration, there is a greater 2 potential for conflict. So it can happen on the issuer side, 3 and as we know, even on the investor side in the fixed income 4 market, there is a large concentration of large investors, so 5 you have to watch for that too. 6 It is critical, really, in the long run -- because 7 we recognize even with an issuer-paid model, our credibility 8 comes from investors. As long as investors have confidence 9 in what we do in our analytics, they will continue to ask for 10 our ratings from the issuers. And so we do recognize we need 11 to continue to look at ways to bring more investor voice into 12 our ratings process. 13 MR. GALLAGHER: Thank you Mr. Sharma. 14 Chairman Schapiro, I think, had a question? 15 MR. GALLAGHER: Commissioner Paredes, let's go with 16 you first. Great, thanks. 17 MR. PAREDES: Thanks. One of the important 18 considerations spurring the entire debate is the effort to, 19 in fact, spur competition, which is an aspect of 20 accountability, but more expansive than that. Which is to 21 say, flipping that around, that it would be unfortunate if we 22 ultimately took steps that had the effect of undercutting 23 competition. This leads to a question I hope you can all 24 speak to specifically, and that is what the competitive 25 consequences would be and consequences more generally in 49 1 terms of public disclosure of ratings histories and track 2 records. 3 MR. GALLAGHER: Why don't we start with Mr. Sharma 4 on this one? 5 MR. SHARMA: We do believe and support public 6 disclosure of our ratings performance, because that becomes a 7 very important element of accountability and an element of 8 comparability. So we do strongly support that and encourage 9 that. 10 MR. GALLAGHER: Mr. McDaniel? 11 MR. McDANIEL: I think there are probably some ways 12 to handle disclosure that satisfy different business models. 13 We are certainly supportive of public disclosure, of 14 information to oversight authorities, and to academics who 15 would be able to compare and contrast ratings quality 16 performance. 17 I think that the application of the current -- of 18 the new rule for 10 percent sample disclosure applied to all 19 ratings agencies regardless of business model probably 20 satisfies both the ability to conduct statistical analysis 21 and to protect the intellectual property of the firms that 22 sell information around rating histories, and I would suggest 23 that as a possible middle path. 24 MR. JOYNT: So I would agree with that position and 25 I think the proposal as made is reasonable, acceptable, 50 1 should be able to be accomplishable by all the firms. 2 I also think that the investor-pay rating agencies 3 in this question seems to identify as a competitive 4 standpoint as the only value is the rating, whereas I would 5 think many investors that use rating agencies, both 6 issuer-paid and investor-paid, find a lot of value in the 7 research and the analysis, not just the letter rating itself. 8 So it would seem to me that you need to keep that in mind 9 when you are talking about just releasing the rating. You 10 are not speaking about all the other material information 11 investors might find value from, I would think. 12 MR. EGAN: We are struggling with that within our 13 firm, actually, how much we give away and maintain our 14 revenue base. There are a number of clients that just want 15 our ratings, and that is it. That gives them enough. Others 16 want our analysis. And the question is, if we make it 17 public, do we grow our market prices that much more to make 18 up for -- we don't know the answer to that. So I think you 19 want to be very careful about mandating that investor-pay 20 rating firms disclose the ratings and other information 21 because you will probably increase the entry barriers in this 22 business. 23 It is also misguided. Some people think -- and it 24 came up with Erik Sirri at the American Enterprise Institute 25 Conference -- he said he can't understand why the timely, 51 1 accurate ratings don't win out. Well, because buyer behavior 2 is such that 90 percent don't want -- 90 percent of the 3 revenues in the business don't want timely, accurate ratings. 4 They want the highest rating possible. So if you are going 5 in the direction and saying "Ah, this rating firm is better 6 than the others," it doesn't matter if you are being 7 supported by issuers. In fact that is a negative in many 8 cases. 9 And there is one poignant example, and that is in 10 the case of the Lehman Brothers index. We have a number of 11 major subscribers. I'm friends with Jack Bogle, sit down and 12 have breakfast with him every once in a while. And I said 13 "Jack, this is incredible. We have this big investor, that 14 big investor, and the other big investors, why don't we have 15 Vanguard?" And so I asked the person and he said "Well 16 because you're not in the index." 17 Well how do you get in the index? You get in the 18 index by supporting the issuers, ironically. Fitch struggled 19 for a number of years to be included in the Lehman index. It 20 wasn't until General Motors helped them out to get in the 21 index because their rating was higher that they were included 22 in it, and then they are relevant. So it is ironic. 23 You have to be careful about how you review this 24 industry. There is a lot of countertrends in the industry. 25 You wouldn't expect that at all. So you are the biggest 52 1 investor in the world, Vanguard, that should want -- that 2 should be pro-investor, and they say "You are not part of the 3 index, get out of here, you are not relevant." And I said 4 "Well, what about this, that, and the other thing?" "So 5 what, you are not part of the index." So basically the 6 timely, accurate ratings are punished, unfortunately, and we 7 are not going to do anything about it. 8 It is also a misnomer to say that the conflicts are 9 manageable. They are not. They are unmanageable, no matter 10 what. If you are getting paid -- Chairman Schapiro said it 11 beautifully just last week, actually, in connection with the 12 CII institute. You said "We all know that compensation 13 drives behavior." It was the opening remark in connection 14 with executive pay, and I would argue it extends to this 15 industry too. 16 MR. GALLAGHER: Thanks, Mr. Egan. 17 MR. DOBILAS: Well we are a subscription pay rating 18 agency, and we are very affordable, and we are happy to take 19 any client that wants to pay our subscription fee, and they 20 can have full access to all of our ratings. I think it is 21 very hard for us to cope with the fact that we need to give 22 any letter ratings away for free in the marketplace, 23 especially at the price point that we are at. 24 I have heard a lot about the reliability of 25 subscriber-based ratings compared to issuer-pay ratings. I 53 1 mean, the data does not show that. The data clearly shows 2 that the ratings shopping effect can be seen in the last 3 three years by all three rating agencies. So I do want to 4 sort of defend the reliability and the credibility of the 5 subscription-based rating agency. 6 There is not much more to say besides customer 7 service is the only thing that keeps us in business and our 8 analysis is the only thing that keeps us in business. Other 9 than that, we wouldn't have much of anything. We don't just 10 come up with a letter rating, it is monthly surveillance, it 11 is monthly analytics, it is listening every year to investors 12 and what they want, keeping a close dialogue, and producing 13 the kind of quality that they want every single year. Thank 14 you. 15 MR. CURRY: Well to the original question of 16 competition, I think that is a difficult problem to solve 17 because the industry is very complicated and there are some 18 important legacy reasons for that. I'm not really sure how 19 you can fix that, but I do think that the key to having a 20 healthy industry is to allow investors to make informed 21 choices. And so to give them open access to the data, to 22 give them access to that analysts, and make their own 23 decision at the end of the day is, I think, a difficult thing 24 to argue against. 25 MR. GALLAGHER: Thank you Mr. Curry. 54 1 I believe Commission Aguilar has a question. We 2 have about 10 minutes left here, so we are going to try to be 3 efficient in the last 10 minutes. 4 MR. AGUILAR: And I will try to ask my question 5 quickly. 6 If I could really start at the other end of the 7 table, because this may be a better question to start with 8 the issuer-paid model. I have heard ratings shopping 9 repeatedly mentioned this morning, and I'm trying to get your 10 thoughts as to what could be done, perhaps, to ameliorate or 11 help in that area. One of the ideas that you read about 12 every so often is an issuer-paid pool where the rater is 13 selected randomly and that that may have its own set of 14 problems, but it may be a solution. I'm not trying to limit 15 you to that, I'm trying to get some thoughts about how that 16 issue perhaps could be addressed. 17 MR. GALLAGHER: Mr. McDaniel, do you want to take 18 that one? 19 MR. McDANIEL: Sure. It goes back to some comments 20 I made earlier. If there is sufficient information available 21 in the public domain to reach reasonable conclusions on 22 securities being offered into the market, that is the same 23 information that allows rating agencies that are not selected 24 to rate a security to act as a check and balance on those 25 that are selected to rate a security. I think it may be a 55 1 difficult solution to implement in terms of changes imposed 2 on the market, but conceptually, I think it is a relatively 3 straightforward action that can be taken. 4 Enough information in the market is beneficial to 5 the -- sufficient information to make the investment decision 6 is beneficial to the investors and to the rating agency 7 industry. It currently exists in the corporate market for 8 public bonds. It does not exist in securitization, and I 9 think that, again, my single highest recommendation would be 10 to alter that condition. 11 MR. GALLAGHER: Mr. Sharma? 12 MR. SHARMA: One suggestion or thought would be to 13 let the issuers disclose as to who they looked at in terms of 14 selection of the rating agency and make that public so that 15 is out in the open and then they can sort of -- investors can 16 determine based on that. 17 You had a second aspect of this, lottery, and that 18 has a lot of unintended consequences to it, and one is if it 19 becomes a random event, then the question becomes how do you 20 foster competition? How do you foster innovation and how do 21 you foster investment? So that is a question to be 22 considered. 23 MR. JOYNT: I agree with the point that would say 24 greater transparent release of information for 25 securitizations would be helpful to many people being able to 56 1 look at and analyze and assign ratings, not just credit 2 rating agencies as recognized by NRSROs, but even new 3 players. So I think that would be helpful and beneficial. I 4 would agree with that point. 5 I think we would remain maybe a little more open to 6 the idea of some kind of pooled concept, possibly because we 7 don't rate 100 percent of all the securitizations and so we 8 don't see the downside quite as much. The points about how 9 that would be implemented are very difficult, as has been 10 pointed out. Who chooses and on what basis, what incentive 11 does it take away for quality work? The concept is a good 12 one, but I grapple with the problems with execution, so I'm 13 not sure I can even think of constructive suggestions to 14 forward that one further, so I guess I default back to the 15 more transparent information. 16 MR. EGAN: In my view, ratings shopping is a 17 manifestation of a flawed system. When you see ratings 18 shopping, you should say "Ah, there is rating inflation." 19 And there is actually no way to prevent it. Investment 20 bankers are creative. They will work out a way to get to the 21 parties that will give them the highest rating possible, and 22 the system will be corrupted over time, not right away. 23 And right now, we are faced with a situation where 24 people don't believe the ratings. The securitization market 25 has broken down because people have been burnt by triple As 57 1 going to D in a period of two days. Rhinebridge, two days. 2 Triple A, thought you were going to get all your money back 3 plus a little tip to nothing back, okay? So that is the 4 ultimate case of rating shopping. 5 How do you solve it? You solve it by making sure 6 that if the SEC is true to its mission of protecting 7 investors, that its agents in the process also protect 8 investors, that there is an alignment of interest. So 9 basically that is a weaning away of any firm that wants to be 10 an NRSRO from the issuer compensation. That is the only 11 ultimate way of dealing with the rating shopping. 12 MR. DOBILAS: I guess the problem with rating 13 shopping is nobody sees it. Nobody sees it happening, it is 14 really based on the honor system. We have participated in 15 it. I used to work at several of the major rating agencies 16 as well as the issuer side of the business. 17 If we make this process a transparent process -- I 18 want to push this point home -- and make it more transparent 19 to investors, let investors see the effects of ratings 20 shopping, let them see the different agencies coming up with 21 their preliminary levels, and they will understand why. I 22 mean we can trace the data back over the last three years and 23 see business share compared to subordination levels, and we 24 have been tracking it for a while. The investors know it, 25 the issuers, and aggregators most certainly know it. I mean 58 1 if we just make that process transparent, that would be a big 2 start. 3 You know, having random rating agencies rate the 4 deals, chosen by the SEC? That is great for business, my 5 business, but it is not going to solve the problem. The 6 quality will suffer. I think making -- looking at the 7 process and making it more transparent is going to be good 8 for business in the long run for everyone. I think the SEC 9 did build a -- or help build Moody's or S&P, their brand 10 recognition. I think subscriber-based rating agencies like 11 Realpoint, we have to sweat it out, we have to fight for 12 every ounce of market share, and we have made a name for 13 ourselves in the secondary markets. It would be great to get 14 some help in the primary markets, but only our analysis will 15 speak for that. 16 Again, I would just point out that transparency, I 17 think, is the key to this. And trying to judge who gets the 18 business share is just going to be a nightmare even though it 19 is good for my business. Thanks. 20 MR. CURRY: Well I think rating shopping is a very 21 difficult problem to solve. The issue of having the issuer 22 or the banker disclose who they showed the transaction to and 23 what their response was can be managed by the banker because 24 rating models are typically available on the websites. They 25 have also interacted with the agencies extensively and they 59 1 know who is relatively liberal for a particular asset class, 2 so they would just naturally gravitate to those agencies. 3 The other issue, which I think you have done some 4 good work on, but the giving other agencies the ability to 5 issue unsolicited ratings. I have done some of those in my 6 career in structured finance, and I have never had anyone 7 thank me for having done that. Issuers find that, obviously, 8 very irritating, and the bankers are angry because it screws 9 up their deal, and the investors were very upset too because 10 they said "I had a triple A rated security that I had agreed 11 to buy, you come in and put a double A on it. Now that 12 security is damaged and it is worth less than I paid for it." 13 So it is a difficult issue to solve. 14 And I think at the core, trying to find a way to 15 break the connection between relaxing your rating standards 16 and driving up your rating revenue is really at the core of 17 the issue that has to be solved. But that is a tough problem 18 to fix. 19 MR. GALLAGHER: We are pretty much done with time, 20 but there is always time for the Chairman. 21 CHAIRMAN SCHAPIRO: Thank you. I know we are out 22 of time. I have actually two questions. One is very quick 23 and very directed. 24 Mr. McDaniel, you referenced the fact that Moody's 25 switched from a subscriber-pays model to an issuer-pays 60 1 model, and I was just curious why. 2 MR. McDANIEL: This was back in about 1970, and I 3 think there were several reasons. First of all, information 4 technology and advances in information technology were 5 increasing what many would call the free rider problem. 6 Xeroxing and faxing -- Xeroxing of information and the 7 availability of information being passed from party to party 8 was increasing the free rider issue associated with 9 subscription sales. 10 I think more importantly, probably though, was the 11 increasing size and complexity of the markets themselves were 12 demanding a model by which additional resources could be 13 funded, more specialization, higher paid people, more of 14 them. And so the decision to move from investor pay to 15 issuer pay was coincident, both with information technology 16 advances and complexity in capital markets and growth in 17 capital markets. 18 CHAIRMAN SCHAPIRO: The last question I wanted to 19 ask is I would hate for you all to escape without having some 20 input on this. As you look back over how ratings performed 21 over the past year or two years or three years in particular, 22 would you conclude that there were instruments that simply 23 should not have been rated? 24 MR. GALLAGHER: Why don't we just go down the line 25 with Mr. Curry, and just keep your responses short if you 61 1 could. Thanks. 2 MR. CURRY: Oh, absolutely. I mean just look at 3 the performance, and there are certain classes of securities. 4 CDOs of ABS are a good example. But I think that the 5 increasing level of complexity became a bit of a smokescreen. 6 Investors couldn't really understand the risks that they were 7 taking, and I think in some cases the agencies didn't 8 understand those either. 9 MR. DOBILAS: Yes, I would agree. Each agency has 10 its specialty, structured finance is no different. CDOs were 11 a big mistake. We made the decision not to rate CDOs because 12 we couldn't understand them. You know, you are out there 13 mixing RMBS, ABS, CMBS all together and then it is a very 14 complex situation. I think more oversight over new creations 15 of structures like that and showing competence and 16 capabilities in rating those securities should be definitely 17 high up on your radar. Thank you. 18 MR. EGAN: There is no question that a number of 19 securities should have been rated, should not have been 20 issued. However, the response is a little bit more nuanced 21 than that, and that is that the -- it really gets back to the 22 investor protection. If there is garbage coming down -- 23 think of rating agencies as being like meat inspectors -- if 24 there is garbage coming down the pike, the rating agencies 25 should identify which things are tainted meat because the 62 1 public can't understand it. Unfortunately, that didn't 2 happen. 3 It also speaks to building out the information 4 infrastructure. Intellectually, there is no reason why CDOs 5 or CDOs or CDOs shouldn't be out there. The problem is that 6 the information isn't -- the infrastructure for evaluation 7 those securities hasn't been built yet. So the rating firms, 8 if they weren't paid, weren't given a million dollars and 9 freedom of speech if they are wrong, if they didn't have that 10 setup, they would have stopped some of this garbage. 11 But I encourage that it helps the economy to have 12 more sophisticated instruments out there. Structured finance 13 is the right way to go. It is a much more effective way, but 14 make sure that there are some safeguards, make sure that is 15 some -- the information out there, and make sure that we get 16 this right, because the securitization market is broken and 17 it needs to be fixed as quickly as possible to get things 18 back on track. 19 MR. JOYNT: I think historically, ratings were used 20 and assigned to term bond financings, and so the assumption 21 that most investors in their association was with some kind 22 of stability, both the bond, the term, and the rating. And 23 then the last five or ten years, the acceleration of finance 24 and the acceleration of trading activities has meant that 25 things that are being issued are tradable items and they are 63 1 being traded actively among investors and not sort of by and 2 hold investors. 3 So I think it is hard for ratings to have the kind 4 of stability when attached items that are tradable and priced 5 so actively -- there is a disassociation there. And I think 6 in the last few years, especially three or four years, and in 7 some securitization markets and with synthetic CDOs, for 8 example, that the derivative instruments became so volatile 9 that it is and was hard to assign ratings that would have 10 much stability to them, and that certainly proved to be the 11 case. 12 The second issue, of course, addresses the how did 13 we rate them and what models were used, were they projective 14 enough, which also I think is giving us some serious 15 re-reflection. 16 MR. McDANIEL: Not to repeat some of the previous 17 comments, but whenever a rating agency is not comfortable 18 that it either has sufficient information or sufficient 19 quality information or a sufficient understanding of that 20 information, it should simply step back and not rate. 21 MR. SHARMA: Clearly when we look at the last two, 22 three years, there are parts of the market where things have 23 been disappointing for us. But as we have reflected on it, 24 we recognize the need for more stability in our methodology, 25 and so we have been making a lot of analytical and 64 1 methodology changes to bring more stability into ratings, and 2 this is a clear recognition that it was needed. 3 The second one was the need for continuing to raise 4 the quality of data coming in, and we have done a fair amount 5 of effort to set the due diligence standards, et cetera, so 6 we know in the structured market that we can raise the 7 quality of data coming into us. That is why I strongly urge 8 the Commission, as we look at new rules and new regulatory 9 proposals, that really the end to end process be looked at. 10 There are things that need to be imposed on the rating 11 agencies, but we need to look at the whole ratings: how 12 ratings originated, the data is originated, how the ratings 13 are done, how they are priced, how they are sold, and how 14 they are bought. So the whole end to end solution should be 15 reconsidered to really bring the confidence back into the 16 marketplace. 17 MR. GALLAGHER: Well, thanks so much. Speaking on 18 behalf of the Division of Trading and Markets and I'm sure 19 for the Commission too, we want to thank the panelists for 20 taking the time to come down here. This was an extremely 21 useful exchange of information, especially in conjunction 22 with your written statements. So thanks once again, and we 23 appreciate your time and effort, and look forward to working 24 with you and your firms as we consider these issues over the 25 course of this year. 65 1 PANEL TWO: COMPETITION ISSUES 2 MR. GALLAGHER: We are waiting for two 3 Commissioners, but the panelists are all here, so I will read 4 you your Miranda rights here. I don't know if you heard the 5 rules of the road from the first panel, there aren't many. 6 The Commission really wants a free flow of 7 information here. I will try to jump in where necessary to 8 keep things moving, but generally speaking, the way we are 9 going to proceed here is I will introduce each of the panel 10 members. And after I do that, I will ask you in the order 11 that I introduce you, from my left to my right, to read your 12 prepared statements. I'm going to allot four minutes to each 13 of you, I know that is not a lot. I will try my best to 14 catch your eye and wave a yellow card when you have one 15 minute left, so if you could look up every now and again, I 16 will do my best to catch your attention. 17 The one thing I didn't tell the last panel is, you 18 know, this is the SEC, we do have rules. I forgot to remind 19 them we have lots of enforcement people upstairs, so with 20 that threat in mind, maybe I will do a little better on my 21 yellow card activity here. 22 So let's go through the introductions here and then 23 we will move on to the prepared statements. 24 First on the panel we have Mr. Ethan Berman, the 25 CEO of RiskMetrics Group. Next to him is Mr. James Gellert, 66 1 Chairman, President, and CEO of Rapid Ratings International. 2 Next to him is Mr. George Miller, Executive Director of the 3 American Securitization Forum. Then we have Professor Frank 4 Partnoy, George E. Barrett Professor of Law and Finance and 5 the Director of the Center on Corporate and Securities Law at 6 the University of San Diego. Then we have Mr. Alex Pollock, 7 Resident Fellow of the American Enterprise Institute. Next 8 to him is Mr. Damon Silvers, Associate General Counsel, 9 AFL-CIO and Deputy Chair of the Congressional oversight panel 10 for TARP. And lastly we have Professor Lawrence J. White, 11 Arthur E. Imperatore Professor of Economics at NYU's Stern 12 School of Business and Deputy Chair of the economics 13 department at Stern. 14 Thank you all for being here today, and Mr. Berman, 15 I will ask that you start with your prepared statement. 16 Again, four minutes, and I will give you the yellow card with 17 one minute left. 18 MR. BERMAN: Great. Thank you very much and also 19 thank you for my father, whose last name begins with a B, so 20 I will get my four minutes in. 21 The ratings industry is dominated by three large 22 agencies that hold disproportional market power. That market 23 power rises from informational privileges and special status 24 of their ratings gained as a consequence of historical 25 regulatory and market actions, giving them unique standing in 67 1 regulation investment guidelines. Certain measures can be 2 taken to promote healthy and fair competition in the long 3 run, but we also believe that intervention will be required 4 in the short term to break the grip of the dominant agencies 5 and jumpstart competition before market forces are allowed to 6 take over. 7 There are four main types of barriers to entering 8 the credit rating business. The first is the regulatory 9 barriers to entry and in particular the requirement to be in 10 business for three years before applying for registration as 11 an NRSRO. The second is the lack of equal access to 12 information by all NRSROs that is necessary to produce high 13 quality ratings. The third is inadequate transparency and 14 disclosure by rating agencies, and the fourth refers to 15 negative competitive dynamics reinforced by the combination 16 of NRSRO status and issuer-paid ratings. Each one of those 17 barriers can be addressed by changes in regulation. 18 Very simply, we encourage the SEC to make the 19 requirements, the process, and the timeframe to obtain 20 registration as an NRSRO as clear as possible to encourage 21 potential new players to enter the space. In setting the 22 requirements, we believe that regulators should stay away 23 from directly regulating ratings methodology because it 24 creates moral hazard for the regulators, stifles innovation, 25 and potentially encourages ratings convergence. 68 1 In addition, regulators should not require measures 2 of market acceptance as prerequisites for registration 3 because they are subjective, difficult to define, and 4 ultimately prevent competition since it is difficult to be 5 accepted by the market when you are not allowed to compete on 6 an equal footing. Instead, regulators should focus on the 7 quality of the ratings process and ex-post-ratings 8 performance as the key criteria for registration. 9 Second, let me discuss unequal access to 10 information. New entrants to the credit rating industry need 11 to gain credibility before they can compete with the 12 established NRSROs. However, it is virtually impossible for 13 new entrants to challenge the incumbents without access to 14 non-public information that is critical to assign a credible 15 rating and demonstrate the quality of the work. It is our 16 view that all NRSROs should have equal access to information 17 from issuers, arrangers, underwriters, and the other parties 18 disclosing non-public information to NRSROs being retained to 19 rate a product. This principle should apply to all asset 20 classes and types of ratings, but it is particularly 21 important for structured products given their complexity and 22 the amount of information required to assess their credit 23 risk. 24 Third, on a lack of transparency. The most 25 valuable asset of a rating agency is its credibility. In 69 1 order to build credibility based on the quality of their 2 work, rating agencies should be given the opportunity to 3 prove and differentiate themselves from their competitors by 4 comparing the relative performance and the information they 5 provide to investors. We view aggregate information 6 performance reports currently published by NRSROs as 7 insufficient to perform an in-depth study of their ratings. 8 In our view, the current market dynamics have to 9 change before true competitive market forces can take over. 10 We are afraid that it will be difficult for new entrants to 11 compete against entrenched players, even in the absence of 12 structural barriers. There are two ways to break the current 13 vicious cycle. The first is to tackle directly the 14 combination of misalignment of incentives in the issuer-pay 15 model and the mandated use of ratings and regulations. 16 To do that, regulators can either mandate a 17 business model where investors decide which ratings are used 18 or completely eliminate references to NRSROs ratings and 19 regulation. If regulators want to spur competition without 20 mandating a change in the business model or the use of 21 ratings and regulations, they will have to directly intervene 22 to give new entrants an opportunity to thrive in the current 23 market dynamics. 24 To this end, we propose two new ideas that could 25 kick-start competition. The first would require issuers to 70 1 obtain a rating from a new entrant for every new issue. The 2 second idea would require issuers to rotate rating agencies 3 every three years. In other words, an issuer would not be 4 allowed to use the same rating agency for a period longer 5 than three years, at which point they would have to use a 6 different rating agency for at least three years. This is 7 analogous to the rotation policy implemented for financial 8 auditors. 9 MR. GALLAGHER: Thank you Mr. Berman. Mr. Gellert? 10 MR. GELLERT: On behalf of Rapid Ratings 11 International, I would like to thank the Commission for 12 including us in this roundtable and commend the Commission on 13 its recent efforts to introduce transparency and oversight to 14 the ratings business. I would also like to say it is an 15 honor to be here with these accomplished panelists whose work 16 I respect. 17 Rapid Ratings is a subscriber-paid firm. We 18 utilize a proprietary software-based system to rate the 19 financial health of approximately 3,500 companies and 20 financial institutions quarterly. We use only financial 21 statements, no market inputs, and have no contact in the 22 ratings process with issuers, bankers, or advisors. 23 Currently, as a non-NRSRO, we view the recent rule amendments 24 and re-proposed rules as one already competing with the other 25 agencies currently from outside the NRSRO paradigm. We 71 1 critically review new regulatory actions and lack of action 2 to understand how they will shape this industry and determine 3 if being an NRSRO adds value to our shareholders. 4 A number of recent SEC initiatives constructively 5 attacked the symptoms, though still not the illness, of 6 issuer-paid agencies' conflicts of interest. But what the 7 SEC has not done sufficiently is address the underlying 8 problems in the ratings business: that the issuer-paid model 9 is clearly conflicted, that accurate ratings are desperately 10 needed but can only come about through more effective 11 competition, and that a reduction to barriers and avoiding 12 new barriers are critical to fostering that competition. 13 Subscriber-paid agencies are the best source for more 14 accurate ratings and competition. Without greater 15 sensitivity to these players, the SEC risks throwing out 16 these solutions with the issuer-paid's bathwater. 17 First, for example, the SEC's new rule require 18 issuer-paid agencies to disclose their ratings history for 19 free is appropriate, but it is counter-productive to apply 20 this rule to the subscriber-paid agencies. For the SEC to 21 propose this suggestion twice now raises concerns about how 22 well our business model is understood. 23 Second, the issuer-paid agencies are claiming that 24 the subscriber-paid agencies are conflicted too. This is a 25 massive red herring that diverts attention from the really 72 1 serious conflicts in our own business model. While a 2 subscriber-paid agency could be influenced by a subscriber, 3 it would be a rarity compared to the endemic conflicts of 4 interest in the issuer-paid model, and to what end? It would 5 be the kiss of death for that company because other 6 subscribers would quickly notice a distortion in risk 7 relativities within their portfolio, as Rapid Ratings' 8 conflict is impossible, because our system is entirely 9 automated. 10 Fundamental to fixing the broken rating agency 11 paradigm and increasing competition is creating a market 12 environment and regulatory structure that fosters effective 13 competition and does not create disincentives to enter and 14 compete. To do this, the SEC must strive to make the NRSRO 15 status desirable for new players. Rapid Ratings is presently 16 skeptical about applying for NRSRO status. There are too 17 many uncertainties over what the NRSRO designation means, 18 notably the direct and indirect costs for subscriber-paid 19 agencies, and to the best of our knowledge, we have yet to 20 lose a piece of business because we are not an NRSRO. 21 The SEC risks making the NRSRO status a 22 disincentive to effective competition, which inhibits the 23 entry of more NRSROs and NRSROs with more accurate rating 24 tools. If the SEC wants to attract more NRSROs, better 25 NRSROs, and promote the delivery of higher quality ratings to 73 1 the market, it must ensure that potential NRSROs see the 2 designation as an asset, not as a contingent liability. 3 Thank you. 4 MR. GALLAGHER: Thank you Mr. Gellert. Mr. Miller? 5 MR. MILLER: Thank you very much and good morning. 6 On behalf of ASF, I greatly appreciate the opportunity to 7 attend this important roundtable. 8 I think it is easier to agree on the goal of 9 enhanced competition and that it is desirable than to agree 10 on necessarily the most effective way to achieve that goal 11 or, for that matter, to determine whether or when an optimal 12 level of competition has been attained. 13 I think there are many different dimensions to the 14 competition issue and what is meant by increased competition. 15 For example, does it mean more credit rating agencies or more 16 NRSROs? And I think there may be different goals relating to 17 competition whether we are talking about credit opinions that 18 are issued generally into the market versus those that are 19 issued by organizations that are recognized and designated by 20 government agencies. 21 Does it mean greatly choice and greater diversity 22 among ratings providers? Does it mean lower costs for users 23 of ratings? Does it mean the production of more accurate and 24 more stable ratings, or for that matter are we trying also to 25 achieve the promotion of confidence and the restoration of 74 1 confidence among credit rating agencies? I think most would 2 agree that by talking about increasing competition, we are 3 probably talking about addressing at least some or if not all 4 of those goals. 5 Assuming that we can agree on the goals of enhanced 6 competition, how do we get there both with respect to actions 7 that credit rating agencies themselves and other industry 8 participants can take as well as the types of regulatory 9 oversight policies that are best designed to achieve goals 10 relating to increased competition? Here again, and I think 11 we have heard that this morning, there are many different 12 prescriptions for how to promote achievement of various goals 13 relating to increasing competition. 14 Among others, those include taking steps to reform 15 or streamline the process for governmental designation of 16 rating agencies as NRSROs, more stringent management and 17 oversight or elimination of conflicts of interest, actual or 18 potential conflicts among rating agencies, steps to either 19 favor or disfavor certain types of ratings models, such as 20 issuer-paid, subscriber-paid, or others, the forced rotation 21 of credit rating agencies among issuers for different 22 transactions, eliminating or reducing the use of ratings and 23 regulation, requiring that some or all information that is 24 provided by issuers or arrangers in certain transactions to 25 engaged credit rating agencies to be provided similarly to 75 1 non-engaged rating agencies, or even scrapping the private 2 rating system entirely and replacing it with a governmental 3 or industry utility. 4 Now here in terms of ways to achieve the goal of 5 increased competition, I think it is also fair to say that 6 there seems to be less agreement, both within the private and 7 public sectors, on the merits of these various prescriptions. 8 Notwithstanding all of those differences of 9 opinion, I do think that it is fair to say that there is one 10 broad area where at least in my judgment there seems to be a 11 greater measure of agreement on how to promote competition, 12 and that relates to enhancing what I will broadly describe as 13 transparency both by rating agencies with respect to their 14 ratings methods, processes, assumptions, inputs, and outputs, 15 but as well by market participants with respect to the 16 transactions that they bring to the market, and especially on 17 the securitization and structured finance markets, the 18 underlying assets, and transaction structures that are the 19 principle determinants of securities' performance. 20 I believe that as a non-exclusive measure to 21 enhance competition, that over time, providing better, more 22 detailed, higher quality, more accurate, and reliable 23 information about the ratings process and the transactions to 24 which those ratings relate will promote more robust 25 competition and choice among credit rating agencies and also 76 1 serve the common interest in helping to rebuild confidence in 2 credit rating agencies by those in the marketplace who rely 3 upon them. Thank you. 4 MR. GALLAGHER: Thank you Mr. Miller. Mr. Partnoy? 5 MR. PARTNOY: I want to thank the Commission for 6 inviting me to speak. I have been writing about the deep 7 flaws in the credit rating system for more than a decade, and 8 I'm grateful for the opportunity to share my views of what 9 should be done to repair this seriously broken system, and I 10 have also written a white paper for the Council of 11 Institutional Investors, which I believe is included in your 12 materials and amplifies my remarks. 13 In my view, the central problem is not that the 14 major rating agencies, particularly Moody's and S&P, have 15 been hopelessly incompetent, inaccurate, and conflicted in 16 their assessments of credit risk, particularly in the 17 structured finance area. All of those points are correct, of 18 course, but these are symptoms, and too much focus on 19 symptoms ignores the underlying disease. 20 Instead, the key issue continues to be 21 over-dependence on credit ratings. I have called the 22 regulatory over-dependence on ratings regulatory licenses. 23 The idea of a regulatory license is just like any license, a 24 driver's license, for example. It is a permit to participate 25 in some regulated activity. Rules that depend on ratings are 77 1 regulatory licenses. They are keys that unlock the financial 2 markets, and there are so many regulatory licenses embedded 3 in our state and federal laws that it is hard even to count 4 or find all of them. 5 The disease has spread like a cancer. Today, 6 private contracts, investment guidelines, and loan documents 7 also depend on credit ratings. Market participants rely on 8 credit ratings even when they know better. The result of 9 over-dependence on ratings is that credit rating agencies no 10 longer play the role John Moody envisioned during the early 11 20th century. They are not information intermediaries who 12 survive and prosper based on the quality of their ratings. 13 Instead, they have shifted from selling information 14 to selling regulatory licenses. They are no longer 15 constrained by reputation, they can issue low quality 16 ratings, and market participants still will pay for them, 17 indeed they must pay for them, because of regulations that 18 depend on ratings. Without a rating, many issuers will be 19 locked out of the financial markets. 20 After the Commission created the NRSRO designation, 21 over-dependence on ratings spread. Again, the metaphor is a 22 cancer. Through our laws and regulations and then through 23 investment guidelines and private contracts, the result is 24 that the credit rating process is not just dysfunctional but 25 diseased. 78 1 So what can be done? There should be two 2 priorities, which I label oversight and accountability. 3 First with respect to oversight. There should be an 4 independent credit rating agency oversight board with 5 authority not only over the substance of the rating process, 6 conflicts, disclosure, and pay, but also the much-needed 7 transition away from regulatory reliance on ratings. This 8 board must be truly independent with separate funding and 9 strict prohibition of rating agency involvement, and it 10 should be charged primarily with encouraging substitutes for 11 ratings, including both market measures and judgment. Market 12 measures are a particularly attractive substitute for 13 ratings. For example, bond credit spreads and credit default 14 swaps have been much more timely and accurate indicators of 15 credit risk than ratings. 16 Second, credit rating agencies must be accountable. 17 Moody's and S&P have used an aggressive First Amendment 18 campaign to insulate themselves from liability for behavior 19 that would have led to damages for any other gatekeeper. I 20 thought it was interesting that we didn't hear much about the 21 First Amendment or free speech from them today. 22 At minimum, credit rating agencies should be 23 treated like other gatekeepers, including bankers, 24 accountants, and lawyers. When any gatekeeper commits fraud, 25 breaches an agreement, or is reckless or negligent in 79 1 conducting its business, it should be held accountable, and 2 unfortunately, Moody's and S&P have been unaccountable, 3 notwithstanding what I regarded as the disingenuous remarks 4 today, the sort of Orwellian doublespeak about lawsuits 5 constituting actual liability. Moody's and S&P participated 6 in creating monstrous structured finance transactions with 7 absurdly high ratings based on models and assumptions they 8 knew or should have known were unreasonable. They helped 9 design, structure, and monitor the instruments at the center 10 of the crisis. 11 Congress already has begun debating the extent to 12 which rating agencies should be held accountable as 13 gatekeepers and courts have recognized the errors in some 14 cases. Rating agencies should not be exempt from securities 15 fraud liability and they should not enjoy any special 16 privilege over gatekeepers in Section 11 of the Securities 17 Act of 1933 Regulation FD or elsewhere. 18 The best solution would be for the government to 19 eliminate all regulatory licenses and leave a competitive 20 credit rating business subject to the same constraints as any 21 other industry. The second best solution would be these two 22 areas of reform, oversight, and accountability. 23 MR. GALLAGHER: Thank you Mr. Partnoy. Mr. 24 Pollock? 25 MR. POLLOCK: Frank, I'm going to sound so moderate 80 1 and middle of the road after you. 2 Madam Chairman, members of the Commission, Mr. 3 chairman of the session. 4 Four years ago, I published an essay entitled "End 5 to Government Sponsored Cartel in Credit Ratings," and that 6 title still summarizes my views. I think if we do away with 7 the government sponsored cartel which Frank calls the 8 regulatory license, a lot of the issues so informatively 9 discussed by the first panel will simply go away. 10 I do think there has been progress in this 11 direction since the Credit Rating Agency Reform Act of 2006 12 and the SEC's implementing actions, but the government still 13 does have a tendency to sponsor favored agencies. For 14 example, the Federal Reserve's recently invented new 15 category, "major rating agencies," simply means exactly the 16 old government sponsored cartel, and further progress toward 17 a competitive sector needs to be made. 18 In my view, there is a fundamental question about 19 the limits of human knowledge involved in all this. Can we 20 somehow assure ratings which are accurate, to borrow terms 21 from a current draft of a bill in the U.S. Senate? Can't we 22 guarantee models which are right? Can't we assure ratings 23 which are stable? The answer is no. Nobody, no rating 24 agency, no regulatory agency, no modeler with however many 25 computers can make universally correct predictions of future 81 1 events. As Isaac Newton wrote after the collapse of the 2 South Sea bubble, "I can calculate the motions of the 3 heavenly bodies, but not the madness of the people," and it 4 is the madness of the people that creates big credit 5 disasters. 6 Now in considering any system, we ought to look for 7 concentrated points of vulnerability to failure, try to not 8 create such points. The old NRSRO system made the dominate 9 rating agencies into just such a concentrated point of 10 possible failure, which then indeed failed in the mortgage 11 securities debacle. So when it comes to opinions about the 12 future called credit ratings, having more credit rating 13 competitors, especially those paid by investors, will in my 14 view be better and will increase the chances that new 15 insights into credit risks will be discovered. 16 Of course, the U.S. government should not enshrine 17 certain opinions as having preferred, preferential, or 18 indeed, mandatory status the way it did with the old NRSRO 19 system. And as we move away from that system, the form of 20 competition from investor-paid rating agencies does seem to 21 me to have a superior alignment of incentives, as has been 22 much discussed today. 23 A frequent objection to competition in credit 24 ratings is that there will be a so-called race to the bottom, 25 but this does not apply at all to the logic of investor-paid 82 1 ratings. So in my view, the SEC should ensure that all rules 2 it adopts are consistent with encouraging competition from 3 the investor-paid model, and indeed, I believe that all 4 regulatory bodies, since there are a lot of regulators whose 5 regulations created the government sponsored cartel, not just 6 the SEC, should find ways to promote the pro-competitive 7 objective of the 2006 act and simultaneously to avoid the 8 creation of a concentrated point of vulnerability to failure 9 which that cartel did indeed create. 10 In any case, successful competitors, probably the 11 ones working solely for investors, will find ways to 12 distinguish themselves by creating more valuable ratings. 13 And in some, greater competition in the rating agency sector 14 remains, in my opinion, not only an essential objective, but 15 also an achievable one. Thank you. 16 MR. GALLAGHER: Thank you Mr. Pollock. Mr. 17 Silvers? 18 MR. SILVERS: Good morning and my thanks to the 19 Commission for inviting me here today. 20 I should begin by saying that my remarks today are 21 my own and the AFL-CIO's, they are not the Congressional 22 oversight panel's or its staff or its chair's. 23 This is going to be a little odd. In listening to 24 my fellow panelists, I realize that I may be as close as the 25 credit rating agencies get to a defender on this panel, which 83 1 is not what I thought I was coming here to do. 2 I want to begin by saying a word or two about why 3 credit rating agencies are useful and why they are 4 particularly useful from the perspective of the retail 5 investor or the small institution. Effectively, they are a 6 way of achieving an efficiency in information economics and 7 they lower the cost of capital in credit markets. Now the 8 question of whether they lower it too much is a question many 9 of my fellow panelists may disagree with. 10 We look at this issue when we think about the 11 investor in a mutual fund -- I'm sorry, in a money market 12 fund, a money market mutual fund. And we think about the 13 public guarantee that we have now extended to those funds, 14 and this lead the AFL-CIO to very strongly oppose the removal 15 of the credit rating NRSRO requirement around money market 16 fund investing that was contemplated by the Commission last 17 year. 18 Now given that importance and understanding that if 19 tomorrow morning the credit rating agency industry vanished, 20 that the cost of investing in fixed income markets would go 21 up significantly for smaller investors. And that perhaps 22 that accounts for some of the enthusiasm among, say, bond 23 market money managers for just such an outcome, because it 24 would turn their business into much less of a commodity and 25 much more of a high value added thing, kind of like being a 84 1 hedge fund that can charge very high fees. 2 Given all of that, AFL-CIO's view is that the 3 question is not do we make rating agencies disappear or shall 4 we open up full competition so that the equivalent of 5 Madoff's accountant in the credit rating agency world can 6 step in and facilitate a money market fund getting yield 7 advantages out of that license that was referred to by 8 Professor Partnoy. That doesn't seem right to us. 9 But on the other hand -- and now I'm going to shift 10 out of my role as defender of the credit rating agencies -- 11 the business model is broken. It is fundamentally broken. 12 It is not a question of a tweak here and a tweak there. The 13 issuer-pays model is not the right way to do this. And the 14 question is what are the alternatives? 15 And I believe it was Mr. Gellert -- but I may be 16 wrong -- went through a list of possible alternatives. We 17 reviewed those in the Congressional oversight panel's report 18 on regulatory reform to the Congress. We looked at a range 19 of options ranging from a strong regulatory model option, 20 which several of my predecessors have described, something 21 like the PCAOB effectively, all the way to essentially 22 turning this function over to a public body. To the extent 23 that we have reliance in our overall financial regulatory 24 system on credit rating agencies, they should be issued by a 25 public body. 85 1 In between are ideas such as mandatory 2 investor-pays, the utility model, co-ops models, models -- 3 when we talk about investor pays, you can either think about 4 that in terms of individual investors seek out individual 5 rating agencies, or you can think about that in terms of an 6 overall assessment across the marketplace that would then be 7 allocated among competitors. Obviously there are competitive 8 implications of each. 9 The point I would like to make to the Commission 10 about all of this is that the one thing that is not an 11 option, in our opinion, is to leave the current unregulated 12 issuer-pays model as is alone in the form it is today. It is 13 broken. At a minimum -- and I think this minimum is 14 necessary whether or not you move further down the line. I 15 mean whether or not you move toward more aggressively 16 encouraging the investor-pays model or whether you move 17 toward other types of approaches analogous to what was done, 18 say, in the analyst area in the post-Enron area -- the 19 minimum is to have a strong independent regulator overseeing 20 the rating agencies, and it is our view that that should 21 either be the Commission itself or a subordinate body that 22 the Commission -- analogous to what Professor Partnoy 23 described, or the PCAOB itself. 24 There are a number of commonalities between the 25 problem of auditor conflicts, the structural embedded problem 86 1 of auditor conflicts, and the problem of rating agency 2 conflicts. I should note the rating agency problem is 3 actually more severe because the absence of anything like an 4 audit committee to manage it as we have in the area of 5 auditors. 6 The question of whether or not -- the idea here is 7 to move to an unregulated world. In my opinion, that is 8 really naive. I do not think that is where we are going to 9 go, and I think it would be harmful to the openness of our 10 credit markets to do that. I think the existence of credit 11 rating agencies is an important factor in the development of 12 the size and sort of universality of our public credit 13 markets. And the challenge is to fix what is broken about 14 that model circa 1970 without reverting to the much more kind 15 of exclusive model of credit markets that would arise in an 16 area where there isn't this essentially public good that the 17 credit rating agencies provide. 18 MR. GALLAGHER: Thank you Mr. Silvers. Professor 19 White? 20 MR. WHITE: All right. First, Chairman Schapiro, 21 fellow Commissioners, thank you for the opportunity to be 22 here and testify on this important issue. 23 As the business school professor on the panel, I 24 had hoped to do a Powerpoint slide presentation, and somehow 25 things have gotten locked here, and I don't know quite what 87 1 to do here. Let me give it a quick effort and if that -- 2 nope. I am at your mercy here. 3 Let me start, and we will see how far we can get. 4 Basically, we went down the wrong path 70 years ago. This 5 goes back to the 1930s, and we had the banking regulators at 6 the time -- here we go. All right, thank you. Great. I 7 don't know, is the Powerpoint slide up on the public screen? 8 It is here. I don't know if it has gotten up to the public 9 screen. 10 Basically, I have provided two quotations, one from 11 the federal code of regulation. I'm not very good at this, 12 I'm not a lawyer, I don't practice law without a license, and 13 there are probably people who could better navigate the code 14 than I. But here is what Frank was calling a regulatory 15 license, what I call the force of law, that regulated banks, 16 savings institutions cannot hold a bond that is below 17 investment grade. And that investment grade, of course, is a 18 rating determination. 19 The other is this disclaimer -- and I'm not picking 20 on you, I'm not picking on Standard & Poor's -- all of the 21 major rating agencies have a similar disclaimer. And I urge 22 you, look at those two statements, and you don't know whether 23 to laugh or cry because there is just such a disconnect 24 between the two. 25 All right, how did we get here? Well, as I started 88 1 to say, we had bank regulators back in the 1930s who told 2 banks you may not hold a bond that is below investment grade 3 as determined recognized rating manuals. In essence, they 4 were outsourcing, they were delegating that safety judgment 5 to a set of third parties. Ratings now had the force of law, 6 and I cannot emphasize the phrase too much: force of law, 7 what Frank has called the regulatory licenses. 8 Over the following decades, that force of law 9 applied to rating agencies' judgments multiplied, expanded, 10 so that we have had that proliferation that Frank talked 11 about. And now, consequently, even unregulated bond market 12 participants would want to pay attention to those ratings 13 because they know the regulated entities must follow it. 14 Again, it is all because of the force of law. 15 Well, this was crystallized in 1975 by the 16 Commission creating the NRSRO category. The Commission 17 unfortunately also became a barrier to entry and compounded 18 by being very opaque in its determinations. Had Sean Egan 19 been here two years ago, you would have heard him talk about 20 all his problems. The major rating firms clearly stumbled 21 badly over the last few years. You know, they weren't 22 perfect before, they had stumbled before, but it had 23 primarily been in terms of their delays in adjusting their 24 ratings, and they have always been slow. 25 All right, where do we go from here? Clearly one 89 1 route, and I can understand the desire, the temptation to 2 sort of grab the rating agencies by the lapels and shake them 3 and say "Do it better. Somehow do it better," and try to fix 4 it by dealing with transparency, dealing with conflicts. 5 That is the route the Commission chose in December of this 6 past year, but I urge the Commission, think hard. Think 7 about the possible pitfalls, the consequences for 8 flexibility, for innovation, for entry in the bond 9 information market and really will this work? 10 Or, there is a second route you could go where you 11 could take the lead, as was proposed back in July 2008 of 12 stop the outsourcing. Cease having ratings having the force 13 of law. The goal of safe bond portfolios should still be 14 there. That is terrifically important. This is not an 15 anything goes. The goal should still be there, but the 16 burden would be on the regulated institution, whether it is a 17 bank, an insurance company, a pension fund, a money market 18 mutual fund, to justify the safety of its portfolio to the 19 regulator. 20 The bond market would be opened up to new ideas, 21 new methodologies, new technologies, new business models in a 22 way that surely hasn't been true since 1975, arguably hasn't 23 been true since 1936. And you would really have a market in 24 bond information and the firms would have to meet a market 25 test. And so, Chairman Schapiro, I urge you, think hard, 90 1 show leadership, follow up with what the Commission started 2 in July of 2008. Thank you for this opportunity. 3 MR. GALLAGHER: Thank you all for your prepared 4 statements. We will move now quickly into the Q&A section of 5 our discussion. We started a little bit late. I don't want 6 to penalize this panel because of the late start, so we will 7 have about slightly under half an hour, so I will ask you to 8 be efficient in your responses. 9 And I know we already have some questions from the 10 Commissioners. In particular, I think Chairman Schapiro has 11 a question. 12 CHAIRMAN SCHAPIRO: It is almost a clarification 13 kind of question. I think I know what Professor White's 14 answer would be to this, but you could eliminate references 15 to ratings throughout the SEC rule set, throughout government 16 programs where they are required, for example, in some of the 17 recent programs. Or you could eliminate the reference to 18 ratings being delivered by NRSROs, which is a pretty small 19 club, as we have talked about. Is there a material 20 difference between those two things? 21 MR. WHITE: I think there is. I mean I think you 22 could say "Look, we want money market mutual funds to have 23 safe portfolios, we don't want anything goes, we don't want 24 them investing in wild and crazy things, so come and justify 25 your portfolios to us," The justification might involve they 91 1 did their own research, or the justification could be they 2 relied on the XYZ advisory firm. And if you want to sort of 3 somehow sketch out in vague, not in precise terms, that it is 4 okay to rely on XYZ but you got to have a basis for relying 5 on firms. So I think get away from NRSRO for sure, but put 6 the burden on the regulated financial institutions to justify 7 the safety of their portfolios. 8 MR. GALLAGHER: Commissioner Walter. 9 MS. WALTER: I would like to ask a follow up 10 question to that to either Professor White or whoever else 11 wants to answer it. It really picks up on a point that Mr. 12 Silvers made, which is the public good particularly for 13 retail investors in giving them a simple shortcutted system 14 which allows direct participation by retail investors, not so 15 much in the structured finance market -- I hope -- but in 16 other areas of fixed income securities. And how do you avoid 17 eliminating that and therefore pushing the little guy out as 18 an investor? 19 MR. WHITE: I'm glad you brought that up. Look, 20 the bond markets for the most part are not a retail market. 21 They are an institutional market, and that is why I think a 22 market in information can work because institutions are going 23 to have memories. They are going to be -- remember that that 24 guy gave me bad advice that year or that guy's predictions 25 haven't worked out all that well, or that guy has a business 92 1 model, gee, I'm not so sure about the conflicts in it. That 2 is something that institutional markets can work out. If 3 this were primarily a retail market, then we are dealing with 4 a whole different level of the need for investor protection, 5 but this is an institutional market. 6 MR. GALLAGHER: Mr. Pollock. 7 MR. POLLOCK: It seems to me, Commissioner, that in 8 the notion that I am thinking of, which is a competition 9 between various models, there is no chance that the 10 issuer-paid rating agencies are going to disappear, and there 11 is therefore no chance that these free ratings for retail 12 investors, to the extent they exist and wish to use them, 13 will be available. I can't forebear mentioning, however, 14 that retail investors, in the form of taxpayers, have become 15 large investors in structured finance assets. 16 MR. GALLAGHER: Professor Partnoy? 17 MR. PARTNOY: Just briefly, I think it is important 18 to keep in mind that ratings themselves, these letter ratings 19 as a mnemonic, as a useful tool, they are meaningless. They 20 don't mean anything. What we actually want investors to look 21 at would be expected probability of default, expected 22 recovery upon default, perhaps correlation and the actual 23 numerical measures underlying them. 24 So if you wanted a shorthand, I think one thing 25 that would be very helpful for the Commission to try to do 93 1 would be to transition people away from reliance on letters, 2 which don't really mean anything, to some other kind of 3 shorthand measure like expected probability of default, if 4 you want to use that. I think there are other shorthand 5 measures that are useful as well, like market measures like 6 credit spreads, and there are other things that don't work as 7 well as mnemonic devices that would be useful as well. 8 But I think we have kind of gotten ourselves into 9 this odd box of relying on letters which don't actually mean 10 anything and vary across categories, municipal, corporate, 11 structured finance. 12 MR. GALLAGHER: Mr. Miller. 13 MR. MILLER: Just wanted to offer a couple of 14 observation about the use of and potential reliance of 15 ratings and regulation. 16 First, it is certainly true that significant 17 segments of the bond market are institutional. There are 18 also significant segments that are not or that have retail 19 components, the municipal market, parts of the corporate 20 bond, and other markets, so it is not a completely clean 21 dichotomy there. I guess just maybe confining my comments to 22 the securitization and structured finance market, I think 23 whether or not ratings are embedded in regulations, I think 24 there is still a danger of overreliance on users of ratings, 25 again, as a result of private investment contracts or 94 1 guidelines. 2 Secondly, and kind of along the theme that NRSRO 3 status and identification of ratings and regulations as a 4 license, I think it is important to view the use of ratings 5 and regulation as a necessary but certainly not a sufficient 6 condition for any fiduciary in the bond market who is making 7 investment judgments. They are required to make those 8 judgments in a way that do not simply rely upon ratings, and 9 so I think it goes too far to suggest that simply having a 10 rating requirement in regulation is as far as anyone would go 11 there. 12 And I think that overall, we do not support a 13 wholesale removal of the use of ratings in regulation. I 14 think that would be a very difficult and intractable process, 15 and we think the better way to address overreliance, if that 16 is the real thrust of those kinds of proposals is, again, to 17 provide more information that would provide users of ratings 18 with the ability to distinguish and make judgments that 19 relate to the accuracy and integrity of those ratings. 20 MR. GALLAGHER: Thank you Mr. Miller. Commissioner 21 Aguilar, I think you had a question? 22 MR. AGUILAR: Well it changed as he was speaking. 23 I'm trying to get back to actually the issue of the 24 panel, which was focused really on competition. We are at 25 the edges of it, but we have strayed a little bit. 95 1 The earlier panel spoke in terms of the fact that 2 the more competition you have, the more rating shopping you 3 have and the more negativity that has involved. At the end 4 of it all, I'm more focused on the accuracy of ratings at the 5 end of the process. And although I have some -- although I 6 do care how we get there, I care less how we get there than I 7 do about the accuracy of the rating process at the end. And 8 I guess I would like some focus on that issue, as to the 9 various suggestions that you made, how that leads us to more 10 accuracy in the ratings and if it enhances competition, so 11 much the better. 12 MR. GALLAGHER: Mr. Berman? 13 MR. BERMAN: Yes, I would agree with your comments. 14 I think that the key to increasing accuracy, I think, is to 15 get more and more people more and more information to enter 16 into this market. 17 I actually think that one of the dynamics that we 18 have talked about, payment models, that sort of goes 19 without -- hasn't been said is around sort of the 20 relationship and particularly around the structured market 21 between the issuers, primarily investment banks, and the 22 rating agencies and the dynamic of the significant number of 23 people who left the rating agencies to go to the investment 24 banks and that you had a drain of talent, I believe, from the 25 rating agencies. 96 1 And so you need to get the business of being in 2 ratings to be attractive from the economic point of view. 3 You have to feel that the people there have open -- 4 transparency is a key word, and I think that there is no one 5 here who is going to question the importance of transparency 6 in that -- and that you have no created some barrier such 7 that high quality talent, which obviously right now in the 8 structured world, there is a lot of high quality talent 9 looking for jobs, who I think can contribute significantly to 10 the analysis of different markets, products, et cetera, and 11 allow them to get that into the marketplace. 12 MR. GALLAGHER: Mr. Gellert. 13 MR. GELLERT: I agree that the core issue is about 14 accuracy of ratings, and I think you get accuracy of ratings 15 when you create a more level playing field that allows more 16 players to come in and compete and provide product, and to do 17 it in innovative ways with new tools, and to challenge not 18 just the supremacy, if you will, of a group of established 19 raters, but also the models by which they create their 20 ratings. And a lot is said about the issuer-pay model, but 21 we need to keep in mind that the issuer-pay product is one 22 that is "rating through the cycle," which is putting a 23 greater emphasis on stability than on accuracy. 24 And to provide something that is adding greater 25 value to the market for all of these purposes that we are 97 1 discussing and were discussed earlier, we need to have 2 ratings out there that are more actionable and that have 3 more -- that have a greater set of uses than just 4 benchmarking a current portfolio, particularly if you may 5 have a preference to have that portfolio to have less 6 volatility than more volatility. 7 MR. GALLAGHER: Thank you Mr. Gellert. Mr. 8 Silvers? 9 MR. SILVERS: I want to try, Commissioner, to blend 10 my answer a little bit with the prior discussion to your 11 question. 12 We at the time seem to be having a discussion about 13 whether the solution to our regulatory crisis at the moment 14 is more deregulation. When people talk about taking some of 15 the current framework for, essentially, licensing, rating 16 agencies off the table, that is what they are talking about. 17 And discussions about the virtues of flexibility and 18 innovation have precisely the same sort of vague optimism 19 that brought us our financial crisis. 20 The question is, in order to get accurate 21 ratings -- or more accurate ratings, you will never get fully 22 accurate ones -- is what is the appropriate mix of 23 intelligent regulation with managed market dynamics? And I 24 would suggest to you that you have to have, in this area of 25 gatekeeping, gates to being a gatekeeper, as we do in every 98 1 other area of gatekeeping in our financial system. You can't 2 just walk into court and say you are a lawyer or walk into a 3 bond deal and say "Hey, I'm a bond lawyer." You cannot walk 4 in and say "I'm an auditor because I say so." We would be 5 behaving extraordinarily recklessly if we created a world in 6 which you could just walk in and say "I'm a credit rating 7 body and we will stop a government guarantee behind a money 8 market fund that I say is safe." 9 That is, I think, a floor to your question. Once 10 you set that floor, then I think a number of the suggestions 11 of my fellow panelists about the notion that the letter 12 ratings maybe aren't that useful and that we maybe have some 13 more sophisticated ways of conveying the kind of information 14 credit rating agencies produce makes a lot of sense to me. 15 The notion that maybe we want multiple measures makes a lot 16 of sense to me. The notion that I think that -- Ethan Berman 17 had a list of items -- the notion that the current way which 18 we keep the gate may be more about kind of special interest 19 stuff and less about a thoughtful way of protecting the 20 public, that may be true as well. 21 But if we take the gate down, we are looking at the 22 giant mess that has just been made and saying "Let's do it 23 again." 24 MR. GALLAGHER: Thank you Mr. Silvers. 25 Commissioner Paredes, I think you have a question. 99 1 MR. PAREDES: A number of folks have mentioned the 2 prospect of overreliance on ratings, and there is also, of 3 course, a lot of discussion on the issuer-pay model and the 4 potential conflicts there. And the question I have is to 5 bring the user side of this into the equation. 6 So much focus is on the role of the regulators and 7 the regulatory regime. A lot of focus then goes to the 8 rating agencies, but there is somebody else who is part of 9 this process, and that is the user. And at least implicitly 10 in much of the critique of the issuer-pay model and the 11 conflicts is I think a presupposition that there has to be 12 some breakdown on the user side of things. 13 And this is a circumstance where there is an oddity 14 for me because the transparent -- the conflict of interest to 15 the extent it exists is transparent. I mean we were talking 16 this morning about conflicts of interest, and much of the 17 discussion in the space centers around that, so it is not as 18 if folks are sitting back saying "There is a conflict of 19 interest. Wow, I didn't know. If I had known, I would have 20 done something different." Rather, it is there. 21 So what is the challenge, is there challenge, is 22 there a breakdown on the user side? I think to say that we 23 have ratings is a little too simplistic, because you may 24 decide to focus on the rating because it serves a particular 25 regulatory purpose, but if it doesn't serve a good business 100 1 purpose, you would think your business judgment may trump. 2 And yet I think a lot of folks are saying that perhaps isn't 3 the case, and so I'm wondering if folks can speak to the user 4 side of things. 5 MR. GALLAGHER: Professor Partnoy, and then Mr. 6 Pollock. 7 MR. PARTNOY: Commissioner Paredes, I think you are 8 absolutely right, and I think one of the keys I hope the 9 Commission will keep in mind is how to work a transition not 10 only on the regulatory side but on the investor side as well 11 to come up with viable substitutes. I think both regulators 12 and users feel like they are in a trap. And for some, it may 13 be that they are not thinking creatively enough about 14 substitutes, for some it may be that they are crippled by 15 their regulatory reliance. But I hear over and over again, 16 and I think you see echoed in the white paper and you will 17 hear from the council later, I think people are looking for 18 guidance in helping the transition away, both on the 19 regulatory side and on the user side. 20 And I think you are absolutely right that we need 21 both of those to move together. So if they could move in 22 lockstep or if there could be a sort of -- this is why the 23 oversight panel might be a way of doing this, to sort of 24 engage in the incremental process of moving away from 25 reliance, not just at the regulatory level, but at the user 101 1 and investor level. I think you are absolutely right. That 2 is a big part of the story about what happened over the last 3 several years. 4 MR. GALLAGHER: Thanks Professor. Mr. Pollock? 5 MR. POLLOCK: Commissioner, one of my old friends 6 said the only people interesting to talk to on any topic are 7 those who have conflicts of interest because they are the 8 ones who know enough about it. 9 So I'm less worried by the conflicts of interest 10 than I am about the creation by regulation of a government 11 sponsored or indeed a government mandated cartel. Once you 12 have done that, that makes the conflict issue much worse in 13 my judgment. And as I look at this situation, as many others 14 around the regulation system, I will draw the opposite irony 15 from that that Damon did. We have all of these problems 16 caused by regulation and we think the answer is more 17 regulation. 18 Deven Sharma, I think in his testimony, rightly 19 said that we "need to avoid inadvertently encouraging 20 investors to depend excessively on ratings rather than 21 treating them as they should, one of the many inputs in 22 decision making." We need even more to avoid intentionally 23 encouraging investors to depend on ratings, and that is what 24 I think the regulations did. 25 MR. GALLAGHER: Thank you Mr. Pollock. Mr. Miller? 102 1 MR. MILLER: Just briefly, I think, Commissioner, 2 you have hit upon a very important point in terms of the user 3 side of the equation and the demand side, and I would just 4 observe very simply that I don't think there has been such a 5 breakdown on the user side that their choice has been kind of 6 effectively prevented from being manifested. 7 And to that end, I think that to the extent 8 investors or other users of ratings would want either 9 exclusively or predominately for there to be subscriber-paid 10 models, I think we would be seeing much more of them in the 11 market than we see today. And I'm not commenting positively 12 or negatively, I'm just saying in terms of steps to either 13 dictate or favor or disfavor particular models, I think we do 14 need to be very sensitive to the self-expressed views and 15 usage of the primary users of ratings, and most prominently 16 including investors. 17 MR. GALLAGHER: Mr. Gellert and then Mr. Silvers, 18 if you could answer very quickly, because I have to move on 19 to one last question from Commissioner Casey, and then 20 unfortunately we are reaching the end here. 21 MR. GELLERT: I will do my best for the 22 Commissioner. 23 I think we have a -- we tend to oversimplify the 24 issuer-pay conflict into being an assumption that if they are 25 paid for by the issuer, the ratings will simply be higher, 103 1 because yes, everybody knows that that is a problem. But 2 embedded in that same system are other biases that users may 3 not be as aware of. The assumption that the companies are 4 too big to fail, that there is a size bias, that there is a 5 brand name bias, that stable ratings should mean that 6 companies shouldn't go too quickly through thresholds like 7 investment grade, either down or up. 8 And there are other issues like issuers can cause 9 their ratings to be withdrawn. So an issuer may very -- a 10 user may very well understand that there is this potential 11 conflict, but the conflict has a ripple effect and effects 12 them in other ways, including a meltdown situation and an 13 issuer pushes to have their rating withdrawn, then that user 14 is out in the cold, regardless of whether they knew there was 15 a conflict or not. So it is a much broader issue than simply 16 the pay structure. 17 MR. GALLAGHER: Mr. Silvers. 18 MR. SILVERS: I would actually echo that comment. 19 That is very on point. 20 I want to make two very brief comments. One is the 21 Commission and the public really can make up their own minds 22 as to whether we are living through a crisis of deregulation 23 or a crisis of regulation, and obviously Alex and I have a 24 different point of view about that. I think I am willing to 25 submit that to the jury of the Commission and the public that 104 1 question. 2 We really are repeating the mistakes of the very 3 recent past. If we look at something like the way in which 4 individuals participate in fixed income mutual funds, 5 particularly money market funds, and assert that that is 6 going to produce the kind of market discipline within any 7 timeframe that we care about, in the absence of effective 8 regulation of the credit rating agency market, it will 9 produce the opposite. If you let this dynamic loose, as 10 members of the panel suggest, the investor is not the 11 investor in these situations. The investor -- it is the 12 public that will move money in the belief they moving it into 13 secure places, and in fact they will be relying on the 14 equivalent of Bernie Madoff's rating agency. 15 MR. GALLAGHER: Thank you Mr. Silvers. 16 Commissioner Casey, I hope by now you have 17 re-formatted your question into a yes or no question. We 18 have about three minutes here. 19 MS. CASEY: I'm going to try as much as possible. 20 I guess I want to kind of step back a little bit 21 and take stock on where we are today, and whether you want to 22 characterize it as a deregulatory environment or a regulatory 23 one. Specifically I want to talk about the law that we are 24 operating under that was passed back in 2006 and that became 25 effective in 2007, and the various rules that the Commission 105 1 has put forward and adopted, some of which have been adopted, 2 some of which are still being proposed and are currently 3 before the Commission for consideration. And I think 4 Professor White spoke about one of them. Some of you have 5 also alluded to some of the rules that have already been 6 adopted. 7 But I guess I would ask two things. First off, 8 what is the appropriate amount of time for us to take a 9 measure of how effective these rules have been or can be 10 given the fact that they have only been -- we have only had 11 the authority for a little over two years and we have only 12 just recently adopted the most recent rules. So what kind of 13 a time table is an effective measurement to determining 14 whether or not we are achieving or seeking to achieve the 15 goals of the law, which is to promote competition, 16 accountability, accuracy in ratings? 17 And then as a follow up to that, given the rules 18 that we have adopted, which have been intended to get to some 19 of the conflict issues here, which have been identified 20 specifically in the issuer-pays model, but also to try to be 21 sensitive to increasing competition by promoting the ability 22 of other rating agencies, particularly subscriber-pays 23 models, to be able to compete effectively, talk about 24 enhanced transparency, which maybe gets to some of these 25 issues as well. If you could identify either a specific rule 106 1 change or a change in the law that you think is necessary to 2 enhance the credibility and accuracy of ratings, what would 3 those be? 4 MR. GALLAGHER: Professor White. 5 MR. WHITE: All right. Let me address both your 6 questions and the previous set of questions. 7 Yes, of course it is all about the users, but in a 8 more open environment, I think we really can rely on 9 competition. We haven't had it, and that is why the problems 10 arose. In a more open environment where users have 11 memories -- remember, this is not a one shot, just this 12 statement, and then everybody disappears. People are going 13 to have memories, they are going to remember that this guy 14 did perform well in the past or didn't perform well, and with 15 those kinds of memories, various models, including the 16 issuer-pay model, can work in this kind of marketplace. And 17 again, it is predominately an institutional model -- market. 18 The one thing, of course, would be pull back, 19 withdraw the delegation, show leadership in this respect. 20 MR. GALLAGHER: Thank you Professor. Mr. Berman, 21 50 words or less. 22 MR. BERMAN: Okay, there are two things I think you 23 need to do. First, you need to have all people's ratings 24 mean the same as others. So if I am competing, my rating 25 means the same as the incumbents. And second, I have the 107 1 same access to information that the incumbents have. If you 2 allowed that to happen, the market will decide who is 3 providing value into the marketplace. I don't think it is 4 who pays, et cetera, as long as those two things are met. 5 In terms of how do you measure that, I don't think 6 you are going to be able to measure ratings accuracy for 7 years. I think that is -- we all know that given what we are 8 trying to do here. I would say much more, what are the 9 dynamics of the marketplace, where is revenue going, and what 10 is the -- has anything changed vis-a-vis who are the people 11 people continue to be using? I do think you are going to 12 hear from investors they want ratings. Again, that is your 13 next panel, we will see what happens, but I do think you will 14 hear that. 15 MR. GALLAGHER: Thanks Mr. Berman. I'm going to go 16 end with Mr. Gellert, Partnoy, and Pollock, you each get 30 17 seconds. 18 MR. GELLERT: How long to assess? I guess I would 19 assess the efficacy of the '06 legislation by asking you the 20 question how many NRSRO applications are before you now? A 21 year ago, there were none, and I would imagine there aren't a 22 lot now. 23 The two things I would do is I would drop the three 24 year wait period because it blocks -- 25 MS. CASEY: Can I -- I don't want to interrupt you, 108 1 but we have approved several since the passage of the law. 2 MR. GELLERT: Yes, you have approved three and 3 seven were essentially grandfathered for all intents and 4 purposes. Three new ones, two of which had been applying for 5 a long time, and one of which was a relative newcomer, and 6 that is excellent momentum, but it is not enough because it 7 is, as I said earlier, there are disincentives to people 8 wanting to be an NRSRO. 9 But I would drop the three year rule because it 10 clearly blocks de novo competition, and I would drop the 11 application by asset class. It directly lead -- it directly 12 contributed to the current credit crisis. 13 MR. GALLAGHER: Professor Partnoy. 14 MR. PARTNOY: To answer your first question, I 15 think without additional changes, we could wait an infinite 16 amount of time and we wouldn't see sufficient reform. And I 17 would predict if there aren't other changes, that within five 18 years, we would see a resurgence of credit rating-based 19 transactions like what we saw in the 1990s, structured 20 note-type transactions. And I don't think it will take long, 21 maybe five years, without the two kinds of more dramatic 22 reforms that I suggested. 23 MR. GALLAGHER: Mr. Pollock. 24 MR. POLLOCK: Since the effectiveness of the rules 25 is a question of the future, you clearly need a rating agency 109 1 to rate your various rules as to how good they were. 2 I would say approving more NRSROs was definitely 3 the right thing to do. Pushing toward making investor-paid 4 agencies publicly disclose their ratings is the wrong 5 direction. And a possible additional direction might be more 6 consultation with other regulators since there are many other 7 regulators involved in these ratings issues than the SEC 8 itself. 9 MR. GALLAGHER: The Chairman trump card has come 10 out again. Chairman Schapiro. 11 CHAIRMAN SCHAPIRO: Sorry, and I will try not to be 12 too long. 13 When the college of regulators, the community of 14 regulators in the U.S. settled the major cases against 15 investment banks for distributing conflicted research 16 produced by their own research analysts who also -- with 17 respect to their own investment banking clients, one of the 18 provisions of that settlement was a requirement that the 19 investment banks, when they distributed their own research, 20 were also required to distribute independent research and pay 21 for that to be distributed. And it created a cottage 22 industry of independent research firms. 23 Would something like that make any sense for us to 24 explore in this context where you would have a requirement 25 for a second or a third rating to be paid for out of the 110 1 money that goes to the primary rater so that there would 2 be -- would it help us jumpstart competition in this space, I 3 guess is what I'm really asking. 4 MR. GALLAGHER: Mr. Silvers. 5 MR. SILVERS: Madam Chair, that I think is 6 something definitely worth exploring, one. And two, I think 7 it points a general direction for the Commission in this 8 area, which is to look very closely at that experience and 9 the experience of the auditing profession since 10 Sarbanes-Oxley because these are so closely analogous to this 11 circumstance. And as I indicated earlier, I think in certain 12 ways this is more serious, but the structure of the problem 13 is very similar, and I think there is a lot to learn for 14 those experiences and I think your inclination to look at 15 that particular learning is correct. 16 MR. GALLAGHER: Thank you Mr. Silvers. Mr. 17 Gellert? 18 MR. GELLERT: I think there are valuable lessons to 19 the global settlement. We happen to be a provider to it, so 20 I am quite familiar with it. I think the risks are that it 21 becomes a regulatory box checking and people say "Fine, I 22 will satisfy this for the period of time in which it is 23 enacted." And yes, it did contribute to the growth of the 24 independent research market, but in July of this year when it 25 rolls off, we are going to see a consolidation in the 111 1 independent research market. I would also point out 2 ironically that possible the largest beneficiary of dollars 3 from that global settlement was Standard & Poor's equity 4 research division. 5 MR. GALLAGHER: Thank you Mr. Gellert. I will 6 anybody else a chance to chime in before we close things down 7 here. Mr. Miller. 8 MR. MILLER: Just very briefly. We haven't thought 9 about the specific proposal that you raise. I think it is an 10 interesting one. I would say more generally though, 11 diversity of ratings opinion available to users of ratings is 12 a good idea. I think that is fundamentally what we -- one of 13 the things we would try to achieve through greater 14 competition. On the other hand, in terms of the particulars 15 of the proposal, I think it would be important to examine and 16 consider closely the benefit cost in terms of the benefit to 17 users and the cost of those who would be paying for those 18 additional services. 19 MR. GALLAGHER: Thanks Mr. Miller. Mr. Berman. 20 MR. BERMAN: Just one last comment on the economics 21 of this. It is in our opinion that the economics of an 22 investor-pay model, where the dollars come out of investors, 23 is significantly different from an issuer-pay model, and so 24 you are going to see, similar to the independent research, 25 where the size of the companies only get so large because the 112 1 investors are not willing to pay significant dollars. They 2 tend to get bundled in other services. I think that will be 3 a similar phenomenon you will have to deal with here, and you 4 won't get the kind of scale I think necessary or 5 sophistication among a very large player to simply to provide 6 to investors investor-paid, unless it is through a different 7 model, such as the other ones that have been proposed. 8 MR. GALLAGHER: Okay, I think we are going to have 9 to close it down. I'm sorry, Commissioner Casey. There is a 10 little bit of mingling that can happen here in the next 11 couple of minutes, but I think we are going to have to move 12 on. 13 MS. CASEY: I apologize, I just have one final 14 question, and it is actually a follow on to a point you made, 15 Professor Partnoy, about short of the short term approach to 16 addressing the underlying problem, which is to have greater 17 oversight and accountability, and to have some sort of an 18 independent board that would achieve that. And I guess my 19 question is, in as much as we continue to look to the audit 20 model as somehow one that we should be considering, and I 21 question what that means for competition at the end of the 22 day. 23 But how do you ensure that even in the short term 24 by having some sort of a stronger oversight board that you 25 don't continue to foster the undue reliance that you are also 113 1 mostly sensitive about, that you are not sending that signal 2 to the market that they should continue to over-rely and be 3 over-dependent on ratings? 4 MR. GALLAGHER: Mr. Pollock gets the nod for 5 nodding. 6 MR. POLLOCK: Thank you. No, no, I would like to 7 answer, Commissioner Casey, because I think you have hit on 8 the key weakness in the proposal, and I think the way to 9 remedy it, and I think it would be absolutely crucial to have 10 the charge of the committee -- and if it is done by 11 legislation or if it is done by ruling, I don't know exactly 12 how it would be done -- the charge of the committee, the 13 primary charge would have to be to migrate away from 14 regulatory reliance to market measures so that the focus of 15 the committee is not so much sort of supplanting in a general 16 oversight role. And in that sense, I think it would be 17 different from PCAOB and other kinds of models. 18 You know, you could have a finite term for the 19 committee. I know that would be controversial, but you could 20 have, for example -- this is ten years. You have ten years 21 to get it done. I know that many people have said we should 22 just say in ten years all these regulatory licenses explode, 23 and now you let 1,000 flowers bloom and try to figure out how 24 to do this, right? Maybe a more gradual way of doing that 25 would be to have an oversight committee that is just going to 114 1 last ten years and has this limited charge, and it is 2 overseeing the functionality, but its primary focus is on 3 making this transition. So maybe that addresses some of the 4 concerns that you have. 5 MR. GALLAGHER: Mr. Silvers, do you want to weigh 6 in on this before we -- 7 I'm going to take this opportunity then to thank 8 you all for being here. On behalf of the Division of Trading 9 and Markets, this has been terrific. We don't often or ever 10 get to speak for the Commission, but I think I can speak for 11 the Commission in thanking you for coming down here. We 12 thought this was a great exchange. We appreciate the time 13 spent getting here, being here. We thank you for your 14 written comments and we look forward to talking to you as we 15 think about these issues over the next year. 16 If the panelists from this panel and other panels 17 could meet down here, that would be terrific. For members of 18 the audience, I would ask you to be back at 1:20 promptly. 19 We are going to get going again on panel three. Thank you. 20 (Whereupon, at 12:44 p.m., a luncheon recess was 21 taken.) 22 A F T E R N O O N S E S S I O N 23 PANEL THREE: USERS' PERSPECTIVES 24 MR. GALLAGHER: Okay, why don't we get started 25 here. Welcome to panel three, user's perspectives. 115 1 My name is Dan Gallagher, I'm the Deputy Director 2 in the Division of Trading and Markets here at the SEC. 3 Thank you for being here today. 4 You have probably already been in attendance for 5 earlier panels, but in case you weren't, just a few rules of 6 the road. Generally speaking, the Commission wants it to be 7 a free flow of back and forth between you and them, question 8 and answers. However, we will start off here after I 9 introduce you with prepared remarks. We ask that you limit 10 them to four minutes. When you have one minute left, I will 11 waive this little yellow piece of paper up, so please look up 12 if you can. I mentioned last time that we have enforcement 13 folks upstairs and it seemed to work, so I'm going to repeat 14 that warning. So please be mindful of the time. After that, 15 we will move into a Q&A session with the Commission. 16 So why don't I get going and introduce the members 17 of the panel. First, Ms. Deborah A. Cunning, Executive Vice 18 President and Chief Investment Officer for Federated 19 Investors Money Market Funds, on behalf of the Securities 20 Industry and Financial Markets Association. Mr. Alan J. 21 Fohrer, Chairman and CEO, Southern California Edison. Mr. 22 James A. Kaitz, President and CEO, Association for Financial 23 Professionals. Mr. Kurt N. Schacht, Managing Director, 24 Centre for Financial Market Integrity of the CFA Institute. 25 Mr. Gregory W. Smith, General Counsel of the Colorado Public 116 1 Employees Retirement Association. Mr. Bruce Stern, Chairman, 2 Government Affairs Committee, the Association of Financial 3 Guaranty Insurers. And finally, Mr. Paul Schott Stevens, 4 President and CEO of the Investment Company Institute. 5 So in the order that I just introduced you, I will 6 ask you to please read your prepared statements, and again, 7 keep it to four minutes. Thank you. 8 MS. CUNNINGHAM: First of all, let me thank you 9 Chairman Schapiro, the Commissioners, and Deputy Director 10 Gallagher for allowing me a spot at the table today during 11 this rating agency roundtable discussion. 12 I am here representing both Federated Investors, 13 which is a mutual fund company based in Pittsburgh, 14 Pennsylvania. We manage a little over $400 billion in assets 15 at this point with roughly three-quarters of those being in 16 the money market fund space. I am also representing the 17 credit rating agency taskforce of SIFMA, which is the 18 Securities Industry and Financial Markets Association, which 19 is a taskforce that I co-chaired along with Boyce Greer of 20 Fidelity Investments. 21 Through my role as Chief Investment Officer of 22 Federal Investors money market funds, I have a rather unique 23 perspective of being both an issuer as well as an investor, 24 which we have been referring here today as either a 25 subscriber or a user in the credit rating agencies process. 117 1 As an issuer, I manage many different money market funds that 2 pay to have various rating agencies rate their 3 creditworthiness. This is done for the benefit of the 4 underlying shareholders and investors in those products. As 5 an investor, I subscribe and I pay quite hefty fees to obtain 6 the credit research and the ratings that are the result of 7 the rating agencies' analysis of the various issuers that I 8 purchase into those same money market funds. 9 Given both of these perspectives, I would like to 10 put forth my first main point, which is that rating agency 11 rules should be uniform, they should be consistent no matter 12 what the business model is of that particular rating agency. 13 The only real way that I, at this point, segregate rating 14 agencies is by their designation, certainly not by their 15 business model. 16 If a credit rating agency has an NRSRO designation, 17 that puts it in a different category, and we have discussed 18 that a little bit in both of the panels this morning. And 19 that is based on the regulatory implications, but certainly 20 not on the quality of the research or the product that comes 21 out of that credit rating agency. Any other type of credit 22 rating agency segregation in my mind just doesn't make sense, 23 and even from a global perspective, the NRSRO versus credit 24 rating agency designations are pretty well understood. 25 As a manager of money market funds, I am required 118 1 by Rule 2a-7 to make a minimal credit risk determination as 2 well as a high quality determination. This has to be done 3 independently for every issuer that I use within the 4 portfolios that we manage. This brings me to my second main 5 point, which is that the credit rating agency research and 6 the ratings are a valuable input into that minimal credit 7 risk determination, but they represent data, they represent 8 reference points in my independent analysis. They are not 9 the end result. This was expounded on the SIFMA 10 recommendations that were brought forth on a credit rating 11 agency basis back in the summer of last year. 12 NRSROs, which are the designated credit rating 13 agencies, are available in this legislation to set a floor 14 for the universe of the issuers that I can choose from when 15 I'm looking to make an independent assessment for those that 16 I'm using within my portfolio, but there are a reference. It 17 doesn't mean I use all the issuers that simply meet that 18 floor. And as referenced in the report of the money market 19 working group in their March report, ratings do provide a 20 clear reference point, but must only be used as input to the 21 independent analyses. 22 This brings me to my third point, which is to 23 reference that ratings should not be eliminated in the 24 various regulations that rely on these ratings to set a 25 floor. We use ratings both from an NRSRO standpoint as well 119 1 as a CRA standpoint as part of our independent review 2 process. 3 My final point in this roundtable emphasizes the 4 need for greater transparency and disclosure, again, points 5 that have been brought up in the other panels, for all the 6 credit rating agencies. In the February 2nd re-proposing 7 release, the SEC proposed requiring NRSROs to disclose 8 ratings history for 100 percent of their current issuer-paid 9 ratings on or after June 26th, 2007 though at a 12 month lag. 10 And while this is of importance to both Federated as well as 11 SIFMA, we actually are looking for increased disclosure by 12 the NRSROs on an analysis basis, rather than on the ratings 13 history, to provide more informational content to the 14 underlying investors of portfolios. 15 Thank you for this opportunity. 16 MR. GALLAGHER: Thank you Ms. Cunningham. Mr. 17 Fohrer. 18 MR. FOHRER: Thank you. I'm one of the issuers 19 here today, maybe only the corporate issuer, which puts me 20 somewhat in a unique position. One of the things our group 21 said is "Make sure, as an issuer, you don't say anything that 22 upsets the SEC." The second group said "Make sure you don't 23 say anything that upsets the rating agencies," so that is a 24 tight line to walk. 25 Just a perspective on our industry and our company, 120 1 we are one of the most capital intensive if not the most 2 capital intensive industry in the country. It takes billions 3 of dollars to satisfy our customer's needs. For example, 4 over the next five years, our capital budget at Southern 5 California Edison Loan is about $20 billion. We also buy and 6 sell about $7 billion worth of power a year in addition to 7 issuing long term contracts to other parties to build power 8 plants for our customers. The bottom line is we live and die 9 on the availability of capital, and it is not something we 10 can do without for any period of time, and it is something 11 that has a profound impact on the cost and reliability of our 12 product. 13 The reason I give this perspective is because any 14 changes that are made that create uncertainty in the 15 financial markets, even for a short period of time, would 16 have a big impact on our customers and our ability to meet 17 their needs. Let me put a couple things in perspective in 18 how we look at ratings. We think ratings is one leg of a 19 three-legged stool. The first is very clear and concise 20 public disclosures. We think obviously the changes that have 21 been made over the last several years in terms of 22 Sarbanes-Oxley and greater transparency of a company's 23 financial statements is a critical part. 24 Second is the ratings where the rating agencies 25 come in. They do an independent analysis of the companies, 121 1 including the use of confidential information that is not 2 public, and they reach a decision. That is a critical tool. 3 And the third is the investor's own analysis. Now we are on 4 both sides of that. We have people that both reach 5 conclusions about whether they will buy our securities. We 6 reach conclusions every day about whether we will use credit. 7 All three of those are important, not just one. 8 The ratings have to be credible and understandable, 9 they have to be consistent. One of the issues we face is 10 that we have to acquire a large amount of capital, as I 11 indicated, and if we are considered an A the one day and a B 12 the next or you see wild changes in ratings, our ability to 13 acquire capital consistently to support the kind of program 14 would be very difficult. So consistency in those ratings is 15 tough. 16 One of the things that has been talked about and I 17 do want to echo is improvements in the rating process. To 18 the extent that the rating process is something of a black 19 box, I don't think it is easy for either the investors that 20 we are trying to attract to bring their money, it is not easy 21 for the companies and how to present their story, it is not 22 easy for the regulators to know how they regulate companies 23 so that we have the ability to raise capital. 24 The next point I want to raise is competition. It 25 sounds good to have a lot of rating agencies competing. One 122 1 of the key things that does make our system work pretty well 2 is the ability to have people that are very knowledgeable in 3 our industry and that we share information with. I don't 4 think we could do that with ten parties. It is hard enough 5 to do with two. 6 And finally, we want to make sure when people talk 7 about liability and imposing greater liability on rating 8 agencies that they don't push the agencies to be so 9 ultra-conservative that our ability to attract capital is 10 impaired. So we think the system can be enhanced. We want 11 to make sure we are focused on what problem we are really 12 solving, and again, thank you for letting me be here today. 13 MR. GALLAGHER: Thank you Mr. Fohrer. Mr. Kaitz? 14 MR. KAITZ: Good afternoon Chairman Schapiro, 15 Commissioners, and distinguished guests. AFP welcomes the 16 opportunity to participate in today's roundtable to examine 17 the oversight of the credit rating agencies and to provide 18 our insights and suggestions from the end user perspective. 19 As the global resource and advocate for the finance 20 profession, AFP serves over 16,000 members who manage and 21 safeguard the financial assets of more than 5,000 U.S. 22 companies and organizations, and I have a significant stake 23 in the outcome of the reform of rating agency practices and 24 regulation. We believe the credit rating process and 25 confidence in those ratings are vital to efficient global 123 1 capital markets, and it is necessary that the ratings 2 produced by the NRSROs are sound and reliable. 3 Unfortunately, we believe that the current credit rating 4 agency processes and business models are neither sound nor 5 reliable. In fact in our view, the system is broken. 6 Since 2002, AFP has been a vocal advocate for the 7 reform of the credit rating industry. Regrettably, passage 8 of the Credit Rating Reform Act of 2006, which we were very 9 supportive of, has in our view not lead to real competition 10 in the credit ratings market. The recent market turmoil and 11 disruptions in the credit markets provide strong empirical 12 evidence that the model and structure of the rating agencies 13 is flawed. We believe the changes alone to the current rules 14 will not adequately address these flaws. 15 I want to say parenthetically in listening to this 16 morning, sometimes when you listen, it is not always in our 17 view what is happening out there in reality in corporate 18 America. And from our perspective, from what we have heard 19 from our members, if they don't have faith in the credit 20 rating system, they are not going to participate in these 21 markets. All you have to do is talk to a corporate CFO or a 22 corporate treasurer and ask them what they are investing in 23 today. It is treasuries. They are going to stay with the 24 most liquid, safe investments right now until they have 25 confidence in our capital markets, and this is I think a real 124 1 reality in the marketplace right now. 2 Today we are going to propose two potential 3 alternatives for just discussion purposes to the current 4 business models for rating agencies that we believe are 5 worthy of your consideration. First, creating a standalone 6 model with the only activity of the enterprise, the rating 7 agency, would be to provide credit ratings, avoid the 8 conflict of interest. And second, implementing rules that 9 direct government support to NRSROs that are an alternative 10 to a market dominated by S&P and Moody's, and a number of 11 those rating agencies were up here today. 12 Our first proposal is to implement a model that is 13 similar in nature to a utility to support rating 14 organizations whose sole business purpose would be provide 15 credit and reliable ratings. These new rating organizations 16 could be financed, as an example, by a transaction fee, which 17 could be leveled both on investors and on issuers, so we want 18 to address that issue as well. Under this model, conflict of 19 interest issues that currently exist could be mitigated. 20 However, the SEC would still have to exercise vigorous 21 oversight of the rating agencies to ensure that they are 22 fulfilling their mission of providing reliable ratings. 23 The second option, very quickly, which we have 24 asked you to consider, would be for the U.S. government to 25 require all federal programs that require credit ratings as 125 1 well as any business that has had a capital infusion from the 2 U.S. government to utilize alternative NRSROs as credit 3 analysis providers. As you know, the TALF program right now 4 at the Federal Reserve has identified trillions of dollars of 5 debt, and they have identified two rating agencies to rate 6 that debt, hundreds of millions of dollars, and quite 7 frankly, in our view, that is rewarding the rating agencies 8 for failure in the marketplace, and we want to encourage 9 other federal government agencies to use other alternative 10 rating agencies in this process. 11 And I know I'm probably hitting the time, so I have 12 given you two we think very solid solutions for your 13 consideration as you move forward. 14 MR. GALLAGHER: Thank you Mr. Kaitz. Mr. Schacht? 15 MR. SCHACHT: Good afternoon. Thank you very much 16 for the invitation to CFA Institute. We appreciate being 17 heard as a voice for users and investors, a voice that would 18 say to you, in four minutes or less, that we are very 19 confused, that we don't trust credit rating agencies very 20 much, and that we too have behaved badly in this crisis. I 21 will talk a little bit about each of those. 22 We are confused because I think we are sort of in a 23 regulatory no-man's land right now. We are neither fully 24 regulated as credit rating agencies, nor have we fully 25 removed from regulatory reference an application of this 126 1 industry. And I know that a lot of our members are looking 2 to you all here at the SEC to sort this out. What is the 3 role, what should be the role? 4 There was some discussion this morning about the 5 pre-1980s model. That has made some sense to us. We have 6 talked a lot about that. It was simple, it was transparent, 7 it was open to competition. Investors were fully on notice 8 that the quality, validity, and accuracy are not examined or 9 sanctioned and there is no sort of regulatory license, as Mr. 10 Partnoy had referred to earlier, free reign in terms of your 11 methodology and your process and your conflicts of interest 12 codes and so forth. So that was one model that we have been 13 talking about in our organization. 14 The second model, if we are going to hold these 15 NRSROs out or this process out as a fully licensed industry, 16 then I think we need to look more closely at the audit firm 17 approach and model for this: a consistent and comprehensive 18 underwriting standard, standard disclosures, consistent 19 disclosures, conflict procedures, comparable ratings 20 methodologies and disclosures of those, and importantly -- 21 this was discussed this morning as well -- that there would 22 be full regulatory and legal consequences for improper 23 behavior. So those are a couple of models. 24 I think the current environment, as I said, has 25 elements of both. It has lead to confusion, it has lead to 127 1 misapplication of ratings, and it has lead to an industry 2 that our members don't trust very much. We did a survey of 3 professional membership in anticipation of today. Nearly 4 1,200 responses that said the following 60 percent do not 5 think that ratings are valid. 60 percent found that they 6 would not find these useful in making investment decisions. 7 70 percent said that they would like to see more government 8 regulation, but in the same survey, 51 percent said they are 9 not sure that they agree with the government approach of 10 trying to remove or deemphasize credit rating agencies in 11 regulation as the approach. 12 I'm not sure quite what this says, but I think at 13 least it says that the trust is still very low even with the 14 reforms that we have seen and that the approach to regulatory 15 reforms is just not very well understood out there. I think 16 investors are just saying "You got to get this fixed and you 17 need to address it." 18 Finally, the credit rating agency industry was not 19 entirely to blame here. They were an easy target, they were 20 certainly a deserving target, but sort of a scapegoat for 21 this entire mess in many respects, and I wanted to just 22 mention that investors and users have to fess up here as 23 well. This gets glossed over a lot in terms of this 24 discussion. Credit ratings generally, but specifically in 25 terms of structured product area, were absolutely misused by 128 1 lots of investors. 2 To Commissioner Paredes' point earlier, users have 3 to understand to some degree at least the basis and details 4 of the ratings process. They need to know what they are and 5 what they are not. They need to do their own analysis of 6 credit and other risks and then match that up with their own 7 investment circumstances. So we really need to break this 8 blind reliance whether there is an NRSRO seal of approval or 9 not. It must never again be acceptable practice for 10 investment managers who are holding themselves out as experts 11 to allow this to happen. Not that we are letting the credit 12 rating agencies off the hook by any means, but our 13 professional can certainly do much better, and I think you 14 would all agree with that. 15 MR. GALLAGHER: Thank you Schacht. Mr. Smith. 16 MR. SMITH: Thank you for the opportunity to be 17 here. My name is Greg Smith, as I was introduced, and I am 18 here on behalf of the Colorado Public Employees Retirement 19 Association, which is a defined benefit pension plan here in 20 the state of Colorado with $30 billion of investments. I 21 also sit on the executive board of the Council of 22 Institutional Investors, which is a consortium of 140 public, 23 union, and corporate pension plans, primarily defined benefit 24 plans, across the country with over $3 trillion under 25 management and tens of millions of members and beneficiaries. 129 1 I'm in a little bit of a unique position, I think, 2 compared to the rest of the panel and the panels that you 3 have seen before this in that we act as a fiduciary on behalf 4 of our members and beneficiaries. We are the sole source to 5 make buy/sell decisions for them. They are not reliant on 6 mutual funds and we are the ones pulling the trigger on 7 behalf our members and beneficiaries. 8 In the previous panel you heard from Damon Silvers, 9 and his statement was that the system is broken, and I am 10 here to agree with that. The issuer-pay model is broken and 11 it is riddled with conflicts, it lacks any alignment with the 12 objective of protecting investors, and it lacks any recourse 13 to the users. 14 I come here from the perspective of a fiduciary 15 charged with investment of assets belonging exclusively to 16 our members and beneficiaries and charged exclusively with 17 the duty of carrying out their interests. I certainly agree 18 with several panel members who state that reliance on ratings 19 and making buy/sell decisions an exclusive reliance on those 20 ratings is an inadequate discharge of the duty of any 21 institutional investor. 22 However, what we do as institutional investors in 23 reality is that we systematically ascertain -- and it is 24 sometimes driven by our needs for returns, it could be drive 25 by a lot of things -- but we systematically ascertain what is 130 1 our appetite for risk? We then ascertain what is the 2 appropriate asset allocation for our funds to reach that 3 appetite of risk. And then we go out in a diversified 4 manner, broad diversification hopefully, to accomplish that 5 appetite and to meet that appetite through our investments. 6 And in doing so, it is imperative that we have access to some 7 type of a measuring stick. 8 Whether it be a mnemonic device or whether it be 9 something else, we need a reliable, credible measuring stick, 10 as does the marketplace. And to say that we can simply do 11 away with ratings is to beg the question, really, what are we 12 going to use in its place and is it going to be credible, is 13 it going to be useful? We have to draw parameters as to how 14 our portfolios, what the universe of those portfolios will 15 be. And if we are going to charge an outside manager with 16 that portfolio, then they need to speak up -- or, I'm 17 sorry -- we need to set those parameters for them. If it is 18 an internal portfolio, we need to set it for those people as 19 well. We need a credible device to do that. 20 We believe that transparency, competition, and 21 accountability in the form of the private right of action for 22 those who are setting those measuring sticks and defining 23 those measure sticks is imperative for our ability to defend 24 its use to our fiduciary beneficiaries. 25 MR. GALLAGHER: Thank you Mr. Smith. Mr. Stern. 131 1 MR. STERN: Thank you to the Commissioners and the 2 staff for having me speak on behalf of the Association of 3 Financial Guaranty Insurers. These are the bond insurers 4 that wrap securities in the municipal and structured finance 5 space. 6 The financial guaranty insurance industry works 7 more closely with the rating agencies than perhaps any other 8 industry. Obligations insured by bond insurers generally 9 carry the bond insurer's ratings. As such, our ratings are a 10 major element of the product we sell and bond insurers are 11 very, very motivated to maintain high ratings. To maintain 12 our ratings, we need to evaluate all our actions and the 13 context of the rating agency impact of those actions on our 14 ratings. Investor confidence in ratings is fundamental to 15 our business and we support the Commission's efforts to 16 restore that confidence. 17 As an industry, financial guarantors seek 18 transparency and responsiveness from the rating agencies that 19 rate us. This transparency and responsiveness applies to 20 both actions needed to maintain our own ratings and the 21 impact of new business on those ratings. 22 In considering changes that might be helpful to our 23 industry we have two suggestions. First, rules applicable to 24 ratings in general should also apply to what we call shadow 25 ratings. Shadow ratings are the underlying ratings on 132 1 insured securities. These underlying ratings are 2 increasingly important to investors in insured securities as 3 they look through the bond insurer to consider the full 4 package of the investment that they are buying. 5 Second, we are troubled by the provisions of one of 6 the rules, Rule 17g-5, that effectively prohibits 7 recommendations to the obligor or issuer of a security about 8 the corporate or legal structure or other aspects necessary 9 to obtain a rating. While we appreciate the concern about 10 conflicts of interest, on balance we believe that the 11 interest in transparency is more important. When a rating 12 agency has a recommendation, it should not be prohibited from 13 communicating that recommendation. 14 Thank you for the opportunity to participate in 15 this roundtable. 16 MR. GALLAGHER: Thank you Mr. Stern. Mr. Stevens. 17 MR. STEVENS: Chairman Schapiro, members of the 18 Commission, thank you very much for the invitation to appear 19 today. On behalf of ICI, the 8,000 funds that we represent, 20 and the 93 million shareholders that they serve, I do commend 21 this continuing inquiry into how to improve the integrity of 22 the rating system. Our funds use credit ratings in a wide 23 variety of ways in the investment process and it is no 24 overstatement to say that they have a vital interest in the 25 soundness of the system. 133 1 Ratings should provide investors with a high 2 quality reasonably reliable assessment of the 3 creditworthiness of a particular issuer or financial 4 instrument. As the capital markets have grown more complex, 5 the need for ratings has grown, and the rating system should 6 have grown more robust to meet that need. Unfortunately, 7 neither is the case. The current financial crisis 8 underscores a stark reality: the long history of poor 9 performance by the rating agencies in analyzing the risks of 10 debt securities. 11 Despite their track record, ratings agencies 12 continue to enjoy virtual immunity from accountability that 13 is unknown to any other participant in our securities 14 markets. Moreover, in many instances investors are not free 15 simply to disregard these ratings even if investors question 16 the ratings accuracy or value. Despite their deficiencies, 17 ratings remain deeply entrenched in financial regulations and 18 in the process for accessing capital. For all of these 19 reasons the credit rating process must be improved. The 20 question is how to do so. The goal of the exercise, it seems 21 to us, ought to be absolutely clear. The Commission, as 22 always, should ask itself what actions best protect the 23 interests of investors in this context? 24 In our written statement, ICI has spelled out a 25 number of steps that we believe will advance investor 134 1 interests. They fall into four key area. First and 2 foremost, investors need better disclosure about credit 3 ratings and the ratings process. The Commission has made 4 strides in this area and we commend you for that, but more 5 remains to be done. Our written submission lays out a number 6 of specific ideas, many of which we have previously addressed 7 to you. Ideas to improve rating surveillance, to make 8 ratings more comparable, to enhance disclosure on rating 9 stability and conflicts of interest, and to make disclosure 10 ratings actions more timely. 11 Second, the Commission should take steps to 12 strengthen the incentives to produce quality ratings because 13 those incentives are simply inadequate today. Credit rating 14 agencies should be required to conduct due diligence on the 15 information they review when they issue ratings and to 16 disclose the gaps in that information. Another way to 17 bolster these incentives would be to hold credit rating 18 agencies legally liable not for the accuracy of the ratings, 19 but at least for the representations that they make to the 20 public and to the investors about their ratings process. 21 Currently, as you know, they are the only experts to enjoy an 22 exemption from liability for statements and registration 23 documents. At the very least, investors should have some 24 legal recourse if a ratings agency issues a rating without 25 following its own disclosed procedures and standards. 135 1 Third, we believe that the rules should be uniform 2 and consistent and apply to all types of credit rating 3 agencies whether issuer paid, subscriber paid, or some new 4 business model. And finally, we recommend that the 5 Commission improve disclosure not just for ratings agencies, 6 but also for issuers of structured products, asset-backed 7 securities, and municipal securities as well. 8 MR. GALLAGHER: Thank you Mr. Stevens. 9 Now we will move on to the Q&A portion. I believe 10 Commissioner Casey has a question. 11 MS. CASEY: I just wanted to follow up on some of 12 the comments with respect to the notion of eliminating our 13 references to ratings. And it is not that we are eliminating 14 ratings because boards could still reference them or rely on 15 them, it is just simply that we would not require them. And 16 I was hoping that you could maybe discuss that or talk to 17 that issue. 18 MR. GALLAGHER: Why don't we move down the line in 19 the order of introduction and then we can mix it up later. 20 MS. CUNNINGHAM: Sure. I think from a Rule 2a-7 21 perspective where money market funds are concerned having 22 some sort of a floor from a regulatory perspective is very 23 helpful for the small investment firms and for firms that 24 don't have all of the resources to be able to do as much 25 independent analysis as is required by many of the 136 1 institutional investors in the marketplace. And by providing 2 this floor, it provides to some degree a safety net. For the 3 past 35 plus years or so in that industry, a good portion of 4 the safety net from a ratings perspective has been intact. 5 And for that reason, although it is an input into the process 6 of independent analysis, by establishing a floor that has to 7 be met as a minimum, I think it is helpful in the context of 8 a very stable and safe market. 9 MR. FOHRER: Frankly from an issuer perspective I'm 10 not sure that fits for us. As someone that manages pension 11 funds and others, having at least that floor rating does 12 provide some perspective. From the issuer, I don't think 13 it's that big a deal. 14 MR. KAITZ: I just want to make sure I understood 15 the question, which was that -- to do away completely with -- 16 MS. CASEY: No, just to the commentary about -- 17 this discussion about eliminating references was somehow 18 doing away with ratings, and all we are simply saying is that 19 we would no longer require that those ratings be used, but 20 certainly boards continue to rely on them and use them as 21 benchmarks. It just wouldn't be a regulatory mandate. 22 MR. KAITZ: The market implication is that you will 23 institutionalize, essentially, two rating agencies in the 24 market, in our view. You might as well -- 25 MS. CASEY: Is that because of the brand reliance 137 1 then, that folks would then just fall back to -- 2 MR. KAITZ: As I said, from a corporate finance 3 perspective, if you are a treasurer or CFO, you are seeing 4 what is happening now because there is no confidence in the 5 market. You are talking about multi-billion dollar companies 6 that might have two or three people in their treasury 7 department. They do not have the time to do the due 8 diligence on these securities. They are going to go with 9 flight to safety. They are not going to invest in anything 10 that is going to put their company at risk in terms of 11 investing the dollars in their company. They are going to go 12 to treasuries. They are putting the money in their bank 13 accounts right now. I mean that is what is happening in 14 reality. 15 So investor confidence right now, from a corporate 16 finance perspective, there is no investor confidence, and 17 that is what you are going to see if you just do away with 18 this regulation. And you are going to end up with two rating 19 agencies. You might as well institutionalize the process. 20 MS. CASEY: What does that say about the quality of 21 due diligence efforts on the part of fiduciaries? 22 MR. KAITZ: They are in the business of -- whether 23 they are making widgets or anything else. And if you are a 24 corporate treasury department or a finance department right 25 now and you are investing in your company, you don't have 138 1 assurances from an independent agency to some extent. The 2 reality is -- all I'm saying the reality is they are not 3 going to invest in those securities. They just don't have 4 the resources to do that. 5 MR. GALLAGHER: Commissioner Walter, did you want 6 to follow up on that? 7 MS. WALTER: I just wanted to follow up on that and 8 press a little bit more on that last point, not just from 9 your perspective, but from anyone's. In using ratings to 10 whatever extent that you do, how much do you rely on the 11 letter grade and how much do you rely on the report that 12 accompanies that? I mean what are people actually looking 13 at? 14 MS. CUNNINGHAM: From our perspective the letter 15 grade is something that is helpful, but for establishing the 16 floor, for meeting the regulatory requirement. That gives us 17 a universe of securities and issuers from which we can 18 choose. The much more important aspect of the rating agency 19 information that we receive is on the actual analysis itself 20 and the information content that is received there, not the 21 actual letter. 22 MR. FOHRER: We take a hard look at the letter 23 rating in terms of how we assess counterparty risk, which is 24 very large in our company when we are buying so much power. 25 That credit rating is a beginning, not an end, to how we look 139 1 at it though. In terms of whether we require a company to 2 post collateral, which can be in the tens to hundreds of 3 millions of dollars, we take a look not only at the credit 4 rating -- and we have a threshold that we would absolutely 5 require collateral below a letter rating -- but we also do 6 our own analysis taking a look at any near term events that 7 have happened. 8 So the credit rating is a tool. It is not the only 9 tool given how we look at the individual entities because 10 things are happening so fast in our market we depend on 11 keeping up with it. 12 MR. GALLAGHER: Mr. Smith, do you want to weigh in 13 on this? 14 MR. SMITH: The concept of doing the due -- I mean 15 any institutional investor, in my mind, that is a fiduciary 16 to members and beneficiaries is derelict in their 17 responsibility if they are not going beyond looking at those 18 letter ratings. There is just no two ways about that in my 19 mind. But it comes into so many other aspects of what we do 20 where it is not being used as a buy/sell decision maker by 21 any stretch of the imagination, but it is creating a 22 universe. 23 And it is also allowing us to create a universe for 24 our managers who we can then benchmark and who we can then 25 track whether or not their performance is appropriate. If 140 1 they are open to the entire universe of instruments and we 2 can't set parameters within which they have to operate, then 3 our ability to track their performance versus another 4 manager's performance that is purportedly in the same 5 universe but may be accessing instruments that have a 6 dramatically different risk rating or risk profile, it 7 eliminates our ability to track their performance and compare 8 their performance on and apples to apples basis. 9 MR. GALLAGHER: Anyone else on the panel want to 10 jump in? Mr. Stevens? 11 MR. STEVENS: I want to go back to Commissioner 12 Casey's question because it is one that we have thought about 13 a lot. When the original proposal came out about stripping 14 the references to ratings out of the Commission's rules, our 15 comments focused on one rule in particular, and that was Rule 16 2a-7. And it may be that that is a reasonable model for when 17 there is a rating incorporate in a rule to emulate elsewhere 18 to the extent there is a need to maintain them outside of 19 that context. 20 What Rule 2a-7 basically says is you cannot escape 21 as the manager of a money fund the job of doing an 22 independent credit determination to determine whether the 23 security presents minimal credit risk, and therefore is 24 appropriate for your portfolio. The analysis begins at 25 whether it has been rated under the current rule, top tier or 141 1 top two tiers, but it does end there. 2 And we sort of think about it as helping to control 3 neighborhood risk, if you will. It establishes a floor under 4 the credit analysis of all the participants in that market. 5 But under the Commission's rules, everyone has a very clear 6 idea, including the fund's board, that there has to be an 7 independent credit determination on top of that. And I think 8 frankly, despite the difficulties that we had in September 9 and October, if you look, as Debbie said, at the broader 10 history of money market funds, that has been a pretty sound 11 approach. 12 I would subscribe to the comments that are made 13 here elsewhere no retail investor, frankly no institutional 14 investor for sure, should ever take the rating at face value. 15 I'm afraid, though, our system has in some ways invited that 16 to be the case, and I think to the extent that it has been 17 overused in regulation or in law, it has only reinforced that 18 temptation. 19 MR. GALLAGHER: Chairman Schapiro. 20 CHAIRMAN SCHAPIRO: I'm not sure this is a question 21 for Paul, but it kind of follows on what you just said. 22 Given what has happened in the last year, I guess for the 23 investors on the panel, has your reliance on ratings changed? 24 Has your due diligence process changed or did you always feel 25 like you were using the rating as the floor and everything 142 1 else you needed to do on top of that? 2 MR. GALLAGHER: Paul, do you want to lead off? 3 MR. STEVENS: My sense, Chairman Schapiro, is that 4 there is a much greater humility that people feel about the 5 ability of ratings agencies to do reliable ratings on highly 6 complex financial structures, and if there has been an 7 internalized lesson, it is probably around that most of all. 8 MR. GALLAGHER: Why don't we move in the opposite 9 direction down, starting with Mr. Stern. 10 MR. STERN: Well as an industry, the financial 11 guaranty industry has always underwritten every risk 12 separately that we insure. Every transaction we guaranty has 13 what is called a shadow rating, as I mentioned before, which 14 is the rating that the rating agencies have applied to the 15 risk. But we separately underwrite every risk and would 16 never independently rely upon the rating agencies for credit 17 judgments. 18 People who don't rely on the rating agencies will 19 really seek a second credit underwriting, often seek bond 20 insurance for that purpose. We are often viewed as sort of a 21 rating agency with a money-back guarantee. We are a rating 22 agency, we do the credit underwriting, we evaluate the 23 credit, but there is full recourse to us as well. We limit 24 our business to investment grade risk. It is supposed to be 25 low risk business, but it is -- maybe what people want to 143 1 push the rating agencies towards when they try to seek 2 recourse against the rating agencies. But I believe there is 3 a different role for a rating agency that just has an 4 independent voice without really any financial recourse as 5 such. 6 MR. SMITH: Well as the lawyer, I'm not sure I 7 could say that we haven't done our due diligence right all 8 along, but even if I could say that, I don't think it would 9 be applicable. I think most of your institutional investors 10 have been more sophisticated than that. They do use not just 11 the letter ratings, but certainly going beyond that to the 12 research. 13 But I will say that I think it has impacted 14 significantly the due diligence process in buying products, 15 and specifically I mean money market types of parking of 16 dollars, or what we use in our securities lending types of 17 arrangements where we are buying into pools or we are placing 18 our money in pools that are represented as being the same as 19 cash. And the amount of drilling into those types of 20 products that was done historically versus the level of 21 diligence you are doing today on liquidity and your ability 22 to know what that is made up of as opposed to just the 23 representation that we are same as cash and just park it here 24 for 24 hours has changed significantly. 25 CHAIRMAN SCHAPIRO: And I guess my question really 144 1 goes to more than maybe just has your due diligence changed 2 but are you looking for more ratings of a single issue or are 3 you using different rating agencies or more inclined to join 4 a subscriber model than an issuer model, so lots of 5 different -- I'm just curious about what we have been through 6 in the past year has changed behavior at every level in terms 7 of the use of ratings and the reliance on them. 8 MR. SMITH: Certainly the legal department has had 9 more meetings with the investment department about their due 10 diligence on bonds and fixed income and so forth than we had 11 previously. In my instance, I'm pleased with the answers 12 that I get, but certainly I think you are seeing that check 13 and balance come into play much more in an institutional 14 investor environment. 15 MR. SCHACHT: I would just add that I think the 16 investor reliance on ratings of structured products is 17 probably pretty close to nil right now. And I would add very 18 quickly this notion that we have -- everyone has been sort of 19 piling onto credit rating agencies, including in the 20 government, in terms of the level of accuracy and their value 21 in this process. And at the same time we have firms that are 22 ensconced in this financial crisis and their ratings remain 23 unchanged and blatantly off track in terms of where they are 24 actually trading because we want to make sure that we don't 25 breach covenance and create any further spiraling down in 145 1 this crisis. 2 And I think that continues to send a very fixed 3 message to the market for sure, and I don't think we should 4 have a situation in this country where honest ratings only 5 apply for firms when government bailout funding is not 6 involved. I think that is a serious concern. 7 MR. FOHRER: Again, our primary exposure is not 8 through investments, it is through counterparty risk and 9 exposure. We have criteria we use on whether we have to -- 10 whether we require people to post collateral that is 11 generally based on being investment grade, and the higher up 12 in the letter, the less collateral you would have to post. 13 We clearly have increased, I would say, our 14 emphasis on our own independent credit analysis because 15 companies are not rated necessarily in real time, but events 16 happen in real time. So we are very cognizant of what is 17 happening to the counterparties and we track them closely. 18 We look at the ratings as a tool but not the sole decision 19 maker and not the sole protection, so it is putting 20 increasing pressure on our risk management staff to make sure 21 that we are following these things in real time. 22 MS. CUNNINGHAM: The only thing I would add to that 23 is that over the course of the past almost 18 to 20 months at 24 this point, we have come to place more emphasis, depending 25 upon the issuer that we are looking at, on specific rating 146 1 agencies. So for instance, if we are looking at a 2 municipality, if we are looking at an issuer that is 3 municipality, we may look at one rating agency's information 4 over another and place more emphasis on the quality that we 5 are getting from that particular one that we feel 6 historically has done a better job on a municipality 7 assessment. It may be a completely different rating agency 8 than if we were looking at an asset-backed auto issuer, in 9 which case we might choose a different NRSRO. 10 MR. GALLAGHER: Commissioner Walter. 11 MS. WALTER: You all to be similarly of the view 12 that people are not relying on ratings the way they used to, 13 that they are much more skeptical about them. And in terms 14 of from a regulatory perspective, increasing the 15 accountability and the reliability of the ratings, which 16 general approaches would you favor? I asked this morning of 17 the rating agencies themselves the liability question that 18 Paul Stevens raised. Is that a viable approach? Is greater 19 transparency principally the regulatory tool that we should 20 use? Which type of approach, if any of them, or should we 21 use none of them, would you favor? 22 MR. GALLAGHER: Why don't we mix it up a little bit 23 and go with Mr. Schacht first on this one. 24 MR. SCHACHT: I'm not sure. And the reason I say 25 that is because if your premise is the system is broken, the 147 1 question is how are you going to fix this? Are you going to 2 put together the pieces that are broken or are you going to 3 throw it out and start over with either a pre-1980s model or 4 are you going to go with a more serious and rigorous audit 5 type model? We have been commenting all along on the various 6 proposals to fix the pieces that are out there, and I think 7 that is probably doable, and we have been watching that 8 pretty closely. 9 One of the things that we were -- everybody has 10 mentioned today is this whole funding model and whether there 11 is a way to bring better conflict protections around it. We 12 are not sure how exactly that works through regulation. 13 We have talked about the elimination of the concept 14 of investment grade. We talked about the concept of changing 15 the rating structure, distinguishing the rating structure for 16 structured products so that there would be a separate 17 designation, a more informative designation for things that 18 are rated that are not the conventional corporate debt. So 19 those are some of the things that we have talked about. 20 MR. GALLAGHER: Mr. Smith? 21 MR. SMITH: Thank you. I think Mr. Stevens put it 22 very well and very consistently with how we view the world on 23 accountability and particularly the concept of a private 24 right of action. We don't expect credit rating agencies to 25 be responsible for the failure of a particular instrument or 148 1 the failure of a product line or of a particular entity. 2 Certainly that is not appropriate. 3 But if there is going to be disclosure and 4 transparency which I think is certainly uniformly pretty 5 clear, then they should be responsible for meeting what they 6 tell us they are doing and meeting an industry standard that 7 I think would be developed over time that I allows uniformity 8 within the industry and within what we expect out of them. 9 And we wouldn't look to them to be the guarantors, certainly, 10 but they like every other service provider, whether you talk 11 about attorneys or auditors or any number of different 12 industries have standards. They should have to meet those 13 standards and if they don't meet those standards there should 14 be liability. 15 MR. GALLAGHER: Mr. Stern. 16 MR. STERN: Well I think there is a very thin line 17 between being a guarantor and having liability, and I think 18 it is going to be very hard to define that line. 19 Transparency might not be perfect, but I think it is the best 20 that can be accomplished here. We just have to recognize 21 that the real issues here maybe come down to credit judgments 22 and all the regulations in the world won't make people 23 smarter. People just have to -- people will learn from what 24 happened in the last couple of years, and I think it is going 25 to influence credit ratings for many, many decades to come. 149 1 MR. GALLAGHER: Mr. Stevens? 2 MR. STEVENS: We are kind of the disclosure people, 3 so it is consistent with the principles that I think our 4 industry is founded on, that we believe disclosure is very 5 healthy. It can be expensive and burdensome at times, but I 6 think it is a regime under which mutual funds have prospered, 7 and I actually do think ratings agencies could as well. 8 I certainly think that we need to get away from the 9 black box, and what that requires is a thoroughgoing regime 10 of information that is available to investors about the 11 ratings process, so that as Debbie said, they can do analysis 12 and determine that this agency is sounder in its approach to 13 this kind of security as opposed to another. It ought to be 14 available to the public as well to a significant degree so 15 that analysts and commentators and others, as independent 16 third parties, could express their own views about the 17 strengths or weaknesses of different approaches. And I think 18 in that environment, the strong, quality raters will out. 19 And so I would say that the most important thing 20 that the Commission could do, and this is very consistent 21 with the testimony I provided in the House and Senate when 22 the 2006 act was being considered, is to establish a 23 comprehensive disclosure regime and then let market 24 participants sort it out after that. 25 MR. GALLAGHER: Anyone else on the panel want to 150 1 jump in? Mr. Kaitz? 2 MR. KAITZ: I would just -- I obviously wouldn't 3 disagree with anything from a disclosure standpoint. I think 4 we still believe that it is a -- the focus is on competition 5 in the marketplace, and what you don't want to do is put a 6 regime that we no longer have people who want to be rating 7 agencies, if in fact that is what the SEC deems that we need 8 to have. 9 And again, I think we have talked about for many 10 years competition in the marketplace. There are thousands of 11 mutual funds, there is thousands of companies, but we have a 12 very small concentration here, and really two that dominate 13 the market. And I would just caution, with all the 14 disclosure requirements and legal liabilities, it is hard to 15 argue against that in this environment, but don't discourage 16 those that want to get into the market, especially 17 entrepreneurs who will -- and we are business model agnostic. 18 We want to encourage people to be in this business 19 whether we have the NRSRO designation or not, I think we fell 20 it is a critical tool, not the only tool, but a critical tool 21 in the whole investment process. 22 CHAIRMAN SCHAPIRO: Can I just follow up on that, 23 because this issue came up a bit in the last panel too. How 24 do we square the desire to have more competition in this 25 space, which I think generally people do agree with, but the 151 1 need to also have quality and some assurance that there are 2 rigorous processes in place within new entrance to the 3 marketplace or do we just completely rely on market 4 discipline to get us over that hump? That is the disconnect 5 that I am having between competition, yet some reasonable 6 desire to ensure that if people are going to rely on ratings, 7 and most particularly if they continue to be contained in 8 regulations and programs like TALF, that there is real 9 quality there too. 10 MR. GALLAGHER: Ms. Cunningham? 11 MS. CUNNINGHAM: I think when you look at a system 12 like Rule 2a-7 which requires the independent assessment 13 based on the information that comes from the rating agencies, 14 the credit rating agencies as a start in addition to other 15 information that can be gathered from other sources, you get 16 to a point where competition and quality basically go 17 together. And as I said before, certain of the rating 18 agencies are better at assessing quality in certain types of 19 issuers and industries whereas others are not as good at some 20 of those. And some have actually made applications and have 21 been qualified as NRSROs in specific sectors of the 22 marketplace. 23 But I think that with that as a basis, a floor for 24 minimal credit assessment and fiduciary duty, going back to 25 one of the things that Greg was saying too, on a fiduciary 152 1 duty basis where you are just not relying on the rating but 2 rather utilizing it as an input to your process, it assures 3 quality from an output perspective by the users. 4 MR. GALLAGHER: Mr. Fohrer. 5 MR. FOHRER: I guess as I said in my opening 6 comments, competition sometimes sounds good and it can 7 provide different things to different people. Let me just 8 take a little bit of a contrarian point of view on quality. 9 Today, we have two primary rating agencies that we 10 use, Standard & Poor's and Moody's. Management spends a 11 substantial amount of time with each independently providing 12 future forecasts, providing all of the public documents, and 13 we spend a lot of time with them. If there were multiple 14 rating agencies beyond this group, I think we would have to 15 have a different approach because the idea of sharing 16 non-public information with anybody that wants to come in or 17 a wide range would make us more nervous about the -- how that 18 information would be used. 19 But secondly, I don't think we as management would 20 have time to meet with five, six, or seven rating agencies 21 and spend that much time. I think by definition what we do 22 is go to a situation where we would provide public 23 information only and they would rate. So what you may pick 24 up by having more people take a look -- a trade off would be 25 less in-depth analysis, less in-depth information. And those 153 1 are choices. I don't think there is one that is right or 2 wrong, but I think we ought to recognize the consequences of 3 those different approaches. 4 MR. GALLAGHER: Anyone else on the panel want to 5 weigh in on that? Mr. Schacht? 6 MR. SCHACHT: I think a couple of things that you 7 could do, whether you -- whether this is self-regulation or 8 not, in this environment there at least ought to be -- if you 9 are going to get into the industry, there at least ought to 10 be some sort of registration requirement with the Commission 11 so that you have the ability to examine what was going on. 12 And I think we should probably require that there is a 13 uniform code of conduct, something like the IOSCO code of 14 conduct. So those two things would play together. 15 MR. GALLAGHER: Before we move on to Commissioner 16 Paredes' question, I will just ask a little follow up of my 17 own. Given that the statute prohibits the Commission from 18 essentially delving into the substance of ratings, do you 19 think there is something there for regulators to be looking 20 at as to the quality that Chairman Schapiro is talking about? 21 Mr. Stevens? 22 MR. STEVENS: As I recall the legislative history, 23 I think there was a judgment that everyone shared that the 24 Commission, and for that matter any governmental regulatory 25 body, was poorly positioned to develop its own ratings 154 1 methodologies and serve the function that ratings agencies 2 currently do. 3 There was also a discussion at that time about a 4 seasoning requirement, and I believe in the statute it was 5 arrived at three years, as I recall. So you have to be a 6 credit rating agency for at least three years before you can 7 then apply for an NRSRO. There was a lot of discussion in 8 the hearings about whether that seasoning requirement was 9 necessary. I would say that it was driven more by the cast 10 of institutions who were seeking NRSRO designation and how 11 long they had been in the business rather than a scientific 12 judgment that three years is the right point in time where 13 you have matured. 14 But I think in response to your question, Chairman 15 Schapiro, if you did have robust disclosure for a new entrant 16 around "Well what is your track record?" "None, thank you 17 very much. We are new to this game," it at least informs 18 participants who might be interested or not in using those 19 ratings and levels the playing field. Presumably it would be 20 to the advantage of established ones, but part of that 21 disclosure should be ratings histories as well, so all of 22 that would be out for the world to see, and I think that 23 would provide a lot of discipline. 24 MR. GALLAGHER: Commissioner Paredes. 25 MR. PAREDES: Two related questions. The first is 155 1 what is the process to decide whether or not to subscribe to 2 the ratings of a subscriber-pay credit rating agency? And 3 then secondly, if you decide that you actually want to 4 subscribe, how to choose among whatever options there are 5 that are available? 6 MR. GALLAGHER: Ms. Cunningham, would you like to 7 start with that one? 8 MS. CUNNINGHAM: Sure. At this point we subscribe 9 to ten different credit rating agencies, seven of which are 10 NRSROs, three of which are not. So they are just credit 11 rating agencies whose independent research we find valuable. 12 Generally what happens on a process basis is that 13 you are approached by somebody from that firm who wants to 14 win your business as a subscriber and they give you a free 15 subscription for a month, three months. Usually if it is 16 month you can talk them into three, if it is three you can 17 talk them into six. 18 But the process is such that they are knocking on 19 your door generally saying "I have a product that is a little 20 bit unique." Generally they want to have some sort of a 21 unique angle associated with it, and this goes back to maybe 22 it is a sector, maybe it is a type of security, maybe it is 23 an international versus -- or global versus domestic. So 24 there is generally some sort of an angle that they are trying 25 to pitch you on in the context of why them versus the other 156 1 15 or 20 that might be out there also knocking on your door. 2 And then from an assessment perspective, it is up 3 to the analysts within the various investment firms that are 4 looking at it, would be utilizing it on that free trial basis 5 and deciding whether or not it adds value. And if it does 6 add value, is it adding value in the context of replacing 7 something that you already have or is in the context of truly 8 adding value in it being unique and new and being additive? 9 MR. GALLAGHER: Mr. Fohrer? 10 MR. FOHRER: Well being a very large issuer, I 11 think a couple of our choices were obvious in terms of where 12 we are -- in terms of the other services we would use for 13 credit analysis for our counterparties. We really leave that 14 up to our risk management group to determine which ones 15 provide us the best coverage of individual companies, and we 16 have made that a risk management committee decision and 17 allowed our people to look at who can provide the best 18 information for a given sector. 19 MR. GALLAGHER: Mr. Smith? No? Any other follow 20 up Commissioner Paredes? 21 MR. AGUILAR: Dan, I really want to drill down on 22 Commissioner Paredes' second question. And Ms. Cunningham, 23 if you could provide a little bit more information on your 24 decision as to which credit rating agencies to use for which 25 products. When they come in to sell their wares to you, are 157 1 you making a decision based on publicly available information 2 of their rating histories and things like that, or are you 3 getting as a preferred buyer some sort of inside look at what 4 their ratings have actually been so you are making it on 5 non-public information, for lack of a better word? And would 6 you think that that information, if it is non-public, would 7 that be useful for the public at large to have access to what 8 you are having access to in making your decisions? 9 MS. CUNNINGHAM: First of all, let me tell you that 10 we did look at five of the major rating agencies that are 11 NRSROs before this discussion today, before the roundtable 12 today. What is publicly available from a credit rating 13 agency perspective for no fee is simply current rating and 14 seven days worth of press releases. So if there is no press 15 releases, you get a single current rating on that issuer. 16 That is all you get on a no subscription basis. 17 So virtually there is no such thing as an 18 issuer-pay only model because issuer pay without subscribers 19 doesn't make any sense. You have to have subscribers that 20 use the ratings then within their investment process. 21 From a standpoint of making the decision, what 22 effectively we are receiving is not just that single rating 23 and seven days worth of press releases. It is the inside 24 information or the informational content that they are 25 comfortable with providing to us on an ongoing basis that is 158 1 the basis for their assessment. So it is the underlying 2 fundamentals of the quantitative and qualitative assessments 3 that they are providing that ultimately end up in the current 4 rating that is then provided in an in-depth form from a 5 credit research perspective that we would use then in our own 6 independent analysis process. So it is certainly not what is 7 publicly available off which we would make these types of 8 decisions. 9 MR. GALLAGHER: Mr. Smith? 10 MR. SMITH: I'm certainly in full agreement with 11 that. Really what we look at is more the quality of the 12 research that underlies that rating. What their track record 13 has been on whether that letter stayed through the course of 14 that instrument or not, got downgraded, whatever, we don't 15 spend a lot of time trying to track that. One because it is 16 very difficult to track and it takes a lot to try and gather 17 that information. It would be useful if that was transparent 18 and I would certainly encourage you to consider that. 19 But really when we are selecting a vendor in that 20 nature, we are looking at the underlying research and we are 21 getting the three months or the six months of free look to 22 see and compare over a period of time over a period of 23 instruments what the quality of that research is as opposed 24 to letter rating itself. 25 MR. GALLAGHER: Anybody else on the panel want to 159 1 weigh in on that? 2 So I have another question here for the panel. We 3 have been talking about the removal of references and that 4 obviously is linked to a Commission rule proposal from last 5 year. At the same time, the Commission also issued a 6 proposal that would call for -- we have been calling this 7 symbol proposal -- differentiation in rating symbols for 8 different products. And many of you or the organizations 9 that you are in have probably commented, we got a lot of 10 comment back in on that. 11 But I wanted to get your thoughts across the board 12 here on the utility as a user of differentiated ratings for, 13 say, structured financial products, munis, or other 14 complicated securities, whether that is something that is 15 useful to you as a user or not. We will start with Mr. 16 Stevens. 17 MR. STEVENS: Yes, I think a cosmetic approach to 18 the issue is tempting, but I think that is part of what has 19 caused the problems that we have. It is working with a 20 different set of black boxes. It may be for retail investors 21 an ultimate trap. What we at ICI fought years ago -- it is 22 15 or so I think -- the proposal that there would be 23 volatility risk ratings on bond funds which would simplify 24 all of an investor's inquiry into some sort of alphanumeric 25 rating. We opposed it. It never really came to the 160 1 marketplace, and I think it would have been very deceptive 2 for retail investors. 3 For the institutional community, I think the point 4 that you have heard here over and over is don't rely on the 5 rating, rely on the ratings process and what it produces. 6 And so I just think that is entirely cosmetic. 7 MR. GALLAGHER: Mr. Stern? 8 MR. STERN: I share Mr. Stevens' view. I have 9 nothing to add to that. 10 MR. SMITH: I agree as well, although unfortunately 11 in our role -- and I will go back to my trying to create the 12 universe for outside managers -- if there is confusion among 13 what these ratings mean and there is overlap when they are 14 using the same symbols, the same rating indicator, whatever 15 that is for more than one type of instrument and therefore 16 having more than one type of risk profile covered in that 17 single letter rating, if you will, that is problematic for us 18 in our efforts to try and use those in setting parameters 19 for our managers. Underlying all that though, they need to 20 be reliable, they need to be accurate, they need to have -- 21 be a credible rating for us to even have any use of them. 22 MR. SCHACHT: As I mentioned, we had -- and I think 23 in one of your previous consultations had supported the 24 notion of different rating symbols to distinguish structured 25 products. Our membership sort of narrowly supports that. 161 1 And whether it is just window dressing or whether it is a 2 meaningful distinguisher for these markets I guess remains to 3 be seen. 4 In talking to a couple of our members about this, 5 they said "You know, I know that when I see a structured 6 product that is rated triple A that it is really going to be 7 trading at A level and I'm hoping to buy it at a triple B 8 price, so I have already factored in this whole notion of 9 having a different nomenclature." So whether or not we need 10 a different nomenclature, they weren't sure about that. 11 MR. KAITZ: I would agree with Mr. Stevens' and Mr. 12 Smith's comments. 13 MR. GALLAGHER: Thanks. Mr. Fohrer? 14 MR. FOHRER: I agree. Let me emphasize I would 15 just be concerned with investor confusion because it is so 16 critical for us to have a constant access to the market. And 17 if we have a period of time that the large investors or the 18 mom and pops don't understand what the ratings mean and pull 19 back or charge more, I don't think we have accomplished 20 something. So I would look at what we are really trying to 21 solve. And just putting a different name on it I see as 22 creating more confusion than helping. 23 MR. GALLAGHER: Ms. Cunningham. 24 MS. CUNNINGHAM: The only thing I would add is the 25 costs associated with changing the systems that monitor 162 1 ratings, that by adding some sort of a subscript or an extra 2 feature to the rating itself probably has a lot of costs 3 associated with it from a systems perspective. You may want 4 to -- you could still keep the ratings, you know, the A, 5 double A, triple A, and then have it as a sector. So the 6 sector is M for munis, the sector is C for corporates, the 7 sector is SF for structured finance or something like that. 8 But to keep it as additive to the standard rating, not the 9 new rating regime. 10 MR. GALLAGHER: Thank you. Commissioner Walter. 11 MS. WALTER: I want to follow up on that and sort 12 of flip it around. There is a flip side suggestion that has 13 been made that what we really should do is mandate a uniform 14 nomenclature. And assuming that we could find one that 15 wouldn't prefer one provider over another or take that as the 16 assumption, would any of you favor that, so that there would 17 be a series of definitions as to what the various categories 18 would mean? 19 I know I have been told by some people that they 20 think none of the rating categories are meaningful because 21 they all are based on comparisons to the other and there is 22 no base category at the bottom that is firm enough for the 23 system to rest on. So would you favor a more or less 24 regulatory defined or industry standard defined set of 25 categories that everybody would use? 163 1 MR. GALLAGHER: Ms. Cunningham, do you want to -- 2 MS. CUNNINGHAM: I think that simplifies things. 3 It makes it easier from an investor and a shareholder 4 perspective to understand what they see in a ratings report 5 or ratings information that comes from various providers. I 6 just think you have to be -- as you said, there has to be a 7 basis. And the basis should be treasuries. It is the 8 riskless security in the market of the riskless issuer in the 9 market, for the most part. And if it is not treasuries, then 10 as a basis, there needs to be something defined as such off 11 which to build the rating scale. 12 MR. FOHRER: I think whether it is an absolute 13 consistency, I think a clear transparency for each agency or 14 each issuer. What does an A mean versus an A minus versus a 15 B? I think that clarity around the process would be 16 critically important. 17 MR. KAITZ: At the risk of never being invited back 18 to the SEC, which was probably the prior SEC for me, but to 19 me, that is window dressing. I mean it is not getting at the 20 core issues that I think we are all trying to address here. 21 I think we have heard that the system is -- however you want 22 to say it -- broken, flawed. We need some real -- and it is 23 all about credible and reliable ratings. All of those are 24 great if we had credible and reliable ratings and we were 25 comfortable with how the system was working. 164 1 So I would just urge the Commission to look at this 2 at a very high level. I know these are very difficult 3 decisions, but I think we are at a point where we are all 4 very hopeful that this Commission is going to take on some of 5 the real tough issues and get us to a point where people 6 really -- everyone in this community wants these ratings to 7 be credible and reliable and we need this as a healthy part 8 of our capital market system. So I really hope we can get 9 there. 10 MR. GALLAGHER: Mr. Schacht? Mr. Smith? 11 MR. SMITH: I think there is some value in having 12 that delineation, in having that set, whether it be by 13 regulator or by statute or however we could do it, because I 14 think that there is a difficulty, again, in having 15 comparability, I guess, and in developing industry standards. 16 If we don't know what a triple A rating is, if it is not a 17 defined term, then how do we know how to hold people to 18 meeting that defined term's definition? And we could let 19 them define them themselves, disclose that, and hold them to 20 their own definition. I suppose that works as well. 21 But as has been said by previous members of the 22 panel, I certainly think that is a tool to getting to where 23 we need to be, a tool to getting to credible ratings, not 24 creating credible ratings in and of itself, of course. 25 MR. STERN: Well I think it is important that the 165 1 rating agencies be transparent, but I do believe that they 2 have to be free to define their own rating parameters and 3 what it means to be a triple A because each rating agency has 4 a different view of a triple A, a double A, a single A. And 5 so long as they define it and people bother to read it, then 6 I think the mission has been accomplished. But I don't think 7 it is helpful to try to pigeonhole ratings for rating 8 agencies. 9 MR. STEVENS: We think standardizing ratings 10 agencies disclosure of their performance measurement 11 statistics is a good idea in general, but if you go to a 12 system where you have proscribed as a Commission the exact 13 structure of what the ratings would look like, that becomes, 14 it seems to me, all the more important. 15 MR. GALLAGHER: Chairman Schapiro, I know you had a 16 question. We probably have time for one last question here. 17 CHAIRMAN SCHAPIRO: It is a little bit of a change 18 of angle, but I asked the first panel this morning, which was 19 largely the rating agencies themselves, if there had been 20 products over the past year, in particular the past two 21 years, that simply should not have been rated. And I was 22 surprised to hear virtual unanimity on that subject, that 23 indeed there were products that should never have been rated. 24 Either they were too complex, there wasn't adequate 25 information, people didn't actually understand them well 166 1 enough within the rating agencies. 2 So given that, would you all have any guidance for 3 us on how we approach this question of ratability at all 4 going forward or whether that is not an issue the government 5 should intervene in? 6 MR. GALLAGHER: Why don't we go with Mr. Stevens 7 and work our way back down. 8 MR. STEVENS: Well there are certainly products 9 that shouldn't have been bought. And I get back to this 10 reality that we live in, particularly with money funds, which 11 is that the ratings is just the place that starts the 12 conversation about whether it presents minimal credit risk. 13 So almost implicit in the question is the assumption that the 14 rating really has some definitive quality to it from an 15 investment perspective. No doubt from a business point of 16 view for both raters as well as for purchasers, it probably 17 would have been better that some of the more complex 18 structures never have been subjected to the process. 19 But whether the Commission ought to be setting 20 standards for securities that are unratable or not, it seems 21 to me a difficult thing for it to undertake. 22 MR. GALLAGHER: Thank you. Mr. Stern. 23 MR. STERN: And I think the rating agencies 24 certainly have learned from their experience. The sector 25 that stands out here are the CDOs of ABS as being too complex 167 1 to rate. You know, I think they recognize that now. And 2 whether it was too complex to rate or too complex to rate 3 highly is maybe another question. They could have rated them 4 all very low and then maybe nobody would have bought them. 5 But it seems -- you know, perhaps the Commission 6 should be looking at the marketplace sometimes. If there 7 were triple A securities out there that just aren't selling 8 and bond insurers won't guarantee it even double the triple A 9 levels, I think that says something about perceptions of the 10 ratings, and maybe it is a warning that should be heeded in 11 the future. 12 CHAIRMAN SCHAPIRO: Because the danger is really 13 not just that the rating doesn't have value, but it could 14 affirmatively be misleading in a case like that. 15 MR. STERN: Right. 16 MR. SMITH: I think that answering that question 17 from the perspective of the NRSROs, really now they get to 18 say "Well, yeah there are some we shouldn't have," but really 19 they were getting paid to rate whatever was put in front of 20 them and they had no accountability for whatever that rating 21 was, so why wouldn't they make the money, why wouldn't they 22 rate the instrument, and why wouldn't they take the money and 23 leave with it? Because they had no downside. So were there 24 instruments that shouldn't have been rated? In hindsight, 25 certainly, but from a business perspective of the NRSRO, 168 1 there is no downside to them, and that is the ultimate lack 2 of accountability that I think has driven them to pursue a 3 business model that is certainly detrimental to the investor 4 interest. 5 MR. SCHACHT: We had taken the position on a number 6 of different consultations that credit rating agencies should 7 not rate new structures. And when you have the circumstance 8 where you have case after case of triple A ratings being 9 placed on securities that have -- there has been no 10 investigation of the underlying portfolio of assets, there is 11 limited performance data in terms of the underlying 12 collateral, and in fact there is really no prior track record 13 at all in terms of default experience on some of these 14 affordability products, how could you actually have rated 15 that? So we had suggested in one of these consults, as I 16 mentioned. 17 MR. KAITZ: I guess we don't see that as 18 necessarily the role of the SEC in kind of the market system, 19 but I think there are a couple of fundamental issues that 20 maybe underlie that. One is the business model, which we 21 have -- which needs to be addressed. One, the conflicts of 22 interest between the rating and the consulting practices, 23 which has been well documented, certainly in this instance. 24 One of the things we have suggested is perhaps you 25 separate the ratings just like the accounting profession, the 169 1 audit from the consulting practice so that there is a clear 2 delineation, there are two separate entities, there is no 3 conflict of interest there. And I think you can address some 4 of those underlying issues, but I don't think we would be 5 supportive of the SEC proscribing business models and how 6 things should be rated or whether they should be rated or 7 not. The market should determine that. 8 MR. FOHRER: Actually I will give you just a little 9 bit of a layman's perspective. I would want to make sure -- 10 I agree with the conflicts of interest. You want to make 11 sure the information you are getting is fair and accurate. I 12 think the transparency around the process and understanding 13 how they went through in assessing the risk, then as an 14 investor if you read that and you still can't understand it, 15 I think that ought to be a real key about not being in it. 16 That is the layman's perspective, but I worry that 17 the process sometimes is not as transparent as it needs to be 18 to understand how do they go through a risk assessment, what 19 risks do they consider, which risks do they not understand? 20 And again, if I called an informed laymen and I read all that 21 and I still couldn't understand it, that would be a pretty 22 good clue, no matter what the rating, not to get into it. 23 MS. CUNNINGHAM: I agree that there were items that 24 were rated that should have been rated lower. I guess 25 probably not that shouldn't have been rated. But basically 170 1 when you look at sort of what I consider a basis of history 2 in any type of an issuer, whether it is a corporate, an 3 asset-backed, a municipality where you want a business cycle 4 basis for that type of issuer or that type of structure and 5 you don't get it, then I think the answer to that is not that 6 you don't rate it, it is that you rate it several notches 7 lower and that you disclose why it is several notches lower, 8 because this is a new instrument, this is a new issuer, this 9 is a new business and there isn't rating history on a 10 business cycle basis off which to judge. 11 MS. WALTER: Couldn't that be misleading as well? 12 If the answer is that there is insufficient information, that 13 is different than there being negative information. 14 MS. CUNNINGHAM: That is true, but with disclosure, 15 you are telling the story, that it is information that is not 16 available yet. You are not saying that it is negative 17 information, you are just saying that it impacts this rating 18 negatively at this point in time on a disclosure basis 19 because it is not available. Which again, going back to the 20 disclosure, the transparency, says something different than 21 if your response to why is this rated triple B instead of 22 triple A is because the history of the rating is such that 23 the underlying collateral is performing in a triple B manner. 24 That is different than if the history of the collateral is 25 such that it doesn't exist, and so we are being conservative 171 1 in that process of rating. 2 MR. GALLAGHER: It looks like we have run out of 3 time here. I want to thank you on behalf of the staff and 4 the Commission for coming down today and spending time with 5 us. Thank you for the useful exchange and your written 6 comments, and we look forward to working with you as we think 7 about these issues this year. Thanks very much, and I will 8 ask the panelists for panel four at this point to come on 9 down. Thank you. 10 PANEL FOUR: APPROACHES TO IMPROVE 11 CREDIT RATING AGENCY OVERSIGHT 12 MR. GALLAGHER: It looks like we have all of our 13 panelists here, so we should probably get going in a minute. 14 Welcome here everybody to panel four: approaches to 15 improve credit rating agency oversight. On behalf of the 16 Division of Trading and Markets and the Commission, we want 17 to thank the panelists for being here today. We look forward 18 to an engaging panel here. 19 By now I'm pretty sure you have heard the rules of 20 the road so I won't beat you into submission here, but please 21 do mind the four minute limit on introductory remarks. 22 Let me go through an introduce the panel. First we 23 have the Honorable Richard H. Baker, former Chairman of the 24 House Capital Markets Subcommittee and now President and CEO 25 of the Managed Funds Association. We have Mr. Jorgen 172 1 Holmquist, Director General of the Internal Market and 2 Services at the European Commission. Mayree C. Clark, Mr. 3 Christopher L. Gootkind, Vice President and Fixed Income 4 Portfolio Manager, Wellington Management. Mr. Glenn L. 5 Reynolds, CEO of CreditSights Inc. And Professor Joseph A. 6 Grundfest, the William A. Franke Professor of Law and 7 Business at Stanford Law School. 8 Welcome everybody here today. In the order that I 9 just introduced you, I will ask you now to read your prepared 10 remarks. Again, four minutes, and I will flash to you a 11 little yellow card if you will be so kind to look when you 12 have one minute left. Thank you very much. 13 Congressman Baker? 14 MR. BAKER: Thank you very much. Madam Chair, 15 members of the Commission, I appreciate the opportunity to 16 visit with you this afternoon. Having survived under the 17 five minute rule for my tenure in Congress, I shall endeavor 18 to abide by the four minute invitation. Having spent 20 19 seconds, I shall proceed forthwith. 20 Let me point out as a co-conspirator on the now 21 infamous 2006 Credit Rating Reform Act, we certainly had high 22 expectations for success, which unfortunately today are clear 23 have not been gained. There are observations I would make as 24 we go forward in this discussion, though, that I think may be 25 of help, some of which is repetitive from your earlier 173 1 panels. 2 Certainly the credit rating agency role is 3 essential, and by result of statutory requirement, firms are 4 obligated to use their services, which is not a bad 5 franchise. It is not too difficult -- different from the 6 Fannie Mae deal, where if you made money you kept it, if you 7 lost it, well it's somebody else's problem. And at the heart 8 of this matter is the moral hazard created by that 9 quasi-governmental stipulation that the agency rating should 10 be a precedent to the issuance of debt. However, the ratings 11 themselves should not be the singular course of action one 12 should take in making an investment judgment, and there are 13 other material areas of market performance where one could 14 turn for appropriate assessment. 15 In our industry -- and by the way, I do not appear 16 here today in my capacity as CEO of Managed Funds 17 Association, but as an individual speaking my own particular 18 perspectives on the matter -- the credit default swap market 19 has become a de facto rating agency in itself. As credit 20 margin yields widen, there is a greater reluctance to engage 21 that particular instrument or corporation in a financial 22 activity. I would hesitate in the current environment to get 23 into a lengthy discussion about short sales, but there is 24 some value in price discovery that those two mechanisms would 25 yield to a person trying to determine creditworthiness that 174 1 would be of value. 2 However, going forward, if there are to be modest 3 changes suggested, it would be to provide disclosure and 4 transparency. As I have learned, the Commission has only 5 recently required the reporting of information in an annual 6 credit report card. I would suggest examination of that 7 disclosure material be broadened to enable a user to 8 determine from a historic perspective the reliability of the 9 particular rating agency's work over some prior period of 10 time, whether 12 to 24 months. 11 And yet to take it just a step further, that the 12 Commission consider what regulatory response there might be 13 when the reported analysis does not meet expectations. Not 14 establishing what those criteria might be, but if the agency 15 is underperforming -- the rating agency is underperforming 16 expectation or aberrantly to others engaged in the business 17 enterprise, there should be some accountability to the 18 Commission to change the modeling or systems whereby the 19 ratings are being determined. 20 And barring that, some disciplinary action reserved 21 by the Commission for those who fail to meet what I view as a 22 fiduciary duty. Bringing back accountability for one's 23 personal conduct I believe is essential throughout the entire 24 financial market, and certainly it is no less important in 25 the credit rating agency world. 175 1 I yield back the balance of my time. 2 MR. GALLAGHER: I don't know what to do with 3 yielded time, but we will move on to Mr. Holmquist. 4 MR. BAKER: We never knew what to do with it 5 either. 6 MR. GALLAGHER: Mr. Holmquist, speak at your 7 leisure. 8 MR. HOLMQUIST: Thank you very much. Thank you 9 Chairman Schapiro and the Commission for inviting me here. 10 This is the first time I participate in anything like this, 11 and I don't think we have quite the same thing in Europe, so 12 it is very exciting for me, actually, and interesting to be 13 here. 14 I am very happy also because we are in the European 15 Commission. When you say Commission here you think of the 16 SEC. When we say Commission, we think of the European 17 Commission. I use that word, the European Commission. We 18 have put forward a proposal for legislation concerning credit 19 rating agencies. That is currently going through the last 20 steps of the legislative procedure and will be put into law 21 quite soon. 22 We see four basic layers that are important. The 23 first one is robust public oversight. The second one is 24 judicious independent internal oversight. Third thing is 25 market discipline, and the fourth thing is effective 176 1 cooperation at the international levels. 2 To go through them very briefly, on the oversight, 3 we see a global consensus that there is a need for robust 4 public oversight, and that came out of the G20 meetings, also 5 the one recently in London. The sense is, I think, generally 6 that self-regulation can no longer be the full appropriate 7 response to the situation as it has emerged. So the question 8 is not actually if but how one should regulate this. 9 And our proposal is that they should be -- the 10 credit rating agencies should be put under supervision of 11 security supervisors. They should be responsible for the 12 registration of CRAs and their ongoing supervision, and if 13 appropriate, they should be able to take supervisory measures 14 and impose sanctions, including the withdrawal of a license 15 of a CRA. This we will be proposing in Europe. The way we 16 see it, it is not that different from what was decided here 17 in the U.S. a couple of years ago. 18 Secondly, on the internal oversight, we are very 19 clearly of the view, like everyone else, I would say, that it 20 is not appropriate for regulators to interfere in the 21 methodologies of rating agencies or impose on the regulators 22 the task of checking the quality of the ratings, which would 23 be an impossible task. They should develop their opinions, 24 the CRAs, in an independent way. 25 And since this is the case, we feel that some 177 1 organization should have the responsibility to have an 2 oversight role. And we have found a solution to that, and it 3 was not an easy discussion, was the non-executive directors. 4 They should oversee the development of the credit rating 5 policy, in our view, the effectiveness of the internal 6 quality control systems, and they should monitor compliance 7 and governance processes, including the efficiency of the 8 review function in the CRAs. 9 Thirdly, on the increased transparency or the 10 market discipline, we are clear about the fact that 11 regulatory reporting and on-site inspections will not be 12 enough. We think that there should be put more emphasis on 13 increased and improved transparency in terms of their 14 activities. This concerns not only their methodologies that 15 should be disclosed, but also the procedures, for example, 16 that are applied when dealing with conflicts of interest. 17 Our belief is that we can get also market discipline through 18 those kinds of measures. 19 Finally, and that is something that has been 20 discussed, of course -- do I have one minute left? Okay, 21 because my watch stopped working when I came here to 22 Washington. It has nothing to do with that, but the battery 23 was out obviously. 24 MR. GALLAGHER: You still have yielded time, too. 25 You are doing great. 178 1 MR. HOLMQUIST: Fine, thanks. 2 So the room for global oversight of the CRAs. And 3 we have of course the IOSCO code, and some people have asked 4 "Why can't you just rely on the IOSCO code?" We don't feel 5 that that is sufficient. It is quite open in many cases, and 6 it has to turn to legal obligations to be effective. And we 7 are actually basing our proposals, what is now being 8 discussed in Europe, on the IOSCO code and also what has been 9 decided here in the U.S. earlier. And if we just relied on 10 the code, then we would be in a situation where we wouldn't 11 be sure that the rating agencies actually implemented it or 12 implemented it fully, and we will be limited to a naming and 13 shaming, which we don't think is enough. 14 On the other hand, we are of course very much aware 15 that these credit rating agencies are very much 16 internationally active and that it is important to have 17 generally applicable international standards. We are very 18 much in favor of further developing the international 19 standards, and we are also very much in favor, for example, 20 of different regulators from different jurisdictions, like 21 IOSCO, jointly inspecting or looking at the work of the 22 credit rating agencies. So we are certainly very much 23 committed to the international aspects and a global approach 24 to credit rating agencies. We have no interest in going it 25 alone, but we have clearly a strong political pressure and 179 1 interest to regulate in Europe as you have done in the U.S. 2 Thanks. 3 MR. GALLAGHER: Thank you Mr. Holmquist. Ms. 4 Clark. 5 MS. CLARK: Thank you and thanks so much for 6 inviting me to be here. 7 Since there is no organization behind my name, let 8 me just take a minute to introduce myself. First, I am here 9 as a citizen. If I have an axe to grind, it is as a taxpayer 10 and as an individual investor. I do have deep experience as 11 a manager of businesses which provide investment opinions. I 12 have spent almost 30 years in the investment business. I ran 13 Morgan Stanley's equity research operations globally for 14 eight years and a private wealth management business for 15 four. I continue to be involved in the institutional 16 investor community through a firm called Aetos, which does 17 not by bonds, and as the Director of Stanford University's 18 endowment. 19 I, like many who have spoken today, am concerned 20 about the state of the capital markets and particularly the 21 credit markets, and I believe that one of the major issues 22 that has contributed to their fragility is the lack of 23 credibility of the ratings provided by the traditional 24 agencies. And the financial institutions, the ratings for 25 the financial institutions which are so critical to our 180 1 society. 2 I have been particularly focused on the 3 misalignment of incentives associated with the issuer-pay 4 model and have become convinced that we can adopt a better 5 model that will align incentives with investors, the group 6 that the SEC is really charged with advocating on behalf of. 7 My colleague Drew Jones and I have explored several 8 alternative models for achieving better alignment, and some 9 of them are outlined in our written statement, but I'm going 10 to restrict my comments right now to just one model, which we 11 call the designation model. 12 Before I launch into that, I just want to take one 13 more second on context. We strongly believe that the credit 14 infrastructure in the United States, and for that matter, 15 globally -- and by the credit infrastructure, I mean the 16 cadre of personnel with requisite knowledge and experience to 17 assess credit -- has failed to keep pace with the complexity 18 of the markets. And that is true for private sector 19 financial institutions, it is certainly true in the 20 regulatory community, and that complexity of course makes it 21 beyond the scope for most individual investors to seriously 22 consider many of the bonds that are offered today. 23 It is for this reason and many others that were 24 mentioned in the last panel that I believe that we cannot 25 eliminate our reliance on ratings in the near term. I think 181 1 that might be a great long term objective and one of the 2 ideas around the model we are going to talk about is to put 3 us on a path to get there. 4 I also very much sympathize with Ray McDaniel's 5 point this morning about why they switched from an issuer-pay 6 model to an investor-pay model. He said two things first. 7 Technology had advanced to the point where you could easily 8 transmit both research and the ratings, and it was very hard 9 to get subscribers to pay for them because of the free rider 10 problem. That is a very significant issue, and you heard the 11 new interests of the market talk about how difficult they 12 find it to get to scale. 13 And sort of related to that point, as Moody's felt 14 and I think all the agencies feel, they need substantial 15 resources in order to invest in the talent required to really 16 do a good job with the ratings. And again, you hear that 17 from both the existing and the new entrants. So the one good 18 thing about the issuer model is that the ratings fees are 19 actually tied to value added, right? Because the value added 20 of a rating is really the difference between the value of an 21 unrated bond and a rated bond. 22 So our idea is to find a different way to make sure 23 that the total pool of rating fees is equal to the value 24 added, but that we let the bonds itself fund the ratings, but 25 we let the investor decide who really gets paid for the 182 1 ratings. So if I can just take a little bit of Mr. Baker's 2 time here, I'm going to take one extra minute -- 3 MR. GALLAGHER: You are going to benefit from my 4 inexperience with yielded time. 5 MS. CLARK: -- to describe how the model works, 6 because I think a lot of what I have said is being listed and 7 we don't take the time to go through it. 8 So we start with a bond that is about to be issued, 9 so we are going to do this through the new issue process. 10 The issuer provides information to all of the rating agencies 11 who want to rate it consistent with the current -- new rules, 12 actually. The rating agency is going to decide if they want 13 to rate it or not. If they do, they actually give -- they 14 publish the ratings, available to everybody, and they give 15 the credit research to those potential purchasers who are 16 interested in the bonds. 17 The rating agency fees are then deducted from the 18 proceeds of the issue, just as they are now. But the rating 19 agency fees are put in a pool and that pool is escrowed, and 20 the allocation of that pool is determined by those investors 21 who actually buy and own the bonds, okay? And they get to 22 designate that in proportion to their ownership of the bonds. 23 The same thing happens along the way as we do coupons, we 24 also deduct rating agency fees, and again, we let the 25 investors who buy or own the bonds designate which rating 183 1 agencies get the fees. 2 And at the end, by the way, we have one last rating 3 agency fee maturity. So if the bond makes it to maturity and 4 there is a payment at the end, then again there is a fee paid 5 and the investor community gets to designate who gets that 6 fee. 7 There are at least eight advantages to this model 8 over the current model, but I'm going to take a second and 9 make sure everybody has had a chance to absorb what I said. 10 Yeah? You have a question already. 11 MS. WALTER: You are assuming that the investors 12 are going to do an analysis of the different ratings 13 available to them and determine which is of the highest 14 quality? Is that expecting too much? 15 MS. CLARK: Well actually, let me come to that at 16 an appropriate moment, but the answer is yes. 17 So let me just run through the advantages and then 18 we can take any questions either in this public forum or at 19 another forum. 20 First, this mechanism really encourages new entry, 21 because now unsolicited raters have a real opportunity to 22 capture revenue associated with the rating. They are more 23 likely to rate the bonds. It encourages competition based on 24 quality, not based on rating shopping. Because the investors 25 who are getting the information are of course going to 184 1 designate the people who have provided the best information. 2 We can talk more about that in a minute. 3 Three, it allows the ratings to be published for 4 the benefit of everyone in the public, including smaller 5 investors and regulators and smaller institutions -- anyone 6 who wants to use the ratings as a screening device, which 7 many institutions do. But, it still causes the amount that 8 is paid -- you eliminate the free riders in the sense that 9 the money that is designated to the rating agencies actually 10 gets to the rating agencies, the entire amount equal to the 11 value added, not just the amount that those good citizens are 12 willing to pay as subscribers, okay? 13 And because of that, it really encourages the 14 buyers to pay attention to the underlying credit work because 15 they get it right along with the bonds. And my hope is that 16 over time that would also incent the infrastructure to be 17 rebuilt. 18 Fifth, it allows the agencies to pay for talent 19 because it forces those who get the benefit of the ratings to 20 pay for them. Sixth, it creates a feedback loop between 21 investors and the rating agencies, which is helpful to 22 getting the rating agencies to really use better 23 methodologies. Again, I can talk about that more in a minute 24 if you are interested. 25 Seventh, this is very important. It is practical 185 1 to implement this. This does not require legislation. We 2 could implement this with the cooperation simply of the 3 people who have been in this room today, including our 4 European counterparts. And eighth, we can start 5 experimenting with this immediately. So in the regulatory 6 community today, we happen to have a few people who are quite 7 influential in the new issue process, which is where you need 8 to start, right? And so via TARP and TALF we could start 9 working with this just as soon as we decided it was the right 10 thing to do. 11 Let me stop. Thank you for the time. 12 MR. GALLAGHER: Thank you much Ms. Clark. Mr. 13 Gootkind. 14 MR. GOOTKING: Thank you for inviting me to speak 15 and participate in the roundtable on the rating agencies. 16 As a credit analyst and research director and now 17 fixed income portfolio manager, I have utilized credit 18 ratings for more than 20 years. I would like to briefly 19 describe what I believe went wrong with the NRSROs and that 20 designation, discuss the misaligned incentives between the 21 rating agencies and the investors, who are the primary users 22 of ratings, and then suggest some changes to the current 23 system. These suggested changes will help achieve the 24 objectives of improving ratings quality and fostering greater 25 accountability, transparency, and competition. They may also 186 1 reduce systemic risk. And I hope to do all of this in less 2 than four minutes. 3 So what went wrong with the NRSRO credit ratings? 4 They failed to accurately rate structured finance securities 5 and vehicles -- we have heard that a lot today throughout the 6 course of the day -- particularly housing related and the 7 CDOs and the SIVs. Critical market participants relied too 8 much on NRSRO ratings and the NRSRO designation should not 9 equate to a government good housekeeping seal of approval, 10 which is what I think it is these days. That has had lots of 11 unintended consequences. 12 The current regulatory requirements involving, for 13 example, money market funds, risk based capital requirements 14 for banks, and insurance companies, as well as broad bond 15 market indices, such as the Barclay's aggregate bond index, 16 formerly Lehman, and most investment guidelines like the ones 17 I work with all reinforce this reliance on the ratings. All 18 this reliance on just a few rating agencies has led to 19 systemic risk, and I think that is an important element to 20 your -- that is maybe more of an issue for the Fed and the 21 Treasury, but I think the SEC should be thinking about that 22 as well, that designation. 23 And also rating changes often severely lag market 24 price changes. We have heard that as well today. They are 25 not adequately capturing market risk, and therefore the value 187 1 of ratings to fixed income investors is diminishing. And I 2 can tell you that I have many bonds in my portfolio that are 3 still rated investment grade and are trading at 50 cents, 40 4 cents, even 20 cents on the dollar. If time permits or 5 during the Q&A, I will explain what I believe to be a flaw in 6 the single point rating system. 7 So let me move to misaligned incentives. Again, 8 probably repeating a little bit of what you have heard today. 9 The issuers incentives are to obtain the highest credit 10 rating and lowest financing cost possible, so therefore they 11 are going to select the rating indices that give them the 12 highest credit rating. And their incentives diminish after 13 the issuance of their debt, they have got their money. This 14 is particularly true if the issuer is not a recurring 15 borrower in the capital markets. 16 The rating agency incentives are to maximize 17 revenue and profits. They are profit-seeking or 18 profit-maximizing enterprise. Most of the fee is paid at the 19 time of the issuance and is contingent upon the issuance. 20 And as we have heard many times today, there is little to no 21 financial consequences if their opinion is wrong, i.e., they 22 are not accountable. 23 Investors, I would argue, desire the accurate 24 rating throughout the life of the bond to assist them in 25 properly assessing the risk and pricing the risk and 188 1 allocating capital, and of course they don't have recourse to 2 the rating agencies for mis-rated securities, as again we 3 have heard a bunch today. So any solution requires balancing 4 the risks and incentives, but bondholders bear the long term 5 risk associated with the rated security and the current 6 issuer-pays model does not align the interest of the rating 7 agencies with the bond investors. I think this has been 8 particularly true with structured finance, but so much so in 9 the corporate area as well. 10 So some suggested changes to the current system? 11 First one, you have heard this one again today, too, 12 investors pay for the credit ratings, and here is where I 13 differ a little bit from Mayree. They pay for it out of the 14 bond coupon and bondholders have the right to select and/or 15 replace a rating agency. This aligns the incentives of the 16 rating agency with the interest of the issuer and investors 17 throughout the life of the bond. Again, they get paid out of 18 the coupon, so as long as the bond is outstanding, they are 19 going to get paid the fee. 20 This promotes accountability, competition, and 21 accuracy of ratings. It also addresses the free rider 22 concerns that we have heard about with some of the 23 subscriber-pay models. I can go into some more detail on 24 that fee during the Q&A if that makes sense. 25 And the other primary recommendation is to 189 1 eliminate the national recognition or NRSRO designation of 2 any credit rating agencies. This would reduce the likelihood 3 of systemic risk linked to rating changes, lets the market 4 determine the rating agencies that are best at analyzing and 5 communicating credit risks. It reduces the use and emphasis 6 of ratings as a substitute for the analysis in the regulation 7 of financial institutions like insurance companies and banks 8 as well as counterparty transactions. 9 And let me pose this to people here. Equity 10 markets involve higher risk than bond markets, and yet we do 11 not have NRSRO designation for equities. What is unique 12 about fixed income securities that requires U.S. government 13 recognition and inclusion in rules and regulations? 14 MR. GALLAGHER: Thank you Mr. Gootkind. Mr. 15 Reynolds. 16 MR. REYNOLDS: Good afternoon. Just by way of 17 background, CreditSights is an independent research firm. We 18 offer fundamental research products and ratings also. We do 19 not manage money and we are not involved in underwriting, and 20 we are a subscriber paid business. We operate in the U.S., 21 Europe, Asia, and we have been able to grow from 8 employees 22 back in 2000 to 120 today, so in a way we are an interesting 23 petri dish to compare on subscriber-pay models. 24 I think the reason we have been able to grow, 25 partly, relative to some research boutiques, is that we have 190 1 not narrowly focused on being an NRSRO as our reason for 2 existence. If we wanted to have a business model that was 3 viable in this market, making NRSRO status a priority for us 4 made no sense. From our standpoint, the NRSRO business is 5 plagued by a stacked regulatory deck and that is still the 6 case today. The way it is set up, a competitor cannot be a 7 meaningful factor except in a small subset of the asset 8 classes. The fact that the SEC is starting to designate a 9 handful of new ones does not change that. 10 The Credit Rating Agency Reform Act was a start in 11 terms of competition, but there needs to be much more follow 12 through. This is especially the case in structured finance 13 and its myriad subcategories of asset classes. The recent 14 initiatives by the SEC are on equivalent disclosures are a 15 great start. The regulatory trail overall in the post-Enron 16 years until now has not been inspirational in any way. 17 We first got involved back in the rating agency 18 reform process back in March 2002 in the Senate hearings on 19 Enron and then we were involved in the SEC roundtable back in 20 the fall of 2002. We got involved in the 2005 ramp-up on the 21 Rating Agency Duopoly Relief Act, which had a catchy title, 22 and then the 2006 Senate hearings and the final bill. The 23 fact that the agencies lobbied so hard to have a high margin 24 structured finance asset class hived off of that bill under 25 the Senate version is probably for me the most important take 191 1 away from the legislation in terms of what their motives are. 2 It certainly stands out given what has transpired since in 3 structured finance. 4 So after the seven year stretch, there is not a 5 great deal to show for it, even if the 2000 act was a good 6 start. We have to say we have been amazed how successful the 7 incumbent agencies have managed frustrated the process of 8 regulatory reform over those years. From our own experience 9 with various parties, we have come away with a few 10 impressions. 11 First, the Treasury Department was not very helpful 12 to the SEC in rating agency reform. Maybe it was the Wall 13 Street connection. The Democrats and Republicans, as usual, 14 go partisan and the House Rating Agency Bill was painful 15 evidence of that. When we look at where we go from here, 16 basically the disclosure and documents have been what we have 17 been focusing on and we have basically referred our comments 18 with the Commission. But the one thing we can be sure of is 19 the duopoly will seek to change the debate again and seek to 20 slow the process down. 21 We may need some additional legislation in some 22 areas, but after the 2005-6 process, I can imagine that is a 23 very daunting prospect. Again, we will refer the Commission 24 to our filed statement. 25 MR. GALLAGHER: Thank you Mr. Reynolds. Professor 192 1 Grundfest? 2 MR. GRUNDFEST: Thank you Mr. Gallagher. I would 3 like to divide my comments into two parts, first briefly 4 discussing what won't work, and I'm afraid that many of the 5 ideas that we have heard about today simply won't work, and 6 then second talk about two sets of ideas that actually could 7 work. 8 First, what won't work. There are five major 9 proposals on the table and I feel that all five of them fail. 10 First, several proposals to end regulatory reliance on 11 ratings. I think it is a great concept, but it is not going 12 to solve the problem. Let's recognize that many private 13 contracts are hard-wired to require extensive reliance on 14 those ratings and those private contracts will continue to be 15 in existence no matter what the Commission does. 16 Second, formally, these requirements also exist in 17 banking regulation, they exist under state law. They are 18 wired into the Basel Capital Adequacy Standards. Even in the 19 SEC waved its magic wand, it would not eliminate regulatory 20 requirements, and we have to recognize that reality. 21 More fundamentally, there is a large and important 22 public demand for free high quality ratings and that demand 23 will continue to exist even if regulatory reliance on these 24 ratings ends. We have to recognize why that is the case, and 25 that is there is legitimate economically rational interest in 193 1 the existence of public, free, high quality ratings in an 2 environment that is subject to extensive economics of scale 3 and very strong embedded reputational capital costs. We need 4 a solution that works in the environment we face today, not a 5 solution that works on some abstract blank piece of paper as 6 though we are coming to the problem for the first time. 7 Second, there is a set of solutions that look at 8 ending the issuer-pay model, moving to a subscription or 9 user-pay model. I think it is great that there is 10 subscription services out there, but they can't solve the 11 bigger problem. The free rider problems are simply too 12 large, there is significant fee allocation issues, and it 13 can't be practically implemented, again, it we want to 14 satisfy the large public interest in the continued existence 15 of free and credible ratings. 16 Third, there are suggestions to create new 17 regulatory bodies to oversee rating agencies. With all due 18 respect to everybody in this building, everybody who has ever 19 been in this building and everybody in the future who may 20 come into this building, the government doesn't know how to 21 rate instruments or how to exercise substantial oversight 22 over the rating process. The danger of political 23 interference in the rating process is huge and only gets 24 bigger by the day. We have already heard of several problems 25 in this regard and why we want to head down a path that only 194 1 makes that clear and present danger only more so is something 2 that escapes me. 3 The fourth idea is to create a government sponsored 4 rating agency. The only good thing I can say about this idea 5 is that it makes the third idea look good. 6 (Laughter.) 7 MR. GRUNDFEST: Last, we have the increase in legal 8 liability, and here, let's recognize that post-hoc liability 9 is not as a practical matter an effective substitute for ex 10 ante incentive compatibility because if legal liability 11 really did achieve that result we would not be in the 12 situation we are in today. 13 Very briefly, what can work? There are two 14 approaches that could work. First, I think Mr. Gootkind and 15 Ms. Clark have already spoken about one set of approaches if 16 you tweak you could work. 17 The other approach is to move to something called a 18 buyer owned and controlled rating agency, or more accurately, 19 an investor owned and controlled rating agency, and I would 20 like to thank Commissioner Paredes for a conversation that 21 elucidated the difference. 22 How would it work? Very simply, a two step 23 process. You define an investor owned and controlled rating 24 agency so that the agency is dominated and controlled by 25 sophisticated investors who are consumers of the rating 195 1 industry's product. The agencies can even be required to be 2 not for profit, they can be for profit. There are many 3 design issues about how you decide who is a sophisticated 4 investor and how you define control. All of these are 5 entirely manageable. 6 Second, most importantly, making the economics 7 work. You require that every issuer paid rating be 8 accompanied by at least one IOCRA rating that is also paid 9 for the issuer. So in other words, although we preserve the 10 issuer-pay model, what we do is we put at least half of the 11 revenue stream under the control of the investors. 12 Now there are six immediate benefits that result 13 from this approach. First, you get immediate new entry by 14 sophisticated new entities. To my knowledge, none of the 15 other proposals that we have heard to date achieve that 16 objective. Second, these new entrants will be able to 17 generate ratings that have a fundamentally different point of 18 view. They will reflect the perspective of investors who on 19 average and over time are both buyers and sellers of the 20 rated instruments. 21 Third, these new entrants will give rise to new 22 innovations that I expect to dislocate the comfortable, 23 dominant duopoly structure that we have in the marketplace 24 today. Fourth, there is no need for a three year waiting 25 period, because in order to succeed, these IOCRAs need not be 196 1 NRSROs. You use your authority over the NRSROs to require 2 parallel credit ratings, which don't have to be issued by 3 NRSROs in order to qualify the NRSRO requirements. Fifth, 4 you don't need any additional legislative authority. 5 Everything that you need is already in the Rating Agency Act. 6 And sixth, most importantly I think in terms of 7 long term incentives, if the IOCRAs screw up, the investors 8 have only themselves to blame. This gives the investors a 9 seat at the table, they will have authority and 10 responsibility, and if they get it wrong, they look in the 11 mirror, they will know that they have got it wrong. 12 Thank you. 13 MR. GALLAGHER: That is terrific. Thank you 14 everybody for your comments. I think we have a little bit of 15 material to work with here for our Q&A session, so I'm going 16 to quickly move over to the Q&A. I will ask the panelists, 17 at any time if you want to be acknowledged, if you want to 18 jump into a question, just give us a look. We will make sure 19 you get recognized. 20 Chairman Schapiro. 21 CHAIRMAN SCHAPIRO: I just want to ask Joe kind of 22 a practical question. Have you talked to institutional 23 investors about whether they would be interested in doing 24 this and why haven't they maybe done it already? 25 MR. GRUNDFEST: Let me not name any names, but I 197 1 have been on the phone with some brand name organizations and 2 I have been invited to call them back, and they say "if the 3 Commission has an interest in pursuing this, yes, we will 4 stand up and say that we would be interesting in actually 5 forming organizations that would qualify and would be willing 6 to enter and compete." 7 As to why they haven't stepped forward to do it 8 now, it makes no sense to spend a lot of money to try to 9 compete with Moody's and Standard & Poor's given their 10 current dominate position in the marketplace. And it is not 11 because they have done anything wrong per se. There are some 12 markets that just naturally evolve into duopoly structures 13 and have very high barriers to entry, and this is one of 14 those markets. 15 CHAIRMAN SCHAPIRO: We talked in an earlier panel 16 about some of the parallels, and a number of things that have 17 been suggested today to the settlement in the investment 18 banking cases and research analyst conflicts of interest 19 where track record publication was made a key part of that 20 settlement, and the provision of independent research by the 21 investment banks when they provided their own research was 22 also a key piece of it, and this is kind of a similar -- it 23 is parallel really to that sort of a provision. 24 MR. GRUNDFEST: There are similarities, but I 25 wouldn't overemphasize the similarities because in this 198 1 situation what you would have is sophisticated investors, 2 sophisticated buy-side investors establishing organizations 3 and making investments in these organizations in such a way 4 that they would be expected to be around for a very long 5 period of time as opposed to a, let's be candid, check the 6 box mentality that has evolved in the environment of the 7 analyst settlements where nobody really expects these 8 organizations to be around once the term of the settlement is 9 over. 10 MR. GALLAGHER: Commissioner Walter. 11 MS. WALTER: And what are the parts of this that 12 will make sure that that happens, that it isn't something 13 that is just a little plant that dies as opposed to growing 14 into a big tree? 15 MR. GRUNDFEST: Well as long as the regulatory 16 requirement is in place that if any -- and I mean here is how 17 the requirement would be written. If any NRSRO issues a 18 rating or an opinion that is paid for by the issuer or 19 sponsor or promoter of the security, then in order to address 20 the potential conflict that exists from that business pay 21 model, there has to be at least one simultaneously issued 22 rating or opinion by an IOCRA, an investor owned and 23 controlled rating agency. 24 And as long as that regulatory requirement remains 25 in place, you are going to have a guaranteed stream of 199 1 revenue to support this alternative voice in the market 2 because you then can't have a nickel go to Moody's or S&P 3 under the current business model without also having funding 4 go to the investor side of the marketplace. 5 MR. GALLAGHER: Mr. Reynolds, you wanted to jump 6 in? 7 MR. REYNOLDS: Sure. We have kicked that around in 8 the past also. We have been approached by investors over the 9 years, and one of the things that we had trouble -- and it is 10 a great concept and it may work. There are a lot of 11 questions of governance, for sure, and anyone who has dealt 12 with a lot of different types of institutional investors, 13 everyone is in charge, you might discover. So you have that 14 problem too around making decisions. 15 The other thing is you still -- I question whether 16 or not you need legislation here anyway because there are 17 still some massive issues around information disclosure which 18 a lot of the committee's hearings earlier addresses. There 19 is a tremendous amount that the market cannot get access to 20 in the area of structured finance, especially in the area of 21 loans. So there is just a lot of practical issues. 22 I still think you are not going to be able to duck 23 legislation in this process if you really want to cure the 24 problem around analysis disclosure, whether it is investor 25 owned or whether it is owned by venture capitalists or a 200 1 strategic -- there is still a lot of wood to chop in that 2 area. 3 MR. GRUNDFEST: If I may, we can solve the investor 4 information disclosure issue in this environment simply 5 through the wording of the requirement because the issuer is 6 going to want the rating by the NRSRO. The NRSRO is not 7 allowed to give the rating unless an IOCRA also gives the 8 rating, therefore the issuer will give the information to the 9 IOCRA. End of story. 10 MR. REYNOLDS: We are talking about structured 11 finance though, and then what you have done by using that 12 model is you have precluded all others. So if we are talking 13 about competition here, you have to have competition that is 14 open to a whole range of models. So if you just -- you are 15 substituting one problem for another in a way, but it is a 16 higher class problem because it is investors who own the 17 assets at least as opposed to someone who is just charging a 18 fee and moving on to burn the next village. 19 (Laughter.) 20 MR. GALLAGHER: Commissioner Casey. 21 MS. CASEY: That was actually my exact question, 22 Glenn, was to appreciate what the implications would be for 23 competition and other models. Joe, do you have a response on 24 how -- I mean is it just you have to have a recognition that 25 it is inherently a duopoly or oligopoly market? 201 1 MR. GRUNDFEST: I think we have to recognize that 2 the marking business for a variety of reasons is not going to 3 be a cottage industry. There are many industries that have 4 natural characteristics which mean they are going to be 5 dominated by a relatively small number of firms. And you can 6 have these large, dominant firms, and at the same time you 7 can have many smaller, successful firms, for example such as 8 Mr. Reynolds that have a different business model, but they 9 will never grow to become the dominant firms. 10 My expectation is that if you were to adopt an 11 IOCRA requirement, you would get no more than at most two 12 dominant IOCRAs as a practical model. It would 13 dramatically -- it would significantly reduce the cash flow 14 to the current dominant firms, so therefore I would expect 15 that Moody's and Standard & Poor's are not going to be great 16 champions of this IOCRA approach. Back of the envelope, you 17 will probably take half the cash flow away from Moody's and 18 Standard & Poor's and move it to the new IOCRAs. 19 And what you then get is a different form of 20 competition. Rather than the notion of lots of shrimpatarian 21 entry and exit, you have two big new sophisticated 22 competitors with embedded incentives to look at the market 23 very differently competing against Moody's and Standard & 24 Poor's, who I expect will continue to exist and continue to 25 be quite profitable. 202 1 MR. GOOTKIND: Joe, if half the fee goes away from 2 Moody's and S&P, then don't you diminish the quality of the 3 research that they have been providing and doesn't that -- so 4 you have this idea that we have to have the duopoly in place, 5 but yet we are even going to diminish the quality even 6 further of the ratings? 7 MR. GRUNDFEST: Well, you know, that is a very, 8 very interesting question because if you take that argument 9 to a logical extreme, the problem is that we haven't been 10 paying Moody's and Standard & Poor's enough in the past, and 11 that if we would have simply doubled the amount that we paid 12 them in the past, we wouldn't have the problem with 13 structured instruments and the life. I think that Moody's 14 and Standard & Poor's will able -- 15 And just as another aside, if you have a look at 16 the financial statements of Moody's, there was a five year 17 stretch when it was apparently the most profitable firm in 18 the United States, and they are not paying their analysts 19 what Goldman Sachs is paying their people. There is a 20 tremendous amount of what economists call rent that Moody's 21 and Standard & Poor's are able to get because they are 22 Moody's and Standard & Poor's and it is rents that are not 23 passed on in terms of salaries being paid to analysts. 24 MR. GOOTKIND: I might argue that the reason they 25 are so profitable is because the NRSRO designation gives them 203 1 that imprimatur of the government and it creates the duopoly 2 and the systemic risk that I discussed earlier. 3 MR. GALLAGHER: Commissioner Casey, you had a 4 follow up on that? 5 MS. CASEY: No. I appreciate that. Thank you. 6 MR. GALLAGHER: Commissioner Walter. 7 MS. WALTER: If the IOCRAs are paid by the issuers 8 as well, why doesn't that recreate the same system because 9 another line of business for those investors, which recreates 10 the same sort of conflict of interest that got us here to 11 begin with. 12 MR. GRUNDFEST: Yes. If I could figure out some 13 way to actually make a user-pay or an investor-pay model work 14 well, I think that would be a great idea. I do think Ms. 15 Clark and Mr. Gootkind have an approach where if you say we 16 want to do the best we can on a user-pay side, I would look 17 in the direction that they have recommended. 18 The bet with regard -- and it is a bet -- the bet 19 with the regard to the IOCRA approach is that the 20 institutions that wind up controlling these organizations and 21 being in charge of how ratings are set and the like will 22 recognize where their bread is buttered and that they have 23 got a much greater interest in making sure that they have 24 quality ratings upon which they and all other members of the 25 investor community can rely than caving in to the interests 204 1 to the specific issuers who are going to be paying for the 2 ratings. 3 CHAIRMAN SCHAPIRO: Just to follow that up for a 4 second. We earlier had some conversation about some other 5 ideas about pooling funding, for example through a listing 6 fee or a transaction fee or creating a rotational requirement 7 such as exists on the audit side. I would love to hear 8 comments from any of the panelists on whether those ideas, 9 which admittedly are more around the periphery than what we 10 have been talking about, make any sense in terms of oversight 11 and breaking the connection between the issuer and the rating 12 agency. 13 MR. GALLAGHER: Mr. Reynolds. 14 MR. REYNOLDS: One of the things we have already 15 tried to focus on is that the markets are blurring across a 16 lot of different business lines. It is really not just 17 ratings. In a way that is somewhat of a Jedi mind trip by 18 the duopoly themselves to try to get people to over-focus on 19 that. 20 Because you are talking about analytics, data, 21 services, consulting. They are buying firms left and right, 22 which is their right, but people who are in those spaces, 23 especially some of the large well-capitalized financial media 24 firms are supposed to be getting in their backyard -- I don't 25 believe that the boutiques are going to cure anyone's ills 205 1 either. You can take -- and we are by far and away I think 2 the biggest boutique out there by a significant multiple, and 3 we feel tiny. So we feel big in a room with boutiques and we 4 feel like the proverbial fly on the elephant when we are 5 standing next to Moody's. 6 But when you all the sudden take a lot of different 7 high information content, high quality institutional level 8 products and roll them up into a financial media firm, 9 whether it be a Thomson Reuters, an IDC, or just picking them 10 out of the air, there are big companies out there that 11 Moody's and S&P compete against. They should have the right 12 to compete against them. 13 And this three year holding period is what has been 14 a real problem, I think, for structured finance. I mean it 15 is easier to raise an armed militia in terms of holding 16 periods, so it doesn't make a whole lot of sense. 17 MR. GALLAGHER: I'm sensing a theme here, Mr. 18 Reynolds. 19 MR. GOOTKIND: I think it is important to note that 20 CreditSights, Glenn's firm, can co-exist with the rating 21 agencies, and I think you have built a fairly successful 22 model, and full disclosure, we are a subscriber to 23 CreditSights. So they are not a rating agency, but they 24 provide independent, high quality credit research, and so 25 that is a free market model that I think can and has worked. 206 1 So I think we can look to them as a model that can work 2 outside of the NRSRO designation. 3 MR. REYNOLDS: What is interesting, though, is that 4 we deemphasize ratings by design because we wouldn't have a 5 revenue model. You need bodies, you need websites, you need 6 lawyers, you need compliance, you need offices overseas, and 7 that takes cash. So as much as people -- the most ridiculous 8 assertion that by taking more rating agencies into the space 9 you are going to increase risks of free riders. You will 10 have no revenue, you will no bodies, you will have no brain, 11 you will have no pulse, so it doesn't work. 12 MR. GALLAGHER: Ms. Clark? 13 MS. CLARK: I mean I think that is so true. I 14 think it is so interesting that you don't go to being an 15 NRSRO and I presume that is because also you would then have 16 to publish your opinion and make it free to everybody. I 17 know that was a constraint for many of the people who were 18 talking earlier this morning. They were quite put off by the 19 need to do that. 20 MR. REYNOLDS: We could post the ratings, but we 21 wouldn't post the information and the analytics, the 22 algorithms, and so on. 23 MS. CLARK: Right. So the idea of setting aside 24 this pool is a good idea. That I think makes a lot of sense, 25 and linking that pool to the value added is a good idea, 207 1 right? The issue is that the way it works right now is the 2 issuer is the one who gets to decide who gets that pool and 3 they are conflicted in every dimension. There is no one 4 inside an issuer, no one, not the audit committee, not the 5 board of directors, who wants anything other than the highest 6 possible rating. 7 The advantage of any of these things we have talked 8 about is they get the investors in the game, because even 9 though they might have some conflicts too, there is always 10 somebody inside that investment institution that really wants 11 the ratings to be fair and also wants good, solid underlying 12 credit research. And the more you can make those rents that 13 you talked about that are associated with the NRSRO available 14 to a group of people who are driven by the investor, then you 15 are going to get the alignment that I think you as an 16 organization really would seek. 17 MR. GALLAGHER: Professor Grundfest? 18 MR. GRUNDFEST: If I may, I do think Ms. Clark's 19 idea -- if the Commission wants to look at going down the 20 pool approach, I think looking at some of the thoughts that 21 Ms. Clark has in her submission are very good ones because 22 what they do is they give the investors control over 23 allocation of the portion of the pot that is going to be used 24 to support the ratings agencies, which in effect could mean 25 that if the investors want to support new entrants, they 208 1 would be able to do it. And it would give the rating 2 agencies a reason to listen to the investor side that they 3 currently don't have. 4 MS. CLARK: And that is a really important point 5 because what you would find is that the rating agencies start 6 competing for that pool of money. And what does that mean? 7 They have got to really focus on the biggest payers of the 8 pools, and who are the biggest payers of the pools? They are 9 the people who have had the best performance in fixed income, 10 so they tend to be the savviest investors. One of the 11 disturbing things I find is often the rating agency personnel 12 don't understand what methodology the most savvy investors 13 are using, and that means they are not really at the cutting 14 edge. And one of the reasons for that is because they got no 15 economic incentive other than the sort of long term franchise 16 value to really be in a dialogue with those people. 17 But if you put this kind of a system in place, you 18 would find there is lots more interaction between the 19 agencies and the investors, and you are going to get work 20 that is actually better. I mean I promise you Glenn knows 21 who the savviest investors are and what assumptions they use, 22 right? He has to because it is his lifeblood. 23 MR. GALLAGHER: Mr. Baker, there has been some 24 discussion about whether legislation would be required in 25 this context. I don't know if you have any thoughts on 209 1 handicapping whether A, it would be required, and B, if so, 2 what the chances would be, and actually C, a historical 3 question here, whether anything like this was discussed at 4 the time of the Credit Rating Agency Reform Act. 5 MR. BAKER: No, a proposal was not that I'm aware 6 was discussed that was similar in the context of the Credit 7 Reform Act, but there is still a potential for legislative 8 modification. I would presume without having examined this 9 that because of all the hardwiring that was referenced in the 10 current statute, both state and municipal, and the 11 requirement for one to engage one of the NRSROs, it is 12 unclear exactly how application would proceed. 13 But assuming that would be manageable, the only 14 other remaining issue of consequence looking at it would be 15 the assurance of ownership interests. And how does one know 16 what constitutes an investor owned entity that would be 17 issuing the rating that would be offsetting to the NRSRO? 18 Probably very simply satisfied, but if there are issues that 19 would have to be examined, and I suspect that Mr. Kanjorski, 20 who is the Chairman of Capital Markets or Senator Dodd would 21 want to fully understand this before saying that it 22 represents the remedy. 23 But I think it is an intriguing idea and it 24 certainly ought to be fully explored to determine 25 applicability. 210 1 MR. GALLAGHER: Commissioner Casey. 2 MS. CASEY: I just had a follow on to that 3 question. In your statement, Congressman, you mentioned that 4 many of the same questions that were asked back in 2005, 2006 5 are now being asked again, and you also mentioned, I think 6 quite rightly, that market realities have caught up with the 7 intent and spirit of the law. 8 So not necessarily specific to the kinds of ideas 9 that are being thrown around here as possible addressing the 10 residual concerns that continue about the quality of ratings 11 and ratings performance, what different questions or 12 different issues, whether it is conflicts, whether it is a 13 question of -- whether you recognize the conflicts at the 14 time, are there different concerns that are being raised that 15 weren't recognized back in 2005 and 2006 versus the remedies 16 and the reforms that we are considering? 17 MR. BAKER: Well the intensity of examination is at 18 an all-time high because of the obvious market conditions. 19 And there was an understanding there was a problem in the 20 oligopoly, as it was delightfully called. It presented 21 challenges to those who wished to enter the market space and 22 found it impossible, literally, to do so. 23 I think in the current environment, having a system 24 that works is the goal everyone is united in achieving, and 25 that means transparency at the very least. It means holding 211 1 one's performance up to public examination at the very least. 2 And I don't believe that prior to the current act that either 3 of those goals was attainable. They were very much black 4 boxes to which questions would be asked and no answer would 5 return. 6 And so I think minimal progress has been made, but 7 clearly, working within the realm of the possible, the 8 disclosure suggestions I have made, certainly are achievable 9 in the short term, would not present any downside risk while 10 the Commission examined the more sweeping measures that might 11 be considered later. 12 MR. GALLAGHER: Commissioner Paredes. 13 MR. PAREDES: Quick question for Professor 14 Grundfest and one for Ms. Clark, and they are related in 15 terms of going into some of the particulars. 16 For Professor Grundfest, if you could just speak a 17 little bit -- and others as well, if they would like -- to 18 what the marginal benefit, if you will, is in terms of 19 changing outcomes versus where we presently are, of having 20 the structure as compared to leaving it to the independent 21 credit analysis that the institutional investors, if they are 22 so inclined, could presently run. I presume there is an 23 economies of scale story, but there are probably other 24 aspects to whatever the benefit is of institutionalizing the 25 function as compared to leaving it to individual investors to 212 1 run their own independent analysis as an offset, if you will, 2 to whatever an issuer-pay rating would be. 3 And then in terms of Ms. Clark's proposal, if you 4 could perhaps go back to, I think, one of the earlier 5 questions as to the kind of ongoing monitoring that would be 6 required by the investors to then, over time, make the kinds 7 of determinations as to how those dollars that are pooled 8 would ultimately be allocated. 9 MR. GRUNDFEST: Well first with regard to the 10 observation about the marginal benefit, I think we simply 11 have to accept the reality that there is a tremendous set of 12 informational economies of scale. This is a traditional 13 information market where once you have one person do the job 14 of doing a good, honest, and credible rating, the marginal 15 benefit to society of having a second person do the job is 16 much, much smaller. There are situations that are sort of 17 unique and boutique-y where it does make sense to put in that 18 additional investment, and that is why you get firms like Mr. 19 Reynolds' and the like to find situations where even if you 20 do a good job on average, there is additional information 21 that can be extracted from the market. 22 Now given that, the question then becomes how do we 23 handle the fact that an issuer-pay model is likely to 24 continue to dominate in the marketplace? That really leaves 25 us only one strategy, and that is to take the market with 213 1 large economies of scale and embedded reputational effects, 2 keep the issuer-pay model, but do it in such a way that 3 governance mechanisms causes the money to support a set of 4 views that more closely reflect investor interests, and that 5 is what the IOCRA proposal is intended to do. 6 It is really driven by the pragmatic nature of the 7 problem that the SEC faces, and if the SEC defines the issue 8 as being how are we going to get new entry that is 9 intelligent and reflects a different point of view in the 10 market space that generates reliable ratings that everybody 11 in the public can look at, and let's try to make it happen 12 within two or three years, I think an IOCRA approach can do 13 that. 14 MS. CLARK: I just have one thing on this business 15 of the marginal benefit of jettisoning the issuer-pay model, 16 if we can call it that, if that's okay. 17 You know, the trouble -- if you think there is any 18 conflict at all in an issuer-pay model, then all the ratings 19 are worthless in this sense. If I buy something that was 20 rated -- or if there is a bond here in front of us that 21 should be rated single A, without any conflict, and then I 22 look at a list of bonds, was that bond that was rated A, was 23 it bumped to a double A, was it bumped to a triple A? Do you 24 understand what I'm saying? Like I don't know how much 25 conflict is there. If I think there is any there, then I 214 1 don't care whether it is triple A, double A, or single A, I 2 don't know what it is, right? 3 So as an individual investor, I have to do all the 4 work from scratch. Similarly, I don't know which bond they 5 fudged. Was it bond one, two, three, four, five, six, seven, 6 eight, or nine, or ten? They probably didn't fudge them all, 7 but if they fudged only one, I got a problem in that none of 8 the ratings are worth anything. So anything that solves that 9 problem has the marginal benefit of really giving the whole 10 framework a much better standing. 11 So now just to -- and by the way, I sort of know 12 this because I was a manager of equity research, and we all 13 know that there were conflicts in that system. I faced those 14 conflicts day after day for years. 15 MR. PAREDES: Before you get to the second 16 question, can I just ask a quick follow up? 17 Is it conceivable, then, taking what you said, that 18 if we have a system with credit ratings that are perceived to 19 be conflicted, that a triple A, as a practical matter in 20 terms of how people trade on it is discounted to, say, a 21 single A? versus, if we had a purer set, non-conflicted or 22 less-conflicted, if you will, set of ratings where a double A 23 actually meant a double A, what you actually have to look at 24 is what people impute to the instrument as compared o what it 25 is rated nominally. 215 1 So even though the nominal rating may be lower, it 2 may in fact be higher than what the actual imputed, if you 3 will, quality of the credit is, so that in a world with pure 4 ratings, even though it is a lower double A as compared to a 5 triple A, it doesn't suffer the additional discount, so it 6 may actually result in a lower cost of capital, not a higher 7 cost of capital. So purity of the rating, accuracy of the 8 rating, even if it is a lower nominal rating, carries a lot 9 of additional weight because there is not the kind of 10 uncertainty that investors have to bring to bear that then 11 causes there to be an additional ratchet down whatever the 12 nominal is. 13 MR. GOOTKIND: I want to get back to -- 14 MS. CLARK: Go ahead. 15 MR. GOOTKIND: Professor Grundfest, you said that 16 there is a natural duopoly here because they have an 17 informational advantage, and I guess I want to challenge that 18 notion that a rating agency has an informational advantage 19 unless it is based on the fact that they have got inside 20 information or non-public information. 21 MR. GRUNDFEST: By informational advantage, I did 22 not mean to say that they naturally have superior 23 information. Rather, an informational advantage in terms of 24 having brand names that are already understood and well 25 established in the market space, and there is a tremendous 216 1 advantage to being an incumbent in one of these informational 2 markets. Forgive me, I should have been more clear in terms 3 of what I meant by informational advantage, then. 4 MR. GALLAGHER: Ms. Clark, did you have a follow 5 up? 6 MS. CLARK: Sure. I just wanted to address the 7 question of how would institutional investors respond and do 8 they have kind of the apparatus to respond. 9 So there are two important precedents for this 10 mechanism that we have described. The first one comes from 11 the fixed income market and the second one comes from the 12 equity market. Historically when new bonds were issued, 13 because many people on the street would be dealing with the 14 institutions, it became useful from all parties' perspectives 15 to let the institutions designate which street firms helped 16 them the most, gave them the highest quality service, through 17 what was called an institutional pot designation mechanism. 18 And so for years -- I mean literally for the last 19 20 years -- the underwriters, the syndicate manager, kept 20 what was called the institutional pot, and institutions would 21 inform the lead manager as to who was to get which 22 allocations, and the whole street was accustomed to that. 23 And originally it was on a piece of paper, and as time went 24 on software developed to make that easier to deal with. That 25 system is not in place today, but that is the history. And 217 1 my point is the mechanism that underwriters are very familiar 2 with it. 3 The second place you will run into this kind of a 4 model is in equity research allocations. And so the major 5 owners of stocks, the institutional owners of stocks, have 6 elaborate mechanisms for designating commissions. Some of 7 them allow a single portfolio manager to operate that 8 designation process. So I look at my portfolio, I decide 9 which agency provided the best research, and I apply 10 accordingly. Some of them concentrate that decision making 11 at an investment committee level or even a higher level. 12 The virtue of that, by the way, is that as it goes 13 up the chain, some of the conflicts that an investor might 14 have tend to go away, because within a large fixed income -- 15 you know, money manager, you have both people who are both 16 buying and selling bonds every day. There is constant 17 transactions. So what they tend to look for who is really 18 providing good, solid work, and they allocate their 19 commissions accordingly. And I think that this same thing 20 would develop. 21 Interestingly, over the break, I was talking with 22 Debbie Cunningham, just explaining to her the process. She 23 said "Hooray, that's great for me," because remember she said 24 "I pay ten different firms"? And she said "My second highest 25 payment is to the rating agencies and the various different 218 1 people who supply the support. If I were actually able to 2 designate that, the rating agency fees that are already being 3 paid by the issuer, it would allow me to spend more money on 4 either my own credit infrastructure or it would just lower 5 the cost of me as a money manager, allowing individual 6 investors to wind up with a lower cost." 7 MR. GALLAGHER: Thank you Ms. Clark. Commissioner 8 Aguilar. 9 MR. AGUILAR: Thank you. I really want to direct 10 my question to Mr. Holmquist, if I may, to try to get an 11 international perspective on the discussions that we are 12 having. As to whether any of these tentative models have 13 been discussed or tried in any of the jurisdictions that you 14 are familiar with, and if we were to go to one in any way, 15 are there any that are more viable for global coordination 16 and a global coordinated regulatory approach? 17 MR. HOLMQUIST: Well, the dominant rating agencies 18 in Europe are the American rating agencies, of course. I 19 find this discussion very interesting. I can't say I 20 understand 100 percent, but it is still very interesting. 21 We have had a bit of a discussion in the 22 Commission, the European Commission, on how do you erase the 23 fees for the ratings, how do you pay for them, and who should 24 pay for them? We haven't come very far on that. We decided 25 not really to touch upon it in our legislative proposal. We 219 1 said that we would come back to it when we evaluate what we 2 have done in three years time or so. So this debate that you 3 are having here, I would say we are behind in Europe on that, 4 but I'm sure that we will come to it. 5 I don't think that there is -- but my knowledge 6 about the rating market is limited -- I don't think there is 7 much that has happened in Europe. There may be other people 8 in the panel that actually know more about that. 9 On the global coordination and global regulatory 10 coordination, we are in favor of this. Obviously the ratings 11 are used worldwide. The firms that are active now are active 12 worldwide, the big ones are at least. And we see a strong 13 interest in discussing it. We would like to develop the 14 IOSCO code. Some people told us "Why are you doing this? 15 Why are you proposing legislation in Europe? Why aren't you 16 just using the IOSCO code?" We felt that it was too general, 17 too open, not enforceable, that it was not sufficient, and 18 that is why we went to legislation. But this does not mean 19 that we don't want international cooperation. 20 We would be very happy if we could develop the 21 IOSCO code further, and we certainly have no interest in 22 itself to be different. We feel that maybe we need to be a 23 bit tougher on certain things, for example, the internal 24 governance in the rating agencies. That has been the case in 25 other jurisdictions for now, but we don't want to separate 220 1 ourselves out. It was rather that we felt we needed to go a 2 bit further. If we could reach international agreements, we 3 would be happy about that. That would be rational. 4 MR. GALLAGHER: Chairman Schapiro. 5 CHAIRMAN SCHAPIRO: Thank you. I actually wanted 6 to ask exactly what Commissioner Aguilar asked. I didn't 7 want to lose the opportunity with Jorgen Holmquist here not 8 to find out a little bit more about what is going on in the 9 debate in Europe. 10 I guess I would ask one specific question, and that 11 is whether, as you have heard throughout the day today, we 12 have talked a lot about the fact that ratings, and 13 particularly ratings by NRSROs are specifically referenced in 14 the securities laws, a number of our regulations, and as Joe 15 and others have pointed out, in private contracts and state 16 law, and investments guidelines, and everywhere. And I 17 wondered if that was also the case in Europe, particularly 18 with respect to laws and regulations, whether a requirement 19 for ratings or ratings by particularly recognized entities 20 was a requirement. 21 MR. HOLMQUIST: Yes, we have that too. When we 22 started to look at the legislation proposal, we looked into 23 it a bit, and we had the impression originally that was there 24 was very widespread -- I don't know now how widespread it is 25 in the U.S. -- there was very widespread in Europe. It is 221 1 mostly, from my understanding, tied into the capital 2 requirements directly through the Basel II. And we were 3 reflecting on it, and we said that we will look further into 4 the possibility of limiting this, of reducing this. 5 At the same time, I think when we thought about it 6 and we discussed with various stakeholders and various people 7 using the ratings and thinking what the alternative would be, 8 it didn't seem to us that there was any obvious alternative. 9 We had to put forward our legislative proposal quite fast. 10 We couldn't haggle all the different issues, the pay system, 11 for example, we did not. This we didn't really touch upon 12 either, but we said that we would come back to it. But to us 13 there was not -- at least working very fast, there were no 14 immediate, obvious solution, but it is something that we have 15 decided to come back on. 16 CHAIRMAN SCHAPIRO: I would say it is clear that we 17 don't have an immediate, obvious solution yet either. Thank 18 you. 19 MR. GALLAGHER: I guess one follow up on that is, 20 in the last three years -- we have been talking all day about 21 since the legislation, the actions, the SEC has taken -- and 22 in the view of the staff, of course, we believe it has been 23 sort of activist, the implementing the statute and then the 24 rulemakings that occurred thereafter. And I think we would 25 say that at that point the U.S. was at the forefront of the 222 1 credit rating agency regulation, in fact it probably still 2 is. 3 Just a question for you, Mr. Holmquist, going 4 forward on this global coordination, how do you see it 5 working day to day? I mean we as the SEC obviously have been 6 tapped with the oversight by Congress of the credit rating 7 agencies. How do you see the best way for us to work with 8 you and others as we kind of craft something that can be 9 implemented globally? 10 MR. HOLMQUIST: I haven't reflected thoroughly on 11 it, not at all. I would assume that we should continue the 12 discussions in the IOSCO, where the European Commission 13 hasn't been very active, but where we expect -- hope to be 14 more active in the future, although the way we participate is 15 a bit complicated. I would also say that it is a good thing. 16 The reason I am here in Washington today or these 17 few days is to talk about the so-called financial markets 18 regulatory dialogue between the U.S. authorities and the 19 E.U., and it is certainly something that we can discuss there 20 too. So far, when it comes to the G20 work, we can see that 21 on financial market regulation supervision. The U.S. and 22 E.U. see things quite in a similar way, at least when we look 23 at the philosophy and the general issues. I think it should 24 be discussed both in IOSCO and in our bilateral dialogue. 25 While I have the opportunity, maybe I should 223 1 mention one more thing. In our original proposal, there was 2 some conditions established that were felt in the U.S. and 3 elsewhere could be protectionist or extraterritorial. We 4 didn't really feel so. It was a question of would third 5 country rating agencies be obliged to establish themselves in 6 Europe and what would we do about ratings that had been 7 established by rating agency entities outside of Europe, 8 could they be used in Europe? 9 This in the legislative process we are processing 10 and we hope that we will find some possibility of equivalence 11 or use of ratings also from other jurisdictions. And I think 12 we will come to a -- this was a concern in the U.S. half a 13 year ago when we put forward our proposal. I hope that we 14 can find a solution to that that will be satisfactory to 15 third countries, because it was not only the U.S. Japan had 16 similar concerns, for example. 17 MR. GALLAGHER: That's great. Professor Grundfest. 18 MR. GRUNDFEST: If I might observe, the IOCRA 19 approach, to my knowledge, would be entirely compatible with 20 all international regimes because it would simply allow the 21 international regimes to continue with their reliance on the 22 NRSROs and it would simply change the environment in which 23 the NRSROs generate their ratings and create more 24 information. So you wouldn't have to worry about changing 25 anything under the Basel accords or changing anything under 224 1 EEC law. You would be able to continue, hopefully in an 2 environment that would be informationally richer and more 3 accurate. 4 MR. GALLAGHER: That's terrific. Chairman 5 Schapiro, did you -- Commissioner Casey. 6 MS. CASEY: I had a question. Chris, I wanted to 7 follow up on a point you made about systemic risk, and I 8 guess the question would be in the absence of making changes 9 to the payment model or removing references or changing 10 disclosure transparency of the information that is currently 11 provided in the market, which of those is crucial to 12 addressing the systemic risk concern, that you have raised 13 that you believe the current approach to rating agency 14 regulation has strengthened and reinforced? 15 MR. GOOTKIND: Well I think we agree that there is 16 a duopoly and maybe Fitch is a third agency that has got some 17 meaningful market presence, at least in certain sectors. But 18 it means that because of the rules and regulations and 19 investor guidelines and bond indices, all of the things that 20 we have talked about, everybody is using and relying upon 21 those two or three rating agencies. That concentrates the 22 risk in the capital markets and the financial system with 23 those ratings. 24 Now it doesn't mean that firms like mine don't do 25 their own independent analysis and we are not forced to buy 225 1 bonds if we disagree with the credit assessment, but -- and I 2 think about counterparty transactions, capital requirements, 3 Basel II, risk-based capital for insurance companies. It 4 means that everybody is using those ratings, and the 5 government having this NRSRO designation as that imprimatur 6 is that these are the right ones that these firms do the 7 right analysis and they are the ones that we should all be 8 using. 9 And I was trying to come up with a good analogy in 10 the equity side and I didn't really have one, but I said 11 Warren Buffett and Bill Miller are known as two very good 12 investors, but we don't all go with what they do. And Bill 13 Miller had a great record, he has struggled in the last few 14 years, and I'm trying to impugn him, he is still considered a 15 great investor, but we don't all go with what those people 16 do, and so the marketplace overall I think can assess the 17 risk and you can look at it, as one of the Commissioners said 18 earlier, in terms of -- whether it is credit default swap 19 pricing or bond pricing -- I mentioned bonds in my portfolio 20 trading in the 30s and 40s, and they still have an investment 21 grade rating -- the market has a very different view. 22 And yet I still need to think about that in terms 23 of what my client guidelines are and how -- the other problem 24 we are having right now I think is the whole mark to market 25 issue. I don't want to get too involved in that one, but 226 1 there are bonds in these banks and in financial institutions' 2 portfolios that have investment grade ratings and they are 3 trading at 50 cents or less on the dollar. And so you think 4 about mark to market, well, it is very reasonable they would 5 say "Wait a minute, the rating agencies say these bonds are 6 still investment grade, why should I mark them down to 50 7 cents? If I can hold them to maturity and they are going to 8 mature at par, why should I have to impair them now?" 9 So there is a big disconnect, and I think that is 10 where this NRSRO designation forces us to all use -- rely on 11 those same ratings, and I think that is causing this systemic 12 risk that we are all living through right now. I'm sorry -- 13 I shouldn't say it's causing, it is strongly contributing to 14 it. 15 MR. GALLAGHER: So I guess as a follow up to each 16 of the panelists, just a question out of curiosity, do you 17 believe that credit rating agencies are systemically 18 important entities as we enter into this discussion about 19 systemic regulators and obviously the systemic importance of 20 some of the firms that failed last year? Ms. Clark? 21 MS. CLARK: Absolutely. And as was pointed out to 22 us today by Larry White, it started in 1935, right? That was 23 the first moment when we started talking about bank capital 24 adequacy being linked to rating agencies. It never quite -- 25 I mean it took a long time for it to become truly systemic, 227 1 but I just don't see how you avoid the conclusion with 2 everything that implies for the regulatory infrastructure, 3 including your role. 4 MR. GALLAGHER: Mr. Reynolds? 5 MR. REYNOLDS: I don't think there is any question 6 about it at this point, but it goes into beyond perception. 7 It gets into just the reality of certain documents as you 8 have seen with AIG and others, cash collateral calls, the 9 ability to access the short term funding markets. And in not 10 quite as stunning a fashion, as you move down the credit 11 spectrum, your loans re-price and so on. So certain 12 parameters are basically inviolable and you can't fight that, 13 and you find ratcheting price risk as you move south on the 14 rating spectrum. 15 And I would say the rating agencies themselves have 16 done their utmost to magnify the impact of the great 17 continental divide between investment grade and 18 non-investment grade. That only increases the power of the 19 brand. In fact there is just one point in the risk/reward 20 continuum, but that regulatory, arbitrary point has always 21 been a focal point for all investors. 22 MR. GRUNDFEST: If you consider the question of 23 systemic risk and you apply a narrow definition to the term 24 and you imply it -- you define it narrowly as meaning that a 25 change in state A for company A has consequences for company 228 1 B, C, D, E, and F in such a way that it has macro effects, 2 then it is almost obvious that the quality of rating agency 3 work has systemic risk in it. You downgraded AIG, you have 4 got all of the consequences we have been talking about. You 5 downgrade a Citigroup, you downgrade Goldman Sachs, it has 6 all of the consequences that we have been talking about. So 7 even though at first blush many people don't look at rating 8 agencies as generators of systemic risk, you stop and think 9 about it for a microsecond, you realize yeah, they are there. 10 MR. GALLAGHER: Congressman Baker? 11 MR. BAKER: The debate on what constitutes systemic 12 risk is gone for many years, but generally it is thought of 13 in the generic terms that if the failure of the enterprise 14 would lead to harm to third parties otherwise not engaged in 15 the risk taking, that that is an ill-advised failure. 16 Therefore whether it is financial, commercial, or otherwise, 17 I suspect in the current environment that Congress will be 18 much engaged in trying to decide what regulatory oversight 19 should be proper. In this case it is not a concentration of 20 financial assets, it is a concentration of financial 21 information. 22 But the contagion is the same, and where they get 23 it wrong, the consequences are just as similar if it were a 24 true economic failure, and the consequences to the broader 25 market are very significant. When does a systemic risk event 229 1 occur, who should decide that it is a systemic risk, and what 2 should be done after it is identified? I don't know, but 3 that is something I'm strongly for when we figure it out. 4 (Laughter.) 5 MR. GALLAGHER: Mr. Holmquist, any thoughts on this 6 as obviously systemic risk has risen to the level of a term 7 of art here in the U.S., but clearly you share it in Europe. 8 MR. HOLMQUIST: No, we are discussing it too. We 9 had this report by the group headed by Jacques de Larosiere, 10 the former head of the IMF, who is discussing this, so we 11 have a discussion of systemic risk and systemic risk council 12 or systemic risk regulator that is quite similar to the 13 discussion in the U.S. We haven't defined the rating 14 agencies as organizations that are important for systemic 15 risk, but the fact that we decided to go from a situation of 16 having them not regulated to having them regulated, I guess, 17 is an implication is that is the way we saw it. 18 MS. CLARK: Could I just add one thought to that? 19 MR. GALLAGHER: Yes. 20 MS. CLARK: I believe it is true that 40 percent of 21 the credit markets, including loans and securitization, is 22 securities. In other words, if you looked at the total asset 23 value of credit-related instruments including the loans and 24 securities, the securities market is a huge portion, and I 25 think that lends -- it adds to this notion that -- you know, 230 1 we don't have any financial institution that is 40 percent, 2 but these guys affect 40 percent. 3 MR. GALLAGHER: Unless there are any other 4 questions that the Commissioners might have or that the 5 panelists might have of the Commissioners, I think we have 6 reached the end of this panel, and indeed the roundtable. So 7 I want to thank the panelists here, the panelists for the 8 three panels also, for your participation, and I'm going to 9 turn it over now to Chairman Schapiro. 10 CONCLUDING REMARKS 11 CHAIRMAN SCHAPIRO: Thanks Dan. Let me also just 12 add the thanks of the Commission to all of you for your 13 participation today and for all of our prior panels. 14 Your written submissions and most particularly your 15 participation directly with us today will greatly inform our 16 deliberations going forward. We have a lot to think about 17 and a lot to contemplate in terms of what actions we might 18 take going forward. 19 I also just want to talk a moment to thank the 20 staff of the Division of Trading and Markets, most especially 21 Dan Gallagher, who so ably led us through the day, but also 22 Marlon Paz, Rebekah Goshorn, Mike Macchiaroli, Tom McGown, 23 Randall Roy, Joe Levinson, and Sandra Seitman. You all made 24 this possible and I think you for that. And I also want to 25 thank Kayla Gillan of my staff for her leadership in helping 231 1 us to develop the roundtable. I thank my colleagues, and our 2 meeting is adjourned. 3 (Whereupon, at 4:05 p.m., the roundtable was 4 concluded.) 5 * * * * * 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25