Select Committee on European Union Minutes of Evidence

Examination of Witness (Questions 1-16)

Lord Turner of Ecchinswell

25 NOVEMBER 2008

  Q1 Chairman: Good morning, Lord Turner. Thank you very much for coming. I need to remind everybody that we are being televised, so it will all appear on the Parliamentary channel. Of course, you do get a draft of your evidence to see before we publish. I have already made a deal with Lord Turner that I will ask him a question as opposed to an opening statement. If I may, I will therefore start straightaway with an enormous opening question. We are very grateful to you for coming because we really need to index this whole subject for ourselves and you are part of us getting a grip on the enormous subjects before us. We would really like to ask you about supervision and regulation, firstly in the European context and, secondly, where you think the proposed global supervision scheme is going to impinge on the European scheme. If I may, I will take it in bits. What do you think the most effective means are to ensure consistent oversight of financial stability in the EU? Roughly then under the headings: are there adequate structures in place to ensure that Member States' regulatory institutions work? What are your views on calls for a single European Regulator? How do you feel about a college of supervisors? All of this is in a European context.

Lord Turner of Ecchinswell: After this extraordinary financial crisis which the world has been through, I think it is very important to focus on what are the most important things to get right for the future and to be clear which of those need to be got right at a national level, a European level and a global level. I would say that one of the most important things to get right is an appropriate regime for the capital adequacy of banks and, in particular, for the capital adequacy of the trading activities of banks. I think there is a very reasonable case that in the past we did not require adequate capital for the trading activities of banks and where with banks in particular, we did not adequately make capital counter-cyclical, we did not require that capital buffers were built up during the good times to be available for the bad times. There are lots of other things we need to do which we can talk about, accounting, liquidity, rating agencies, et cetera, but if I take that one central point, the issue then becomes who has to design counter-cyclical capital requirements. I think the first thing to say is ideally this is agreed at a global level. Banking is a global business, many of our institutions operate across the globe, therefore a capital adequacy regime should ideally be agreed at a global level. The key mechanisms for agreeing a capital adequacy regime are in particular the Basle Committee on banking supervision, which played a key role in the existing capital adequacy regime in Basle 2 and also the Financial Stability Forum, which has been charged initially by the G7 but now the G20 to develop ideas on this. There one is trying to get a global agreement on a set of rules. Where does Europe come into that? Europe comes into that because these rules receive a legal expression in Europe through the Capital Requirements Directive, so our law, as it relates to the capital adequacy, comes from Europe in that respect. The fundamental concepts of capital adequacy would ideally be agreed at global level. There is a legal expression through the Capital Requirements Directive, but the fundamental concept has to be global. How it is put into place, actually enforced, gets to the issue of supervision, because it is the nature of capital adequacy that you cannot simply express it as a rule and just get people to obey the rule. It is a very complex thing which needs to be subject to very careful and in some cases discretionary supervision. At the moment that is done at a national level and I believe for the foreseeable future that is only going to be done at the national level. I do not think there really is a realistic possibility or a desirability of a single European Regulator; I do not think it would necessarily help us. The way we fundamentally work at the moment is that the capital adequacy of a bank group is determined by the supervisor of the home state of its major legal entity, so the FSA is responsible for HSBC and Barclays, and BaFin and the Bundesbank (the German authorities) are responsible for Deutsche Bank. The reason why I do not think we are going to move away from that system at least for many decades is that ultimately if you get it wrong, banks are bailed out by governments. Unless you are suggesting that when we get it wrong banks are going to be bailed out by combinations of governments which agree in a combined fashion to bail them out, ultimately the supervision has to be organised around a national level as well. However, that does not mean there is not a major role for colleges of supervisors. Again, I think one has to be a little bit clear here about colleges of supervisors at European level and at global level. There are already colleges of supervisors organised at global level. For instance, we have had one for quite some time for banks like HSBC or Standard Chartered, which are somewhat informal mechanisms and have no legal status. They have not been established by any category of treaty-based law, but they are fairly effective mechanisms for the relevant senior supervisors of global groups to share perspectives and share information. Those are going to be given a greater formal definition and extended to a wider set of global financial institutions by proposals which are being put forward by the Financial Stability Forum; that is part of the agenda of the Financial Stability Forum. There is then a separate set of colleges that already exist within the Capital Requirements Directive because if you are to agree with a European bank to allow it to use the internal ratings approach under Basle 2, that is a decision which requires the joint decision of many different supervisors across Europe. That is the core of what that sort of college presently does, it makes that specific decision. At the moment we have a global set of colleges which are informal, non-treaty based and focusing on a very wide set of supervisory issues bringing together the three or four major supervisors of a group like HSBC or Deutsche Bank or UBS. We also have a specific set of colleges, what you might call colleges or combinations of supervisors, set up under the Capital Requirements Directive to make the specific decisions which are required as to whether banks can use the internal ratings approach under Basle 2. How do we see this is likely to develop? We believe the most important thing to develop in the immediate future is to extend the global approach to colleges, which has been driven by the Financial Stability Forum, to a further stage of formalising them and making sure they are put in place for all the major global financial institutions. At the moment we would see that global aspect as the primary bit where we should develop the role of colleges. Having said that, again I think one has to be realistic about what colleges can achieve. They can achieve sensible sharing of information and perspectives on capital, liquidity, accounting and risk management. But I do not think they can cut across the fact that we have to have a clear sense of who the home country supervisor is, who is ultimately responsible for the overall capital adequacy of a group. There are also very important issues, which I think we should come back to, in relation to the supervision of subsidiaries and branches, what we call "host company supervision", where you are looking at the supervision of an individual part of a global entity.

  Q2  Chairman: Thank you very much, Lord Turner. We shall come back to that question. I have got a supplementary on that. What happens if there is a disagreement between members of a college? We do not have much experience of that, do we?

  Lord Turner of Ecchinswell: The answer at the moment is, I think, one again has to understand these different approaches. Colleges as understood at a global level, which until now have been entirely informally developed entities but are being given a bit more formalisation through the initiative of the Financial Stability Forum, are not ultimately decision-making processes, they are information-sharing processes. That is still incredibly valuable. It is very valuable for our supervisors at the FSA to be sitting down every six months with the supervisors of, for example, HSBC. For the US, it is actually the Chicago Fed in that case, because HSBC happens to be registered as a holding company within the Chicago Fed rather than the New York Fed. The college also includes the Hong Kong Monetary Authority, and other major supervisors. We compare notes on our understanding of all the aspects of HSBC's management. We do that for that bank in particular on a six-monthly basis with a fairly formal agenda et cetera. There are also similar mechanisms which have been in place for some years for Deutsche Bank. What is happening within the Financial Stability Forum is the definition of about 33, I think it is, global financial institutions for which such colleges should be in place. There is a process of making sure that each of the major home countries' supervisory bodies reports back by the end of this year whether those colleges are in place and, if not already in place, what the plans are for doing so. Those are not ultimately decision-making bodies. Decision-making, in the sense of what is the precise capital requirement, is fundamentally done by the individual supervisors of legal entities. It is done by home country supervisors in relation to group capital adequacy and host countries in relation to the local subsidiary. What one will typically have is a capital adequacy requirement or a liquidity requirement, both in relation to the group and in relation to the specific legal entities within the group. Again, I cannot see that we are going to divert from that. Indeed, I have to say that some of the experiences of this year will probably make host company supervisors more determined than before to focus on the capital and liquidity of the local entity for this reason. It goes back to my point about where the fiscal resources ultimately come from to save an entity. Let us take the case of Lehmans, for Lehmans to have been rescued depended on the action of the American Government. It is not going to be the case, particularly post Lehmans, that any supervisor around the rest of the world says, "We are willing to cede the supervision of that to the home country supervisor and rely on them at a global level to get it right". I have to say, post Lehmans there is going to be a greater focus than before in making sure that your particular legal entity within the total group has adequate capital and liquidity separate from the group. Until and unless you imagine that in conditions of crisis the US Treasury Secretary is going to ring the Chancellor of the Exchequer and say, let us jointly organise a recapitalisation of Citicorp's, as over this weekend, or RBS for us. At the end of the day people are going to focus on legal entities because they are not part of the decision-making process about what we do in extremis and what we do in rescue. We have to be careful with the assumption that we can proceed to the "global" supervision of banks, absent the creation of a global government. Until and unless we have a global government with global fiscal resources which makes joint decisions on behalf of all global taxpayers as to whether you bail out a bank, the supervision of banks will have a very significant national element. You cannot cede the ability to make those decisions about supervision away from the legal entities in your particular part of the world unless you are also participants in the decision as to whether to bail it out. I do think this debate about national or European or global decision-making needs to be rooted in a realistic understanding of what happens in extremis and the fact that in extremis, emergency liquidity is provided by central banks and emergency fiscal support is provided by governments.

  Chairman: Thank you, Lord Turner. That is very clear indeed and I can see it has sparked a row of questions on my right. Lord Trimble?

  Q3  Lord Trimble: First just to comment before I come to my main point. I entirely take the point you have just been making at length in the last sentence there. Can I say that you are going to find it easier to make that argument if you drop this word "college" because college carries with it the connotation of an institution and what you are actually describing is not an institution at all but simply an informal way of sharing information to enable other people who have got an institutional role to take whatever are the important decisions. If you use an institutional term such as college, particularly a global college, then some people are going to run away with the idea that you are creating a new global authority which is going to rule the world or rule bits of it and, as you say, that is quite unrealistic. Coming back to more fundamental matters, you were saying that the key issues from your point of view were capital adequacy, accounting requirements, liquidity requirements and so on and, as you pointed out, these are rule-making matters, these are matters of law, they are quite different from supervision. What I find rather puzzling in this is you are approaching this as if it was a matter of regulation and rule-making, whereas the real problem is supervision which is a different function entirely. To give you the basis of this, this is a point which Lord Lawson has made a couple of times in debates on this issue. They are saying that banking supervision is a fundamentally different thing from market regulation, and the points you are making about accounting liquidity or capital adequacy requirements, that is market regulation. Banking supervision is looking over someone's shoulder to see whether they are being sensible or not in how they operate their business.

  Lord Turner of Ecchinswell: Let me pick up both points. First of all on the point of colleges and languages, I have come into a job where I have inherited a certain language which is fairly dominant throughout the world and I will probably live with it and try and make sure it is understood. I am not sure what people take from the word "colleges". I entirely understand the point, but it is a word which is out there on a fairly large scale at the moment. On your substantive point about rules and supervision, I think there is a greater overlap to this than you might be implying by that. If we look at capital adequacy under the Basle 2 arrangements, the way that it works is to try and have a quite fine-tuned definition of the capital that it requires a bank to have related to the risks that it runs. The way this works is a bank which is allowed to use what is called the "internal ratings approach", which is pretty much all the big sophisticated banks, is developing what is called an "individual capital assessment" where they are, within the guidelines of the rules, coming to their supervisors, so in the case of UK banks this is the FSA, and saying, "Here is our proposed bank capital adequacy plan" and these are 200-page documents with layers and layers of detail beyond them that are saying, "We are doing mid-corporate lending in India. Our models of the past loss rates that have emerged on mid-corporate lending in India suggests that over time it might produce this loss rate on average, but in extreme circumstances it could produce this loss rate. When it produces these defaults, this is the loss given default et cetera, et cetera, therefore we believe that we need this amount of capital in order to support that slice of lending". For a large global bank there will be hundreds of these individual statements as to the risk characteristics of this bit of lending, this bit of deposit taking, it will specify their operational risk, it will specify their trading risk, out of which becomes a capital adequacy plan. Therefore, and this is an absolute core role of what an FSA supervisor of a bank is doing, we are assessing that capital adequacy plan, challenging those models, saying are those models good models, comparing the model which Bank A is using with the model which Bank B is using and saying, "No, we think you are making optimistic assumptions compared with your peer group", looking in the trading area at whether their operational risk control is adequate.

  Q4  Lord Trimble: I understand the point you are making.

  Lord Turner of Ecchinswell: Therefore, the end point is the application of the rule of capital adequacy and, indeed, the determination of how much capital they have is supervision. I really disagree with either yourself or Lord Lawson that there is a clear distinction between rule setting and supervision.

  Q5  Lord Trimble: Can I make the comment that what you are describing, where people are making an assessment of risk and putting in its place whatever they think is necessary for that risk and then you are assessing their judgment on risk compared with what other people are doing in the same sort of market at the same sort of time, is fine when everything is good, but will not work when things turn bad because then the risk suddenly changes. You are talking about a system which is not designed to deal with circumstances when things go bad.

  Lord Turner of Ecchinswell: Yes, but I am not sure there is any simple rule that can.

  Q6  Lord Trimble: That is what I was quarrelling.

  Lord Turner of Ecchinswell: Where we started was the distinction between regulation and supervision and I am suggesting there is no distinction between regulation and supervision. In both regulation and supervision we need to make sure that we have systems which are robust for systemic events, events where the whole system goes in one direction.

  Q7  Lord Trimble: Clearly we have not had that.

  Lord Turner of Ecchinswell: We have not had that and we have got to get that right.

  Q8  Lord Trimble: Can you do that as a matter of rules which apply in all circumstances? I doubt it.

  Lord Turner of Ecchinswell: No, but I think that makes my point, that to do this you need rules which are closely supervised, which is why the distinction you were drawing between regulation and supervision is not one which I recognise. I think that is a different issue from how we get the regulation and the supervision correct to deal with systemic risk, but the distinction between regulation and supervision is not one which works in this area.

  Q9  Lord Woolmer of Leeds: On that very point, Lord Turner, you came into the job not after the event but well into the event. In practice, the public out there across the world see an almost complete failure in the system of supervision. You set out very clearly and coherently all the theory but in practice, all these colleges of supervisors, host countries and home countries and so on, did not supervise the system well enough to prevent a meltdown. It would be interesting to me to hear from you in this still relatively early stage what is it that is going to change to restore the credibility of supervisors across the globe which will give people the confidence that things really have changed. Nobody in the supervision world has said, "Sorry, we really absolutely got it wrong". It would be helpful to hear in practice what you think is going to change. Secondly, on that theme, again you talked convincingly about the fact that host countries are not going to ignore what is going on simply because of home country regulation, and you gave the example of Lehman Brothers. The UK is host to a lot of financial institutions, many of whose home country is the United States. What is it precisely that the FSA is going to be able to do which regulates and supervises American institutions operating out of London in a way that it did not before?

  Lord Turner of Ecchinswell: First of all, let me say that I accept entirely your first point that the system failed and people are right to look at the totality of the system of the world regulation and supervision combined of the global financial system and say pretty clearly something went wrong. It would be pretty bizarre if after having spent $500 billion of public money in the US and £37 billion of public money here we did not say something had gone wrong. Secondly, I do not entirely accept that nobody has said sorry. The FSA in its internal audit report on Northern Rock—this was before my time—was exemplary in putting its hand up and saying, "We got a set of things wrong". We got it wrong in terms of not paying enough attention to when a whole business model was risky, we got it wrong in some of the details of supervision having too rapid a turnover of the individual supervisors of Northern Rock so that they were not close enough to the company et cetera, et cetera and we are focusing on putting right the failures which were identified in the Northern Rock internal audit report. I think they were identified in far more detail and far more effectively by a document which we produced and published than by anybody else's commentary on it. We are putting that right through what we call the "Supervisory Enhancement Programme", which are improvements in the number, quality, training and processes of our key supervisory staff of key systemically important institutions. We are doing that. Having said all that though, within the debate about how we prevent this occurring in the future, I think it is very, very important to identify the things which are really, really important and will make a big difference. I think the improvement in our supervision is important, but even if the FSA had been exemplary in its supervision of Northern Rock as an individual institution and similarly around the world, we might still have had a world financial failure if we had not had a couple of big things right. The first, as I have stressed already, is a different approach to capital. We have learned that there was simply not enough capital held against risks in trading books. This gets to a debate which I know you have later on your list of questions to do with gross leverage ratios et cetera, but I think in retrospect—and this was not the FSA alone, here we were part of a global rule-setting process, we were part of the rules defined by the Capital Requirements Directive, rules worked out by people who thought they were very sophisticated and the best experts in the world—we did not put enough capital against risky trading activities in the capital markets and trading activities of banks. We did not have a system of capital adequacy of banks which was counter-cyclical, indeed it may in some ways have been pro-cyclical. Across the world there was inadequate focus on liquidity and the supervision of liquidity. In a sense regulators across the world moved to a belief that as long as people were operating in a traded market, there would always be liquidity because you could always sell things. They moved away from some of the classic old beliefs of banking that you actually had to have short-term absolute liquid assets, things which were short-term to maturity rather than things that you could sell. Even more important, what we are doing in our Supervisory Enhancement Programme is important, but if people want reassurance that we will significantly reduce the likelihood of this occurring again in five or 10 or 15 years' time, I would look to the new rules which are going to be brought forward by the Basle Committee and the Financial Stability Forum and others on the capital adequacy of banks, both in terms of total amount, the amount specifically related to the trading books and the extent to which it is counter-cyclical, that is what will really make a big difference. I would look to regulators throughout the world paying much more attention to liquidity. We have a consultation paper about to come out on that. There is new Basle Committee guidance on that, but that is where we need to make sure we get it absolutely right. Finally, at national, European and global level I would look to a much more structured focus on what is sometimes called "Macro Financial Stability Analysis" and "Macro Prudential Analysis". In retrospect, I think in the UK the biggest failure which occurred was not a failure of a specific supervisor of an individual bank but a failure to look at the whole system and to realise that there should have been warning signals that the whole system was risky. It should have gone as follows. We were running a very large current account deficit. We had very, very rapid growth of mortgage lending, including to people who had not been in categories of people who had borrowed for mortgages before. We had a set of business models, by which I mean the new go-getting mortgage banks, Bradford and Bingley, Alliance & Leicester, Northern Rock, which were rapidly expanding beyond their deposit base and relying on the ability to sell retail mortgage backed securities. That was the flip side of the current account deficit, that was part of how the current account deficit was being financed. Those were being sold to, for instance, SIVs and conduits. These were new forms of shadow bank which were taking very significant forms of maturity mismatch risk. They had short-term liabilities and much longer-term assets, as indeed had US money market funds. I think the big failure was not made at the level of the individual supervision of an individual institution because I do not think you can see the pattern at that level, it was the failure to put this whole story together and say there is something fundamentally risky about what is going on here. Therefore, we either ought to be pulling monetary policy levers which lean against the wind, as the phrase goes, macro-prudential levers, such as capital adequacy requirements which lean against the wind. One of the most important debates to be had at global level—and I have to say, I think it is more important than the debate about college of supervisors, though I think that is a useful thing for us to be doing—is how we, Europe, the world and the IMF get a better ability to see the overall pattern of the risks which are coming at the total level and to be taking appropriate measures to lean against the wind. That then takes us back to the capital adequacy requirements because one of the ways you lean against the wind, and it may be a much more effective way than classic monetary policy, ie interests rates, is through increasing capital requirements, either on a rule-driven basis or a discretionary basis, when it is obvious that we are in the middle of an unsustainable boom.

  Lord Trimble: My Lord the whole function of the central banker is taking away the punchbowl when the party gets too frivolous.

  Q10  Lord Haskins: Alan Greenspan worked, of course, on the assumption that good corporate governance would avoid all these problems and the boards would see the issue coming and act accordingly and it would not be left to the regulator. Have you anything to say about corporate governance and the way it has failed during this crisis?

  Lord Turner of Ecchinswell: The idea that good corporate governance at an individual institution level could avoid the risks which eventually occurred is wrong because I think we have a fundamental collective action problem here, the fact that we are dealing with an issue which is systemic. I think it is very difficult for an individual institution or an individual set of managers to see the overall pattern of what is unsustainable. It is also the case that an individual institution is caught in a problem whereby if they are the first to abandon this risky activity, they perceive that they lose market share without necessarily themselves being protected against the systemic problems when they occur. They might to a degree, and clearly there is a differentiation around the world of those banks which were most cautious and those which were least cautious, but the share prices of good banks have fallen dramatically as well as bad banks. Although there is a role for good governance and good management, and I think there are lessons, there is a collective action problem here which has to be dealt with through Central Bank action and through regulatory action. Let me give you one example, the issue of remuneration, which, again, is on your list of questions. There are issues about the way remuneration is structured which can either encourage excessive risk-taking or not encourage excessive risk-taking. If you are paying people off the immediate one-year profit and loss of a trading book and saying, "It looks as if we made a profit, therefore I'm going to give these traders a whacking great cash bonus", you will probably tend to produce more risky activity. If you say, "Well, there may or may not be real profit here. I'm going to give you a bonus, but it's going to be in shares deferred for five years with the ability if things go wrong to take it away from you", you are probably likely to produce activity which at the limit is slightly less risky and pays more attention to some of the long-term challenges of the firm. However, it is very difficult for an individual remuneration committee or board to be disciplined about this because what you get are proposals of "We need to get Manager X", you say, "Okay, we'd like a sensible remuneration structure for Executive X from this other bank. We'd like it to be deferred. We'd like it to be paid in shares" and they say, "That's all very well, but our competitor has offered this sum in cash and immediate". This is where one of the things that regulators can do, by setting a set of principles and making it plain that remuneration practices are one of the things we are going to look at and supervise and put in our risk mitigation programme, is create some space where the individual remuneration committee feels more empowered to do what it wanted to do in any case because, again, it is one of these collective action problems. You need things which make it easier for that remuneration committee to know what it knew in its heart of hearts was the sensible thing to do in the first place. There are lots of things that good governance can do, but we have systemic problems which require good macro policy and good overall supervision policy in order for us effectively to guard against the risks.

  Chairman: Thank you, Lord Turner. I am trying to exhaust our questions on regulation and supervision because I need to go on to the more provincial considerations of what we are going to do about the EU directives. Lord Moser?

  Q11  Lord Moser: I was going to ask a supplementary very similar to Lord Trimble's because you started your opening answer on capital adequacy, in relation to trading activities of banks and this is incredibly important. I was simply going to ask at that point whether you would be satisfied just to deal with a sort of ratio or whether you would go into details of the trading activities, but then when you answered Lord Trimble, you talked about the detailed supervision or monitoring that you would go in for at the FSA and that all focused on risk management. In other words, you get these big documents from a bank and the risks they are going to take and you ask is the capital adequate for those risks. I understand the structure, the worry expressed in my case against a background of total loss of confidence in banks—I come from a banking family—is that risk assessment is not a science, risk assessment is a very risky business. When these banks give their risk assessments, I do not believe in them basically because they want to do business. The centre of this whole structure still worries me. I understand totally that capital adequacy is the beginning of your monitoring but it ends up with the banks saying, "This is okay because we've assessed the risks, it's fine". I do not believe it.

  Lord Turner of Ecchinswell: I think the difficulty is if you do not believe it, you would probably end up believing that banks are rather risky things whatever we do, which may have an element of truth. What you suggested there touches on a crucial debate as to whether we try and regulate banks' capital adequacy through simple ratios, such as a gross leverage ratio, or by what are called "risk sensitive" or "risk precise" type measures, such as Basle 2 attempts to do. I think there is a pretty strong emerging argument for doing both in a belt and braces approach. Each of them has problems. The problem, you are right, if we take the Basle 2 approach, which is the individual capital assessment that I described earlier, where what one is saying is the principle is we want to hold the amount of economic capital which is really required for the risk, so we want a granular analysis by the bank of what the risks are in Indian mid-corporate lending or prime mortgages in the UK or trading activities and it is built up in a very sophisticated form. Getting that right depends crucially on both the quality of what the bank does in its analysis and also depends crucially on the quality of the challenge from the supervisor. That is the point at which the supervisor can say and does say, "No, we disagree with that, we disagree with your assumption about what the default risk on this category of bonds will be" and we can do that and we do do that. It is a very detailed process and it could be that in that detailed process people still get it wrong and they simply underestimate the amount of risk in an activity. Therefore, people hanker after a simple rule which will control this, such as leverage again, your capital as a per cent of your total assets should be five per cent, 10 per cent, whatever. The difficulty is that, of course, the simple rules are what we used to have and what we realised was that any simple rule will produce activity which maximises the risk underneath it. If you had a simple rule which says capital to total assets will be five per cent, a bank has a clear incentive to put on risky assets, poor quality mortgages, not high quality mortgages, because in your simple rule you are basically saying, £10,000 of mortgage is £10,000 of mortgage, you are not differentiating. Whereas, the first approach, which is the principal approach, allows us as the supervisor to say, "Hang on, I can see that your mortgages appear to be riskier than the average mortgages, therefore I want you to be having seven per cent, even though I am demanding four per cent off the other guy". So the simple rule also has its problems, and indeed America had a simple rule on leverage; it had that before this crisis. A simple rule has to work on equity to assets. What is an asset? People said, "An asset is a thing which is on the balance sheet, and these things which were off the balance sheet in our SIVs and conduits were then not on the balance sheet, so they are not an asset, so I do not have to hold equity against it." So for any simple rule that you have which you write down with a legal definition, some clever lawyer will find a way of doing some other economic activity which does not count as an asset within that rule, whereas the first approach, which is the principles and discretion approach, enables us to say, "I know this is not an asset but it is still a risk, so I want some capital against it." I think what we have learnt is that there is quite a strong argument emerging, and this may well be where the world ends up, for a belt and braces approach. If we look at the way, for instance, the Swiss have now decided to have a gross leverage ratio. It is quite interesting why they ended up with that. It was heavily to do with the large trading books of Credit Suisse, to a degree, but UBS in particular. UBS, which has recently been subject to a re-capitalisation and bad asset scheme by the Swiss National Bank and Government, was reporting right up to the time when problems emerged very, very strong ratios on the sophisticated risk assessment process. It was arguing that was how the previous rules had said, "Everything we do in this whacking great trading book is secured, based upon collateral calls, based on marginal calls, therefore it is very low risk, therefore in our sophisticated weighted risk asset approach we do not need much capital against it, therefore we have a high capital ratio." But things still went wrong, and they went wrong because, despite the fact they had a high capital ratio against risk weighted assets, they had a huge balance sheet. So the Swiss National Bank and the Swiss regulatory authorities have essentially said, "We do not trust either yourselves or ourselves to be sophisticated enough to work out precise risk in a whacking great trading book, we just have a generic feel that if it is huge in some way it is risky, and therefore in addition to the Basle 2 type looking-at-your-individual-risks approach, we are going to have an aggregate leverage ratio as well." As I say, I think we will probably end up as a world regulatory community believing we need both. If we just rely on gross leverage ratios, people will be finding all sorts of ways round the rules, and if we just rely on the detailed analysis of the banks and our ability to challenge that, we will sometimes get it wrong in a way that a single rule would have managed to capture. So that is why I think we will end up with a belt and braces approach.

Chairman: Thank you very much. I am now going to move this discussion, enormously useful as it has been, from the high level to the things which are actually going to have to go through ECOFIN next week, and ask you for a quick opinion on all of them. What is going through ECOFIN is the CRD, and an opinion would be enormously useful. Lord Woolmer?

  Q12  Lord Woolmer of Leeds: In the Capital Requirements Directive there is a proposal, as you know, for a 5 per cent quantitative retention requirement of the total material share of the risk for the originator in securitisation deals. It has been said that this may encourage banks to move offshore in European terms. What do you think of that fear? Given that G20 is going to look further at things, does this imply that the Capital Requirements Directive may be re-visited again later next year?

  Lord Turner of Ecchinswell: This issue relates to what is sometimes called "skin in the game", as people call it, that when you originate a loan and then you distribute it, should you be required to at least keep some exposure to it. I think in the FSA we have been less convinced than the European Commission that this reform is a crucial reform, in the following two respects. First of all, many people who have got into trouble on securitised lending in the big trading books and have made large losses were actually maintaining skin in the game; the fact they were maintaining skin in the game did not necessarily stop them getting into trouble. Indeed, if you look at the total system, you could argue that the problem with the way we run securitisation is actually that the banks in total kept far too much of the risk, not too little. When securitisation was originally described and grew up in the 1980s, people said, "Here's the plan, we don't want a West Texas Savings and Loans institution to carry a non-diversified book of exposure to West Texas mortgages, because that means that if the local economy goes down we will have householders going down and a bank going down; much better to turn it into a security and sell it on a diversified basis to a diversified set of investors throughout the world." So the concept was, deposit-taking institutions, banks or savings and loans institutions, will end up holding less of this long-term risk which will be held more by pension funds, insurance companies, et cetera, who are natural hold-to-maturity investors of diversified holdings of assets. If you then look at the figures in the latest IMF Global Financial Stability Review of where the losses on securitised lending have been, about 70 per cent of them have been in the trading books of the banks, not in the pension funds and life companies. Having developed this idea which was meant to originate and distribute risk, at the level of the overall banking system the system was not distributing at all, it was originating. You were often distributing your own originated security but then your traders were, as an exercise in position-taking buying somebody else's securities. So at the total system you could say the problem has been not too much originate and distribute but the fact there has been far too much originate, distribute and then buy it all back in order to indulge in trading activities. If you look at the specific issue of the thing which I myself originate, should you continue to hold a slice of that, in many cases as I have said that already occurs, in many cases you could say it is an unnecessary restriction because you want people to be able, once they have originated a risk, to get it fully off their books if necessary to get the risk away from the banking system. You also have to think about how you are going to implement this because of course you can lay off an economic risk by taking a position in the credit default swap market even if you have maintained that risk on your books, so you have to think about the operation of it. So there is a set of reasons why we have been less convinced that this is the core of the issue but it is a proposal which in its current form we can live with, we think it is reasonable in its current form. The likelihood of a significant movement offshore set at the level it is at the moment? No, I do not think that is very large.

Chairman: If we could ask Lord Moser to ask a question on credit rating.

  Q13  Lord Moser: Some people in the blame-game point at the ratings agencies. This is a straightforward question, what is your feeling about future regulation of the agencies, what is necessary, what can be done, and what should be done in the European context? They have made their proposals, are they adequate?

  Lord Turner of Ecchinswell: I think the over-use, or in some case the misuse, of credit ratings is a significant part of the story of what has occurred. I think it is reasonable for us to register and oversee rating agencies, but I think I would be very cautious in believing that the regulation of rating agencies will be a crucial bit of what gets it right in the future. I think actually more important is the fact that there is going to be natural market reaction which is more sensible about the use of ratings in the future than has been in the past. Let me explain what I mean by that. If you look back through the history of rating agencies, for many years they rated the individual securities of major corporates and they ranked them as AAA, AA, et cetera, et cetera, and my understanding is that if you look over the decades this was done reasonably well in the sense that the test of this is an empirical test, was it the case that the rankings AAA or AA or BBB in 1980 were a reasonably good predictor of whether they did or did not default? If we want to do the analysis of whether they did their jobs well, that is what we are essentially saying. What happened I think in the last 10 years is that ratings were extended to a wider and wider set of securities, including to some very complicated, structured securities, where it was actually much more difficult to work out what the true default risk to maturity would be, because these things had not existed before, so you did not have a track record; they were inherently more complex and they did not have a track record. The rating agents were too willing to say, "I put my rating on this CDO" without really working out whether that rating could be considered of the same quality as the rating which had previously applied to a General Electric straight bond. Secondly, I think probably too many investors relied entirely on ratings. It is very difficult for many investors not to rely on ratings. Corporate treasurers throughout the world depend on ratings to say, "I am only going to put my cash with a AA rated bank". It is difficult to imagine a system which does not have some reliance on ratings, but the system relies on a balance between a reliance on ratings and other people doing fundamental research, and probably too many people ended up saying, "Because this has a rating, therefore by definition it is good." Thirdly and I think crucially, people implicitly fell into the belief that the rating might be telling them something about the mark to market value of this instrument, ie what I would be able to sell it for in six months' time, and of course it is not going to tell you that at all, it is meant to tell you what is the probability of default if held to maturity. When you get changes in the liquidity of a market, you can have a major collapse in a price of an instrument even though the probability of default if held to maturity has not changed. So there was an over-extension I think of the science of ratings to things on which it was not good science on which to rate them, there was an over-reliance on ratings in general and in particular there was an over-reliance on the idea, implicit in some people's behaviour, that they told you something about liquidity and market value rather than just telling you what they meant to tell you, which is what is the percentage probability of this thing defaulting if held to maturity. There are then a set of ways in which ratings also played a key role in what occurred because of the way ratings end up as triggers within the system; the loan covenants will have a ratings trigger in it and if one thing changes other people automatically change their behaviour. So all of that is a problem. There were also clearly within the ratings areas issues of conflicts of interest which may have encouraged that tendency to rate things for which there was not a good basis to rate. There is a problem which needs to be carefully managed in governance terms, that rating agencies are paid for by the people who are issuing the stock, that is who pays for a ratings agency, and there is an implicit danger of conflict of interest there which needs to be carefully managed. All of which says in a whole variety of ways that it was a significant part of the problems which have occurred. Therefore it is perfectly legitimate to say, "We want to be able to register them, we want a regulatory oversight of them, and we want to at the very least make sure we have governance procedures that make sure they have appropriate skills, they have appropriate procedures for making sure there are not conflicts of interest." What I do not think one can envisage however is that the nature of regulation is ever going to involve us as the FSA saying, "No, we disagree with the rating that you have given to Company X." It is quite difficult for us to be second-guessing the science of ratings without us becoming the rating agency, and I think all regulators are going to be very, very wary of that. That is why I say, although there is something you can achieve with the oversight of rating agencies, probably the most important thing that is going to occur is a market reaction. I think investors are going to be much more wary in future of relying entirely on what the rating says. I think they are going to be particularly wary of it where this is a rating of a complex and recently innovated structured security with some alphabet soup name. Indeed what I think is actually going to happen is a lot of those things are simply going to disappear from the financial markets and the securitisation model is going to get back to the basics, back to what it was originally meant to be about, which was the much more straightforward and transparent definition of securities to a much greater extent sold through to the end holders of those securities, rather than sliced, diced, structured and traded en route. So, yes, I think it is a legitimate part of it, but I think the most important thing that is going to happen in the role of ratings and in the whole structure of securitisation is a market response which is going to stop people using ratings in an inappropriate fashion.

Chairman: Thank you very much. I think we just about have time to ask Lord Haskins to talk about deposit guarantee schemes, since this is critical for ECOFIN.

  Q14  Lord Haskins: You said earlier with regard to Lehman Brothers that the host country in the future may take firmer action than in the past; that there was a tension between home and host country. That is okay if there is an American bank involved but if it is a European bank, a German bank, do you think that sort of firmer action by the host country is compatible with the single market?

  Lord Turner of Ecchinswell: I think the issue of the interface between the single market and European supervision and European deposit protection is a fundamental issue. At the moment we, as a regulator, are required under single market rules to accept the passporting rights of a branch; a subsidiary requires its own authorisation. As long as something sets itself up as a branch in the UK we are not the prudential supervisor of it. That is true not only from elsewhere in the European Union but from elsewhere in the European Economic Area. So Landsbanki was a branch of an Icelandic bank, not a subsidiary of an Icelandic bank, and we were therefore not the supervisor of the capital adequacy of Landsbanki. We were required by the European single market legislation to rely on the Icelandic authorities as the home regulators of the overall entity, Landsbanki. This was something which caused us considerable concern over the last year, a concern as to whether we were happy with the supervision and the support which was there and also, in retrospect, we have learnt questions about whether ordinary people understood that distinction. It was there on the websites, the websites did say, "This is regulated by the Icelandic Regulatory Authority" but I think we realise that ordinary people tend to think a bank is a bank is a bank and do not make those distinctions and we have to get that communication clear for the future. It does raise this very clear point, if we have a branch from another European country, EU or EEA, which has a banking system which is so big relative to the fiscal resources of that country in extremis—which was clearly the case in Iceland, you had a country which simply did not have the ability, as we do, to prevent the bankruptcy of banks because it simply has not got enough money as a country to do it—should we have to recognise the branch under a single passport? I think you have to go one of two ways. You either have to say, "We cannot have that very straightforward single market rule in terms of the recognition of that branch" or you have to go the other way and either have pre-funded deposit insurance, which makes sure they have put enough money aside outside their small country in order to be able to pay depositors in the event of a default rather than relying on the fiscal resources, or you do have to have some sort of not European supervision but European supervision of supervision; some process of saying, "Is the quality of this small country supervisor up to the quality which we want in order to require everybody else to recognise those branches?" It is not an easy issue I think to know what is the sensible approach. The UK political approach to this, and I think the FSA approach to this and the Treasury and all the political parties in the past have tended to say this in the UK, is, "We want open competition and single market rules, and therefore we support this passporting of branches because it is good competition and services, but we are against European supervision of supervision or European supervisory rules", and I think we have realised that there is a fault line in trying to say those two things simultaneously. It is not obvious which way you go, and one way to go would be to give us the power to say, "We do not have to recognise the branch of a bank in a small European country which does not have the fiscal resources to support it"; that in those circumstances we would have the right to say, "If you want to operate in this country, you are going to have to operate as a subsidiary", at which point we can make sure there is adequate capital and liquidity here. But it is a crucial issue which needs some thinking about.

  Q15  Chairman: Thank you very much. I would love to ask you one more question but I expect you are running out of time.

  Lord Turner of Ecchinswell: I am, yes.

  Q16  Chairman: I feared as much. Thank you very much for coming. It is much appreciated. We might, please, ask you to come back again towards the end of the inquiry when we all know a bit more and we have seen what is happening in Europe.

  Lord Turner of Ecchinswell: Yes, that is fine.

  Chairman: Thank you very much indeed for coming.

previous page contents

House of Lords home page Parliament home page House of Commons home page search page enquiries index

© Parliamentary copyright 2008