To be published as HC 1866 -i

House of commons



Treasury Committee

Inquiry into Credit Rating Agencies

Wednesday 29 February 2012

Malcolm Cooper, Georg Grodzki, Mark Hyde Harrison and John Grout


Evidence heard in Public Questions 1 - 138



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 29 February 2012

Members present:

Mr Andrew Tyrie (Chair)

Michael Fallon

Mark Garnier

Stewart Hosie

Andrea Leadsom

John Mann

Mr George Mudie

Jesse Norman

Teresa Pearce

Mr David Ruffley

John Thurso


Examination of Witnesses

Witnesses: Malcolm Cooper, Global Tax and Treasury Director, National Grid plc, Georg Grodzki, Head of Credit Research, Legal and General Investment Management, Mark Hyde Harrison, Chairman, National Association of Pension Funds, and John Grout, Policy and Technical Director, Association of Corporate Treasurers, gave evidence.

Q1 Chair: Thank you very much for coming in this afternoon. This is a very important subject, and we are looking forward very much to hearing what you have to say. Can I begin by asking, Mr Grodzki, whether you have taken any actions as a consequence of Moody’s decision to adjust the ratings of a number of European countries? Does it matter to you?

Georg Grodzki: We were expecting that, so any adjustments to the portfolio had already taken place by the time the rating agencies, or Moody’s, executed the action. The market was already pricing in a high likelihood of rating change. The market was already pricing in a higher rate of risk on the grounds of certain economic data, fiscal data, so one cannot say that the announcement of the rating change itself was an earth-shattering event. If anything, it is the evolution of the crisis as such, way before the rating agencies took note, which has resulted in a number of portfolio adjustments to the effect that we reduced exposure.

Q2 Chair: So it is the press and the public who take far more notice of these things than the organisations that are acting on their decisions.

Georg Grodzki: Absolutely, because it is so much easier to communicate a rating change from triple A to a lower rating than a credit spread level. It is very difficult to explain to the average person what a credit spread is and why it is important that it is not 50 but 200.

Q3 Chair: In which case, setting aside the commercial side for a moment, on the sovereign side of your business-the sovereign side of risk rating-what does it matter really? Why are we all worried about it? Why are we doing all this regulation?

Georg Grodzki: There are a number of funds that can only invest in issues and bonds that carry a certain minimum rating. For example, when we invest, we have to make sure we are in compliance with such provisions if they do exist-

Q4 Chair: Those are regulatory provisions?

Georg Grodzki: Not necessarily. They are provisions put in place via agreement between the customers of ours, whose money we manage, and ourselves. The customers usually have a certain degree of risk appetite, and typically that is being expressed by references to ratings in the so-called IMA, the Investment Management Agreement. Therefore, we have to make sure we are in compliance, which sometimes can force us to sell securities if they are downgraded below a certain level, or buy them if they are upgraded. Also, even if we are not directly affected, we need to anticipate the market’s response due to such rating references. We need to anticipate the market’s response to a certain rating action because the market response could mean prices move, and of course we want to be on the right side of the next pricing move and that means we have to look at them.

Q5 Chair: On the technical side, looking at the output, do you think that the rating agencies do a better job than you do internally?

Georg Grodzki: They do not do a better job as such, but the fact that they are doing a job that is normally in our world, which is corporate credit ratings, reasonably is positive because the market benefits from a diversity of opinions. We use them as a reference point. We use them as a discussion partner, as a source of information, so the market’s thinking and reflection of credit quality is enhanced by institutions that regularly produce verdicts on credit quality.

Q6 Chair: But if they do not bring superior wisdom to the piece, and in certain cases you are forced to sell to take account of some arbitrary rules that have been put in place, some targets-they are not completely arbitrary, but some fairly rigid targets-in fact, they are having the effect of distorting optimal capital allocation, not facilitating it, aren’t they?

Georg Grodzki: They could if they were to systematically overrate or underrate securities. Indeed, that would be the outcome. In the world in which most of our funds are invested, there is no evidence that the majority of bonds being rated are either overrated or underrated. I would hazard a guess that, by and large, two thirds of the bonds carry a rating that reflects the market consensus opinion or our own opinion. In sum, the rating agencies are not necessary; the markets would work without credit ratings, but they can be quite useful as long as they are not too powerful. It is dangerous if the rating agencies and their ratings gain too much influence in the market. It hampers their own flexibility in adjusting ratings and it reduces the efficiency of markets if they are too powerful.

Chair: Others may want to come in, in a moment, on some of the points that have already been raised with which we have an interest but, in the meantime, I am going to pass questions over to David Ruffley.

Q7 Mr Ruffley: Most of the credit rating agencies sleepwalked into the 2008-09 financial crisis, inasmuch as they mispriced risk, did not identify it adequately with mortgage-backed securities, CDSs, CDOs, and so on. What I would like to hear from each of you, is for you to tell me whether you thought there was any problem with the ratings that most of the agencies were giving; some of these very exotic and-as we found out to our cost-very risky products. Perhaps each one of you on behalf of either your company or indeed your members.

Georg Grodzki: I would agree that the rating agencies collectively failed.

Q8 Mr Ruffley: No; I am asking you, in relation to Legal and General. What were you thinking when you saw the rating agencies giving you assessments of X, Y, Z mortgage-backed security?

Georg Grodzki: We were not ourselves investing in any meaningful size in those securities. I can say, with some satisfaction, that at the time the crisis gained momentum and escalated, we were not overly exposed to the toxic parts of the market. Whether this was smartness or luckiness, I leave to you. We were not surprised to see, in 2008, the dominoes falling, starting in the US and then continuing in Europe, because we felt what was completely missed in a way, by the agencies as well as by the market, was the enormous leverage the system had built over the years, which was based on the assumption of markets remaining liquid and that debt could be rolled over and refinanced at any point in time. As soon as that assumption was shaken and called into question, that house of cards was bound to wobble, if not collapse. So the market’s focus on the trees rather than the forest-i.e. risk-weighted capital ratios rather than unadjusted capital ratios, if you look at the bank balance sheet-was its downfall.

Q9 Mr Ruffley: Association of Corporate Treasurers, your members; were any of them just taking these rating agency assessments and ratings without any questioning? Did you rely on them? Did your members make mistakes by relying on the rating agencies’ ratings?

John Grout: The ratings you are referring to were those of the structured finance industry. Companies are not investors in the way that a fund like L&G is. They are investing structural or temporary surplus funds. It is not their business. It is just money they need, so what they do tends to be very simple, and I would be very surprised if any company had certainly any significant investment in any of those sorts of activities. What you are suggesting is that users may pay too much attention to ratings as an indicator as such, rather than using other indicators as well. Absolutely right. Companies, particularly smaller companies, do not have staff devoted to that kind of credit assessment. They will tend to look at the credit rating of a Government or another company, which wants to deal with them, and certainly of a bank counterparty. Our own guidance note on the subject says, "Look at other indicators. Look at credit default swap spreads, since they came out, look at other factors", but if you are a small treasury-I was talking to a treasurer of Interpol yesterday. He has three people in his treasury. They have a structural finance surplus, and they invest using the ratings as their principal trigger. They read the press, they listen to other indicators, but their principal trigger is the credit rating, and I would say that would be true of most companies using it as the start. If you are a really small company, you probably don’t look beyond the rating.

Q10 Mr Ruffley: I am guessing, Mr Cooper, that your answer will be similar to National Grid?

Malcolm Cooper: Yes, it is. National Grid did not invest directly in mortgage-backed securities. We could have had an exposure via our pension funds. We have some £20 billion of pension assets. Our UK mandate specifically precluded structured finance. In the US, we had a tiny exposure via an aggregated fund, which was allowed to invest in mortgage-backed securities.

Q11 Mr Ruffley: National Association of Pension Funds, what is your answer to the question?

Mark Hyde Harrison: The UK pension industry did not have any material exposure to those types of instruments. I think our members could see in the marketplace that the yields on those sorts of instruments were higher than other high creditworthy instruments, which indicated that they were risks, and the way in which risk could be assessed was difficult to assess; the cliff-edge nature of default rather than the gradual decline, which tends to happen with corporate debt. One of the concerns that we would have had is that the credit rating agencies, in giving a rating to different structured types of instruments, sometimes present a backdoor to allow our investment managers to invest it if we are not careful. It was alluded to earlier that our members give investment management agreements and use credit rating agencies, and the ratings given on bonds, to describe the portfolios-the riskiness of the portfolios that our pension funds want to have. So if they want a very safe portfolio, they will say, "We want triple A-rated paper". They are willing to be much more exposed to credit risk. They might go down to triple B paper; that is entirely fine, they expect to make a greater return.

What is difficult for them is if a different structure appears in the market, which then gets rated as double A, and if they have many managers and the manager says, "My investment management agreement allows me to invest in this structured product", and does not come back to the client and make clear what they are doing. Therefore, what we would say is that we are quite alive to new types of debt structures being created and we are wary about them getting credit ratings, which might allow them to access our members’ funds where there has not been an explicit agreement between the investment manager and the client to allow that type of new structure to be invested in.

Q12 Stewart Hosie: Mr Harrison, that is very interesting. There seems to be a huge reliance on the rating itself. Is there a way of avoiding or minimising the use of ratings by investors, by pension funds? Are there alternatives to the formal scoring of the big three credit rating agencies?

Mark Hyde Harrison: I suppose what I would want to make clear is that the ratings are used to describe the portfolio. You select the investment manager, based on their ability to do credit analysis on individual debt issues and their ability to find bonds that they think they are going to make a suitable return on and avoid defaults. So you are not relying on the credit rating agency to do the credit analysis; you are relying on them to describe a type of riskiness of asset. You would hope that double A bonds, as a whole, generally have a certain risk return profile and, as was alluded to earlier, that the credit rating agencies’ credit ratings-double A bonds in aggregate-have tended to follow those risk default parameters.

Q13 Stewart Hosie: So you would not see a reduction in the use of the credit rating in the way you have just described it?

Mark Hyde Harrison: It is a way to describe the riskiness of a portfolio, to give the mandate to someone else; otherwise, it is quite difficult to describe how risky you wish the portfolio of corporate debt or sovereign debt to be, unless you have some third-party way of describing risk.

Q14 Stewart Hosie: Can I ask you, Mr Grout, if there was a move to reduce reliance on ratings, would that have an impact on the ability of companies who were seeking to raise finance through debt issuance rather than, say, from the banks, or would the people who were buying that debt still look at the same sort of credit rating that they are currently looking at?

John Grout: The experience we saw prior to credit ratings becoming prevalent in investments was that for companies like National Grid, companies that are large and therefore a significant part of the market, companies that are household names-the Tescos of this world-it does not make a lot of difference. The smaller the company is, perhaps the worse the credit rating would be, and the more significant the study of the individual company becomes. It is more of a store-y credit. Smaller investors of any kind have fewer resources to do that kind of analysis and the credit rating of a company, that type of store-y credit, is a very good first indicator of, "Am I interested in this or not?" So it can be used for screening, and we would see a lot of companies that use it for that purpose. They use it as a starting point and then they read around: are they comfortable? A change in the rating, or a change in an outlook, is going to trigger to them a review of what they have done. Companies, in the sorts of investments they make-we are talking about cash investments, effectively-are very conservative. If an industrial company loses money on an investment of that kind, the shareholders are not going to be very sympathetic; the chief executive is not going to be very sympathetic to the treasurer, so they are out as soon as it gets beyond their comfort range.

Q15 Stewart Hosie: So, in the sense of those scores being used even to screen, these big agencies have the whip hand in that sense. They might actually stop someone even looking into a company to see if it was viable and profitable because it was below a certain score?

John Grout: Yes. The gathering of information about unrated companies-if you look at mid-sized companies in the UK, that would be the huge majority-is quite expensive, and if you look at a big investor, their cost of discovery in deciding this is a smaller unrated company that you even might want to invest in, is high.

Q16 Stewart Hosie: It is helpful to hear that. Mr Harrison, can I just come back to you on a slightly different question? Companies with pension funds and pension funds themselves will pay into the Pension Protection Fund. The amount a company or a fund or trustees pay is dependent, in large measure, on the score from a credit rating agency. Bear with me. The data at the beginning is clear, the score at the end is clear but as I am sure you know, the bit in the middle, which turns the data into the score, is obscure to the nth degree. Is this an issue that concerns you, and is there something that could be done to open up the calculation-the black box-to show those who are paying the money how the raw data on the fund, or the company, becomes the score that then determines the amount they pay?

Mark Hyde Harrison: The Pension Protection Fund uses Dun & Bradstreet to provide a default score on companies for the levy. Since it started, it is been much more transparent in how that score is calculated. Inevitably, it covers all companies that have pension funds in the UK, not only those who issue bonds, which are then rated themselves, and quite often there is a disparity between the ratings that credit rating agencies give to debt issuance and the Dun & Bradstreet score, which is given for the Pension Protection Fund. Because it is being used for all companies, they are being much more sensitive to things like county court judgments or the number of directorships held, which would not be such important factors in a credit rating agency looking at a major bond issue. I think they are very different things and they are not quite the same.

Q17 Stewart Hosie: But in general terms, the means of turning the data into the score, is that too opaque, whether it is for pension protection or more generally, or is it right that it should be a black box because it is the company secret?

Mark Hyde Harrison: It is not quite a black box. I think you can understand the factors that have gone into the score, and then there is an appeal process with Dun & Bradstreet, which has now been put in place, to argue the case so you can get better judgment where the mechanical nature of the score production, which has happened in Dun & Bradstreet, can be amended through an appeals process. So it is not quite as black box as I think you are implying for the PPF levy.

Stewart Hosie: I may come back to this.

Q18 Andrea Leadsom: I would like to talk about conflicts of interest, potential conflicts. Mr Cooper, could you tell us what is National Grid’s rating right now?

Malcolm Cooper: We have a variety of ratings. We have around 15 rated entities in the group. They range from triple B-plus to the highest rating we have, which is A2, or A flat.

Q19 Andrea Leadsom: How do you decide which rating agencies to commission to undertake a rating?

Malcolm Cooper: For many years we have used S&P and Moody’s. About 10 years ago, we added Fitch to the list of agencies we used. Our very strong view is that having more than one agency gives investors more comfort in the quality of the analysis. The reason we added Fitch is that at that stage we had a very wide disparity between ratings. We had a four-notch differential-the notches are the levels of credit rating. There was a four-notch disparity between the two agencies, and we felt that was probably disadvantaging our cost of borrowing, so we brought Fitch in and asked them to rate us as well to get a fairer assessment of what the ratings were. It was many years ago; the company has changed since then and the ratings are now broadly in line across all three agencies. In addition to those three, we use two other agencies. We use an agency called A M Best who rate our insurance company. We have two captive insurance companies, and A M Best are actually better known for rating insurance companies than either Moody’s or S&P, so we have dropped Moody’s and S&P from our captives and put A M Best in there. In addition, we use an agency called JCR, the Japan Credit Rating Agency. If you wish to issue debts directly into the Japanese market in straight yen form rather than euro-yen form, you really need to get Japanese investors on board. You need a JCR rating as well, so we have five agencies in total.

Q20 Andrea Leadsom: When you were using S&P and Moody’s and you had a four-notch differential between the two ratings, where was your cost of funds coming out; at the lower of the two or at the higher of the two?

Malcolm Cooper: Nearer the higher than the lower.

Q21 Andrea Leadsom: So you were given the benefit of the doubt by the-

Malcolm Cooper: No. I am sorry; I said that the wrong way round. Nearer the lower rating, so I apologise.

Q22 Andrea Leadsom: Yes. So you had a higher cost of funds?

Malcolm Cooper: We had a higher cost of funds, yes.

Q23 Andrea Leadsom: Yes, that is what I would have expected. So really, the question is do you have any idea of how much you pay them to rate you?

Malcolm Cooper: Yes. Yes, I do. I mean, we pay-

Andrea Leadsom: Do you know it offhand?

Malcolm Cooper: It is several million pounds a year.

Q24 Andrea Leadsom: Is there is big differential between the costs of one rating agency versus another one; say, JCR versus S&P? I imagine there would be.

Malcolm Cooper: JCR would be a lot cheaper, mainly because they only rate a single entity. We only issued one debt from one entity into the Japanese markets. So JCR and A M Best are much cheaper. Moody’s, S&P and Fitch are broadly the same, although we have had an indication that one of the agencies is likely to put their fees up considerably this year.

Q25 Andrea Leadsom: Would you be in a position to tell us, for example, if S&P and Moody’s are messing about with a huge differential between where they rate you and you ask Fitch to come in, does Fitch come in on the promise of a significant six figure sum per annum-a new client-on the expectation that they are going to sort that problem out for you?

Malcolm Cooper: Absolutely not. I don’t believe any rating agency would come in and I don’t believe you could pay for the rating you wanted.

Q26 Andrea Leadsom: Obviously, that is what I am getting at. There is that risk in the market, isn’t there, and, human beings being human, it is highly likely that other less honourable companies would say, "We’ll pay you to rate us, so long as it comes out closer to double A than A".

Malcolm Cooper: I think a rating agency would refuse to rate you if you had that conversation to begin with. They also have quite a wide process in place to make sure that the ratings are done accurately. You typically have an analyst that will analyse the company and a back-up analyst that reviews that data in addition. There is a credit lead and, in addition, there is a committee of five people-usually five-that review the rating. So you would need to have eight people who said, "The fee of £500,000", or whatever the number is, "is sufficiently large that I am going to give them a rating far better than they deserve". I really do not think that would happen.

Q27 Andrea Leadsom: Why would that be the structure? Why is it that the issuer pays rather than the investor pays?

Malcolm Cooper: My personal view is that the investor pays would not work. First of all, you would have a number of investors that would say, "I don’t need credit ratings", so the big investors, Legal and General, may well say, "We are capable of doing it ourselves and we are not going to use it". Smaller investors would then have to bear the wider costs. Credit ratings are typically public and, in fact, John reminded me there is a requirement in the US for ratings to be public. In addition, ratings are not used just by investors; they are used very widely. As a corporate, I use credit ratings everyday for various business decisions. As John outlined earlier, when we are making investment decisions-where do I manage my spare cash-rating is an important category.

Q28 Andrea Leadsom: Yes. If I can just ask Mr Grodzki to comment on that, because I think that is a very important point: actually what happens is that once there is a rating out there, there is instantly that group-think, and notwithstanding that Legal and General may have their own credit teams doing their own extra analysis, nevertheless, if National Grid is rated triple B-plus, your analysts are unlikely to come out and say, "No, as far as we are concerned, they are triple A, so we need to have 20% of our portfolio in National Grid". That is very unlikely to happen, isn’t it? So there is that leaning towards the credit rating.

Georg Grodzki: There is. None the less, it was recently the case that we very strongly disagreed with a public rating and we refused to buy the bond, and within reason, it happens. A very crass difference in opinion is rare. It could happen as well, but it is rare. Let me add, until the mid-’70s rating agencies were only paid by the investor. For the first 40 years of their existence, investors paid for the rating agencies, and only when the rating agencies became powerful enough to ask the issuer to pay-and they were powerful enough in the 1970s-did they ask the issuer to pay; because they were basically the gatekeeper. You could not access the market in the US without paying the rating, so they monetised the clout that they had over the market. They were the regulator of the market basically; it was the entry fee, which the issuers were happy to pay because it resulted in market access, lower funding costs-as we have just heard-and the business model changed.

You could imagine a world without issuer fees-you definitely could imagine that. The rating agencies would have to reduce their staff levels. They probably would have to reduce their services and would shrink to the core of their business, which is providing rating opinions with the analysis that is necessary. At the moment they are doing a lot of things that are useful but definitely not necessary, strictly speaking, to assign a rating opinion. The alternative model would work, and you could imagine a world without ratings as well. We had this discussion: what proxies would the market look at in order to gauge, at least in some form, the riskiness? You could look at the size of companies. Definitely there is a high correlation between default risk and size; size measured by turnover, balance sheet, market cap-not without flaws but definitely with some merit. You could look at credit scores. There are some models out there that have some validity, fact-tested. You could look at the market’s opinion, which is the credit spread, the aforementioned. If credit markets are liquid enough, the credit spread gives you a good indication of the market’s perception, which can be wrong-there is a subjective element-in the same way as the credit rating can be wrong because that is a highly subjective element, but you do not need to pay Moody’s to get a subjective opinion of the credit quality. You could just look at the credit spread; even though it is not pure, it tells you something about the market’s perception of the credit. The market would still work in a slightly different way, and I would not dare to guess whether it would be more efficient or less efficient.

Certainly, in the last couple of years, there are places in Europe where bond issuance emerged without ratings, and obviously many institutional investors, such as ourselves, are not in a position to participate because our mandates restrict us from buying non-rated securities, but other investors, especially retail investors, were able to buy and did buy in big numbers. Whether all this will end in tears in a few years’ time, when some of these issues default, and then the blame game will start, we will see, but a world without ratings is perfectly conceivable and is happening in parts of the world.

Q29 Andrea Leadsom: There is one other question I want to ask. Mr Grout, Fitch, Standard & Poor’s, Moody’s, they would all potentially be clients of yours, as the ACT, would they? They are all treasurers and so they would have companies, so they would have corporate treasuries.

John Grout: No. They are French or American, so they would not come under this necessarily.

Q30 Andrea Leadsom: They would not come under it-okay. Well, my question is this. Are there not issues, then, of competition? You said yourself, Mr Grodzki, that they effectively are the gatekeeper, and you are constrained because if something does not have a rating you simply cannot touch it now. So in effect, they have created a market and they are making out like kings and obviously everybody is hostage to their accuracy. Specifically, of course, with the asset securitised loans, blatantly they proved not to be triple A rated in spades and yet the whole world had bought them because they were rated triple A. Is this not a large risk? I don’t know whether any one of you has discussed this very concern and come up with conclusions on how to improve the way that ratings are managed.

John Grout: If I can pick up the start of that question? The situation is that the rating agencies add a lot of convenience.

Andrea Leadsom: Of course.

John Grout: There are other people who will publish credit opinions. There are some very good firms that do this. They do not rate the full universe that the others do because they only rate people they think their potential customers will be interested in, so there is a sort of self-fulfilling prophecy. But there are other people that provide credit ratings. These are unsolicited by the issuer and, therefore, there is no transfer of information directly from the issuer to the credit analysts, which, in theory, should mean they are not so good. That is the first thing, and it is easy to envisage a world in which there are purely public information ratings, particularly as we get on with more computer system based analysis.

The thing that differs with a credit rating agency-not Dun & Bradstreet, of course, which is not a credit rating agency but a credit scorer-and the credit rating agencies as we normally talk about them, is that the credit rating analysts there are interested in more than just the historical numbers that have been published. They are interested in policy strategy and, particularly nowadays, risk management within the company. That is the non-public bit. There may be other things, but that is the non-public bit, which I think is important to the rating agencies. It is not what this year’s profit is going to be, which is perhaps what it might be assumed to be, but that is much, much less relevant to a credit analyst than it is to an equity analyst. The kind of conversation that goes on would normally drive an equity analyst potty because it is about how you deal with long-term risk and that sort of thing.

Going back to the usefulness of credit rating agencies, in the 1970s-before there were rating agencies widely used in the United States, and they were not used at all over here-I was responsible for a pot of several hundred million pounds of a large British company, which was difficult to invest. There were no credit ratings. You struggled to get a sensible set of criteria. There were not even the kind of credit analysts that exist today that you could subscribe to. The rating agencies are a huge convenience and for smaller companies than the one I was working for, that convenience is not substitutable by saying, "Employ more people or pay large fees to other people." If you are a medium-sized company here in the UK, it is a very useful thing to do.

Malcolm Cooper: May I just go back to the issuer pays point for one second? There is one very important point that John has just touched on. With an issuer pays model, I have a contract with a rating agency. That contract includes a confidentiality agreement and, under the terms of that CA, I am willing to give them very confidential, forward-looking information that would never go to any agency if I didn’t pay them. So if you had an investor pays model, there is no way I would give that information to an agency I did not have a confidentiality contract with, so the quality of information available to the rating agencies would be significantly lower and, in my view, would make the quality of the end product significantly inferior.

Q31 Chair: Are you required to supply all information?

Malcolm Cooper: You can clearly refuse to give them information but they would mark you down for that.

Q32 Chair: How would they find out about it?

Malcolm Cooper: If you withheld information that was material and they were unaware could be there, they would hold it against you if they found out afterwards. For example, very simply-

Q33 Chair: When the auditor comes by, and he is demanding this, he is backed by various forms of corporate governance and, in some cases, statute, and you have to provide it. When the rating agency comes by, what is the obligation; you have to give them a favourable picture?

Malcolm Cooper: The obligation is that we will be open and transparent with them, but probably, more importantly, it is very forward-looking information. The kind of thing that a rating agency would be very interested in is if you are about to go out and make an acquisition for debt and that would clearly damage your credit profile, if you met the agencies, did not tell them you were going to do it, and then did it, the next time they come to have a meeting with you, they would clearly count that against you.

Q34 Chair: But we are all human beings, so there must be huge optimism bias in the information passed to the rating agencies in these circumstances?

Malcolm Cooper: There may be optimism bias in there, but their job is to assess the level of that optimism bias, so they will sit in a room with a chief executive and a CFO and push them quite hard on the detail you have provided. The optimism bias, for a company like National Grid, is possibly slightly easier, but if you were something like a farmer or in consumables, your sales projections could be quite severely mis-stated, but the agency would need to do sensitives off that. The key thing they are really interested in is whether you are committed to your current credit profile and are you going to do anything that is going to materially damage your credit profile. That is a question of looking the chief exec, the CFO and the treasurer in the eye and actually asking that question, and building up trust in the relationship with them so that they will believe the information you are handing over.

Q35 Chair: You are not filling me with great optimism. Can I ask, Mr Grodzki, if there were no rating agencies at all, would you expect more volatility in credit spreads?

Georg Grodzki: Not necessarily. We should not forget, compared to the 1970s, there is a lot more transparency about corporate financial affairs around. It is much easier these days to analyse a company without the privileged access to proprietary information that the agencies sometimes dwell on as their competitive advantage. The reporting patterns are shorter and the level of disclosure is better than it ever was. Therefore, the market is less disadvantaged these days, relative to the agencies’ privilege, than it used to be 30 years ago. While small investors probably would still struggle more than others, they are definitely not completely in the dark, provided the company is meeting its disclosure requirements, which normally it has to do in order to be stock market listed, in order to issue bonds. Therefore, with the level and timeliness of availability of information we have at the moment, I think the rating agencies have far less extra value to add than they used to in times when corporates were opaque-much more opaque than they are these days. That would mean that the net effect of a withdrawal of ratings would be far less pronounced. The liquidity could be different. The liquidity levels, i.e. the ability to trade in and out of bonds because there are market makers out there, would probably be somewhat affected, because liquidity is usually provided by market makers, banks in particular, and they have to hold capital to underpin the trading positions and again that capital is tied to the credit quality of the bond score.

Q36 Chair: Yes, but that goes back to the regulator.

Georg Grodzki: Absolutely correct.

Q37 Chair: So it is the regulator who is creating the need for the rating agencies?

Georg Grodzki: Absolutely correct.

Chair: Right, okay. We are getting nearer the heart of it now. Andrea, you wanted to come back.

Q38 Andrea Leadsom: Yes. It was just on a very brief point, again, Mr Grodzki, because you are saying things have changed dramatically. Twenty-five years ago, I was selling synthetic floating rate notes to investors, and they were literally buying any old thing so long as it had an interest rate swap with Barclays’ name on it and it was triple A rated. That is in complete contrast to what you say-that investors do their analysis. Would you say that has changed and I am simply reflecting a very long time ago, or is that still a risk?

Georg Grodzki: Some investors still go by name recognition and-

Q39 Andrea Leadsom: No. I am saying rating recognition. In other words, they care that the interest-rate swap counterparty had a triple A rating and they care that the bond-whatever it is, the company name that they have never heard of-has a triple A rating, but they are happy to buy that package regardless of the analysis that they have not done of the company, on the grounds that the package is triple A rated.

Georg Grodzki: As Mr Grout was saying, the fact that some investors for convenience sake just make do with the rating and do not look any further, does not mean that they could not look further and develop their own opinion. They just decide that is good enough for them, so they still have that naive, or not naive, trust in the ratings. It saves them time and good luck to them. Some did that with Lehman Securities. Therefore, that does not contradict the point that there is now a wealth of information out there to analyse Barclays. Barclays’ interim report is now well over 100 pages and, on the point about management meetings, we do meet the CEOs and CFOs of the companies we invest in as well. We make a big point of that.

We are not allowed-that is a difference from the rating agencies-to become privy to non-public information, but we do have very lively conversations with them about their strategies and about their risk appetite and about their contingency plans. Therefore, big investors-I admit it is not available to everybody-do have access to boardrooms and are able to enrich, so they are investment decisions with those insights, even though, again, they are all within the confines of public information.

So there we are. The world has moved on since ratings were invented in the 1920s and, therefore, the ability of investors to source information on the internet, to look at a wide variety of opinion providers-not always called rating agencies: credit research firms, independent credit research firms and equity research-is massively improved. The timeliness of the information has improved. Again, the absence of insight into corporate plans can be a disadvantage but the agencies obviously did not make much use of their insight into corporate projections when they were rating Northern Rock or Lehman. So it did not help them either, because corporate projections can miss the point by a wide margin.

Chair: Lots of people want to chip in. Mr Grout wants to say something and then I am going to bring in John Mann.

John Grout: We started a little while back talking about whether the rating agencies may be influenced by the fees they earn, from corporate ratings in that case. I think it is very important that there is no corporation-even one the size of National Grid-as an issuer; fees being related to the amount in issuance is a material part of a credit rating agency’s income. Where it is different is if you look at structured credit ratings. Rating agency business is slow growth. Structured credit provided a growth opportunity. This growth opportunity was controlled by a handful of large banks. Therefore, it is very clear that in that market individual customers or individual decision-makers, because the nominal customer was the sponsoring bank for the decision-maker, had enormous influence over that section of the credit rating agency’s revenue. It is very important to distinguish corporate ratings certainly from those structured credit ratings.

Q40 John Mann: I just want to clarify one point first with you, Mr Cooper. When you are planning an issuance do you ever make changes in order to achieve a better rating?

Malcolm Cooper: In vanilla debt, you typically don’t. If you have a company that is issuing just standard bonds then that company will have an issuer rating, and that will determine the rating for that debt. If you are looking at structured finance then, yes, you can do. You can change the rating by changing the terms of those bonds, typically by either giving it greater security by moving it closer to the assets, or giving it first right of claim over the assets, or by moving it away. You can subordinate it to other debt and then reduce the rating. So, yes, you can change the rating structurally by changing the terms of the bond.

Q41 John Mann: I want to know whether changes in ratings lead market opinion or follow market opinion.

John Grout: Rating agencies are lagging indicators. Anybody that is surprised by a change in credit rating, unless there is something exceptional, just has not been paying attention. The rating agencies have addressed that a little bit with the greater emphasis on outlooks. If an event takes place, they might announce something is on credit watch with positive or negative implications, or the equivalent terminology of other agencies. But the use of outlooks is an indication that-maybe not because of an event but because of events in general and what they know about a company or the issuer-they think there will be a change within so many months, or there may be a change; not that there will be.

It is unusual for a rating to change without there having been some signalling in advance by the rating agency. The putting of an issuer on an outlook of that kind-an outlook with whichever implication-is what people try to follow, rather than the rating itself.

Q42 John Mann: If there is a lag and we hit a major financial crisis, then it follows automatically that those who are relying solely or primarily on the rating agencies are going to get their fingers burnt more than others.

Georg Grodzki: One could say, though, that the agencies rate companies on the basis of the most likely scenario, which does not always materialise. So it is not evidence of flawed judgment or a flawed system that, from time to time, ratings need to be adjusted because the actual development is much worse than the issue with the highest probability development. Therefore, rating agencies never claim to be perfect; what is important is that they spell out the assumptions in terms of growth, for example economic growth, what their ratings are based upon, so that one understands their ratings better.

On the subject of whether the rating is always lagging the market, it tends to be more often than not. None the less, when ratings change they can still move the market, which would contradict on paper what we have just been saying. But it does not necessarily contradict because rating agencies sometimes decide to disagree with the market and to assign rating actions that are in contrast to the market consensus.

Sometimes the market ignores them. Sometimes the market pays attention. That is good. Rating agencies should not just echo, reflect or duplicate what the market consensus is, and I personally would be worried-I have been worried in many instances-if rating agencies are simply saying back to the market what the market is saying to them. That used not to be the case but it has been the case in a number of instances, because the agencies have become more self-aware and more concerned about being too far away from the market consensus, but that is dangerous because we need plurality, a diversity of opinions.

Q43 John Mann: To whom and how should the credit rating agencies be accountable?

Georg Grodzki: The investor. Moody’s was once called Moody’s Investor Service. Sadly, they dropped the "Investor Service" bit, but that is where it is coming from because they used to make their money when they were founded in the 1920s.

Q44 John Mann: That is the who; but how?

Georg Grodzki: In terms of the investors-

John Mann: How should they be accountable?

Mark Hyde Harrison: If an investor lost confidence in a credit rating agency, they would no longer say that they wished to use that credit rating agency’s ratings for those bonds in any way. The issuers of debt would find that they did not get access to the buyers of their bonds if they solely relied on that rating. Trust with the investor is absolutely the core of their business model.

Q45 John Mann: We have heard that there are not many of them. We have heard from you that they are rather similar. You have indicated that. Therefore, how can they then be accountable to the investor if they get it wrong?

Mark Hyde Harrison: I suppose an analogy I would give is that I view credit rating agencies a bit like a listing authority for equities. They rate a bond and, therefore, make it available to an investing world. If they decide to change a credit rating it is a bit like dropping out of the FTSE 100 index. Everyone could see a company was getting smaller, but when it drops out people trade on that equity because it is in or out of an index. If it drops a credit rating people will trade it because they no longer want to hold triple B debt; they want to hold double A debt or whatever it might be.

They serve a very important purpose in defining a market and there is a question of whether they have the right regulatory controls, like a listing authority, in terms of making sure who has access to the market. If you lose that trust and you list everything, or give credit ratings to everything, your investors will no longer invest in that market. If the London Stock Exchange or the UK Listing Authority allowed too many equities on that people did not trust-if they said, "These aren’t proper companies"-people would stop investing in that market.

Q46 John Mann: That does not answer the question of how they can be held accountable. Let us just take the example of the Icelandic banks, where there was a lag with the credit rating agencies and people continued investing and got their fingers burnt-local authorities are an example. There is no way that I can see currently that those local authorities can hold, in any way, any of those credit rating agencies accountable, even on the margins. They are, are they not, unaccountable?

John Grout: Can I just chip in there with the observation that I don’t think that any significant corporate losses were suffered on that Icelandic front, and this comes down to what a credit rating is. If it was true that the credit rating of some of those Icelandic banks was still A but that was for their obligations in Icelandic króna, if you are a UK investor you should not be looking at that rating. You should be looking at their foreign obligation rating, and you should be looking at the discussion in the rating agency reports of the willingness and ability of the Government, in the jurisdiction in which they are incorporated, to support those banks. In Standard & Poor’s case that is a BICRA rating. There are other terminologies in other rating agencies and they publish in different ways and at different times. But if you were keeping up, that is to say you were reading all those reports and not relying on the headline rating, you may not have been in the position of suffering those losses.

The trouble with the headline rating is that it is plain, simple and easy. It appears to mean the same thing between different rating agencies, but actually it means different things. For a weak credit, an S&P rating is a default rating and they publish a separate recovery-given default rating. Moody’s notches their default rating to produce a combined rating of the two. This temptation to just follow the letter ratings, rather than read the reports and think what they mean, and not to read the reports on the relevant sovereign, which is very important-sovereign ratings are very important when it comes to rating banks and corporates-that is where you get the problem.

Q47 Jesse Norman: Two very quick questions. For you, Mr Cooper, how much is a notch worth in your cost of borrowing?

Malcolm Cooper: Ten or 15 basis points; so 0.1 to 0.15%.

Q48 Jesse Norman: What is your current outstanding debt?

Malcolm Cooper: About £25 billion. It is a huge number.

Q49 Jesse Norman: So the incentive to get the right rating from an agency, potentially by playing them off each other, is enormous in terms of lowering your cost of borrowing. You do not have to respond. That is just an observation.

Chair: Please do not contradict him, Mr Cooper.

Jesse Norman: No, no, I am so sorry, Mr Cooper, if I have said something that is untrue, absolutely contradict me but, otherwise, I am-

Malcolm Cooper: I don’t think it is possible to play agencies off against each other. What you have to do is make sure the rating agency fully understands your business and gives you an accurate rating reflecting your business.

Q50 Jesse Norman: Under the remote hypothetical that it was possible to do that there would be every economic incentive. That is all I wanted to get to.

Mr Grout, why should banks be able to have their issues rated at all, given that there is every incentive to stuff them with-goodness knows what-junior subordinated rubbish, and pretend the portfolio effect will bail you out with a triple A rating?

John Grout: Why should banks be rated?

Jesse Norman: No. Why should it be possible for banks to go to rating agencies and have their new issues rated?

John Grout: New issues of their own obligations?

Jesse Norman: Or of obligations that they are packaging up, of CDOs or whatever it might be. Should that not be stopped?

John Grout: I have no idea of the answer to that question. The social welfare benefits of rating packaged securities is not my field, but I would say that if you-

Q51 Jesse Norman: But there is every incentive to try to dupe the rating agency by packaging securities.

John Grout: Certainly, and the rating agencies need great skill in analysing that. It is very important, particularly for smaller companies-we will see that more as private placements become more significant in the UK as banks withdraw from industrial finance-that the packaging of securities is allowed and is supported by necessary infrastructure, otherwise companies will not be financed.

Q52 Mark Garnier: Can I quickly turn to competition? Perhaps if I can go through two questions with each one of you in turn, to be clear. First of all, to start with you, Mr Grodzki. How many credit rating agencies does Legal and General use?

Georg Grodzki: Three.

Q53 Mark Garnier: You pay for the detailed report from them?

Georg Grodzki: We pay all of them.

Q54 Mark Garnier: Which ones are they?

Georg Grodzki: Moody’s, S&P and Fitch.

Q55 Mark Garnier: Okay. You are what I would refer to as a consumer? You use them to assist your investment decision in terms of investing on behalf of your-

Georg Grodzki: Absolutely. Let me say it is not only the rating opinion, which we would probably be able to find out without having to pay the rating agencies because it is public information these days, but it is also the underlying literature that they produce and the analysis of the companies, which is helpful, too. It is the whole package.

Q56 Mark Garnier: Which is very useful, but the key point is, from your point of view, that you are looking at this as somebody who looks to them to enrich your decision-making process as opposed to being-

Georg Grodzki: Absolutely, stimulate our thinking. That is right.

Q57 Mark Garnier: Absolutely. John Grout, you speak on behalf of an association, so perhaps it is a little more difficult for you to answer, but if you were looking at your members, generally speaking, what agencies would they use?

John Grout: They would tend to use Standard & Poor’s, Moody’s and Fitch in that order, unless they are in a particular industry where there are specialised rating agencies that have a more important position, or if they are going into other geographies. Those are the ones with the biggest market share.

Q58 Mark Garnier: Unless it is a specific thing, okay. Again, you look at it in terms of an investor? You are talking about investors?

John Grout: As an investor, yes. Companies will themselves often subscribe to, for example, insurance company credit raters. Because companies have big liability insurance, they want to know it is going to be there.

Q59 Mark Garnier: Sure. Malcolm Cooper, you have already sort of answered this question, but just to confirm, you are an issuer so obviously you pay several million pounds, you said, to S&P, Moody’s, Fitch, A M Best for your insurance company rating and JCL for your one Japanese yen bond issue thing.

Malcolm Cooper: That is correct, yes.

Q60 Mark Garnier: Mark Hyde Harrison, you are from the National Association of Pension Funds. Again, can you give us a flavour of what type of-

Mark Hyde Harrison: If our members were investing in rated debt they would generally describe that to investment managers using either S&P, Moody’s or Fitch to describe the portfolios, but they may give mandates to investment managers to invest in unrated debt as well, so bank loans and things like that.

Q61 Mark Garnier: Would your members subscribe to the rating agencies? Would they write a cheque out to get the full detailed reports or would they tend to use what is published?

Mark Hyde Harrison: No, they would not subscribe to a rating agency as they would normally employ investment managers to do the credit analysis on their behalf.

Q62 Mark Garnier: Again, typically, when you are looking at your members, would they tend to use the big three like everybody else?

Mark Hyde Harrison: I would say they do not use the big three, except to define the market in which they invest. They use investment managers to do the detailed credit analysis and do not rely on the credit rating agencies generally to determine what they buy.

Q63 Mark Garnier: What I am trying to get to the bottom of is the competition in this, because the numbers tell us that 95% of people who use credit rating agencies use the big three. I think there are 15 that are registered in Europe and then there are a number in Asia, the Middle East and the US. Yet it seems the vast majority of people tend to stick with just the big three. Is this apparent lack of competition bad for the market? Just throw in your answers as you have a thought.

Georg Grodzki: For us, the rating agencies are helpful-not necessary, but very helpful and useful-because their opinions certainly stimulate our thinking and so does their analysis. They would be less helpful in that respect if we knew they were financially weak because there were very many of them and they had to fight for business, which would compromise their ability to assign lower ratings, which the company would not pay for, or to downgrade companies, taking the risk that they lose the rating.

Q64 Mark Garnier: That is an incredibly important point. I am sorry to cut across you, but we have heard other people asking about it: can an issuer have a conversation with a credit rating agency and suggest perhaps that they can get a better rating? Malcolm Cooper, I think you said, "No, under no circumstances would that happen", and yet you are directly contradicting that with the smaller end of the marketplace.

Georg Grodzki: There have been suggestions in the past with certain rating agencies-there used to be two, there are now three-who were trying to win business, and it is a balancing act for them because companies would only pay them if, in return, they get better market access and possibly lower cost of funds. But if these agencies are not known yet, and are not respected yet by the investor, issuers would not be able to reap that benefit from it-a third or a fourth rating.

It is a tricky situation. One could say that when Fitch arrived in the European marketplace clearly the agencies became much more aware of market share, of issuer relations and of the impact on the market. I do not have evidence that a single rating was compromised, but clearly there was heightened awareness about the risk of maybe losing your second spot if a company only wanted two ratings and was contemplating replacing the second rating by another rating. So competition in laxity is dangerous and would certainly make rating agencies less useful for us, if we had the impression that they were assigning friendly ratings in order to gain market share and in order to improve their own financial performance.

Q65 Mark Garnier: So it is a real problem. In order to get credibility with the investor in the marketplace you need to be a very tough rater, but in order to get credibility with the issuers you need to be a very soft rater.

Georg Grodzki: You can only be tough if you are strong-if you have financial resources that allow you to lose business, because your ratings are not nice enough but you still prevail.

Q66 Mark Garnier: John Grout, you were the one shaking your head.

John Grout: I shook my head because there is an assumption that companies want a higher rating. I used to be Director of Treasury at Cadbury Schweppes-now departed-and when I was there we were a strong A and we were a fairly strong company. We had discussed with the rating agencies our long-term plans, which meant two very large acquisitions that would have meant more leverage and would take us down to triple B.

To go from anything above single A to triple B might frighten the horses, and they invented a new rating agency called the European Credit Rating Agency. They were doing unsolicited ratings and they published a double A rating of us, which is perhaps what we might have been if you only looked at the numbers and did not have that strategic access, which you would not have, of course. I spent a lot of time over several years persuading them that, while we might look okay, we were not that okay, and eventually persuaded them to drop us down to A. It is not to be assumed that companies wanted a higher rating.

Q67 Mark Garnier: Why would you do that?

John Grout: Because the population of people that invest in your company, whether that is in its securities or its shares, is determined by the risk profile they think you have. You can change the risk profile of the company dramatically. If Cadbury Schweppes had announced a division on nuclear engineering that might have caused people to change their view of the risks. You would expect to see a change in your investor base from funds like your existing profile to funds like the profile you were going to, and you would have upset considerably-particularly, if you go to a type of risk that is less welcome-your existing investor base, and you want to do future issues and you do not want to upset that sort of investor base. So if you are a strong A that knows it is going to be going downwards but cannot tell anybody, you do not want people raising you higher, because to be downgraded by not just notches but whole rating steps really is a significant thing for investors.

Q68 Mark Garnier: Were your other rating agencies getting it right and it was just this European Credit-

John Grout: No, the other rating agencies weren’t getting it right because they were being clever. They were getting it right because we were telling them what our plans were.

Q69 Mark Garnier: We come back to Mark Hyde Harrison’s point, which is that it is market confidence in the rating agency that is key to this whole point. If you had the other three presenting what was ostensibly the right rating, and this one agency that is out there, which presumably has to publish the fact that it does not have the inside track, I suppose-

John Grout: It didn’t in those days.

Q70 Mark Garnier: It didn’t have to in those days. So it was pretending it was as good as the others. What has happened to it now? I should know this.

John Grout: It is part of Fitch.

Mark Garnier: It is part of Fitch. They took it over. Thank you very much.

Q71 Teresa Pearce: I am quite interested in what has been said about how times have moved and, historically, rating agencies were much more important than they are now. I think it was you, Mr Grout, who said, "Anybody who is surprised by a downgrade obviously hasn’t been paying attention". People do have access to information now. If we were starting now would you invent credit agencies or are they superfluous?

Georg Grodzki: You probably would not invent them because it takes 20 or 30 years to make that business model profitable, if you are lucky. You would have to have very modest return expectations and you would have to be almost feeling like a missionary. You would not do it for the money. You would just do it for the good of the nation.

Q72 Teresa Pearce: That brings me back to the point about competition. New players to the field, it will be very difficult for them because they do not have the reputation to compete. Do you think if we had mandatory rotation it would increase the ability for new players to come to the field?

Georg Grodzki: First of all, is it desirable that a lot of new players come in? We have a number of non-rating agency credit opinion providers who make a living because they have developed a proprietary scoring model using corporate financial data, and they can show that if you had used that model over the past 20 years you would have done well. Obviously that does not necessarily mean it is going to work going forward, but for some clients that is good enough. Therefore, there is already quite a microcosm of information and opinion providers out there, which are not regulated or not recognised in the same way as the established rating agencies, but that are being paid attention to by the market. Therefore, to the extent they gain traction, they may start rivalling them.

Coming back to whether we would invent them, there is already a wealth of information out there. There is already a wealth of opinion out there. We should not forget that nowadays there are research departments in banks, reasonably well supervised in terms of not just talking the book of banks, which provide investors with their opinion. You don’t have to look very far as an investor to find somebody you could talk to about the prospects of a certain credit, which, again, in Europe was not the case 20 years ago. Banks did not have those departments. In a way the world of bonds has become a lot more like the world of equities, where you have brokers providing information, providing opinions and recommendations. Again, many of them are often wrong but at least they give you information and they stimulate thinking.

The agencies are, therefore, not in the same solitary, almost exclusive, position as they used to be 30 years ago when they were the only port of call for anybody who wanted to have some information on the credit prospects of a certain bond issuer, because you could not find it from the banks and there was no such thing as an independent credit research group you could have turned to. Therefore, the current set-up-let me stress this-works pretty well in the industrial corporate bond world.

The agencies clearly have made mistakes in structured securities. You can also question their record very much in the banking world. But as far as industrial, i.e. non-financial, corporate bonds are concerned, I do not recall such a high incidence of blunders or failures beyond the normal errors of judgment, which sadly do still happen, over the last 10 or 15 years in Europe and in the US, that I would complain and call for a massive overhaul of the system or much tighter regulation than we have so far.

Chair: Teresa is coming back with another question, then we must move on.

Q73 Teresa Pearce: Just one last thing: the criticism is lack of competition, dominance of the market by three main players and too close a relationship with customers. Those are the same criticisms that were levied against the big four auditing companies pre-Enron and pre the collapse of Andersen. Do you see any parallel here or any lessons to be learned at all?

John Grout: No. I would say none. The relationship is different with the rating agencies and the auditors. But, in particular, there are only three-there are only two and the third one was not that handy, although you found it useful in your particular case but it was just another mouth to feed. The expensive part of credit ratings is the time your senior management, including the treasurer, spends with the rating agencies and devotes to preparing stuff for them. Senior management time is the scarce resource in a large company, much more so than cost. If you are told there are now going to be more rating agencies and you have to train up a new rating agency, this is bad news because you have to get face time with them, with your chief executive, if you are a large company, with divisional heads and certainly with your finance people. That is why there is only three. The other reason is Legal and General only subscribe to three.

Q74 Chair: I am sorry, Mr Grout, you also were nodding your head, or at least there were some visible signs of non-oral communication going on, when the suggestion was made that industrial companies had been basically rated okay and it was just the financials that had gone wrong.

John Grout: Yes. It was when you said you couldn’t remember any scandalous misjudgments. There were, of course, cases when things that were highly credit rated proved not to be quite of the substance demanded. One thinks of Parmalat in Italy. One thinks of the scandals in the US. On the banking side you can think of Johnson Matthey Bankers. The day that Johnson Matthey Bankers went bust, the head of what was then IBCA, now part of Fitch, just happened by chance to be in my office when it was announced that Johnson Matthey had disappeared in a puff of smoke, and I said to him, "Did you spot that coming?" He said, "No, they told us the same lies they told the Bank of England". The rating agencies are not auditors. The rating agencies are analysts, based on what they are told. If what they are told does not make sense they are supposed to have enough sense to query it, and if they do not like the answer they are supposed to stop rating.

But that works two ways. When you sign a credit rating agency agreement you agree that if you stop paying, the rating agency will still go on rating you as an unsolicited rating from public information. So if they have downgraded you, you do not have the threat of cutting them off. You might stop their cash inflow but it is not going to change the bad news that has got out.

Chair: We have one more area that we particularly want to cover. I realise we are running overtime and I apologise to the witnesses for the next session who may be waiting.

Q75 John Thurso: I have one quick question for all of you. It is the same question, but can I just come back to a point that you made earlier, Mr Grout? It was regarding the Icelandic banks, when you said we need to read the reports of the sovereigns behind the banks. This morning I asked Adam Posen, when he was in front of us, what he thought about credit rating for countries and whether it made any difference. Without wishing to put words in his mouth, he more or less said he did not really care a great deal. I am indebted to a colleague for some figures that show different credit ratings for different countries and their cost of borrowings. There appears to be little or no relationship between the cost of borrowing, certainly at the five-year bond level and the actual ratings. How much faith can we put in the credit rating of sovereigns, if that is the sort of information there is and that is what people think?

John Grout: If you look at a website called Wikirating, you will find they put down ratings from the major agencies alongside those of agencies like Gadong, the Chinese agency, and their own Wiki ratings, which are based on published methodology and a sort of poll rating from people that happen to come across the website. You will see that there is a huge range of ratings on most countries; you will see that Gadong’s view of the US is not very high and so on. When you are looking at credit ratings, you try to understand the methodology and positions of the credit rating agency that are relevant and you form a view that takes into account what you have just learned from that credit rating agency, as you were explaining; but they are not determinative. They are just opinions-another opinion to be considered.

Q76 John Thurso: Where, for example, you have the five-year level for the US at 0.8, UK at 1.0 and Germany at 0.9, but you take a country like Canada that has three triple A stables and it is 1.4, what that says is the market takes a view of the country pretty well irrespective of what the rating agency says.

Malcolm Cooper: There is probably a liquidity premium in there as well. If you take US treasuries, it is the most liquid financial instrument in the world. For Canadian bonds, I have no idea where that ranks but it is going to be a hell of a lot lower than T-Bills.

Q77 John Thurso: The question people ought to be asking is, "Can I dump this quick if I need to?"

Malcolm Cooper: That would be one of the questions.

Q78 John Thurso: Which brings me to the question I want to ask. I will start with Mr Harrison. Could give me a reasonably quick answer to this? It is the same question for everybody and it is on the proposals put forward by the European Commission, CRA3. Do we need it? Do you agree with the need? Do you agree with the proposals, broadly, and what are the cost implications?

Mark Hyde Harrison: Our view is that CRA3 most probably goes too far over CRA1 and 2. We do not believe that we need to have the proposal to rotate the credit rating agencies every few years. It will add costs to companies in doing that, in terms of getting new credit rating agencies up to scratch to understand them and create instabilities in credit ratings on issues. On the proposal to make credit rating agencies liable to investors if they get it wrong, I think we believe that also goes too far, in terms of producing a lot of financial risk for those entities in way that is unhelpful for the market. So we are not supporters of CRA3. In terms of costs for the pensions industry, there would not be many costs.

Q79 John Thurso: Okay. Does anybody disagree with anything there or want to add anything there? You don’t need to repeat it.

Georg Grodzki: From our side we do not see any benefit and we would not agree that the various elements like rotation would make things better. In fact we are concerned they would make life more difficult for us as corporate bond investors because of the noise of rating agency changes. The costs would go up for us as well and the timeliness of ratings would certainly suffer, too.

Malcolm Cooper: I agree with those points. There are two things I would like to add. As an issuer, the cost to us would go up enormously; and secondly, the rotation is completely unworkable. We are a frequent issuer and I would need to change agency every year. You cannot reappoint an agency within four years. So at the end of three years I would have gone through Moody’s, S&P and Fitch and would not have a rating agency to go to.

John Grout: The position is that we have not yet seen the full effects of CRA1 and 2, and I would suggest that the effects of some of the provisions in CRA3 are simply that solicited ratings would disappear. The internal cost from the point of view of time, and so forth, plus the higher cost that Malcolm referred to, and the sheer grief of it all, would probably mean that you would see the disappearance of solicited ratings. You would go entirely to unsolicited ratings, and I am afraid you would have to subscribe to a lot more than you currently subscribe to.

John Thurso: In summary, a nil point.

Chair: We got a pretty clear message from you there, and in fact we received some very interesting answers throughout the afternoon. Thank you all very much for coming along. If you feel that we haven’t covered something you wanted to say when you came in, please jot it down. We would like to see it in writing. Thank you very much.

Examination of Witnesses

Witnesses: Steven Maijoor, Chair, European Securities and Markets Association, Verena Ross, Executive Director, European Securities and Markets Authority, and David Lawton, Acting Director, Markets, Financial Services Authority, gave evidence.

Q80 Chair: Were all three of you in for the last few moments of the last session? I would like to begin by asking whether you agreed with the conclusion that everybody came to, about the latest efforts of improving the regulation of the sector. I will start with Mr Maijoor and move to his right.

Steven Maijoor: There are a number of elements that are positive in the CRA3 regulations. One relates to the fact that there is regulation now proposed regarding the ownership of CRAs. I think there have been a couple of cases where a very clear position was taken; a rating decision was taken by the rating agency itself and there was an interaction with the ownership of the rating agency, for example, by owning the security that was rated. I think that this should be arranged and should be regulated. Another point relates to improved disclosure; for example, the disclosure that improves on aligning loans in pools. Pools of loans and some more information on that is helpful. Finally, I think, there is an argument for some liability for CRAs.

Verena Ross: Yes. I would add to that probably the general greater wish to create transparency and disclosure, making ratings more comparable. I think the earlier session talked about having different types of ratings, and often investors not understanding how the different ratings relate to each other from different rating agencies. I think that is an additional area that could be beneficial.

David Lawton: We would support the stated goals of CRA3, in terms of diminishing the risk to financial stability, enhancing transparency and trying to promote competition, but we have concerns that it comes too hard on the heels of a very significant introduction of new regulation in CRA1 and 2 and that, with some of the measures that are designed to achieve those goals, it is unclear whether indeed they will succeed.

Q81 Michael Fallon: Mr Lawton, you have registered five credit rating agencies, as I understand it, under the ESMA umbrella. What is the assessment process?

David Lawton: Each rating agency had to apply for registration in all the jurisdictions where it had an office. The registration process, therefore, involved us assessing compliance with the regulation domestically, but there was an overlay of working collegiately with counterparts around Europe, and that arose for two reasons. One, because obviously a number of agencies wanted to register offices in several countries, and we wanted to make sure that we were approaching the registration process in a converged way; and secondly, under the regulation other competent authorities had the right to veto our own registration decision if they did not agree with it.

Q82 Michael Fallon: How many did you turn down?

David Lawton: We registered all the ones that applied with us. Five applied and we registered five.

Q83 Michael Fallon: Did you come close to turning any of them down?

David Lawton: We did not come close to turning any down but during the course of the registration process, as often happens with entities seeking authorisation or registration, we insisted on a number of changes to procedures and systems in order to make sure that the regulation was fully complied with.

Q84 Michael Fallon: In what kind of areas?

David Lawton: I will just give three or four examples. Making sure that the boards of the subsidiaries that we were registering had the appropriate degree of expertise; ensuring that the boards were getting the right level of information; ensuring that the independent compliance function was appropriately resourced; and ensuring that the ratings approval process was sufficiently independent from the people who had prepared ratings.

Q85 Michael Fallon: This is governance and transparency, essentially?

David Lawton: Governance and transparency, but also resourcing and systems and controls.

Q86 Michael Fallon: How is this going to work, going forward, once the rating agencies come under more direct European supervision? What is the role of the FSA then?

David Lawton: ESMA will be the direct supervisory authority, and I am sure that Steven and Verena can talk a bit about that. In the supervision context, the regulation provides for ESMA to draw on FSA resources to carry out tasks, but the supervisory judgments would remain those of ESMA. In the policymaking space, we would continue to advise the Treasury on technical matters in relation to credit ratings, and participate in the ESMA board of supervisors as the collective in taking policy stances as ESMA would take policy stances.

Q87 Michael Fallon: Yes. But I want to be clear: who has the responsibility in future? It is ESMA, not the FSA. Is that right?

David Lawton: For supervising agencies according to the regulation, that would be an ESMA responsibility. That is correct.

Q88 Michael Fallon: If ESMA then calls on you, how is that resource paid for?

David Lawton: There is an ability for us to seek recompense via ESMA for the costs that we incur.

Q89 Michael Fallon: Does that fee get levied on the agency eventually?

David Lawton: I assume ultimately it would get levied on the agency.

Q90 Michael Fallon: But not by you?

David Lawton: Not by us.

Q91 Stewart Hosie: I will start with you, and then ask Verena or Steven to comment on the same question. Does the expanding scope of the credit rating agency regulation risk transferring over-reliance on ratings to over-reliance on regulation?

David Lawton: You are talking about the new proposal, CRA3?

Stewart Hosie: Yes.

David Lawton: There are probably a couple of elements in CRA3 that speak to that point. The first is a proposal that all future changes in rating methodologies should be approved by ESMA; and secondly, that ESMA should establish a harmonised rating index that would give a single rating, combining all the views of the different rating authorities. I think on both of those points we would have two concerns. The first is that there is an element of moral hazard in ESMA being seen to approve changes in methodologies; and secondly, that by creating a single ratings index we are encouraging greater reliance on ratings, when in fact we think the appropriate public policy goal should be to go in the opposite direction and seek to remove reliance on ratings through the regulatory system.

Q92 Stewart Hosie: Before I ask you to comment on that question about the risk transfer-over-reliance on ratings to over-reliance on regulation-on your point, David, you said you had asked a number of those five agencies to make changes, and you spoke about the harmonisation of the index by ESMA. Is there not a very real risk, both in the action you took and the approach that you are taking, that we make it so homogenous that there is no agency on the outside thinking differently; that the risk of a group-think scoring, which then blows up catastrophically, is exaggerated by the changes that you have asked agencies to make and the approach that ESMA have taken?

David Lawton: The changes that we asked agencies to make were not around the details of their methodologies-just to make that clear. But I agree with you that I think it would be a weakening of the system if regulators drove all methodologies to be the same. One of the elements about credit rating agencies that we need to preserve is the diversity of methodologies, and that is because there is no single methodology that is the right view.

Q93 Stewart Hosie: That is helpful. Who would like to answer the risk transfer question?

Steven Maijoor: As a start, two elements. First of all, in addition to CRA3 there are other movements where we would reduce the reliance on ratings. This relates, for example, to CRD4, where we reduce the reliance and move to a model where there is a combination of using ratings and internal systems and internal models, as discussed in the previous session. I think that is an important aspect related to CRA3. It is not only the regulation itself but in addition to that, a movement of reducing reliance on the various regulations to make sure that we get less dependent on the rating agencies. On the part about the approval of the methodologies, my personal view is that there are definitely tensions there.

One of the strong points of CRA1 and CRA2 is that, at ESMA, we can supervise governance, we can supervise internal controls, we can supervise transparency and independence, but we cannot rate the ratings themselves. We cannot interfere with the ratings themselves, and I think it is very important to make sure that the rating agencies can do their work in independence. Moving to the new CRA3 has indeed the tension that we, as ESMA, become involved in the rating methodologies. There is clearly a tension there with the strong points of CRA1 and CRA2 that we should not interfere with the ratings themselves. An alternative, of course, would be where the CRAs would be required, when there is a major change in their methodology, to report it to ESMA. Just like in a regular relationship between supervised entities and their regulator, if there is a major change in their operations they would typically report it to the regulator, but it would not be approved in advance.

Q94 Stewart Hosie: That is helpful. Just a final question because I know we are short of time. What progress has been made with regulators in de-linking the ratings from the regulatory framework itself? I think that is quite important in answering some of these questions. What is the progress on that de-linkage?

Verena Ross: There are a number of different initiatives under way to try to create that de-linkage, and that is a global process, not just a process in Europe. In Europe itself, the key measure is-as Steve already mentioned-the CRD4, so the banking capital requirements, where clearly the linkages are the ones that are most concerning people. But at the same time CRA3 itself has two parts to it; one part of it is a reduction in the reliance on ratings in the fund management area, for example. Clearly all those measures are ultimately trying to achieve the goal that the Commission has set out, of reducing reliance on ratings while at the same time making sure that, through regulatory requirements, you also create ratings that can be trusted and investors can use.

Q95 Stewart Hosie: Just as a little final question then, if we reduce the reliance on the ratings, and that is a good thing, there will be a requirement for more independent analysis of companies. I am just wondering, in general terms, how that will be facilitated in terms of regulation and supervision so we do not replicate mistakes between ratings agencies and other analysts. What is the shape of the regulation and supervision to make sure that independent analysis of organisations and businesses is where it needs to be?

Verena Ross: That is clearly a big challenge because ultimately, as we heard earlier, there are some companies who have the capacity to make that type of judgment themselves. They have credit parts to their business where they can make those kinds of assessments, and they can now rely on a great range of publicly available information to make those types of judgments. But there are a number of smaller companies who just do not have that capacity and capability to do so, which means that ultimately in reducing reliance you need to have some calibration and some possibility that, at least for some of the companies or end investors who are less able to make the real assessment themselves, they have a range of different opinions and different types of scoring methods and ratings assessments to call upon to make their judgments.

Q96 Mark Garnier: Verena Ross, I think you were in the room a bit earlier when we were hearing from the other witnesses. You heard that there was a predominance of reliance on just the big three ratings agencies. How do you think greater competition could be fostered between the rating agencies, in particular bringing in the smaller agencies?

Verena Ross: It is quite difficult-how that can be done in practice and, in particular, asking that of a securities regulator rather than a competition regulator. The key issue for me is that you have a level playing field where at least all the people who want to enter the space and want to work enter it under the same conditions. There are clearly some smaller credit rating agencies who are keen to grow their business, to become more involved, and I think there it is important to see also whether they want to do that in a niche and concentrate on one area, or whether they are keen to broaden out into other aspects. But it is clearly not something that you can do overnight unless you take quite fundamental competition-type measures of breaking up industries or doing things like that.

Q97 Mark Garnier: Do you think that the measure under CRA3, in terms of the rotation of rating agencies, is a good one and might address this competition problem?

Verena Ross: That is clearly the intention of the proposals that have been put forward. From a purely personal point of view, I think rotation can be a good and valuable tool to try to encourage other players to get a chance to play in the playing field and get to have a go at rating some of the issuers they might not otherwise have a chance to rate. At the same time, it clearly potentially introduces a risk that some of the players are not yet able to provide the same quality ratings that some of the bigger and more experienced players currently do, and that is something that needs to be considered in terms of cost benefit analysis.

Q98 Mark Garnier: David Lawton, do you have any views on competition from the FSA point of view?

David Lawton: I think this is a difficult challenge. This is, by its very nature, an industry that is going to be concentrated-just the economics of that. I think it is really important that steps that might be taken to promote competition are not at the expense of quality, and the concern I would have about the rotation proposal is that it would diminish quality. One of the things that the rating agencies need is a long track record of data to be able to produce their ratings. One of the risks that you run by rotation is that you lose that continuity and you diminish the incentive to do that.

Q99 Mark Garnier: What you are saying is that when you are not the current rating agency under this proposal you are not getting those interviews and that can-

David Lawton: That is correct. Some of the things that have been done by way of regulation, perversely, may help in the competition space. One of the issues that a new rating agency would face is establishing credibility quickly. One way that that credibility could be demonstrated is to pass the registration test and to be supervised by ESMA, and that is a reasonably quick way, potentially, of establishing credibility compared with building up the business over a number of years. Investors already have, in some sense, a mark of quality.

Steven Maijoor: Also, I would not rule out that there might be an interest from organisations that already have an international reputation but are not currently active in the CRA space. You can imagine that there are, for example, some consulting firms or even some auditing firms-in the past they have expressed some interest-because they would have less trouble getting this worldwide name recognition. Competition might not only come from the smaller ones but also from competitors in sectors that are close to the rating agency business.

Q100 Mark Garnier: Just one last question. This is more to do with the issues on sovereign ratings. Do you think that the regulation proposed by CRA3 is going to provide very good oversight of sovereign ratings without necessarily allowing political interference to come into play?

Steven Maijoor: First of all, for ESMA itself as an organisation, it is extremely important that our decisions as a supervisor are taken independently. That is what we are used to as a national supervisor, and it is also embodied in CRA1 and CRA2. We have already described the tensions in the fact that we would have the role of approving methodologies, and we have to be very careful that ESMA itself is not getting into the business of rating. We need to be independent from that process, and the tension that we referred to has that risk.

Q101 Mr Ruffley: Ms Ross, on CRA3, as we know there is a proposal to introduce liability on ratings agencies where investors can demonstrate loss. Her Majesty’s Treasury have said that placing unlimited liability on rating agencies, "could drive CRAs to limit their liability through withdrawing certain ratings or issue more conservative ratings or seek to limit their liability in other ways". What is the answer to that objection?

Verena Ross: It is absolutely a valid risk to identify. It is quite hard to know exactly how a market will react to the introduction of a liabilities regime.

Q102 Mr Ruffley: Do other jurisdictions have a similar law making CRAs liable for economic loss or is this just a European idea?

Verena Ross: I do not know exactly what the liability regimes are like in other countries outside Europe. I think the aim is clearly to try to improve the quality of the ratings that are produced, and by putting a more direct link between the credit rating agencies needing to be standing up for their rating-

Q103 Mr Ruffley: I think we all understand that, but the Treasury have said, in terms, that the CRAs might try to limit their liability by withdrawing certain ratings. So that is not going to help anyone, is it, if they just withdraw from the difficult areas where they think there might be a risk and they might cause loss if they get it wrong? It seems to be pretty much a knock-down argument, doesn’t it?

Verena Ross: There could be a problem with introducing a liabilities regime. On the other hand, it is very hard to judge. It might not happen. They might not withdraw from key ratings. I just do not know the answer.

Q104 Mr Ruffley: We understand that the Treasury argument here is that a minimum of six CRAs may be required by 2015 to meet the demand of large issuers of structured products. Do you think six is a realistic figure or do you think the Treasury is wrong on that?

Verena Ross: There are certain rotation proposals, obviously, in CRA3 in terms of the timing, and it will depend a lot on how many structured ratings are issued how quickly that rotation would kick in. Also, in the structured finance area, there is a proposal in CRA3 that there would be more information about the underlying asset pools made available to the general public and to any rating agencies that would rate. In that context, the issue that David raised in terms of having sufficient background and information to do the next rating might be slightly ameliorated.

Q105 Mr Ruffley: The Treasury put it more bluntly, don’t they? If it is going to be a minimum of six and, therefore, implicitly, perhaps more than six, they say, "In the absence of any eligibility performance criteria, CRAs would be assured of a market share, irrespective of the quality of their rating, perhaps resulting in a deterioration of ratings quality". Again, this could be quite a flaw in the CRA3 regime, don’t you think?

Verena Ross: I think I said earlier that the rotation policy has at least that risk-that there could be deterioration in the quality of the ratings.

Q106 Mr Ruffley: You are going to have to implement some new regulatory standards. The Treasury also raise questions about this, for instance, in relation to the establishment of EURIX, based on a harmonised rating scale. That is an example of a proposal that would in their words "pose significant technical challenges, in respect of both Information Technology infrastructure and expertise in credit risk modelling". Do you think you have sufficient resources at your disposal at ESMA, as the prime regulator here, to manage those challenges? What is your budget?

Verena Ross: We currently have a budget of about €16 million and staff of around 70 people. Clearly, for CRA3 we would require additional resource to fulfil the requirements in CRA3.

Q107 Mr Ruffley: Could you give us an order of magnitude? Twice your current budget or-

Verena Ross: As ESMA we have broader responsibilities than just CRA supervision and regulation. Currently we have about 15 staff working on CRA matters, mainly supervision, some on policy issues, and we think if CRA came into place, the way it is currently proposed, we would probably need an additional 20 to 25 people to fulfil those requirements-obviously, as you rightly say, also some IT budget to actually build the type of systems that are envisaged.

Q108 Mr Ruffley: That is very helpful. Perhaps Mr Lawton might want to also answer this question, but Ms Ross, could you tell us what further opportunities you see the UK having to influence the final shape of CRA3? What stage are we at in the negotiations and what scope does Britain have to press home, and I am presuming will be pressing home, some of the Treasury objections that I have just been reading out?

Verena Ross: I will leave the question on what the UK will do to David, but where we are in the process at the moment is that the proposals have been put forward by the Commission and are currently in the negotiation phase between the European Parliament and the Council. Basically they are both looking at the proposal, proposing revisions to those proposals, and ultimately will vote respectively on their preferred scenario. Then that will come together in the so-called trialogue.

Q109 Mr Ruffley: Mr Lawton, does the British Parliament have any say on this?

David Lawton: Of course, the Treasury is participating in the negotiations in the Council, as it would for any European regulation, but this will be a co-decision between the Council and the European Parliament. It is at a relatively early stage. There have been three Council working groups and the European Parliament have just produced a first draft report. My sense is that a number of other member states around Europe share some of the concerns that we have been discussing this afternoon.

Q110 Mr Ruffley: What is the timescale, in terms of months-this year, next year?

David Lawton: That would, in part, depend on the ambition of the respective presidencies, but I think that the aim is that this is something to be completed within the course of the next 12 months.

Q111 Mr Ruffley: Just one final question. Issuers rated by two rating agencies under CRA3 may retain one agency for six years-am I right in that?-but the other must be dropped after a three-year period. Is that correct?

David Lawton: Is that for structured finance products?

Mr Ruffley: Generally under CRA3, as I understand it. Can you help me here, Ms Ross?

Verena Ross: There is certainly a distinction. If you just have one credit rating agency or if you have already two rating you then there is a different rotation policy, but I have to admit I cannot exactly remember the dates now.

Q112 Chair: Can you remember the rotation policy?

David Lawton: No.

Q113 Chair: We have three regulators here who are not sure what the rotation policy proposal is.

David Lawton: The basic rotation policy is rotation after three years or after you have rated 10 instruments.

Q114 Mr Ruffley: Yes, we have that here. Is that all?

Chair: Is that for structured products?

David Lawton: No. That is for all products.

Chair: All products?

David Lawton: Yes.

Q115 Mr Ruffley: That is for everything, and then one rating agency may be retained for six years. What happens to the other one?

David Lawton: I can’t quite remember the rotation periods for structured products, but for structured products the proposal is that you must have more than one rating.

Q116 Mr Ruffley: But one agency can be retained for six years, yes?

David Lawton: I am sorry. I am afraid I can’t remember whether that is the precise proposal.

Q117 Mr Ruffley: There is some confusion on your part, so it is not surprising that we are confused. But the suggestion seems to be that whatever the rotation policy and whatever the specifics, it could entrench the established rating agencies in their dominance. Do you think that is correct or not?

David Lawton: As you were questioning a few moments ago, I think the challenges with rotation are two or threefold. The first is, will it increase competition? It is not clear that rotation among the big three would increase competition, for precisely your point: it would tend to give you a minimum market share come what may. Secondly, in respect of those niche agencies that specialise in particular sectors, there is a limit to what you can be rotated into. If you specialise in insurance companies, being rotated into structured finance or sovereigns is not ideal from a business model point of view. The third challenge is the point about capacity, because if the goal is that rotation generates more credit rating agencies, but the rotation periods are set very short and those agencies do not appear by the relevant dates, the market comes to a complete standstill. I think the Commission, in making the proposal, to my view, have not completely convinced on those three points.

Q118 Chair: Judging by some of the things you have said, "has not completely convinced" is one of the great understatements, isn’t it, Mr Lawton?

David Lawton: I am a regulator, I am afraid. I don’t do overstatement.

Q119 Chair: You do understatement. Okay, so we should aim off for these statements. If you do not think that competition is going to be increased, we are going to find the risk of greater reliance on rating agencies, because they have an endorsement from regulators, with no offsetting competition benefit. Correct?

David Lawton: Through rotation or just generally?

Chair: Generally; rotation and generally.

David Lawton: The public policy challenge to focus on is to work to reduce the mechanistic reliance on ratings in the regulatory system.

Q120 Chair: So we need to downplay the importance of these things?

David Lawton: We need to downplay the mechanistic reliance. It is unlikely we will ever get to a situation where they play no role, but they need to be just one of a number of indicators of creditworthiness that we need to rely on.

Q121 Chair: Is that what ESMA are trying to achieve?

Steven Maijoor: It is clear, as was said earlier on CRA3, that in addition there is indeed reduced reliance on ratings. I think that is broadly supported, even by the rating agencies themselves, as a result of all kinds of changes in the past.

Q122 Chair: I am just asking if ESMA are trying to find ways of reducing reliance on ratings.

Steven Maijoor: Yes. We have recently taken a decision on this, and also in our own regulations and guidelines where we refer to rating agencies and, indeed, changed the wording there to make sure that you do not only rely on ratings but also rely on internal investments. But the biggest news on ratings is not at the level of guidelines but at the level of laws and regulations, like the CRD4 regulation.

Q123 Chair: You are not trying to create a stamp of quality that can then add value and, therefore, enhance reliance on the-

Steven Maijoor: No. I think what we are trying to do is look at existing laws and regulations. The legislators are changing the existing laws and regulations and reducing their references to ratings. That is a very logical process. At the same time there is a limit to how far you can go because ratings themselves still have an important economic function. They summarise information. Rating agencies have information on issuers that others don’t have. So we should move in the direction of reducing reliance on ratings. At the same time there will be a limit. Those limits relate to the fact that they, indeed, have privileged information that other market participants do not have; also some of the smaller financial institutions that would have to do the internal modelling do not themselves have the resources to have solid internal modelling and internal own-assessment of credit risks. In addition, there is the problem, of course, that when you do your own assessment of credit quality that the independent element is not there.

Q124 Chair: Earlier you were talking about the cost benefit analysis of this whole thing. How developed is your cost benefit analysis of your proposals?

Steven Maijoor: Well, the-

Chair: Sorry, I was asking Verena Ross that question because she used the phrase "cost benefit analysis" earlier.

Verena Ross: For any proposals that the Commission put out they have to propose an impact assessment, which I believe they have also done in this case. As far as ESMA itself is concerned, where we create regulation at a European level as ESMA, where we basically draft technical standards, we have to do an impact assessment.

Q125 Chair: I know you do. But what I am asking is what have you done so far on this proposal?

Verena Ross: On this proposal, we have done nothing because it is not our proposal. It is the Commission’s proposal.

Q126 Chair: Have you looked at what the Commission have done?

Verena Ross: I have looked briefly at the impact assessment they have provided, yes.

Q127 Chair: It all gets me very worried. ESMA looked briefly at the impact assessment. Wouldn’t it be helpful if you went through it with a fine-tooth comb?

Steven Maijoor: I think it is important to realise that ESMA is obviously not a legislator, and the process is now in the hands of the European Commission, together with the Council and the European Parliament. Obviously it is relevant for our planning and to assess how our work will change in the future, but the decision-making on CRA3 itself is obviously not in our hands.

Q128 Chair: So you have not been through it. You have had a cursory look, but you hope that there is an impact assessment out there. You reckon that the benefits might outweigh the costs-I hope I am not summarising the position too unfairly-and you are determined not to allow this to develop into a stamp of quality. You are trying to downplay the ratings. Have I summarised what you are trying to achieve?

Verena Ross: Ultimately, we will be the receiver of whatever legislative proposal is agreed by the Council and the Parliament. We will then have to clearly implement that legislative proposal and regulate the credit rating agencies to the regulations and legislation that has been provided.

Q129 Chair: These are organisations that people say played a huge role, an important and damaging role, in the development of the crisis that we are all now in, the whole world is now in, and particularly in Europe as it happens. Convince me that I should not be concerned by what I have been hearing about the lack of clarity about even what the proposals coming from the Commission are, which are going to be passed on to you-the rotation and the vagueness about the cost benefit analysis among other things.

Verena Ross: Maybe we should draw a distinction here between what we are currently doing, which is basically already supervising and regulating credit rating agencies to the current credit rating agency regulation, and the future proposals that might come forward. We are very much engaged at the moment on making sure that supervision of credit rating agencies to the current regulation happens. Indeed we have been very active on that over the last few months since we took on that responsibility, but that is separate from the future development of further regulation.

Q130 John Thurso: You would not want to be given a future regulation that you felt was unworkable, so you will have some input on the future regulation. Mr Cooper from the National Grid-I think you were all here and heard-told us that he used three agencies, and that seemed a very sensible business model. If he is required to rotate and he wants to have those three established quality players, who are largely the only three established quality players in the market at the moment, then presumably what he needs to do is go out and rapidly find another agency he can rotate so that he can hang on to them. What he has are the three he has now, and he has Mickey Mouse SARL from somewhere else to be his rotator-I have just invented a word-to be the person who is rotated. Is that seriously the way we want to drive better regulation?

David Lawton: I have already expressed my reservations about the rotation proposal as a means of driving competition at this point.

Steven Maijoor: We expressed concerns that there would be, as a result of the rotation mechanism, smaller entities getting the possibility to rate issuers. The question is whether they have the right capacity and the right internal control systems. Do they have the right quality to make sure that this is a high-quality rating? On the other hand, we should realise there are already rotation requirements internally within the credit rating agencies. So there are possibilities already and requirements already within the organisations themselves to rotate.

Q131 John Thurso: So possibly the rotation should be within the agency rather than of the agency?

Steven Maijoor: As we just expressed, this would be a movement to go to a rotation requirement at the level of the rating agency itself, and we have pointed out the risks of such a policy.

Q132 John Thurso: One last point, which is how Europe deals with countries outside Europe. Obviously the vast bulk of the requirement for rating or a large bulk is in the US, and the US is unlikely to obey anything we send them particularly, unless they feel like it. How are we getting on in discussions with the US about bringing in new impositions?

Steven Maijoor: Maybe I can start with the mechanism, and then Verena can add to that in terms of where we are with the discussions with the respective regulators. The model in place is a model where we say that whatever rating we use in Europe, whether it is issued in Europe or coming from outside Europe, in principle needs to meet the same requirements, of course. Otherwise, if you did not do that you would have for the investor different qualities of ratings.

So, in principle, the model is that before a rating can be used from outside the EU it needs, in principle, to meet the same regulatory requirements as in Europe. Of course, we are in a transition period-there has been a transition period of two times three months and we are now in the middle of the second of the three months-and I think Verena can express where we are in terms of the various discussions with the various regulators.

Verena Ross: We have to basically make an assessment in two ways of non-EU regulators, and that involves, on the one hand, working out whether the regulatory regime they have for credit rating agencies provides as stringent requirements as we have here in Europe under the current regulation. Then we also have to make sure that we have a good co-operation agreement with them in place. On the basis of those judgments, we can then basically make a judgment that they should be endorsable and that the ratings can be used here in Europe. We have had intensive and productive discussions with a number of countries that we hope will get us, in the course of the next months, to a position where basically the vast majority of ratings that are currently produced outside the EU are endorsable and therefore usable in the EU. Indeed, we already have Japan and Australia in that position, and we are currently actively working on a number of other countries.

Q133 John Thurso: You are referring to ratings that are produced outside the jurisdiction and their applicability here.

Verena Ross: Yes.

Q134 John Thurso: I am also concerned about issues made outside the jurisdiction, because it would be possible, would it not, eventually for people to say, "There is no point in issuing anything in Europe since it is such a pain in the neck". You have Singapore that says it is going to be the next greatest financial centre. You have Hong Kong, which has pretensions to that already. You have Shanghai, which thinks it is going to get there in five years, and about three other cities in China that believe they can also aspire to it, and you have New York. Is there not a danger that people say, "Well, we will just go and do it there and we will just give Europe a miss as a place to issue anything"?

David Lawton: The equivalence in endorsement decisions is important because they unlock the ability of EU-based financial institutions to use those ratings for regulatory purposes. So it is not the place of issue that drives this point; it is the fact that EU regulation has embedded within it the reliance on ratings, and it is the qualifications those ratings have to have in order to be relied on.

Q135 John Thurso: But what do we do if the non-EU players, who are going to be the dominant ones in the future, decide it is still too much of a pain in the neck? That then means that the EU funds are pretty well restricted to something in the EU, and it quietly dissolves into a minority sport.

David Lawton: I think that is a particular point in relation to CRA3 that we have not talked about this afternoon: the assessments, which Verena was talking about, are in relation to the CRA1 plus 2 position. If CRA3 came in that would shift the goalposts for the equivalence decision further and the assessments would have to be done again. As you say, if other jurisdictions decided they did not want to-

Q136 John Thurso: The unintended consequence of all of this is that we are wonderfully regulated but do nothing because everybody has gone everywhere else, and that is back to the impact assessment that the Chair was referring to. Sometimes you can be brilliant but do nothing, whereas we do need to have some of the activity here in Europe.

Steven Maijoor: I must say, looking at where we are now with the various discussions, I am optimistic that the main other markets, in addition to the US, will be part of the assessment and will have the same requirements as we have in Europe regarding credit rating agencies. In the end, of course, if we think that CRA1 and CRA2-I think there is quite a lot of support for these measures-are needed to make sure that we have high-quality ratings in Europe, if you then opened up a weaker system outside Europe allowing these ratings to be used in Europe, you would get un-level protection of investors and also an un-level playing field for issuers. Obviously we need to avoid a situation where other important financial centres would have different standards but, in the end, if you support CRA1 and CRA2, it would not be a good idea simply for that reason to say, "Let’s go for lower standards outside", because if you believe in the quality and the effectiveness of CRA1 and CRA2 you also need to have these requirements for ratings issued outside Europe but used in Europe by European investors.

Q137 Chair: I am still not quite clear whether you are trying to boost the ratings quality, or whether you are saying it is not your business.

Steven Maijoor: Our business, in the same way as inspections or supervision of other financial institutions, is to inspect these CRAs. We go there-we have already completed the first round of inspections at the larger CRAs-and look at whether they meet the new requirements regarding governance, internal controls, independence and transparency. So these are the type of things you can inspect. Do they have the right counterbalancing powers internally? Do they have the right requirements in place regarding independence? These are the type of things you can inspect, but we cannot-

Chair: Given there are only three-

Steven Maijoor: As I said, we will not interfere with the ratings themselves or with the methodologies.

Q138 Chair: Given there are only three isn’t it going to become a box-ticking exercise pretty quickly? They know what you are going to come and ask. They know how to feed you with the right information and you are going to go away happy.

Steven Maijoor: I think it is a regulatory model that you see also in other parts of the financial markets. It is the model where you focus on processes, internal controls, as we are doing at financial institutions, as we are doing at auditing firms. So in that sense it is a similar type of model.

Chair: Let us leave it there for the time being. As I said at the earlier session, if there are things that you have heard this afternoon-and thank you very much indeed for coming-which you feel we haven’t given you the opportunity to air, please do come back to us in writing. We would be very interested to see it. This inquiry has some time to run. Thank you very much indeed for coming.

Prepared 6th March 2012