CORRECTED TRANSCRIPT OF ORAL EVIDENCE To be published as HC 821-iii

HOUSE OF COMMONS

HOUSE OF LORDS

ORAL EVIDENCE

TAKEN BEFORE THE

PARLIAMENTARY COMMISSION ON BANKING STANDARDS

SUB-COMMITTEE E

PANEL ON REGULATORY APPROACH

MONDAY 21 JANUARY 2013

ANDY HALDANE

USE OF THE TRANSCRIPT

Oral Evidence

Taken before the Parliamentary Commission on Banking Standards

Sub-Committee E-Panel on regulatory approach

on Monday 21 January 2013

Members present:

Lord Lawson of Blaby (Chair)

Mark Garnier

Mr Andrew Love

Lord McFall of Alcluith

Examination of Witness

Witness: Andy Haldane, Executive Director, Financial Stability, Bank of England, examined.

Q154 Chair: May I warmly welcome you, Mr Haldane? I am conscious that this Commission is making exorbitant demands on your time. It is very good of you to come and see us this afternoon, in addition to the other times that you have come to see us and, in a previous period, the Treasury Committee of the House of Commons; but I hope that you will feel that the work of this Commission is worth while. Is there anything you would like to say before we ask questions? We are going to divide this into two parts, without an intermission, because we want to make the best use of your valuable time. We will talk about regulatory matters in the first half and tax, audit and accounting issues in the second. Do you have any kind of opening remark to make?

Andy Haldane: Good afternoon. I very much welcome the opportunity to come back before the Commission to give some more evidence on the more specifics of tax and auditing and regulation. Perhaps the only thing I would say by way of introduction, Lord Lawson, is to note that in the Parliamentary Commission’s initial report, with pre-legislative scrutiny of the Vickers Bill, I found a lot that was very encouraging. In particular, the proposals around so-called electrification of the ring fence were a very positive step forward in reinforcing the Vickers proposals. What was said about the leverage ratio was also very encouraging. Overall, my impression of your interim report was a very positive one.

Q155 Chair: Thank you very much. I am pleased that you feel that.

May I ask a general question, which bounces off your reference to the leverage ratio? It is already clear, certainly to me-I think you have made some remarks in this vein yourself-that Basel III bears a lot of the marks of Basel II. It is not really a different kind of animal. It may be less bad than Basel II, but it has many of the same defects.

Incidentally, I notice that one of the members of the Financial Policy Committee-I think it was Michael Cohrs-said that he was not sure whether Basel III would ever come into operation anyway. Even if it does, it is too complicated and has too much box-ticking, an inadequate leverage ratio, and too much attention to risk-weighted assets; you know how meaningless in practice they are.

If we cannot persuade the rest of the world to change all that, to what extent can we and should we, as a major global financial and banking centre, with banking a higher proportion of our GDP than that of the United States or any other major country, do what we think is right, even if it differs from Basel III? If we can, what would you suggest we should do?

Andy Haldane: I very much share your prognosis regarding the embedded problems, which started at Basel II and have been followed through to Basel III. We made a grave error as regulators in the 1990s, not just in accepting what in the event turned out to be very fragile models as a basis for setting regulatory standards, but, more fundamentally, in moving to a regulatory system that was effectively self-regulatory. We vested in banks the decisions about how much capital to hold, based on their own models; we were asking people to mark their own examination papers. The consequences of that have been in some ways entirely predictable. We have seen banks holding progressively less capital against their exposures.

If I were to compare the situation today with the situation, say, 20 years ago, banks are holding on average half as much capital for their assets as they were back then. Grade inflation is the result of banks marking their own exams. This is a problem for regulators, who cannot really police this complex beast. Increasingly it is a problem for investors in banks, too, who cannot make any sense of the published capital and regulatory ratios.

What is to be done? One thing we can do-this has just commenced-is try to convince the rest of the world that we may have taken a false turn in the road. I have been encouraged to see, over the course of the last few months-perhaps the last six months-that there is an increasing awareness among international regulators that we may have, indeed, taken a false turn. There are moves now afoot within the Basel committee to seek ways of simplifying and streamlining, and to move to a proper, regulatory-rather than self-regulatory-edifice. That may take some time.

In the UK, we can do a couple of things in the interim. First, we can insist on a far greater degree of transparency regarding the inputs into these various models and what they are quite telling us. That greater disclosure is the least that we can expect; at present, they are a complete black box. Secondly, where we as regulators have misgivings about the robustness of these models, we can disallow them; we can require something simpler and more robust to be put in their place. In one or two cases right now, that is just what the FSA are doing, so, to give one example, for commercial property portfolios where the internal models have been found to be subject to very considerable error, the FSA are requiring firms to move to a simpler, more robust, more prudent benchmark. There is no reason why they could not do that across a wider set of portfolios than just commercial property. This could be the imposition of floors, for example-levels below which capital is not allowed to fall, even if the model says that it should. That move towards floors in the capital regime would prevent some of this grade inflation that I mentioned. There is nothing legally that prevents us, the UK, imposing such floors and regulatory discipline on these models.

Ultimately, I would like to see regulators putting far greater weight on the leverage ratio that you mentioned, which is not reliant on any one model, complicated or otherwise. It provides a robust back-stop that is less susceptible to arbitrage. Personally, I would be setting that leverage ratio at a somewhat higher level than that currently prescribed in Basel III. As you know, it is set at 3%, or 33 times leverage. During the crisis, we found that those levels of leverage were often a recipe for failure and, in the longer term, there is a strong case for thinking about a leverage ratio north of that 3%, and one that is not reliant on risk weights. All those are things that we, the UK, can do today to lean against the problems that you mentioned up front, which I think are very real.

Q156 Chair: On leverage, Vickers recommended 4%. You said something like "At the end of the day", but is there any reason why we should not go to 4% straight away? That was, I think, the intention of the Vickers recommendation. Would you favour that?

Andy Haldane: I would. This is something that I have thought a little about personally, but it has also been discussed by the Financial Policy Committee. Our view was that the Vickers 4% was a more prudent number for the ring-fenced bank, where we would want a higher degree of assurance of its safety. The 4% was a step in that direction; there are analytically coherent reasons for thinking that even that may be somewhat on the low side, but it is a better starting point than the 3% that is baked into Basel.

Q157 Chair: You would like to go beyond 4%? What do you think is appropriate?

Andy Haldane: I would not want to shout out a magic number today. I think that 4% was a more prudent, conservative starting point for the ring-fenced bank under the Vickers proposals. Looking into the middle distance, there is a lot to be said for thinking about whether that gives us sufficient assurance of the robustness of banks. The evidence during this crisis was that 4% would not have been sufficient for some of the banks to prevent their failure. Longer-term, there is a programme of work to be done to explore whether even the 4% is sufficiently prudent.

Q158 Mark Garnier: In the past, you have spoken on a number of occasions about the common utility IT infrastructure platform, based, in the broadest sense, on a common platform that can provide account number portability. One of the things I have noticed when speaking to some of the banks, in terms of their IT specialists, is that there is a very hard push back against this, and a great reluctance to go towards a new account number portability type of regime. But when the argument is presented in a slightly different way-first of all, it improves competition; secondly, account number portability and an improved utility platform could help with resolution in the case of a failing bank; thirdly, it could potentially help with transparency for the FPC; and fourthly, it is coming in a reasonable and timely way, given the fact that their legacy systems date back to punch-card systems in the ’50s-do you think that, given all of those together, there is a compelling argument to move to a new utility platform?

Andy Haldane: I do. I am strongly of the view that we need a thorough, objective, arm’s length evaluation of whether such a utility model-lots of models are possible, as you know-might be in the long-term interests of both the efficiency and stability of the UK financial system. One of the attractive things about this proposal is that it appears, on the face of it, to hit two birds with one stone: it helps on the dimension of efficiency, competitiveness and lowering barriers to new entrants. Just as importantly, I think, it would help safeguard the safety, security and resilience of the payments mechanism. We do not have to think back too far-just to last year-to think of the problems that RBS encountered in its making of payments in the UK, in Northern Ireland in particular. That, at the end of the day, was down to that self-same legacy system problem.

On the stability front, I also think that lowering barriers to new entry would do a little bit to chip away at the "too big to fail" problem. The inertia embedded within current accounts is an ingredient of the "too big to fail" problem, and competition is one prospective solution to that problem. A central utility could help in that endeavour.

I am aware, in saying all this, that the precise model needs to be articulated. A number of models are possible. Some are pretty straightforward: they would just be about making common a language by which accounts talk to one another. Others are more ambitious-more Railtrack-type options, where you have a central utility into which others then plug and play. Those should all be evaluated properly. I think this is a place where the UK could put itself on the front foot and could be seen as being genuinely world-leading in the payment services it may provide. The legacy position is not positive; people are locked into primitive technologies. Now, as you mentioned in your introduction, is as good a time as any to take that big leap forward. One reason I am hesitant about the proposal on the table from industry right now is that it is not a great leap forward; it is a small step forward.

Q159 Mark Garnier: When you say that, do you mean-

Andy Haldane: I mean the seven-day account portability proposals from the Payments Council that come from the Vickers proposals. I think they are incrementally and directionally right, but really, for me, lack a degree of ambition. We, as the UK financial services sector, can and should aim higher. IT is fundamentally transforming the payments services industry, in front of our eyes, day by day. It would be very nice if we jumped ahead of where the frontier of technology was.

Q160 Mark Garnier: I am sure you agree that the seven-day payment does not necessarily address the problems of transparency in the system, or resolution, either.

Andy Haldane: No, it doesn’t, especially in respect of the big banks.

Q161 Mark Garnier: The banks are arguing that this should be given a chance to bed down and for people to see it, and that it should be measured by its success, but they then go on to say that nobody really knows quite what the measures are to prove that it has been successful or otherwise. Have you had any thoughts about their proposition that it should be allowed to run for a bit, and secondly, what a measure of success actually is?

Andy Haldane: I can see the case for putting in place a plan for getting to the seven-day deadline. I would not want to do anything that retarded progress in that direction, because it does help a little bit for resolution of the smaller and medium-sized banks. At the same time, I would not want to use that as an excuse for not thinking ambitiously about the long-term model and for not putting in place some of the foundation stones for that.

What would be the success criteria for such a project? I think an absence of complaints by new banking entrants about their access to the payments system would not be the worst such criteria. You know that there are some such complaints right now from those new entrants about the cost and difficulty of plugging into payment systems and often having to do it indirectly via existing banks. Ultimately, I think that we ought to see the charges that are levied on customers for the broadly based provision of banking services coming down, if such utility were to deliver the efficiencies I mentioned earlier on. I think we would see much greater account switch than we have seen historically. That has picked up pace over the last few years, but generally speaking, there is still a hefty degree of in-built inertia. There is nothing like as much switch as we see, for example, between mobile phone producers or utilities providers. These are top-of-the-head suggestions, but I think it would be too difficult to define a set of objective criteria for what counts as a contestable market. We are some way short right now of having a market for current accounts that is truly contestable.

Q162 Mark Garnier: But you wouldn’t wait for a year or two after this thing comes into place before we start having a proper commission to look into the feasibility of an alternative system.

Andy Haldane: No, there is absolutely no reason so to do. Having looked at this payments industry now for a good number of years, there is a heavy, in-built inertia within it that certainly cannot be justified on operational or technological grounds. In some ways, the UK payments industry did a good job in delivering the faster payments service that now operates. But let us be clear: first, that was done through gritted teeth, as a result of a hefty regulatory push, and secondly, it was by no means fast in its delivery. We should be aiming higher, and if that requires a regulatory or indeed political nudge, I think we should provide it. The payments industry internationally is moving forward at a rate of knots. If we sit still for 12 months, we will be even further behind.

Mark Garnier: That is very interesting. I am slightly conscious of time, so I have one question on a slightly different subject, if I may.

Q163 Mr Love: Could I ask something on this before you change the subject? It is only one point, which relates to two aspects of the platform that we have not discussed so far. One is those civil libertarian concerns that there might be about amassing all this information in one place-how do we address that? The second is, of course, the security aspects: if all that information is held and something was to happen, we could really be in difficulties, so how do you address those two?

Andy Haldane: Those are both perfectly legitimate concerns and considerations when we are building this, if we were to build such a centralised platform. On the first point, I think it is possible to build a degree of redundancy into a system, even if it is centred around a given model-to have a fall-back site, so in the event of the central platform hitting problems, you can revert to the back-up site, or the back-up model, to pick up the slack. In other industries that have a similar single-point-of-failure problem, the solution that you typically find is having one or often two back-up sites to which you can revert if the first one were to fall over.

On your second point about security and safety, it is absolutely the case that we want to set as high a threshold as possible for the security and resilience of this thing. I think that is made easier by the fact that it is a single entity. One of the difficulties right now is that we have a very distributed and diverse set of systems. Basically, every bank has its own bespoke system, and it is quite difficult to evaluate across banks what is their degree of safety and soundness. At least with this common central mechanism, there is a way of ensuring that the whole system maintains a high degree of safety and soundness. We want to set the very highest set of safety and soundness standards. Both of your considerations are completely fair challenges.

Q164 Mark Garnier: I have one quick question on remuneration. In recent speeches, you have talked about banks switching their model of remuneration from a return on equity to a return on assets, and about that being a much more effective way of demonstrating that, in terms of the performance of senior managers and the bank and also in terms of bonuses. Do you want to expand on that and fill in more of the thoughts?

Andy Haldane: I am happy to. On remuneration policies and remuneration practices, the emerging evidence base is that they did have a significant bearing on pre-crisis risk taking. I know that that sounds completely obvious now, but for a long time it was hard to forge a link between the two. I think that the link has now been forged and underscores how important it is to have remuneration policies and remuneration codes that are sufficiently prudent. Progress has been made on remuneration codes internationally and domestically over the past three or four years. For my money, they still fall somewhat short of the standards of prudence that I would wish to see embedded to alleviate risk-taking.

Let me mention three specific areas where those remuneration codes can and probably should be tightened up, relative to where they are right now. The first-the one you mention, Mr Garnier-is the set of performance metrics that we use when setting salary levels. It is deeply irresponsible to be using performance metrics that fail to take adequate account of risk. The reason why the return-on-equity metric that you mentioned is a problem is that it can be easily gamed by risking up the system through leverage. I think that in the remuneration code, we should require reference to a set of performance metrics that are not susceptible to that problem; they are risk-adjusted metrics, unlike a return on equity.

The second place where the remuneration code can and should be tightened up is in respect of the horizon over which performance is evaluated. Right now, there are provisions in the code to allow the deferral and clawback of bonuses. Typically, those clawback or deferral periods are roughly three to five years. For me, that is far too short to capture the cycle in credit, the cycle in the financial sector. We had roughly a 20-year boom in the run-up to this crisis, so measuring performance only over a three or five-year window is far too short.

Q165 Mark Garnier: On that point, I do not know whether you noticed the evidence that was given by the UBS executives a couple of weeks ago. There was the so-called trader A who generated $260 million in profitability from LIBOR, but he stayed with the bank for just three years and I think went to Citigroup after that. I mention that because he was on a three-year deferred bonus. Presumably-we need to take evidence on this to confirm it, but you may have an opinion-he would not have moved to another firm if the deferred bonus that he was losing was not compensated in one way or another. Presumably, the deferred bonus, in terms of crystallising it by moving shop, can, arguably, encourage worse behaviour by encouraging more people to move around. If they feel they have a problem coming round the corner, they can get someone to buy out their deferred bonus and move on to the next business before the previous time bomb goes off. Is that fair?

Andy Haldane: I can see this will need to be designed in a crafty way to prevent that gaming of the system. I have not given thought to how precisely that would happen. The example that you give, if I heard you correctly, would require the firm to which this person is moving to buy out the amount forgone in moving prematurely. It is quite hard to lean against that in individual cases. If all firms are subject to the same rules, it becomes a little less likely. None the less, it is hard to lean fully against that.

On the issue of time horizons, deferral and claw back, there is a case for horizon lengthening. There is a case-

Q166 Chair: To what? How many years?

Andy Haldane: I would be looking somewhere close to five to 10 years, rather than three to five. Actually, I find the HSBC model very attractive, where you are effectively locked in until resignation or retirement. One reason I like that is that it gets you as close as you possibly can get to the old partnership and unlimited liability model, where you are on the hook up until the point where you walk out the door. There is room for improvement there.

For completeness, the third area where there is room for improvement is on the form of the instrument used to pay out to people, where we have moved somewhat away from cash and towards shares. There is a case for thinking about whether we would not wish to remunerate to a greater degree in debt instruments of various kinds, because then you get less of the upside from gambling and risk-taking, but still suffer the downside in the event of things going wrong. Sometimes the incentives created by paying in shares are every bit as great as the incentives created by paying in cash. Debt or subordinated debt or bail-in debt or CoCos are ways of adjusting incentives in positive ways.

Q167 Chair: I want to ask two quick follow-up questions on remuneration, because it is an important area. You have indicated that you would like to see bonuses paid in subordinated debt, rather than in the form of equity. You have also indicated that there should be a longer lock-up period. Do you or does the PRA have the power to impose that on British banks or banks banking in Britain? That was question one. Question two is: if it does not have the power, should it have the power? Should we legislate for that, or does it have the power already? Finally-a sort of question 2(a)-if we were to go that route, do you think that bankers would vote with their feet and go elsewhere?

Andy Haldane: To your first point, I am not a legal expert on this. It is certainly the case that the UK is bound by what is a European-level remuneration code. That does set the broad parameters within which the UK must then comply.

My understanding is, however, that there is discretion. There is room within that to be more prescriptive. How much room, I would probably need to go away and explore a bit further. My sense is that there is some room, but not too much, to be, if you like, super-equivalent. I would hope on this front, actually, that we might ourselves be able to convince the rest of Europe that a somewhat tighter remuneration code might be in their interests as well as that of the UK. Of course, it is some of the other countries in Europe that have been keenest within the context of the capital requirements directive to narrow down-to reduce-the amount that is paid in bonus relative to fixed salary. So my sense is we might be able to make some headway, even within Europe, in reforming the existing remuneration code. On top of that, I think we have a degree of national discretion. Quite how much, I’d probably need to go away and check for sure.

Q168 Chair: Could you let us have a note about that?

Andy Haldane: Yes, I’d be happy to.

To what extent would this lead to financial sector workers upping sticks and off? Well, despite the fact that salaries within the financial sector have come down somewhat over the last two or three years, there is still a very healthy wedge between how much your average investment banker is paid and how much a similarly skilled lawyer, doctor or accountant is being paid, so, although there has been some southward movement in salaries, the levels of those salaries still provide a very strong incentive, I think, to stick rather than to twist.

Actually, I think that some rebalancing of the labour force might even be-could prospectively be-in the long-term interests of the UK.

Q169 Chair: I am sure it would, but the question is whether those bankers leave the UK altogether. That is the issue.

Andy Haldane: I think this does speak to trying to make inroads to alter the remuneration code as it applies at a European-wide level. Of course, that doesn’t stop people moving to New York, and we should try internationally too to put that international code perhaps on a stronger footing. For me, given the starting point, this isn’t a risk I would weigh particularly heavily just at the moment; but it certainly bears some watching.

Q170 Mark Garnier: Just a quick supplementary, if I may, on the common utility platform. You talked about the fact that it would be very good if we, as a country, were taking the lead. Do you think that would also give us a competitive lead as well, in terms of international banking, if we had a common utility platform?

Andy Haldane: I think it could. Some of the bigger benefits, I think, would probably accrue to domestic residents here, in the payments service that they were being offered; but I don’t doubt that if this could be shown to work well it would put UK institutions on a strong footing internationally, when making and taking international payments, which is a big part of the business of some of our bigger banks.

Q171 Lord McFall of Alcluith: On the point Lord Lawson was making-just to take that forward, Mr Haldane-we have seen over the past few decades the best and brightest being attracted to the financial services industry, and some would say that that isn’t good for the economy as a whole, generally; but maybe the reason why the best and brightest are attracted to it is that we have complexity at the heart of our system. We have just come from the panel looking at the mis-selling of PPI, where I was taken by a comment made by one of the witnesses that the regulators, and the FSA in particular, have been captured, cowed and cornered, in that they will always be chasing the banks. Any idea we have of a regulator or politicians changing the culture of banks when we have the present system of complexity and incomplete information is really pie in the sky. From our perspective, we do not understand the incentives that are driving these people.

How do we get to a system where we flip the responsibility? Rather than the regulator chasing the banks, as we have seen in the case of PPI, where over an 18-year period the banks were hitting back, threatening legal action and doing whatever else, how do you have a strong regulator with simple rules, where we understand the incentives, and they have not just to comply but to obey? The word compliance indicates an agreement. How can we get to a situation like that? Because otherwise, we politicians and regulators are kidding ourselves that we have this laundry list of problems that we are going to fix.

Andy Haldane: I think you have raised a tremendously important issue. I do not know that I have a magic bullet solution to it, Lord McFall, but let me say one or two things. First, I very much buy your prognosis that unnecessary complexity is a recipe for rent seeking and ripping off. We have plenty of evidence of that having been the case, not just in the selling of products to customers, but, going back to my earlier discussion with Lord Lawson, in the pulling of the wool over the eyes of the regulators about how much risk is actually on the balance sheet, through complex models. How to lean against that, from a regulatory incentive-that is a theme that might run through what I am going to say today. As you say, it is about providing a different set of up-front incentives to the bank to do the right thing. How can we as regulators begin to do that? How not to do it is to engage in very intensive and intrusive regulatory scrutiny of every product that passes or every asset on the balance sheet. We have been down that road-it is what Lord Lawson called the box-ticking culture in his introductory remarks-and it has not served us well.

What are the alternatives? I am attracted to the idea of what you might call a spot-check approach to supervision and regulation, which is not to have an army of regulators and supervisors peering down every rabbit hole, looking for problems, but instead to pick an asset or product at random, through a spot check, and assess whether it is doing all that it says on the tin. Is it really too complex to be sold to the public? Is the asset really doing what we think it is doing. In the event that it is found not to do what is said on the tin, I would impose a pretty sharp set of incentives on the most senior level of management in the firms.

I think back to the approach to supervision and regulation when it was last done in the Bank of England. At the centre of that approach was a conversation at the highest level between the Governor of the day and the chief executive or chairman of the bank. That conversation was effectively a spot check on the adequacy of the senior management in understanding its business. In the event that the senior management was found to understand its business, that was fine. In the event that they were found not to understand it, there would be a change of management. The sanction would be felt by the highest level of the firm.

I think that an approach of that general type, updated to the 21st century, could have some relevance today, and that the supervisor’s role would not be to peer into every nook and cranny but to selectively-on a spot-check basis-assess whether the assets and the products are fit for purpose. If they are not found fit for purpose, ultimately the sanction is felt by the CEO, or the chief risk officer, or the chief operating officer.

Q172 Lord McFall of Alcluith: I was discussing this with someone else in the last week or two, and about something along these lines-I am just thinking off the top of my head-whether you call it a health check or a state of the union check. If we have individual accountability what we have seen with PPI and other aspects, such as LIBOR, is that the senior management knew nothing about it; we had UBS executives in here, not knowing anything about it. Is there a case for saying then that every year or so we have the chairman, chief executive or whatever, with the regulator-the Bank of England, the FCA or whatever it is-and there is a thematic discussion taking place about the bank, in terms of its products, culture and standards, and therefore people know, in a public sense, that there is that accountability to the regulator? Is that something that is worthwhile exploring?

Andy Haldane: I think it is worthwhile exploring. However, I think it needs to be accompanied by a very clear understanding, on the part of both the regulator and the regulated firm, about what the ultimate sanction is. Not knowing cannot be a legitimate excuse. If it was made clear that, whatever the product or whatever the asset, if it is not doing what it is meant to be doing, the sanction will be meted out at the highest level of the firm, and that those incentives would run down the core of the firm from the top, that would help. I think that if the CEO, or the chief risk officer or the chief operating officer, knew that their job was on the line, their behaviours would then rub off all the way down the organisation. You would find fewer of these products being sold in the first place; you would find fewer of these assets finding their way on to the balance sheet.

A Roosevelt-type approach of speaking softly, but carrying a big stick, would leave us much better equipped for next time. Right now, we have a high-cost, low-incentive approach to supervision and regulation. It is high cost, because there are lots of supervisors and regulators around; it is low incentive, because the sanctions are never imposed. Everyone is "fit and proper" all of the time.

I think that the regime we have been talking about would be lower cost, selective and "thematic"-to use your word-but it would carry stronger incentives, because the sanctions would be imposed at the very top and the incentives for everyone to do the right thing would trickle down.

Q173 Lord McFall of Alcluith: Sir David Walker’s report mentioned that with bank boards the problem relates more to behaviour than to organisation. It is obvious in the panels that I have been on that there is a deficiency in the standards and culture of the companies, and it is to do with culture and ethics: culture being the behaviour; and ethics being how they resolve the conflicts of interest in that behaviour. I have asked a number of participants if the banking industry can change and do it on its own, and largely speaking-whether it is for effect, because I have asked the question, or not-they have said that they can’t do it on their own. So, if they cannot do it on their own, there must be a need for outside scrutiny to assist them in that.

One of the suggestions I had was a standing commission, independent and working alongside the banking industry, so that it could be complementary to what the industry is doing and complementary to what the regulator is doing, and not interfering. But that would hopefully satisfy in many ways, or start to satisfy, the societal obligation of banks. Again, is that a runner, or is it something that you would consider?

Andy Haldane: Without having given it a lot of thought, it sounds-on the face of it-attractive, Lord McFall. It has some parallels, I think, with the proposal that the BBA came forward with during the last couple of weeks. I am not sure that there is any one single solution that will change all of this. I think that your idea is one well worth pursuing. I would hope that the Vickers proposals for structural reform will have a bearing on culture and risk-taking.

There is more that could be done on the regulatory front, through auditing practices, through tax practices. It has to be a shared endeavour to try to shift the set of incentives that exist throughout the system. There were so many places, pre-crisis, where the incentives were in the wrong place that we should seek to correct them one by one. This would be an important element of that.

Q174 Lord McFall of Alcluith: Too big to manage, too complex, too big to fail, too big to prosecute: too big to regulate? Is there a regulatory argument for smaller banks?

Andy Haldane: Yes.

Q175 Chair: May I follow that up? Obviously we would all agree that there is no single, silver bullet. There are a number of things that it might be sensible to do. The fact that none of them is a silver bullet is not an argument against doing it. Following on from what Lord McFall has just said, we are now, not out of deliberate policy but as a consequence of events, in a position where the state has an 80% share in the Royal Bank of Scotland and a controlling interest in the Lloyds HBOS banking group. Do you think it would be in the public interest, in the interest of having a healthier, stronger banking system in this country, if the Government were to, certainly in the case of the Royal Bank of Scotland and possibly in the case of Lloyds, as a means to an end, acquire 100% so that they can then have a break-up?

The break-up would be in two ways. First, the classical good bank, bad bank break-up. The bad bank would be a run-off of the bad bits which, as the Governor has pointed out on a number of occasions, are substantial. Nobody knows how big, but certainly the so-called forbearance makes banks’ balance sheets something of a fiction. So there would be that good bank, bad bank division, and possibly splitting up the good bank so that there was more than one. So you really have more competition. You address to a considerable extent the too-big-to-fail problem, which you mentioned and which Lord McFall was mentioning. This is something that we can do. It would probably make privatisation quicker too because it would be easier to sell off to the private sector these smaller banks. Do you see that as a desirable way forward?

Andy Haldane: Taking your last point, Lord Lawson, as a starting point, I think the task here could be put as, what would it take to get the two partially state-owned banks back into the private sector, given their starting point, to rehabilitate them? What actions would be necessary to put their balance sheets in a position where they could attract private capital and stand on their own two feet? As we have said at the FPC, we think, looking across the system, that there is a need for action, and, for some banks, that action may take the form of a restructuring of their balance sheet or the selling off of assets to put them in a position whereby private capital would re-flow. In the case of the two partially state-owned banks, they are both in a position where it would be difficult for them to attract private capital without some rehabilitation, some restructuring of their balance sheet. It is probably not for me to say today quite what form that would take. Were you to do that, what you might hope would emerge from it would probably be a somewhat simpler, slimmer, good bank, purged of its bad or suspect assets, possibly in a ring-fenced form and therefore able to serve the needs of the wider economy.

Such a slimline, simplified bank, perhaps as part of a ring-fenced group, would have a realistic prospect of reattracting private capital and having a life of its own, getting the Government off the hook for providing more capital. The trajectory you set out has some attractions. It is probably not for me today to go into detail about what form that takes.

Q176 Mr Love: To follow on from that, Michael Cohrs in a previous session said that the banks were too big. I think you are on record as suggesting that the benefits of size reach an upper limit. You have talked in response to Lord Lawson’s inquiry about RBS. If we decided as a matter of policy that the banks were too big, for the reasons that you and Michael have stated, how do we go about the process of bringing them down to the size that you would envisage as being in their interest?

Andy Haldane: There is quite a lot already in train to try to get our arms round these mega-banks and that is still working its way through the system. One thing we intend to do internationally over the next few years is require these bigger banks to hold bigger slugs of equity than the smaller banks, recognising their system-wide importance. Those so-called systemic surcharges are not yet in place. They are directionally very sensible. Personally I think the extra chunks of capital being required will not by themselves be sufficient to solve the too-big-to-fail problem.

There is a second strand of work. As you know, we are requiring all banks, in particular the biggest banks, to put in place plans to enable them to be wound down safely. It is early stages for putting in place such plans. We are not yet at the point where such plans exist, where we could credibly say-hand on heart-that we would know how to wind down a big bank in the UK or anywhere else right now. However, it is an aspiration towards which we should keep heading over the next few years, and hopefully that will make some headway.

The third thing is Vickers, Volcker and Liikanen in Europe. That, too, will hopefully, make some headway in reducing the implicit subsidy to the biggest banks, raising their cost of funding and causing them to shrink somewhat their balance sheets.

All three of those go in the right direction. None of the three is yet in place. We should see how far that takes us. If it does not take us sufficiently far, we are into the realms of thinking about rather more simple but robust means of keeping a cap on these banks. One that has been discussed internationally, certainly in the United States, over the past few years is placing an explicit cap on size, whether measured in terms of total assets or shares of GDP in a country.

There will be a lot more work to do to make such a plan truly operational but, if the three initiatives I mentioned-capital, resolution, structure-did not take us sufficiently far, that would be the next step.

Q177 Mr Love: Let me just ask a question in relation to that. Michael Cohrs was very much of the view that we ought to set a size limit based on the American experience. Let us assume that the issues you talked about have a slightly longer time scale. Currently there is an investigation going on into personal current accounts, and it is possible that there will be a Competition Commission referral. We were told by a former director of the Competition Commission that they have the powers, and we saw that with the British Airports Authority. Is there any merit to using that mechanism to address the issues of size?

Andy Haldane: I think competition policy, whether it is policy in the legal sense or more in the sense that I discussed with Mr Garnier earlier about the proposal, can and ought to play a role in eliminating this too-big-to-fail problem. If we have greater entry and a more competitive, contestable banking market, that ought to chip away at some of the subsidies I mentioned earlier. That could include formal inquiries and it could include lowering regulatory barriers entry, which I think will happen. It might include-I mentioned this in previous testimony to you-placing, as the US do, an explicit cap on the market share of different institutions. In the US, there is a 10% cap on the deposit market share. All those serve some double duty. They are there for competitive reasons, but may, as a by-product, have some benefit in chipping away at too big to fail.

Personally, I am hopeful that the banking market is moving. There are some new technology companies moving into the banking space. Things like peer-to-peer lending and crowd funding are disintermediating banks to some degree. They are pretty small scale right now, but the process of technology generating creative destruction, which has been true in lots of industries, is beginning to nibble away at banking too. That is something we should all applaud.

Q178 Chair: We must move on soon to the important questions of the tax system as it affects banks and the currency system. Before we do that, I have just one question, which to some extent links to what you were saying earlier. We were set up to look into the problems with banking standards and banking culture in the light of the LIBOR scandal. The LIBOR scandal was, par excellence, a scandal of proprietary trading. The culture that you have mentioned is one of prudence in the service of clients, whether they are individuals, SMEs or larger corporations. In the light of our terms of reference and the need for prudence in banking-and you mentioned Volcker a few moments ago-do you think it would be helpful, in the interests of what you would like to see and what we would like to see, to prohibit banks, or at least banks in a ring-fenced group or groups that have ring-fenced banks in them, from engaging in proprietary trading? If you do, do you think it is workable?

Andy Haldane: The last part of your question goes to the heart of the issue. As a matter of principle, my mind is very much where yours is, Lord Lawson. In principle, there ought to be a sharp distinction between directional bets made on one’s own account and the provision of liquidity services to customers and clients. Any commingling of those two sounds as if is in some ways best avoided.

The practical question is, is it possible to differentiate those two things as cleanly as I just have in words? Having looked a little at how market making works these days, I can see the awkwardness in differentiating those two sets of activities. In some ways, what I rather lament is that we have moved away from the sort of market-making model that existed 10 or perhaps 20 years ago. Back then, market making in many markets was a designated activity, and the job of the market maker was to be there to supply two-way prices, whatever the weather. In some ways, what a designated market maker was permitted to do was charge a spread, which would give it a profit in normal times and act as an insurance premium in stress times, thereby enabling it to provide liquidity services and make markets, whether the weather was fair or foul.

We have progressively, over the course of the last 10 or so years, moved away from that model to one where, effectively, almost anyone can be a market maker; anyone can, in principle, make two-way markets. That has had the benefit of shrinking the spread-the bid-ask spread-but one consequence of that is that, in situations of stress, we find too few participants willing to make two-way prices. The market-making community has become somewhat fair-weather, which means we have had these episodes, over the last few years, of liquidity being entirely absent from markets. That was one of the reasons why we had the infamous flash crash in the United States in 2010, and we have had several such mini-flash crashes since.

I say all that as a slight digression to underscore the fact that we have moved to a market-making model which is radically different from that which we had 10 or 20 years ago and which conflates the act of making markets with the act of making money on a proprietary basis. I see no realistic prospect of rolling that back very quickly, which leads us to the problem that it is very difficult in practice to differentiate these two sets of activities. That is a problem the US authorities have found when they have tried to make practical the Volcker rule. As to what they have come up with, I am genuinely not sure whether it will work or not. Were it me, I would like to take time to see how the US experience pans out with their Volcker rule. Down the line, if we have a review of Vickers in four or five years’ time, as you proposed in your earlier report, we might see whether imposing the Volcker rule on top is a practical proposition, or whether the US finds that the problems I have mentioned are insuperable.

Chair: Thank you. It is a fascinating question, and that is a helpful answer.