To be published as HC 1866-ii

House of commons



Treasury Committee

Inquiry into credit rating agencies

Wednesday 7 March 2012

Paul Taylor, Frederic Drevon, Dominic Crawley and Alan Reid

Evidence heard in Public Questions 139 - 269



This is an uncorrected transcript of evidence taken in public and reported to the House. The transcript has been placed on the internet on the authority of the Committee, and copies have been made available by the Vote Office for the use of Members and others.


Any public use of, or reference to, the contents should make clear that neither witnesses nor Members have had the opportunity to correct the record. The transcript is not yet an approved formal record of these proceedings.


Members who receive this for the purpose of correcting questions addressed by them to witnesses are asked to send corrections to the Committee Assistant.


Prospective witnesses may receive this in preparation for any written or oral evidence they may in due course give to the Committee.

Oral Evidence

Taken before the Treasury Committee

on Wednesday 7 March 2012

Members present:

Mr Andrew Tyrie (Chair)

Michael Fallon

Mark Garnier

Stewart Hosie

Andrea Leadsom

Mr Andrew Love

Mr Pat McFadden

John Mann

Jesse Norman

Teresa Pearce

Mr David Ruffley

John Thurso


Examination of Witnesses

Witnesses: Paul Taylor, Group Managing Director, Fitch Ratings, Frederic Drevon, Managing Director, Head of EMEA, Moody’s Investors Service, Dominic Crawley, Managing Director and Head of Financial Services Ratings, Standard and Poor’s, and Alan Reid, Managing Director Europe, DBRS, gave evidence.

Q139 Chair: Thank you very much for coming to give evidence to us this afternoon in this inquiry into risk rating agencies. We asked the banks, after this crisis, if they had realised the enormity of the blunders that they had made and the price that millions of people have had to pay as a consequence. Have the risk rating agencies apologised? For example, Mr Drevon, you are from Moody’s. Has Moody’s apologised?

Frederic Drevon: We have indicated publicly that we are not satisfied with the performance of our ratings that were associated with the US mortgage housing market.

Chair: That goes without saying, doesn’t it? But I am asking you just a very simple question, whether you have apologised to the millions of people who lost out as a consequence of this.

Frederic Drevon: Again, I think we clearly indicated we are not satisfied with the performance of our ratings.

Chair: Is that an apology or not?

Frederic Drevon: It is exactly what I have indicated; we are not satisfied with the performance of our ratings.

Chair: So we should take that as "no" then?

Frederic Drevon: We should take it as I have indicated: we are not satisfied with the performance of our ratings related to the housing market. I think it has to be put back into context where our ratings address credit risk. They are an instrument in the context-

Q140 Chair: Mr Crawley, what is S&P’s view of this?

Dominic Crawley: Mr Chairman, talking specifically-and I will talk more generally but specifically-about US subprime, we have repeatedly, over a number of years, expressed our regret about the assumptions that were built into our subprime models. We, like many others in the market at that time, did not foresee the severity of the downturn that we saw both in the economy in the US and as regards asset values. More generally, we take what we do in the markets extremely seriously.

Q141 Chair: But I just want to be clear on this point. I did not think it would take this long to get through the preamble, but you are saying sorry to the people who got hit? You regret it and you are saying sorry?

Dominic Crawley: We have already gone on record saying that we regret that the assumptions that we used did not hold true, based on what actually happened in the market.

Q142 Chair: Well, that is a step forward. What about Fitch, Mr Taylor?

Paul Taylor: It is a very leading question, of course, and I know why you ask it. On structured finance transactions, where we saw many, many problems, we have apologised. We have not apologised for much of the rest of our work because it has not gone wrong. So, why would you apologise? For structured finance and the aspects that went wrong, absolutely, we have apologised but that does not cover all our activities.

Q143 Chair: But the question related to the enormity of the blunders that had been made in risk rating, which have contributed to this crisis. It did not strike me that I have covered all your activities but only a small proportion of them, perhaps a large proportion of them. In fact we might ask that as the next question. What proportion of your income, Mr Drevon, in 2005-I have a book here that purports to tell me the answer-came from the structured financial sector?

Frederic Drevon: I do not have the specifics in front of me, I am afraid, but it was a significant part of our activity.

Q144 Chair: Well, I have read that it was almost 50%, nearly half your total income. Do you know what your total income was, say, 10 years earlier?

Frederic Drevon: No, I am afraid I do not have the figures in front of me.

Q145 Chair: What proportion of that came from the structured financial sector, which was in its infancy in those days?

Frederic Drevon: I assume it was a lower number. The markets in structured finance had not really been developed outside the US, so it was certainly a smaller percentage.

Q146 Chair: People close to this market have told me that this was a revenue source that was "just too good to be true". Do you recognise that?

Frederic Drevon: We are in the business of providing credit opinions to the marketplace where there is a need for providing those opinions. That is true for different types of issuers. It is true for corporate issuers, for banking issuers; it is true for sovereign issuers. It was also true in the case of structured finance where there was an interest on the part of investors to obtain independent credit opinions by ratings agencies. So, to the extent that there was an interest by the marketplace for our views, we provided those views to the marketplace.

Q147 Chair: In your written evidence, Mr Taylor, you told us that the European proposals, CRA 3, "could have a severe impact on the real economy", by reducing the ability of businesses to issue debt as an alternative to borrowing from banks to raise finance. What is the scale of the problem for UK institutions from CRA 3?

Paul Taylor: I am not sure we can size it exactly in the way you are suggesting. We think it would lead to less efficiency in issuing fixed income bonds and, by implication, that means that there will be a cost to the market as a whole. So we think the markets will just work less efficiently than before the proposals would have been implemented.

Q148 Jesse Norman: Mr Crawley, you said it was a matter of regret that you had got your structured finance assumptions wrong. It is a source of regret if there is a crisis in the economy or subprime, but it is not something that anyone takes any personal responsibility for when they say it is a regret. Does that mean you don’t think that your organisation was responsible for those and, therefore, you don’t want to apologise?

Dominic Crawley: The last four years have surfaced a lot of new information for everybody in the market, including the rating agencies. We are very ready to acknowledge that and clearly, from the information that you gather in the market, you revise and adjust criteria to reflect changing market circumstances. What I can say to you very clearly is that at Standard & Poor’s there is a very high level of individual ownership and responsibility for the roles that people carry out in their jobs and there is a very strong collective corporate responsibility about what we do. We take what we do extremely seriously. We are very ready to learn and build our knowledge based on what we learn from the marketplace and revise our criteria accordingly. The market is not a perfect place and we all know that the last four years has presented to everybody-not only the rating agencies but investors, regulators, central banks-some major challenges in the way that the interdependency of markets has developed.

Q149 Jesse Norman: It is good to hear that people are held personally responsible for these decisions. How many people have been fired as a result of this for cause-because they were incompetent?

Dominic Crawley: That is not what I said. There is a personal sense of responsibility. We rate according to our criteria, beginning, middle and end, and that is the way that we go about assigning ratings. Ratings are not assigned by individuals. Ratings are assigned by committees according to criteria that we have spent a lot of time working on with our regulators and our supervisor, ESMA, to make sure they are transparent, public and clear.

Q150 Jesse Norman: It sounds like personal responsibility is really committee responsibility. You are not apologising and no one has been fired for cause, for making this egregious set of errors.

Dominic Crawley: I have to say to the Committee you have to go back to the base of what ratings are. Ratings are about opinions. They are about forward-looking opinions about future relative credit risk. These are opinions.

Q151 Jesse Norman: Okay. Mr Drevon, in 2007 Moody’s downgraded to junk status 89% of its investment-grade bonds. Was that because they had all got much worse than you thought or because there might have been some mistake in evaluating their original rating?

Frederic Drevon: I assume your reference is to structured finance instruments?

Jesse Norman: Yes

Frederic Drevon: I am sorry, I do not have the relevant information in front of me. A very high percentage of investment grade ratings in relation to mortgage and housing in the US have been downgraded, in particular those transactions that originated in 2005, 2006 and 2007. I think we realised that there was a high degree of correlation in the performance of these different types of transactions because they were exposed to the same underlying risk. So at the point in time where we had underestimated the magnitude of the potential change in the US housing market, it affected a very high number of transactions in that segment and, therefore, the high number-

Q152 Jesse Norman: But it has been an open secret for many years that there was an enormous profit stream to the US banks from taking goodness knows what quality assets, bundling them together into structured products and then asking you guys to bless them, and you played along because the money is so good. What I want to know is, given that that happened, as we all know, and given that you should have known about it at the time if you had been doing your credit analysis accurately, why have you still not yet apologised?

Frederic Drevon: We did the work we had to do based on a very professional approach with high ethical standards at Moody’s. We developed the methodologies. We included in our methodologies an expectation that these were loans of average to very poor quality, and transactions were in fact structured in a way to address some of the risks that we and the rest of the market had anticipated but we underestimated the magnitude.

Q153 Jesse Norman: So you don’t share my view that this was a racket?

Frederic Drevon: Certainly not.

Q154 Chair: Mr Crawley, you said a moment ago these are opinions. Does that mean that people should not take too much notice of them?

Dominic Crawley: What I would say very clearly is that we recognise that these opinions are important in the market, but we always say that our opinions are but one piece of information that informs credit risk decision-takers about whether they decide to invest, hold or sell. Our opinions are not involved in any kind of advice or anything like that around investing, buying or selling. We are providing credit opinions about relative credit risk, the probability of default, and there are many other pieces of information that any decision-maker, when they are taking on a credit risk position, will take into account before coming to that decision. So it is a piece of information but it is one tool that decision-makers will use among many others in the market.

Q155 Chair: The important thing is not to take too much notice but to look at these other pieces of information in the market?

Dominic Crawley: It is down to individual credit risk-takers to decide how much use to make of what we provide. We strive very hard to ensure that our ratings are of good quality. We provide additional research, opinion and analysis to support it.

Q156 Chair: So it was a mistake of the investors in taking too much notice that led to this crisis?

Dominic Crawley: I said it is entirely up to the investors to choose whether they make use of our ratings or not.

Q157 Chair: I am just trying to be clear whether you think you got it wrong or whether they got it wrong and what the proportion of responsibility is between the two of you.

Dominic Crawley: I understand that, Mr Chairman, and I know there is, quite understandably, a lot of attention specifically going back to US subprime, and we have spoken about that already. In Europe we rate over 2,000 entities and we have about 100,000 individual ratings of securities. The track record of our ratings when compared to actual defaults over a long time series is extremely good.

Q158 Chair: But you got it totally wrong in this case, didn’t you, in the SIVs and the CDO case, in the structured products case?

Dominic Crawley: In US subprime, as I have explained, there were assumptions that did not hold up based on actual evidence. They were-

Q159 Chair: So people should not believe in the models?

Dominic Crawley: As I said, the assumptions that we used, which at the time were broadly regarded as conservative, did not hold up based on the severity of the downturn that we saw. As I said, we all learn as more evidence is built up.

Q160 Chair: I am just trying to ascertain whether you are saying we should take less notice, the same amount of notice or more notice of the credibility or the robustness of your methodology than we did prior to this crisis.

Dominic Crawley: I am not trying to be difficult here. I am trying to be as clear as I can. The methodologies across the broad range of entities that we rate have been built up over a long period of time. There is very good evidence of the performance of those ratings. We regularly review those ratings based on changing market conditions. The focus has been on looking specifically at subprime, and we have talked about the assumptions and the difficulties with the assumptions.

Q161 Stewart Hosie: Mr Taylor, Mr Crawley said that they score on the basis of criteria that are transparent, public and clear. Fitch has already told the Committee in writing that your rating process is transparent. You have clearly documented criteria that describe a framework that is freely available on your public website. So you obviously agree that your process-your techniques-are transparent. Were they equally transparent before the crisis when you were, I think, Global Head of Structured Finance Ratings?

Paul Taylor: I was Head of Corporate Ratings before the crisis. I became Head of Structured as the crisis hit, for my sins. The answer is yes, they were-the criteria were clearly laid out. The problem, of course, is that the assumptions in the criteria were not robust enough to deal with the situation that unfolded. So the nature of how we did the analysis was clearly laid out. The assumptions were clearly laid out. The assumptions themselves were wrong.

Q162 Stewart Hosie: What changes have you made in terms of assumptions and methodologies since the crisis?

Paul Taylor: Particularly to structured finance?

Stewart Hosie: Particularly, yes.

Paul Taylor: Without trying to become too technical, structured finance covers dozens of different asset classes in different countries. Each one would have a criterion. There are literally dozens of criteria pieces explaining how we do things, what we look at and what sort of weight we attach to those different aspects. With any kind of situation that we have been through over the last four years or more-and when we talk about the subprime crisis it is assets that were being written in 2005, 2006, so it is even further back than we are suggesting here-you learn from your experience, of course. If you look at our US mortgage criteria, as the best example, now compared with what they would have been like in 2003, 2004, there have been fundamental changes to things like correlation of assets, the assumption about how far house prices can decline, how many people may default and a range of issues like that, which have been adjusted and adapted based on experience.

Q163 Stewart Hosie: Significant changes to the criteria on a whole range of asset class in the structured finance product range?

Paul Taylor: Yes.

Q164 Stewart Hosie: Can you quantify that in terms of the size of the change of assumptions in criteria you made? Is it 10%, 20%, 50%, 70% of the assumptions in criteria? What scale of changes did you make to the SFP range of products?

Paul Taylor: You can’t generalise because it varies between asset classes. Maybe a good way of thinking about this is when you think about the US subprime assets, which is the heart of the original problem. In the US housing mortgage-backed market you had three tiers of types of asset. You had prime mortgage, the best quality mortgages; you had subprime, which were known to be to lesser quality creditors; and you had a piece in the middle called AltA. Before the crisis, the amount of credit enhancement attached to these different transactions varied already. You may have had 5% for a prime transaction and you would have had 20% for a subprime transaction. Back in 2004, when we first saw deterioration in the US mortgage market, we adjusted criteria for subprime. We would have moved it from something like 20% to maybe 23% or 24%-very general numbers. Of course, what we should have done, with hindsight, is move them to 50%. So the market was worse than we had expected.

Q165 Stewart Hosie: That is the question. Why didn’t you do that? In fact, why did none of the agencies do this properly? Did you lack the skill set to properly assess these structured financial products? Were you in denial, thinking that it can’t possibly be as bad as that? Was there group-think going on?

Paul Taylor: There was certainly market group-think going on. Obviously there were many commentators on the market. It was not just credit rating agencies. I think no one expected the magnitude of the hit in that market. It was unprecedented. It was as much about behavioural change in the US as anything. People handed back keys if they went into negative equity. We did not allow for that on the scale that it happened.

Q166 Stewart Hosie: We asked the bankers to convince us that they understand what they are doing now. How can we be confident you now have the people and the skills and are prepared to make the right calls if you assess a risk like this in the future? People are depending on your scores before they take investment decisions for their futures, for their pensions, for everything. How can we trust you?

Paul Taylor: It is too simplistic to say they are depending on our decisions to make their investments. We are one input into that process. We do the job to the best of our ability. We have a particular job to do and we try to do that job. If the concern is about conflicts of interest and that we wouldn’t say no because it would lose us business, maybe what is going on in Europe at the moment is the best defence I can give of that without going into the technicalities of how we manage the conflicts that we have. I am not sure-

Q167 Stewart Hosie: I am less concerned about conflict of interest. Have you brought on more people? Have you brought on more analysts? Have you brought on better skilled people? What have you done so that you don’t misjudge this sort of thing in the future?

Paul Taylor: There is no guarantee that we would not miss something again in the future. We don’t have a perfect crystal ball. We think we look at the things we need to look at and we try much harder to look for shifts in the curve. The thing we missed wasn’t just a movement in the curve, it was a complete shift. We have put a structure in place internally. We have a team of around 40 people, senior people, who are away from the day-to-day task of assigning ratings, to look for these bigger trends that are going on, these bigger shifts in the curve that could happen. So there are structural changes that have been put in place in addition to just criteria changes of tightening up some of the standards on criteria.

Q168 Andrea Leadsom: Gentlemen, do any of your organisations charge sovereigns for ratings? A yes or a no.

Frederic Drevon: Yes.

Dominic Crawley: I believe we do, but I don’t know because I am on the analytical side of the business so I don’t have any sight of that.

Paul Taylor: Yes.

Alan Reid: I believe that we do, but I am on the analytical side of the business and that disclosure is not made available to me.

Q169 Andrea Leadsom: Okay, thank you. So the chances are you probably all charge sovereigns. Is there a concern that in a regulatory sense there is an over-reliance on ratings? Clearly, in terms of Basel III liquidity requirements, in terms of the liquid assets that banks have to hold, there is a requirement for rated assets and, in fact, the good news for you as commercial companies is that Basel III requires two ratings, not just one, to give assurance of the credit standing of a sovereign. Does that concern you, bearing in mind that you have been found out before and you have all said to the Chairman and to Mr Norman that you feel it is important that investors do not just rely on ratings?

Dominic Crawley: We have been very clear for quite some time now that we would like to see this mechanistic inclusion of ratings within regulation to be removed. I know back in October 2010 the Financial Stability Board published this principle of reducing authorities’ and financial institutions’ reliance on ratings, and that subsequently was endorsed by the G-20 at Seoul in the following month of that year. That is now being followed through on a globally consistent basis by the EBA, by ESMA and also by IOPA where they have been instructed to review and, where appropriate, remove ratings from their regulations and guidelines. That is something that we have very publicly said that we support. Ultimately, in terms of our reputation and position, we want to be judged not by some mechanistic requirement but by the market in terms of the quality of what we are providing.

Q170 Andrea Leadsom: Mr Reid, as you haven’t said much yet, would you comment on how you think that we can reduce regulatory requirement for credit ratings?

Alan Reid: Credit ratings are in the fabric of the global capital markets. There is clearly a concentration risk there, given the amount of agencies that are active in the market. What are the credible alternatives to credit rating agencies? What can we use instead of them? We share the same view as our colleagues from S&P about taking ratings out of regulation, but they are part of that fabric.

Q171 Andrea Leadsom: How can you do that, then? Is there an answer? What can investors use, if not credit ratings, to assess the creditworthiness of bank liquidity assets and so on?

Alan Reid: I think your starting point is the amount of rating agencies that you have. You have a concentration of rating agencies at the moment and if there could be some diversity of opinion, there would be less concentration and perhaps less risk.

Q172 Andrea Leadsom: DBRS is the smallest, I am guessing, of the ratings agencies here today?

Alan Reid: DBRS is the smallest but we have been in business since 1976. We are a global credit rating agency. We have been in the United States since 2003 and we recently entered the European market. The management team is highly experienced. We have all come out of Moody’s, S&P, Fitch, as well as the global investment banks and commercial banks.

Q173 Andrea Leadsom: Is your answer to this then just to have more credit rating agencies so that DBRS gets a piece of the action as well; maybe that you have three or four ratings for every piece of bank liquidity asset so that you all get paid for rating an organisation? I can quite see there is a strong argument for that when looking at your own business model.

Alan Reid: You make some very fair comments. It was not us who asked for rotation or us who suggested it. It was something that was proposed through Europe and has been supported by the Treasury here in the United Kingdom. We would like to be part of the solution. We have invested a significant amount of capital here in Europe. We have gone through a registration process with ESMA, the same registration process that my colleagues here have gone through. It took 13 months and we submitted over 2,000 pages of information. Other agencies have gone through a similar process and, having gone through that process, if you are looking for competition, you have to extend the amount of agencies that are being used.

Q174 Andrea Leadsom: Thank you. Mr Drevon, the last question is would you like to see investment mandates remove the specific explicit requirement for rated assets?

Frederic Drevon: We support in general both investors and regulators using ratings as an input into their decisions, but we are concerned by any mechanistic reliance on ratings because it removes the responsibility of the end-user and it does not promote ratings quality. We believe that it is particularly relevant when we think of the regulatory environment to change some of the rules proposed but we think it should be up to each investor to make their own decisions on how they want to use ratings, whether it is through a ratings mandate or any other process.

Q175 Andrea Leadsom: You don’t think that people should rely on ratings under any circumstances?

Frederic Drevon: No, I don’t believe that is what I have said. I think we believe strongly that we provide high quality and useful ratings to the market. They should be an input into any decision made by an investor among a variety of different sources. We believe that it adds value to have fundamental analysis and research about credit risk. It should be one input among a variety of different sources of information, one where there is track record and transparency, but we certainly realise that that decision should be made by each individual investor.

Q176 Andrea Leadsom: Just to be very explicit, as you well know, there are many investors out there who say, "I only buy AAA or AA" and then, once they have found a AAA- or a AA-rated credit, they do not look beyond that. They do not even know what the name is of the issuer. Are you saying that that is a very bad policy on the part of investors?

Frederic Drevon: It certainly is not a policy that I would see as the appropriate one. I think that every investor can rely on our ratings if they believe that they are useful-

Q177 Andrea Leadsom: So that is a yes, it is a bad policy?

Frederic Drevon: -but they should also look at research we provide, absolutely.

Q178 John Mann: If I could just interpret what you are saying, for example, to local authorities investing funds in this country. Is it is not sufficient solely to rely on your ratings?

Frederic Drevon: That is correct, yes.

Q179 John Mann: That is very clear. Who are you accountable to?

Frederic Drevon: We are accountable at different levels. We are first accountable to our board of directors where we have independent board members, certainly in the UK. We are accountable to regulatory authorities such as ESMA, which has the authority to supervise our work, or the SEC that supervises our work on a global basis. We are certainly accountable to investors. As we publish the performance of our ratings, we are very transparent about how different categories of ratings perform over time, and it is open to debate and criticism. We are also subject to liability in various European jurisdictions.

Q180 John Mann: If we look at the Icelandic bank situation, of course you have all been keen to try to pigeonhole the problem as being the American subprime housing market, but Landsbanki had no exposure whatsoever to that. In terms of your failure to catch up with the real world there, is it not true that you were responding at the very last minute to what was going on and that, therefore, investors who were not following things, certainly month by month, were going to get their fingers very badly burned?

Paul Taylor: We can talk about Icelandic banks and you would find that the ratings work was quite good if you read the research, quite bad if you followed the ratings. There was a bit of a contradiction.

Q181 John Mann: I have just looked at an article in the financial press from 23 February 2008 that points out and states the high ratings that you were all giving in February 2008, using your ratings as the basis, reassuring people that even though there had been Bear Stearns and there had been Northern Rock the previous autumn, people need not worry because they could compare those ratings with institutions such as Standard Life, specifically, and feel reassured, using your ratings.

Paul Taylor: Yes, so a newspaper article-I think there is a fundamental lack of understanding of ratings in the comment you have just made, which I would like to try to explain if I can, without showing any disrespect. We and other agencies rate thousands of companies around the world and if you look at the nature of the ratings we expect a certain number of defaults to happen over the course of one year, five years, 10 years. It is impossible to predict which rating is going to go bad and when. There will always be some that go wrong that we do not spot in time. You can use Parmalat, Enron, there are lots of examples-MFC Global, maybe.

Q182 John Mann: What percentage of the ratings would you deem would or could go wrong?

Paul Taylor: We publish extensive default statistics. We publish transition statistics as well-that is ratings moving between categories. Fitch’s default statistics would tell you that between 1990 and 2010 for non-structured transactions-we can talk about structured, this is non-structured-for BBB ratings that started the period at BBB, just over 5% would have defaulted by the end of the period. For BBB it was about 5.2%, for A it was 2% and then it becomes something like 0.5% for AA and-

Q183 John Mann: I am asking a slightly different question. Within those parameters, how often are you getting it wrong with your ratings? Is it beyond that margin of error? What margin of error would be acceptable beyond that?

Paul Taylor: The statistics-the detailed table-tell you exactly that answer. What the statistics do not tell you, of course, is when we get it right. There have been hundreds of defaults during that period of time we have talked about and the vast majority would have been down at the B, CCC level.

Q184 John Mann: I am trying to get a picture of the margin of error that one could anticipate with yourselves; of whether that is under 2%, under 1% or as high as 10%?

Paul Taylor: If you had 100 BBB credits, after five years 2% of those would have gone bust; two out of the 100 would have failed. Over 10 years, five of the 100 would have failed.

Q185 John Mann: That is your working assumption, not looking backwards but looking at the next five years? What you are telling me is history. I am interested in the assumptions based on going forward, because you are predicting and investors are predicting.

Paul Taylor: Yes. I would assume we would not be too far away from those numbers.

Q186 John Mann: So, 2%?

Paul Taylor: You are being too specific with something that does not work that way, but over time, over the past-

Q187 John Mann: If I am trying to calculate that risk as an investor, looking at the percentage would be a good parameter for me. If you are 2% wrong, I want to know about it. If you are 10% wrong, I want to know that. I am trying to ascertain what percentage of error one can anticipate.

Paul Taylor: That is why we published the default statistics-that is your answer.

Chair: Yes, I think we have understood the point.

Q188 Mr Ruffley: Could I ask Moody’s first and then S&P, of the investment grade ratings of A or higher in 2006, what percentage of those in the structured finance instrument market did you have to downgrade subsequently?

Frederic Drevon: Again, I am afraid I do not have those specific data in front of me. If it is the US housing market, it is going to be a very significant percentage; if it is the European market, a very small percentage.

Mr Ruffley: The US market?

Frederic Drevon: Yes.

Mr Ruffley: What percentage? Your numbers-what are we talking, 80%, 90%, 100%?

Frederic Drevon: Again, I am afraid I will have to-

Mr Ruffley: You do not know?

Frederic Drevon: No.

Q189 Mr Ruffley: S&P, do you have any clue at all, or are you similarly clueless? The same question that I asked Moody’s.

Dominic Crawley: I do not have those numbers, but we can very happily provide them.

Chair: Fair enough. Where you have said you do not have data to hand it would be very helpful if you could provide it.

Q190 Mr Ruffley: It should be burned on your minds, given that you got it so wrong. There are two possible interpretations; one is incompetence and the other is that the conflict of interest affected judgment. You said, Mr Drevon, that it was not individuals who were culpable or accountable, there were committees. How many people on average were sitting on the committees that were doing ratings for the US market for structured finance instruments? Are we talking 20 people, 10 people?

Frederic Drevon: I believe the comment was made by my colleague Dominic Crawley that in practice the average size of a credit committee would have been anything between five and 10 persons typically.

Q191 Mr Ruffley: Five and 10. I am assuming it is the same in S&P, Mr Crawley; is that right?

Dominic Crawley: It is broadly that and it is not a surprise that I would not have figures about US structured finance committees to hand. Again, if you would like that information, I can provide it to the Committee.

Q192 Mr Ruffley: Yes. The next question I have, of course, is that there were a lot of people on these committees, all of whom got it wrong. You have admitted that there were lots of downgrades. Are all those people still working in Moody’s, Mr Drevon, or have any of them been laid off?

Frederic Drevon: Can I just comment on one thing?

Q193 Mr Ruffley: I would like you to comment on the question.

Frederic Drevon: I will, but downgrading does not mean there is an error or fault. Downgrading will happen if there is a financial crisis and in general we would expect ratings to get downgraded in such an environment. It does not indicate that our ratings are wrong or right, just to be absolutely clear on that. I do not have any specific numbers on people who left the company around the subprime crisis or afterwards, but the company certainly reduced in size because of reduced activity.

Q194 Mr Ruffley: It was reduced in size, okay. What about S&P? Were any people dismissed for lack of competence in pricing risk on the structured finance instruments in the US market, Mr Crawley?

Dominic Crawley: We rate structured finance instruments. We rate corporations, whatever it is. We rate according to our criteria and the committee will sit down and review information.

Q195 Mr Ruffley: We have heard all that. I am asking whether anyone has been dismissed following the disastrous mispricing of risk in relation to structured finance in the US market. I am asking you the question: what happened subsequently when the shambles unravelled at S&P, Mr Crawley?

Dominic Crawley: S&P’s ratings are not about pricing risk. They are about doing assessments about the relative creditworthiness of individual instruments and that is what the committees would have sat down and-

Q196 Mr Ruffley: Did you fire anyone?

Dominic Crawley: I have absolutely no idea. The point I am stressing to the Committee is that rating committees will conduct their analysis according to the criteria, and that is the way that rating committees work.

Q197 Mr Ruffley: Right. In both your case, Mr Crawley, and in your case of Moody’s, you do not think there is anything to apologise for. Is that right, Mr Crawley, you have nothing to apologise for? You have no regrets?

Dominic Crawley: We have already been through this discussion.

Q198 Mr Ruffley: No, you have not given an answer. You have nothing to apologise for; is that what you are saying to the Committee?

Dominic Crawley: I have said that we regret that the assumptions that were used did not prove to-

Q199 Mr Ruffley: Let me move on very quickly, because there is an important point to all this, isn’t there? It is suggested by the European Commission and others that the lack of competition in the credit rating agency market means, frankly, that you are not accountable to anybody, are you? The market can’t really punish you. You can’t lose a lot of business, because there are not enough competitors. You would agree with that, wouldn’t you, Mr Crawley?

Dominic Crawley: We are very accountable. We are accountable every day to the market. As I said, we rate such a wide variety of instruments and entities. We interact with a vast array of investors every single day. We are regulated. We are regulated and we are accountable to our supervisors.

Q200 Mr Ruffley: But there are not many people who can leave you if they are not very happy with your performance. Let me bring in DBRS here, because I am very interested to learn what the barriers to entry in this market are for smaller players who obviously want to grow their business. What are the barriers that your outfit experienced, because you are in a position to make it clear there is a lack of competition?

Alan Reid: Sure. Regulation, perversely, is a very significant barrier to entry because there is a very significant cost associated with that. From our investment here in Europe, a significant amount of money has been spent on meeting the regulatory requirements. We have been in business, as I said earlier, since 1976 so we have an infrastructure in place. We have credibility in the market for the services that we are offering. We have demand for our ratings so as a result of that we are able to generate revenue and, therefore, support the cost of this regulation. But it is very expensive.

The second barrier to entry is the recognition of the larger names. Moody’s and S&P, for example, are synonymous with rating agencies, so you think, "rating agency", and you think, "Moody’s and S&P". Obviously the market has to learn that there are alternatives out there. The third area is the actual distribution of ratings, and the primary distribution of ratings through the global capital markets is through Bloomberg. We had to lobby Bloomberg for two years back in 2004 and 2005 to get them to accept our ratings. A bunch of different things makes this a very challenging industry to enter.

Q201 Mr Ruffley: Do you think the relative lack of competition hampers a proper dissemination of information? Isn’t it the case that there should be more players here? Do you understand why there are not more new players entering the market?

Alan Reid: I think I have answered the question with regard to why new players can’t enter the market or why it is challenging. A very significant capital outlay is required. I think you have a very significant concentration in risk in any industry if you are just looking at one or two players. I think that diversity of credit opinion is beneficial. As we have been discussing earlier today, you have talked about the issues that there have been in the market. I think the more opinion that is out there, the better it will be for avoiding crises like that in the future.

Q202 Chair: That is very helpful. Thank you very much. I just want to clarify one point. You said there are significant costs associated with the regulatory requirement. Is that over and above what you would have to do anyway? In other words, it is a pure regulatory cost?

Alan Reid: Historically, in the rating agency business, the methodologies were prepared and agreed upon by the people who covered those specific industry sectors. Now we have checks and balances in place where we have an independent review function who will look at the methodologies that we are preparing, and this is also elevated to the board. We have also had to employ-

Q203 Chair: I am trying to clarify what costs you can attribute to the regulator and what costs would more reasonably be attributed to a function you would perform anyway.

Alan Reid: We established our business here in August 2010 and I would say to you that probably, and I am guessing, maybe as much as 50% of our first year’s costs were related to regulation.

Q204 Chair: I think this is extremely helpful. It would be very valuable if you could drop us a line with that. That is the first year "get yourself accredited" number, by the sound of it, but then there will be an annual running cost?

Alan Reid: There is an annual running cost.

Chair: Rather than give it to me now, go and have a look and try to produce your best effort at a figure. I would also be interested to see the same figure from the other three agencies here today, if that is okay.

Q205 Michael Fallon: Mr Taylor, if we could turn to the European proposals, what are the weaknesses in CRA 1 and CRA 2 that CRA 3 is needed to address?

Paul Taylor: I am not sure what weaknesses there are in CRA 1 or 2 because they really have not been implemented yet. They were passed quite recently. We are only just beginning to go through the process of ESMA coming in to review us. They have just finished. We have not seen how those two first rounds of regulations will work.

Q206 Michael Fallon: What is the answer to my question? You are saying we do not need CRA 3?

Paul Taylor: No, I haven’t got to CRA 3 yet. I am saying with 1 and 2 it is just too early to say what impact it is going to have and how useful or not it is going to be. I think, on balance, with 1 and 2 it is going to add to the transparency of what we do, so it is probably okay; we just have not seen it in practice yet. There are some aspects of it as well that we do not agree with and we think are not worthwhile. We have talked about the regulatory cost issue, but we have to comply because we are regulated.

Moving on to CRA 3, there are 32 provisions in it. A handful of them are very damaging. Many of the others are fine and, in fact, some of them will be quite helpful, mostly again with regard to transparency and visibility of what we do and how we do it. Most of the discussion around CRA 3 comes down to rotation, the ability of ESMA to approve methodology, and then the last one is the legal liability regime. They are the ones we have concerns about.

Q207 Michael Fallon: Sure. Let us ask Mr Reid about rotation, then. Is it your view that mandatory rotation will increase competition?

Alan Reid: It is my view that mandatory rotation will increase opinion, provide diversity of opinion, and will reduce concentration risk in opinion.

Q208 Michael Fallon: Does that mean it will increase competition?

Alan Reid: It should increase competition, yes. It will create a more level playing field because at the moment that playing field is dominated by a couple of agencies. As you open it up, as mandatory rotation takes place, it allows more to come in and play in that particular market or provide opinion in that market.

Q209 Michael Fallon: Are you aware that the Treasury evidence to us is that mandatory rotation "May incentivise rating agencies to issue favourable ratings in order to gain favour with the issuers, generating the same, or even stronger, conflict of interest that the regulation is intended to avoid"?

Alan Reid: I think if you look at ratings across the globe that have been provided by the four rating agencies you will see some differences in opinion. I am responsible for the financial institutions business; I am responsible for the sovereign ratings business at DBRS. We track our performance of our ratings versus those of the other rating agencies and we hold different views. Again, I come back to the point, when you have a concentration in an industry, the market, the investor base, is predominantly looking at two names.

Also, your comment about the Treasury-I have a quote here from them where they say that they hope the rotation requirement would ensure that credit rating agencies and issuers do not develop a long-term relationship that can potentially lead to more favourable ratings for issuers. I think the Treasury in certain respects agree with the opinion that the greater the diversity of the relationships, perhaps the better quality of the ratings.

Q210 Michael Fallon: Let us try Mr Crawley. Why wouldn’t you be in favour of rotation?

Dominic Crawley: I think there are probably three principal reasons. One is I think it will run the risk of generating volatile ratings because there will be replacements-new ratings will come in, so there will be volatile ratings. I think because of that the second big concern is that the capital markets are now international capital markets. We do not just have a European capital market and a US capital market; we have global capital markets. I think the ability for UK businesses and for European businesses to raise money on capital markets will become more difficult as a result of rotation. I think access will become more difficult and, even if they do get access, the cost of capital will go up. Thirdly, it is a blatant impingement on normal competition to impose this.

Q211 Michael Fallon: Are there other examples of sectors where the big three have welcomed the entry of numbers four, five and six?

Dominic Crawley: I can’t comment. I don’t know.

Q212 Michael Fallon: You do not. You are the big three; why would you favour competition? This is self-interest, isn’t it?

Dominic Crawley: We have been very clear and very public over an extended period of time that we welcome competition. Ultimately, we are going to stand or fall by our reputation in the marketplace and the value that we can offer. We have no problem with competition whatsoever.

Q213 Michael Fallon: Why wouldn’t rotation give us some competition?

Dominic Crawley: You asked me what I saw as the risks and I am looking at it from the point of view of the market-the risk to the market of rotation. One will be volatile ratings and another one we see very likely to be problematic in terms of companies’ access to funding.

Q214 Michael Fallon: You are sure you are not just looking at it from your own position? The Treasury seem to believe that if you accept the Commission’s proposals you need a minimum of six agencies by 2015 and that might be quite difficult to get. Is it possible, Mr Drevon?

Frederic Drevon: It is very difficult to speculate how the market would develop. I think one of the important things about the value that a rating can provide to investors is it has to be, firstly, a global tool, so available to investors all around the globe. It has to be comparable, so established on standards that make it useful for an investor in Asia to consider, for example, a UK bond. It will take some time to develop this type of new competition. You need also to establish a track record and, therefore, publish your statistics about how well your ratings will perform over time. We have to say that we also consider that our competition is beyond just the rating agencies around this table. We compete also with other providers of research about credit risks such as the banks but also the market. Credit defaults are also a measure, in fact, of credit quality, which is very much relied on by the market and we also have to compete with these different tools, which may be more volatile but may be also quite accurate in the short term.

Q215 Michael Fallon: In your normal line of work if the big three become the big six, you are going to suffer commercially, aren’t you?

Frederic Drevon: Again, it is very difficult for me to speculate on how this market will look in many years but I think we have been quite public in saying that we support competition if it develops ratings quality and diversity of opinion. I think those two conditions are very important. As we understand it and as we analyse the rotation proposals, we see those proposals going against developing ratings quality because you would be chosen by default after every rotation, irrespective of the quality of the work you do, and you would reduce the number of opinions because we would be forbidden to provide an opinion. It does not seem the right way to develop competition as opposed to other measures that could be taken. Those other measures could be looking at reducing the reliance on ratings in a regulatory framework. We could also look at improving the disclosure framework that exists in the marketplace. It is important for existing but also potential new rating agencies to have better quality access to information. There are a number of different channels to try to achieve that that would probably be more successful than what the European Commission is suggesting at this stage.

Q216 Chair: Mr Reid, you wanted to add something?

Alan Reid: Yes. I do not agree with these opinions on ratings quality or volatility of ratings. We are regulated by the SEC-"we" being DBRS-and my three colleagues here are regulated by the SEC. DBRS and the three rating agencies here went through the same registration process with ESMA, as did a number of other rating agencies. We have been recognised by the Federal Reserve Bank in the United States for TALF for the liquidity programme. We also have ECAI status where our ratings are used. So there are a number of credible regulators, central banks, that are using our ratings and using the ratings of other rating agencies. I think there is a perception that other rating agencies provide quality ratings and that it does not create volatility in the market. I would argue in many respects that if you look at where we have had concentration of rating agencies and one of the larger organisations makes a rating move, a negative one, inevitably the other organisations may make a rating move as well, which can result in some negative things happening in the market. Therefore, the more agencies you have, the more opinion there is out there, and the less volatility there should be.

Q217 Michael Fallon: That is group-think.

Alan Reid: Group-think? We all bring different things to the table. If you look at our methodologies, they differ. One of the things that was talked about with the ESMA proposals was harmonisation of the rating scale. We look at different factors in addressing or coming to our ratings decisions so I do not think you end up with group-think.

Q218 Chair: Do you have anything to add to that rebuttal of the point you made on volatility, Mr Crawley?

Dominic Crawley: I think it is a matter of fact that if ratings are going to change more frequently, in terms of being provided by different providers and changing, there will be more volatility. I look at it from the perspective of the market, and I think about investors both in Europe and investors outside Europe who will be looking to invest and will be, say, taking into a portfolio a security with a particular rating on it that then disappears, say six months later, and is replaced by another rating at a different level. I have a real concern that it could create dislocation in the market and make it more difficult for entities to raise finance.

Alan Reid: I think that is an opinion we would agree with to a degree but-sorry, I have lost my track.

Q219 Chair: Don’t worry. I will have a go at this volatility point. When a stock has a larger number of buyers than sellers it does not tend to increase the volatility; it tends to decrease, doesn’t it?

Dominic Crawley: The specific point I am making is about access to funding and to capital markets. I am thinking particularly about primary markets here, the ability of entities to raise capital in the capital markets when they first go to the market. If there is going to be uncertainty in the market about whether a rating that is there at the start is still going to be there in six months’ time, whether it is going to be replaced by somebody else, whether not only the rating but also the research and analysis and opinion that supports and is provided by a particular agency is no longer going to be there, that will also be-

Q220 Chair: That is not the same point at all that you made to begin with, which was to suggest that the quality of capital allocation would decrease with the consequence of an increase in volatility in the market because more ratings were available.

Dominic Crawley: I think ratings will become more volatile and I think that, therefore, will impact upon the volume and the willingness of investors to invest and that will result in a higher cost of capital.

Q221 Chair: The markets already price in a given level of volatility, don’t they?

Dominic Crawley: There are other factors that go into-

Q222 Chair: As a consequence of the fact that there is more than one regulator, if we go from three to six, they will price in a higher level, the market will aim off for the fact that there is now a higher level of that type of volatility in the market, won’t they? Therefore, there will be no effect on the efficiency of capital allocation.

Dominic Crawley: But that in a way speaks to the point because if you have an investor down in Asia who is trying to decide whether to allocate investments to the US, Europe or somewhere else, they will take that into consideration.

Q223 Teresa Pearce: We have three credit agencies that dominate the market. We have heard that the individuals take great responsibility for the work they do, that they are just opinions, that they are highly trained, highly experienced, highly paid, yet at the same time you are saying that this is just one piece of information investors should rely on and they should not rely on your opinions alone. Are you saying that investors should do their own modelling, the same modelling that highly paid, market-dominant analysts do, to make a decision? Are you saying let the buyer beware? If that is the case, who can rely on your opinions?

Dominic Crawley: I am sorry to sound like a broken record, but I am going to say what I said at the start. Ratings are opinions. They are one piece of important, valuable information that is available to the market and to investors. But there are many other pieces of relevant information around credit decision-making that are also available elsewhere in the market. We have been absolutely clear that we do not expect an individual investor or, for that matter, at the other end of the spectrum, a sophisticated asset manager to rely solely on the tool that we provide in order to come to a credit decision.

Q224 Teresa Pearce: But do you not think that, because you dominate the market and are highly paid, people would have some reliance on your opinion?

Dominic Crawley: We go to extraordinary lengths, in terms of time, effort and money, around the way that we dialogue with the market to explain to them what a rating is and what a rating is not. Notwithstanding that very often we are unpopular in terms of what we do, at this moment in time we have never had more what I would describe as reverse inquiry from the market generally wanting to know more about our opinions, research and analysis in our ratings. We spend an enormous amount of time interacting with the market explaining precisely what a rating is and what it is not and how it can be used and how it should not.

Q225 Teresa Pearce: Mr Taylor, what do you think?

Paul Taylor: Ratings addressing long-term fundamental credit risk through an investor in securities-there are other factors you need to take into account as well. Probably the best example of that recently was that S&P downgraded the US and more investors bought US securities. They went completely against the direction of the rating change. They did that was because they were looking for liquidity, a safe haven. People do not blindly follow the ratings, often because they have their own credit opinion that is different from ours; that is healthy, but also because there are other factors they are thinking about: liquidity, pricing, positioning their portfolio, where they are not as concerned about the absolute credit, fundamental long-term credit risk, which is what we are opining on.

Q226 Teresa Pearce: You are obviously saying the same as you said before. I did get it clear before that that is what you were saying. I will move on to a point about accountability and if I could ask you, Mr Drevon, what role do your non-executive directors take in holding you to account? Has that changed at all since the crisis or is it still the same?

Frederic Drevon: In fact, it was one of the, I think, quite positive changes that was introduced by the first round of regulation, CRA 1, which introduced a number of changes around the governance of rating agencies. In the context of European-registered rating agencies, one of the requirements was to create a board with the inclusion of independent board members. These independent board members have some very specific responsibilities about monitoring the activities of the rating agencies, ensuring that we have adequate controls in place and then providing opinions to the board about their findings and their views about the activities of the rating agencies. It adds another element to the controls we have in place by bringing in independent directors in that role. Certainly, I think, a very useful and positive change.

Q227 Teresa Pearce: Mr Taylor, what do you think?

Paul Taylor: It has been a good change. We did not have non-executive directors in the ratings business. We had a group company that owns other subsidiaries where we had independent directors or shareholders. In the ratings business itself, we introduced non-executive directors, I guess, 18 months ago. My initial view, to be perfectly honest, was it was going to be a burden, and was going to be an admin burden. It turned out to be anything but that. In fact, the two people we have are incredibly useful and helpful in terms of testing us and investigating the business, and they are getting into the details as well. It has been a healthy and positive change.

Q228 Teresa Pearce: If I could ask Mr Drevon and Mr Crawley a question. In the submission from Fitch, the requirement for ESMA to review changes in rating methodologies had cold water poured on it somewhat. Does your organisation agree with that?

Frederic Drevon: There is, as part of the first round of regulation, a requirement that rating agencies, at least on an annual basis, revisit their methodologies. We have at Moody’s hundreds of methodologies, but it ensures to the market that these methodologies are not stale and that we update them on a regular basis. The responsibilities that ESMA had under the first round of regulations was to ensure we had adequate controls in place to make this review as effective as possible.

I think what is being suggested by the third round of regulation is something quite different, where ESMA would pre-approve or post-approve methodologies. That brings you into a completely different field where you are, to some extent, substituting the opinion of the rating agency with the opinion of the regulatory body, because approving really means interfering with the independence of the rating agency. The concern about that is, first, that if there is a loss of independence, there is a loss in the value that the rating may provide to the marketplace. It also creates, over time, naturally, a system whereby the majority of rating opinions may converge towards a unique view of the world. At Moody’s we believe that we should promote diversity in opinions and not a single view about credit risk, because it is an imperfect science. We are looking at future events. We are doing our best to try to be as accurate as possible, but there is some degree of uncertainty in the work we do.

Dominic Crawley: Yes, analytical independence is a cornerstone for us, in terms of the way that we function. The existing regulation spells out that analytical independence is a principle that it upholds. I know that the FSA quite recently has gone public saying that the principle of the independence of analytical approach should be upheld. I think this goes to discussion and questions earlier during the Committee hearing around diversity of opinion. If this was to come to pass it would, over a period of time, remove that diversity of opinion. So, as has just been expressed, I think that would be a negative.

Q229 Andrea Leadsom: On one of your answers, Mr Taylor, to Teresa Pearce, I am slightly astonished that you would say that, in the case of the United States when it was downgraded, more people bought the stock. You are not seriously suggesting that if you were to, say, downgrade the National Grid from its investor BBB-plus credit standing to CCC, which would not be investment grade, that people might buy that debt on the basis of your downgrade? You took the one exception that proves the rule, didn’t you, to infer from it that people do not take credit rating agencies literally? I would challenge you to think of many more examples.

Paul Taylor: There are lots of examples. The point you are making is a good one, which is the cliff risk between investment and non-investment grade. France was another good example just in the last month. The point I was trying to make is it does not always follow that the market follows what a rating says. It is simply not the case. In fact, if you look at the eurozone at the moment, the market is completely not following what the rating agencies are saying.

Chair: We are going to come back and take a look at sovereign risk at a subsequent hearing so maybe we should leave that for a later time, if that is all right, Andrea.

Q230 Mr Love: Coming back to a point raised by Mr Mann earlier on about banks, can I ask you-I will start with Mr Crawley-is the rating of financial institutions as reliable as the ratings of corporate or other types of ratings?

Dominic Crawley: Yes, I believe it is. It has one layer of added complexity in it that, for the vast majority of, say, industrial corporations does not exist. That is that financial institutions, regulated banks are highly leveraged entities and they manage to operate with ongoing system support, access to liquidity, funding from central banks and so on, plus, depending on the individual financial institution and how important it is to the system, it also gets support from governments, in terms of the expectation of future government support. That is an added layer of complexity around the way the banks are analysed. But, absolutely, and we have been very conscious of the need to provide greater clarity and transparency to the market around how we see what we describe as the stand-alone credit profile of the bank, standing alone without government support and how much government support we would factor into a bank’s rating.

Q231 Mr Love: Mr Taylor, would you agree with that? Perhaps you could focus on this issue about implicit government guarantees, because there seems to be some ambiguity about exactly how to quantify such a thing. Does that make it more difficult to rate financial institutions?

Paul Taylor: It is not easy to rate financial institutions. I guess it never has been. It is a very good example of something; it shows how ratings work. During the financial crisis we saw many banks fail so they needed support. The level of bank failures to, say, corporate failures is something like 10 times the level. You do not see it in the debt markets because the banks did not default, they kept paying their securities. Even though they were failing, they were still paying their debt instruments. One of the things we have been wrestling with over the last couple of years, as the debate has happened among policymakers, is that we think in many markets, the level of central support will come down for financial institutions. Therefore, you are more reliant on their stand-alone strength, the strength of an individual institution.

How do you build that into the ratings? How do you take away the support that would have been there in the past and look at them more on a stand-alone basis. If you look at our financial institution ratings-if you look at all the financial institution ratings over the last three years or so-they have been coming down, not down into non-investment grade but down by a couple of notches, maybe. So it shows ratings adjusting to the new reality of less support assumed going forwards. But it is different in different markets. We would not have assumed less support in China, for example; we assume less support in the UK.

Q232 Mr Love: That does not sound too scientific; a recognition that there is less support might lead to a downgrade. It is how you measure, evaluate that downgrade. I submit to you that we had the Independent Commission on Banking, which had one figure; the market had another figure; all the research institutions had different figures. Nobody could quite quantify and, therefore, I wonder is your evaluation of financial institutions as solidly based as it is for others?

Paul Taylor: The stand-alone financial strength can be assessed. Obviously at the same time they are losing support, their capital is going up and their asset base is getting stronger, so you can assess that and measure it. It is never going to be an exact science of knowing exactly what is going to happen. It will come down to the specific situation, the decision of governments. We can’t predict that. We can put our best judgment in place and explain it but we can’t predict it.

Q233 Mr Love: Can I ask you, Mr Drevon, when Moody’s downgraded I think it was 12 UK financial firms, Lloyds commented that Moody’s was reflecting what was already understood in the market. I think they were suggesting that the credit reference agencies were a lagging indicator in that regard. Is that often the case? Do you just reflect what the markets say?

Frederic Drevon: When you look at the level of support from a systemic point of view for financial institutions, there are good examples where we did not like the market and, in fact, had quite a different opinion. Before the start of the financial crisis, we had taken the view that systemic support was going to be generally quite high for a number of countries and we had assigned ratings that included that level of systemic support. That assumption held very well during the financial crisis, but obviously over the last two or three years we have, as with every other market participant, understood that this was an evolving situation, that each country was approaching this in a different way with, in general, a reduction in the level of systemic support to be expected. So we have reflected that in our rating on an ongoing basis, not only for the UK but also for a number of other European countries, differentiating what we would view as very large financial institutions, which are more likely to receive some degree of systemic support, and smaller financial institutions.

Q234 Mr Love: Mr Crawley, Lloyds were making the point that the downgrading would have minimal impact on their funding costs. Has that proved to be correct so far? Have the financial institutions, as a result of the downgrading of the credit reference agencies, had a material increase in their funding costs?

Dominic Crawley: I would like to be helpful. I do not have that information to hand but it would not surprise me at all if it had not had a material impact.

Q235 Mr Love: So, in effect, you are a lagging indicator? The market was already pricing in what you said latterly.

Dominic Crawley: The pricing of credit risk at any one time is affected by a whole variety of different factors: vagaries, market sentiment, supply and demand, many other factors.

Q236 Mr Love: Mr Reid, perhaps I will ask you. It is an entirely different question. The European Commission has proposed that the issuer should be informed of rating changes in advance and also that announcements should be published outside market hours. How will that affect you?

Alan Reid: The logistics of that can be very challenging. At the moment in Europe we have a 12-hour rule, so you inform the issuer and then you have to wait 12 hours before you publish the opinion that you have or the rating change. That 12-hour time period obviously leaves a window for potential market abuse of that information. There have been some proposals that those timeframes become longer, which would obviously heighten those risks. Historically, we would reach out to an issuer, communicate the change to them, and within the space of about an hour we would communicate that change to the market, which seemed fairly effective.

Q237 Mr Love: Mr Drevon, what difference will it make in terms of the issuer? Will it give them more time than is absolutely necessary to prepare their response?

Frederic Drevon: From our point of view I think we were not in favour of the 12-hour rule. We felt that it would create potential market situations where information that should not be disclosed would be disclosed to the marketplace. We have worked generally within the principle that when there is information that should be made available, it should be made available to the market as soon as possible. So I think that the principle is if it does not go in the right direction then arguably 12 hours is sufficient for any issuer to review a potential press release and comment on how we are correct in terms of the facts or if we are not providing the market with any confidential information. We do not need 24 hours or 36 hours to do that.

There is a separate appeals process that is available to any issuer who disagrees with a rating decision, and we have a specific separate process that ensures this appeals process provides the issuers with the adequate process in terms of us revisiting and reviewing a rating.

Q238 Mr Love: Finally, Mr Crawley, why would the European Commission suggest it? I am thinking particularly about the 12 hours. It seems rather strange to me, but perhaps you could enlighten us as to why they think it is necessary in the circumstances.

Dominic Crawley: Sadly, I can’t, and I do not know what has driven it. I would just make one observation that, before the 12-hour rule, we operated with a much shorter timescale and that worked perfectly satisfactorily at the time. The current 12-hour requirement is something that is logistically very challenging but also leaves open this potential for selective disclosure and subsequently market abuse.

Q239 Mr McFadden: Mr Crawley, I am afraid I want to go back to structured finance packages and subprime mortgages in the United States. Looking back at that period, to what extent did the business model for rating these, with banks negotiating with rating agencies for each issue, encourage a form of rating shopping? Did they try to structure these packages to maximise the ratings?

Dominic Crawley: I am going to be helpful to the Committee, but there are two observations I have to make. One is that this is five years ago now and we have made various representations over the years around this. The second point is we are talking about US structured finance, which is an area of the business I have never been involved in so I am not familiar with it. But I would say this, and I will repeat what I said earlier on in the hearing: we rate according to our criteria. Whether it is structured finance or rating a company or an insurance business or a bank, we have our criteria, we have our information requirements. We take that, we work against the framework of our criteria, and we then take it through the committee analysis and decision process. Structured finance in the US would have been no different. We rate according to our criteria.

Q240 Mr McFadden: Mr Drevon, perhaps I should ask you then if Mr Crawley says he had nothing to do with this. Do you recognise this term "rating shopping", and did banks issuing these securitised packages negotiate with Moody’s to try to shape the packages to get the best rating they could?

Frederic Drevon: I think the concept of rating shopping certainly happened. I think it is well known that this was one of the issues with the way the US structured finance market performed. In fact, it is one of the side-effects of having very transparent methodologies, because as soon as you publish something that has very precise criteria to the market, underwriters will read this very carefully and try to optimise the structure in response to those criteria and therefore achieve the highest possible rating.

So that is a side-effect of transparency and in that context you had what is so-called rating shopping where underwriters would approach different rating agencies and, to the extent that one or two rating agencies were not in a position to achieve the rating desired, they would not work with that rating agency. You could argue that rotation, as proposed by the European Commission, may achieve the same thing, because every three or four years it would be open to each issuer to re-tender for a rating and in all likelihood they would want to choose the highest rating.

Q241 Mr McFadden: So you are confirming that rating shopping took place-people tried to game the agencies to maximise their rating. Going back to the initial question that the Chairman opened the session with, in those circumstances what steps did someone like Moody’s take, given that the fees being paid were very lucrative and were becoming a growing part of your income? What steps did you take to make sure that the ongoing commercial relationship did not pressurise analysts to, if you like, inflate the ratings?

Frederic Drevon: That is probably one of the most significant changes, which came at the initiative of the rating agency, and I can speak for Moody’s, but also what is subject to regulatory oversight: the creation of a complete separation between the analytical side and the commercial side. I think we have seen around this table we have certain individuals who are focusing purely on the analytical side and others who have responsibilities that may include commercial aspects.

So an analyst at Moody’s will have no information about the level of fees, will not be involved in any commercial discussions, will not be involved in any contractual discussions, and will have obviously an additional requirement in terms of not holding any security off an issuer that that analyst would potentially rate. So there are a number of ways to manage this and it will now be subject to supervision. ESMA will come in and check that we are operating under those rules. We believe we are, but there will be oversight by a supervisory body to do that.

Q242 Mr McFadden: Are you saying there is a sort of internal Chinese wall between those doing the analysis and those negotiating the fees?

Frederic Drevon: Yes, absolutely.

Q243 Mr McFadden: Moving on to you, Mr Taylor, I want to ask you a question about the general way that rating agencies are paid. In the early days of rating agencies it was the investor who paid. The investor would say, "I’m thinking of investing in this product, would you rate it for me? What is your assessment of the strength of this product?" It is not like that any more. It is the issuer who pays the fee. It is the seller of the product who pays the fee. Does that not create a conflict of interest with the seller paying the person who is expressing a rating opinion on his product for the investor market?

Paul Taylor: Yes, it does create a conflict of interest. I think the key thing is to recognise that conflict of interest and to manage it. Any other business model you could come up with would have a conflict of interest, so if investors pay they have an interest in whether we are rating up, rating down, or keeping it the same, because there are many types of investors. Many are short-term trading investors, so there could be a conflict with investors. If some kind of central government body is assigning ratings, there would be a potential conflict there; would it downgrade the nation’s banks when it needs to?

So there will always be a conflict in whatever model you choose. You have to manage the conflicts. I think with the issuer pays model, it has been scrutinised so much over the last seven years in particular that there are a whole raft of issues put in place now to manage and to disclose those conflicts.

Q244 Mr McFadden: Explain to me why there is a conflict if the investor pays.

Paul Taylor: I own a bond and I am thinking of selling my bond, so if there is going to be a rating action on that bond it could influence the price of it. So if I can influence the decision as to whether it goes up, down or stays the same I may have an interest in it.

Q245 Mr McFadden: Across the industry, if you recognise that there is a conflict of interest in the issuer pays model that you all use, has there been any cross-industry discussion or is there a deep internal discussion, say in your firm, about, "Could we come up with something else? We admit there is a conflict of interest here. Could we come up with something else that would give greater market confidence about who pays us and how we earn our fees?"

Paul Taylor: We have had those discussions internally. We can’t come up with a solution that allows us to keep the scale of activity we have, to have the breadth of coverage we have, and to have the independence that we have. The issuer pays model is understood, it is accepted. When you talk about improving market confidence, I think the market still has confidence in credit rating agencies in the work they do. We are still used by people; they still talk to us on a regular basis. I think they are a bit wiser now in terms of using other sources as well, there is less blind faith, but we still have perfectly good discussions with the market.

Q246 Mr McFadden: You must accept the reputation of the rating agencies took a very substantial hit after the whole American experience.

Paul Taylor: Yes, it did. We are still talking in sweeping statements about the work of the rating agencies, which is how we started. We are now talking about subprime structured finance and back to the point that there will always be some individual mistakes. If you are an investor in corporate or FI securities, you are probably quite happy with the work that has been happening for the last 10 years.

Q247 Mr McFadden: I will come back to you in a second, Mr Crawley. Mr Reid, just coming to you on this issuer pays model. What is your view of this? Do you think it is the right thing going forward? Do you question it at all?

Alan Reid: We have spent a lot of time considering the model as well. This model has been in place for about 40 years, so it is not a new change that has been made. One of the benefits of the issuer pays model is that ratings are widely available and they are free of charge to anyone who wants to use them.

Q248 Mr McFadden: They are not free of charge to the issuer.

Alan Reid: They are free of charge to the investor, so they are publicly available. It seems to be the most effective model.

Q249 Mr McFadden: Mr Crawley, you wanted to come in.

Dominic Crawley: Thank you, yes. Two observations: one is that we do have parts of our ratings business that is an investor pays model; the other point was already made, which is the key thing about the issuer pays model is it allows very broad dissemination of ratings at the same time across the market, so it creates the widest market scrutiny of those ratings. The issuer pays model is a model that, as was just said, has been around for 30, 40 years. Originally rating agencies had an investor pays model and it was the market that required that change because of that ability to get wide dissemination.

Q250 Mr McFadden: Thanks. Mr Chairman, it does seem to me something we need to reflect on, that we have had a broad admission that there is a conflict of interest here and yet the attitude seems to be, "We don’t really know what to do about it so we’ll carry on as we are".

Dominic Crawley: There is a potential conflict of interest. It is about how it is managed, and under existing regulation and supervision with ESMA, it is part of the work that we do with ESMA all the time around demonstrating that we have appropriate processes, procedures, firewalls and so on, to manage that potential in a satisfactory way.

Alan Reid: I would argue that the issuer will be incorporating the price of the rating within their overall sales price of the security.

Chair: Thank you very much for coming in this afternoon. It has been bumpy at times. I thought you didn’t want to come in, Mark.

Mark Garnier: No, I do.

Chair: I am sorry, Mark. Mark wants to come in. It may be a bit more bumpy in a moment.

Q251 Mark Garnier: Yes, it may well be. Thank you, Pat, for asking helpful questions. That is very good of you. Mr Crawley, a bit earlier you said to Mr Fallon that you stand and fall by your reputation. How have your gross sales been affected since 2008? How has your income for Standard and Poor’s been affected? You said that you stand or fall by your reputation. Your reputation in 2008 was not the best. How has the market reacted to that in terms of your ability to sell your-

Dominic Crawley: I do not have access to that information so I can’t tell you. What I do know generally is that our model is fairly significantly correlated to activity in the market, debt issuance in the market, and therefore issuers’ desire to come to us for ratings to access the market. Obviously, over the last year, 18 months, two years, we have seen a very significant downturn in activity, but I do not have access to that information.

Q252 Mark Garnier: My whole point behind this is I have heard it before that rating agencies stand or fall by their reputation, yet what I am trying to get a handle on is whether issuers turned away from you at all? Notwithstanding the fact there has been a reduction in the issuance, have they turned away from all of you as a result of the fact that clearly you got it wrong in 2008?

Frederic Drevon: I can certainly respond to that question. We have not seen any unusual degree of withdrawal of ratings. Certainly in the European context there is an existing market of unrated issuers. Those are typically smaller-sized companies who have determined that they have a sufficiently strong brand and determined that they do not require ratings to be successful in the capital market, and that certainly is an option. But in general we have not seen a change in the degree of interest both from the investor and the issuer side.

Q253 Mark Garnier: So nothing has changed for the rating agencies? You still have the same market share, you are still doing the same level of business, and your reputation has not fallen at all?

Frederic Drevon: That is correct.

Q254 Mark Garnier: Okay, fair enough. Mr Drevon, you also answered questions by my colleague, John Mann, a little bit earlier about the local authorities who were looking at the rating agencies’ ratings on particularly Icelandic banks, and I think you said that they should not rely on rating agencies. An awful lot of people relied on them. Wyre Forest District Council, which I represent, lost £9 million on three Icelandic banks. They would argue that they were relying on you to come up with some ratings for security and now you are saying that they were foolish to have listened to you.

Frederic Drevon: I don’t think they should rely exclusively on the ratings we provided. I think again it should be-

Q255 Mark Garnier: Who should they rely on?

Frederic Drevon: They should rely on their own research. They should look at the ratings, they should look at the underlying research we provide, and they should look at additional research that is provided by the marketplace.

Q256 Mark Garnier: They can’t; they are a district council. Let us be serious about this: huge numbers of people are out there who rely on some sort of advice. A district council or district councillors are elected to go along and do a whole number of different things, planning, licensing, all sorts of things. They also have a certain amount of cash. They go out and they buy in investment advice. Those investment advisors are relying on you. There is a huge amount of money. Even the smart guys, even the Audit Commission, lost a huge amount of money as a result of taking the advice of rating agencies, and those guys are boffins.

Paul Taylor: They should have had a diversified portfolio.

Q257 Mark Garnier: They did; they had three banks.

Paul Taylor: So they should, therefore, understand that some ratings will always prove not to be the accurate assessment of what happens in the future. There is a built-in level of loss assumption in an investment portfolio.

Q258 Mark Garnier: You said a bit earlier, I think, what is it, a 2% failure of AAA-rated-

Paul Taylor: BBB-rated.

Q259 Mark Garnier: BBB. Well, these were AAA-rated banks.

Paul Taylor: No, they were not.

Q260 Mark Garnier: What were they?

Paul Taylor: Single A probably.

Q261 Mark Garnier: Okay, they are A-rated banks, three of them. All the banks in Iceland went bust. That is pretty amazing. So from now on people should not rely on rating agencies?

Paul Taylor: No, that is not what we are saying. We are saying it is a statistical-

Q262 Mark Garnier: There is an awful lot of very, very upset people out there, particularly taxpayers in some of these local authorities that John Mann and I represent, who are furious at this, absolutely furious. You are talking about libraries being closed down as a result of people relying on you guys.

Paul Taylor: You need to understand the product and how it should be used. There is always going to be a degree of-

Q263 Mark Garnier: Who would you blame?

Paul Taylor: There is no blame. We have to distinguish between the subprime crisis where there was a failure of the market, including the rating agencies, to identify what was going on there versus-not a normal situation, but a 10, 20-year situation where you should assume that some investment grade ratings will fail. That is what the default statistics tell you. So if you are telling me I have to predict every default that comes up with 100% accuracy, that is impossible.

Q264 Mark Garnier: All the banks in a country.

Paul Taylor: We can’t do that. There are thousands of-

Q265 Mark Garnier: You can’t? Okay. The other point is that you are talking about rating obviously debt issuance, but in fact in the case of banks people are looking at them in terms of getting an idea of whether they should put money on deposit or not. Are you saying that they should absolutely not use the rating agencies for that type of advice at all?

Paul Taylor: No, we are saying the rating is a good indication of long-term credit quality. There will always be exceptions to that being a black and white definitive because we can’t predict the future perfectly.

Q266 Mark Garnier: Getting back to conflict of interest. Thank you, Mr McFadden, for asking so many of my questions. In 2008 the Chartered Financial Analysts Institute did a survey in which 211 of its 1,959 members surveyed reported that they were aware of credit rating agencies changing a rating as a result of pressure by an investor, issuer or underwriter. When asked to specify how this pressure was manifested, over 50% stated that it involved a threat to take future ratings business to other CRAs and 17 mentioned the offer to deliver more business to the agency. Is that not quite a damning indictment that issuer pays has quite a strong conflict of interest?

Paul Taylor: I do not know the piece you are referring to and how it was created.

Q267 Mark Garnier: 2008 Chartered Financial Analysts Institute report. But it is a big issue, is it not? People have been talking about the fact that issuer pays has a substantial conflict of interest.

Paul Taylor: We are back to the point we talked about earlier, which is there is a conflict with issuer pays, as there is a conflict with any other business model. You have to identify and manage that conflict.

Q268 Mark Garnier: How would a new rating agency break into the market? If we look at the experience of Kroll, who have been going around trying to work on the model of investor pays, they have been going round and eventually gave up, so now they are looking at a model of issuer pays. But how does a new entry come into the market?

Paul Taylor: It takes a long time and a lot of investments. DBRS are doing what we were doing 10 years ago. There have been a couple of comments from people earlier about the three dominant players; that is wrong as well. We have a 17% market share. S&P and Moody’s about 40%, so there is not a dominant three. There are two and there is-

Q269 Mark Garnier: Is that 40% each?

Paul Taylor: Yes, of business. I am talking business, not number of ratings, because you can have unpaid ratings out there. So, market power, influence, business-there is a difference. You have to spend a long time knocking on doors, talking to investors, explaining your methodology, introducing them to your analysts, convincing them that what you are saying has some value and that they should listen because it may give them an insight into the future credit quality of an entity. There is no short-term solution to that. You just have to pound the pavements and do it and invest the money to do so, invest money in people. You are doing that now, we are still doing it. It takes a long time. There are new players in the US who are doing it. Morningstar have just announced a move into corporate ratings. They have a large parent behind them to fund it. But it is a long-term process; there is no short-term solution.

Sorry, just to finish on one comment: we think rotation will limit competition among European agencies. We think it will have the complete reverse effect of what the Commission is intending to do.

Chair: Would any of my colleagues like to ask any more questions? We will have another opportunity to examine some of these issues because we are going to be holding a hearing specifically on sovereign risk, and we will be sending invitations to a number of you to attend that. I have to say, on the basis of the reaction and questions from a number of my colleagues this afternoon, it appears that you have left the Committee unconvinced that many of the problems that attached to risk rating agencies in 2008 have yet been adequately addressed. I think there is deep concern in this Committee that these problems simply have to be tackled, not least the conflict of interest, lack of competition and the fact, to some degree, that a number of you do not even feel you have anything to apologise for. I think that is deeply concerning. But we will return to this at a subsequent hearing, and thank you very much for coming today.

Prepared 15th March 2012