Treasury Committee
Oral evidence: Independent review of the Co‑operative Bank, HC 2084
Tuesday 30 April 2019
Ordered by the House of Commons to be published on 30 April 2019.
Members present: Nicky Morgan (Chair); Mr Steve Baker; Colin Clark; Mr Simon Clarke; Stewart Hosie; Catherine McKinnell; Wes Streeting.
Witness
I: Mark Zelmer, Author, Independent Review of Co-operative Bank Supervision.
Examination of witness
Witness: Mark Zelmer.
Q1 Chair: Mr Zelmer, thank you very much for coming to the Treasury Select Committee session this morning. We appreciate that you have flown over especially from Canada as well. Perhaps you can introduce yourself and say a little about your background in this sector, before we get the questioning under way.
Mark Zelmer: Thank you very much for inviting me to be here. It was a real honour and pleasure to be asked to conduct the review of the prudential supervision of the Co-operative Bank. My background is more than 30 years of experience dealing with financial sector policy matters of one form or another, most of the time with the Bank of Canada. I spent a few years with the International Monetary Fund and the last five years as the deputy superintendent of financial institutions in Canada with what would be the Canadian equivalent of the Prudential Regulation Authority.
Q2 Chair: I thought you were going to say it was an honour and a privilege to be in front of the Committee; that is not what people normally say. Perhaps we could start with the bigger picture of the report you have prepared. Based on your experience, could you give us a feeling for how you would describe the regulation of the Co-op in the period that you reviewed?
Mark Zelmer: The bottom line is it really was a work in progress. To be perfectly blunt, prudential oversight in the United Kingdom at the beginning of the period, really the beginning of the crisis, did not resemble anything that I would call prudential supervision, at least back in my home country. It was a primitive state in terms of supervisory practice relative to what I was familiar with, but it was rapidly evolving through the period in the wake of the crisis, while at the same time they were in the midst of fire-fighting on a number of fronts, as you folks know far better than me. They were dealing with the Co-operative Bank issues at a time of extreme stress in the United Kingdom, in the midst of a major reorganisation. It would have been a rather chaotic environment to be trying to build a prudential supervisory business while, at the same time, dealing with some very troubled institutions.
Q3 Chair: We will come on during the session to explore how some things have changed since—lessons learned from this and broader issues as well. Your inquiry was based on a direction from the Treasury. Did you feel that direction asked all the right questions? Did it allow you to cover all areas that you felt needed to be covered? Was there anything you did not get a chance to look at during the review or are there any gaps that the Committee should be aware of?
Mark Zelmer: No, I thought it was a complete direction, to be honest. I felt I was able to make the observations I wanted to make, and I had also set aside the final chapter of the review as a means to bring forward thoughts that came to mind in the context of doing the review that went, strictly speaking, over and beyond the points of the direction. I felt that I had the scope necessary to explore the issues at hand.
Q4 Chair: Are there any particular areas that you feel this Committee or the Treasury should look at further, and things that remain undone?
Mark Zelmer: The point of the recommendations I made in the report, as well as the observations in the final chapter, is to signal the areas that I feel the authority should pay attention to, going forward. I deliberately kept them at a reasonably high level, because I do not live in the United Kingdom, so I do not know all the intricacies of how things work here. I wanted to give some flexibility, so that how one pursues the recommendations and observations could be tailored to the United Kingdom context and is not overly influenced by my Canadian background. They were meant in the spirit of what I perceived to be the gaps that exist today and issues you might want to focus on going forward.
Q5 Chair: Did you get all the assistance that you asked for? Were there any people or organisations you wanted to speak to as part of the review who you found it difficult to contact or speak to at all, or get answers from?
Mark Zelmer: No, I was fortunate. The bank was extremely helpful in giving me the resources that I felt I needed to conduct the review. It gave me access to the individuals who I needed to talk to. I would have been jumping up and down pretty fast if I felt I was being hindered in any fashion.
Q6 Chair: The PRA and the Bank have responded to your report. Were there any parts of their response that you found disappointing or incomplete?
Mark Zelmer: I consider the response to be in the spirit of a holding response, saying basically, “We received the report. We are sympathetic to many of the recommendations and we are thinking about it.” That is how I would characterise it. Rightly or wrongly, that is how it came across to me.
Q7 Chair: You would expect or hope for a further, more detailed response to the recommendations or even, going further, “We accept this, and this is what we are going to do to change things”?
Mark Zelmer: I would suggest that it would be reasonable for you folks, in your role, to expect them to give you some concrete ideas of what they plan to do and over what timeframes, bearing in mind the other demands on their time.
Q8 Chair: Before I hand over to Steve, is your contact with the PRA over now?
Mark Zelmer: Yes, it is.
Q9 Chair: That is it; there is no further involvement on your part?
Mark Zelmer: No, there is none.
Q10 Mr Baker: Good morning. The Bank and the PRA agreed with your recommendation that they should continue to evolve their stress tests to encompass a broad range of risks to which banks are exposed. What risks in particular are you keen that they evolve to cover?
Mark Zelmer: Stress-testing is more of an art than a science, at this stage of its existence. The road has been well-trodden in looking at traditional balance sheet risks, such as credit risk or exposure to interest rates. The risks arise in the operations of a bank and related issues about conduct, which you folks have seen many incidents of in recent years. That terrain is less understood, not just here in the United Kingdom; it is a global issue.
Frankly, UK stress-testing practices are very much among the leading edge of global practices. My thoughts in this regard are not to suggest any shortcomings in the UK relative to its peers. Quite frankly, there are none. I suggest it is at the frontier. It is just that trying to model, in a statistical sense, the risks associated with conduct issues or, more generally, with the basic operations of banking is at a much earlier stage of development. I feel those issues are important, and it is important that they continue to carry out work to flesh out their stress-testing in that regard.
Q11 Mr Baker: Could you give us some indication of the scale of those risks that are in early-stage development? I think you said operational risks and so on.
Mark Zelmer: A good example was in the experience of the Co-operative Bank, where there were several hundred million pounds of losses associated with conduct issues. Those are one example of the stresses that can happen that, as shown with the Co-operative Bank, can be quite material to a bank, going forward. If you look at the various conduct issues associated with the LIBOR scandal and others, you are talking of billions of pounds of fines and penalties that can come back to haunt banks if they do not manage these risks properly.
Q12 Mr Baker: To what do you attribute the earliness of the development of these factors? Is it because the Bank concentrates on factors that are easier to quantify, or are there other factors at work?
Mark Zelmer: It is very much the case that stress-testing around the world has focused on what people perceive to be the traditional risks of banking—extending credit, and managing the matching of assets and liabilities of the bank. Because of the minute-by-minute data that people rely on to do stress-testing, with some reasonable statistical tools, it is natural to go there first. The operational issues tend to be the kind of things where you may have lots of data points, but they tend to be in a more mundane space. The really big issues that could threaten the existence of a bank are few and far between, so it is hard to model them in a rigorous mathematical way.
Q13 Mr Baker: You have pre-empted a question I want to clarify. Is it that the risks that are less well developed, such as conduct and operational risks, do not correlate to the major systemic economic events for which the Bank is primarily stress-testing?
Mark Zelmer: They do not naturally, no. They often appear at the same time, largely because—to quote somebody whose name I cannot remember off the top of my head—when the tide goes out, you see a lot of crap. Often it looks like they are correlated because they are uncovered in a crisis, but the issues are often lurking in good times and bad. They do not necessarily correlate with the cycle.
Q14 Mr Baker: You noted that a number of factors “are always likely to make stress test results more benign than outcomes”. You then recommend that the authorities “consider how best to incorporate the inherent uncertainty that would prevail as a stress scenario unfolds in real life”. How easy is it for the authorities to incorporate that uncertainty? Presumably it is very difficult.
Mark Zelmer: I think it is very difficult. In their response they made some commentary on this issue, but I am focusing on the fact that, when you give banks a stress scenario to run, they see the whole scenario. That is not how real life unfolds. In real life, you go point by point, and do not know how the rest of the event will unfold. I would submit that the behaviour of institutions in practice is different relative to what they think it would be by running a stress-test exercise, because they do not actually know how things will unfold. Yes, you try to make the best inferences you can based on past experiences, but do not underestimate the uncertainty that would exist in real life and how it could potentially affect their behaviour. I suspect it would make the second-order issues that people talk about far more important than we realise in practice.
Q15 Mr Baker: Given what you have told us, what is the role of assessing management capability in a stress test?
Mark Zelmer: I make a clear distinction between stress-testing the vulnerabilities that you see in a bank’s operations or balance sheet versus what I would call the mitigants that are applied to manage those risks. Let me give you an example. The risks or vulnerabilities are things like credit risk, a bank’s exposure to movements in market prices and the like, but how those risks are managed is a function of the quality of management, the quality of governance in general within the institution and the quality of their controls. To my mind, you do not try to embed them into the stress-testing framework. You use the stress-testing framework to illustrate what you think are the vulnerabilities, then you use a separate process to evaluate the capacity of the bank to manage the risks. The combination of the two should inform what you feel is the appropriate amount of capital or liquidity that the bank should be carrying.
When I was working full-time, I was quite nervous about the prospect of publishing stress-test results for individual banks. I understand the context of why it was done within the United States and the European Union but, from a Canadian perspective, I was concerned about publishing the results. I was concerned they could tie the hands of regulators. The more capital requirements appear to be publicly linked to stress-test results, the less the degree of freedom you have to use capital and liquidity requirements to effect change to management practice and to governance and controls. To my mind, there is clearly a role for stress-testing, but I would be very cautious about drawing too close a link between stress-test results and capital, liquidity or other prudential requirements. You need to be able to use those requirements as a lever or club to encourage the institutions to address any shortcomings that may exist in their capacity to manage the risks.
Q16 Mr Baker: You have made it clear that you see the regulator adjusting capital requirements based on its assessment of what the management of the institution is capable of. That is something we will need to reflect on. Can I turn to complexity and incentives? You argued, somewhat worryingly, ”The PRA should continue to pay close attention to any attempts by banks to circumvent regulatory and supervisory requirements and focus on the economic substance of transactions, not their accounting treatment or how they are funded”. You possibly will not know that I have gone on about this issue in the past for some years. Do you think incentives have increased for banks to circumvent those requirements, particularly in relation to the way they use accounting treatment?
Mark Zelmer: Yes, I do, because banks’ perception of risks and the public sector’s perception of risks are two different things. If you ask a bank what the costs of a crisis were for them, they would think in terms of what the impact was on their own institution. They would not think about the impact of the crisis in terms of job losses and the disruption to society more generally. The consequences of bank stress are much broader than fall within the narrow remit of the institution, so it is not surprising that a bank’s perception of how much capital it should carry to manage the risks it faces is likely to be lower than what the public authorities think is appropriate, bearing in mind the potential economic fallout from a crisis.
Over the past 10 years with capital requirements on the rise in the wake of the crisis, there was understandably an impact on banking profitability in some countries. To the extent that the banks feel that the capital is not warranted by their own perception of risk, their incentive is to try to minimise the amount of capital they need to carry. It is a little like income taxes that nobody really likes to pay; some people are more aggressive in trying to manage their tax liabilities.
In that context, the incentives to try to manage their capital positions become greater. When the rules become more complex, the probability that mistakes are going to happen will also rise. I find it intriguing that, when banks make reporting mistakes, they tend to be on one side. To my mind, the two are somewhat correlated. From that perspective, I believe that risks are on the rise from the prudential regulator’s perspective. It therefore needs to be even more vigilant in making sure that banks are calculating their capital requirements and liquidity requirements appropriately, and to shut down attempts to go beyond tax avoidance into tax evasion.
Q17 Mr Baker: To what extent are you concerned that the treatment of derivatives under International Financial Reporting Standards is leading banks to underestimate the capital they require?
Mark Zelmer: It is not an issue specific to derivatives. Derivatives can help them to manage risks in some respects, but this is a broader issue than derivatives per se. It can pop up in many guises and, in this particular case, we saw it pop up in the context of IT expenditure. It is much broader than capital market transactions.
Q18 Mr Baker: Where else should we be looking for areas of business where accounting treatment could cause banks to miscalculate their capital?
Mark Zelmer: From a United Kingdom perspective, I would pay close attention to structural reform. I do not know the issue as well as you do, but it is leading to the creation of separate banks for different activities. That area could open opportunities to manage capital very efficiently between the various organisations within a group. The more complex you make an organisational structure, the more opportunities there are that could give rise to looking for ways to artfully manage the capital liquidity requirements. That is one example. Again, it is a little like income tax; it is a profession looking for novel and wonderful ways to minimise liabilities. My usual rule of thumb is that anything new will present opportunities. Anything that looks lucrative probably means something is going on behind the scenes somewhere.
Q19 Mr Baker: Is this kind of complexity inevitable in the modern financial system?
Mark Zelmer: Regrettably, it has been. I wish we could come up with a framework that takes a different approach. One of the things that I am personally concerned about is that it is a little like tax when, over time, you go down a path of increasingly more complex requirements, trying to nail those who wish to evade their obligations inappropriately. It is the same with capital and liquidity. It would be better to think of a system that makes it more in the banker’s interest to pursue what is perceived to be the public good, as opposed to using increasingly more complex rules and requirements to try to enforce behaviour.
Q20 Mr Baker: Serendipitously, I believe there is an Adam Smith Institute paper out tomorrow on just that subject, but perhaps that is a story for tomorrow. Could you offer us any advice on the way that the rules could be written differently to narrow the gap between the letter and the spirit of what they say?
Mark Zelmer: I would not focus as much on the rules as making the supervisory process as effective as possible. You can only go so far with rules that are being applied to the industry as a whole. The issue is how you tailor the rules for individual bank circumstances, which is the role of supervision. If I look back over the past 10 years, there has been a great deal of focus, in the wake of the crisis, on shoring up the regulatory rule framework. That is a necessary precondition, to be clear, but there needs to be more attention now on how you shore up the supervisory practice and make sure the supervisors have the necessary powers and support to be able, nimbly and artfully, to apply those rules in accordance with the unique circumstances of each institution.
Q21 Mr Baker: Finally, you made some recommendations about third-party reviews of prudential information supplied. To what extent do you think that banks, and possibly regulators, might be flying blind without adequate information?
Mark Zelmer: That risk has grown over time. With the introduction of more complex rules in the wake of the crisis, the risk of mistakes in the practice of banks has probably grown and is probably harder for the regulators to detect, because the calculations are more complex. In many respects, those calculations are grounded in the banks’ own internal risk models, as opposed to filling out the appropriate boxes of the tax form. It would help for bank regulation to borrow a practice from insurance regulation, which has third parties that can complement the regulatory review by giving independent attestations. It is done in insurance, because of the complexity of insurance with actuarial calculations and the like, and it might be something to consider on the banking front, as it becomes more complex.
Q22 Colin Clark: Good morning. Thank you for joining us. One of your recommendations is around the balance between depositor or creditor interest and the PRA objective of safety and soundness, with a focus on financial stability. You note several potential remedies for this, from an explicit warning to depositors to a new mandate, with options in between. What is your preference?
Mark Zelmer: Before I give you a direct answer to that, I need to admit a bias. I come from a jurisdiction where the statutory mandate is focused on protection of depositors and creditors from a prudential perspective. What was striking to me from looking at the Co-operative Bank situation was that, as they were trying to figure out how to deal with the stresses of the Britannia Building Society from a systemic perspective, I did not see much conversation about the repercussions of any prudential actions on the depositors and creditors of the institutions themselves. To me that was shocking, given where I come from.
That begs the question in my mind of whether the public fully understand what the statutory mandate for the prudential regulators is. Do they understand that, strictly speaking, they are there to protect the stability of the system as a whole and not necessarily individual classes of creditors, whether they are retail depositors, bondholders or the like? I noted with interest that, on the insurance side, there is a requirement on the PRA to pay some attention to the implications for policyholders. That led me to wonder, given the system that I come from, and seeing that on the insurance side here and what had happened to the Co-op Bank experience, whether there should be more focus on or more consideration of the implications of prudential actions on different classes of creditor.
I realise, and the PRA and Bank are keen to note, that there is the Financial Conduct Authority. Presumably it has an interest in that territory, and I know the two agencies talk to one another, but it still begs a question about whether there should be something more formal and explicit. As I note in the report, that is not the only way you can do this. There is a spectrum of approaches that you can take, and the answer to the question in the context of the UK has to be a Government decision about what exactly society expects from the service of its prudential supervisors.
Q23 Colin Clark: Did the bankers understand the risk and were they driven more by other drivers, such as profit, rather than the risk to depositors?
Mark Zelmer: Do you mind rephrasing that slightly? I am not sure where you are coming from.
Q24 Colin Clark: In the answer you just gave, you said that people were not really asking questions about the risks, but did the bankers understand the risks? Did the complexity of what they were doing mean they were not asking the obvious questions, so you, when you reviewed what they were doing, thought there was a gap?
Mark Zelmer: I do not think the Co-operative Bank, for example, when it was merging with the Britannia Building Society, understood the risks that existed in the Britannia Building Society whatsoever. Clearly the regulator had a different appreciation of the risks than the Co-operative Bank itself did.
Q25 Colin Clark: Do you see the way in which the UK has split conduct regulation from prudential regulation as a structural weakness?
Mark Zelmer: No, and I think it has some advantages. Again, I come from a regime where there is a split. It is an unavoidable split given our constitutional framework in Canada, but it is also helpful. To some extent, when you are carrying out prudential supervision, the discussions between the supervisor and the bank need to be pretty private, because otherwise you run the risk of bringing forward a problem you are trying to avoid. On the other hand, conduct authorities tend to come from a history of promoting investor protection and the like. They are very much in favour of public disclosure as much as possible, so there is a tension between the two business models of prudential supervisors versus conduct supervisors. I guess there are some advantages in having the two of them separate.
Q26 Colin Clark: Does it risk the PRA losing focus on depositors?
Mark Zelmer: That depends on how good the working relationship is between it and the conduct authority. Different countries have different arrangements for this, and some have combined them. Japan is a case in point. Others like us have kept them apart. We did not have a choice. Any variation can work, but a lot depends on the working relations between the organisations and how you structure information-sharing and the like. I do not think there is a right or wrong answer to this, to be perfectly frank. You want to be sure, though, that, if you are going to keep them separate, the conduct authority looks beyond purely conduct from an investor protection perspective, but also thinks about investor protection from a prudential perspective.
Q27 Colin Clark: I will move on to open banking and sticky deposits. Your report questions whether open banking may make deposit flighty. How far is this an open banking problem, rather than an internet banking problem? By the way, can you explain the risk of sticky deposits as opposed to flighty deposits, and why this is going to change because of open banking?
Mark Zelmer: As the report indicates, my premise in this area is that, when you put a third party between depositors and their bank, the incentives of the third party are different. Really, you are taking what used to be retail deposits, so they start to behave more like wholesale deposits. Wholesale creditors tend to be more market-savvy and inclined to disappear in a hurry at the first signs of a bank coming into stress. In my mind, there are clearly lots of advantages to be gained from open banking. I am not against it, but one needs to recognise that it could lead to behavioural changes. What you thought were core retail deposits that will stick by you through most states of the world will probably start to behave more like wholesale funding, which banks use at the margin. The deferential treatment given to retail or household depositors in prudential requirements at this point in time, especially on the liquidity front, probably needs to be revisited. Indeed, the Canadian experience of the Home Trust episode led to a change in prudential liquidity requirements in Canada when it comes to the treatment of third-party deposits.
Q28 Colin Clark: Most people, particularly when using an app or the internet, will choose a bank for its interest risk, not its solvency. I think you mentioned the risk of people perceiving a solvency issue. Is there a risk that open banking will make solvency an active decision? How can the regulator stop the third party involved making that an issue? Will it create a systemic threat to the banking sector? I suppose the FCA is not going to allow somebody to be involved in open banking if it may do that, but is that a risk?
Mark Zelmer: Think about it from the standpoint of a third party. Say you have a situation where, as a result of open banking, third parties start to offer customers more centralised information on their financial affairs, drawing on data from different financial institutions. It does not stop there. The next obvious route they take is to offer to manage the deposits for the depositor, which is when they suddenly become more like wholesale funding. In that case, the interests of the third party are reputational. It does not want to have a conversation with the depositor about, “Why am I seeing this bank in the news? My money is with that bank.” It does not matter how good the deposit insurance system is. They are going to cut and run, because they do not want to have that conversation with their clients. To me, by all means pursue open banking, but make sure you understand how the behaviour of household depositors will change because of the third parties that are now managing money for those depositors.
Q29 Colin Clark: In 2016, the Competition and Markets Authority said that open banking would bring competition and innovation. It was trying to make it more dynamic. Imran Gulamhuseinwala, who is a trustee of OBIE, said it will “make Britain one of the best places in the world to bank”. As third parties will be involved, does this not suggest a level of sophistication in wholesale banking, as you have already said? Financial crime is already a massive problem. Are we exposing consumers, supposedly to bring greater competition? It poses a risk to depositors and to the banking sector? Are we trying to run before we can even walk after the financial collapse?
Mark Zelmer: The risk can be managed, provided the prudential requirements adapt accordingly. Most of the discussion you are alluding to has been on the potential competitive benefits to be gained, which are to the good, frankly. I am just suggesting that some attention should also be given to how the prudential framework should evolve, because I have not heard much conversation about that when people talk about open banking.
Q30 Colin Clark: Is it possible we assume consumers are more sophisticated than they necessarily are? Is it not the nature of open banking that it is open to the industry to suggest that consumers are sophisticated and what they are offering is justifiable because they are making a free choice?
Mark Zelmer: If I understand you correctly, you are starting to head into the territory of what kinds of consumer protection issue arise when you have third parties basically acting as the agent of the consumer when it comes to managing their money and banking affairs with institutions. To be honest, I am not an expert on consumer protection policy.
Colin Clark: But there is a risk?
Mark Zelmer: I certainly think there are bound to be issues, as there always are when you have people acting as agents for others. You have to be careful to make sure that they are incentivised to operate in the interests of the consumer and not in their own interests. That begs issues such as about the extent to which they should have more formal fiduciary obligations imposed on them, as opposed to the lesser kind of obligations one normally sees in the banking world these days.
Q31 Stewart Hosie: Can I go back to conduct risk? How should banks quantify conduct risk? Is it increasing or decreasing at this point?
Mark Zelmer: In recent years, conduct risk has crystallised with very expensive fines and the like, so it has become a much more visible issue for banks and their supervisors. Looking forward, it is bound to continue to be a risk and is likely to manifest itself in different ways. We have seen issues associated with questionable behaviour, with things like the LIBOR scandal and rigging in other markets, but I suspect conduct and operational issues will also be associated with IT platforms and the like. Issues about privacy of information start to look like conduct issues, as well as operational risk issues more generally. To the extent that you see banks far more engaged in offering a broader range of services than the traditional lending/deposit activities, as you get into wealth management and the like, it may not bring balance sheet risk to the banks, but it is likely to expose them to a wide range of conduct issues.
Q32 Stewart Hosie: On balance sheet risks, do you think bank management actually understands that conduct risk, let us say mortgage mis-selling, can very quickly morph into systemic or prudential risk?
Mark Zelmer: They might understand it at a conceptual level, but I do not know to what degree they have thought about it. I am not in a position to comment on the extent to which British bankers have thought about it.
Q33 Stewart Hosie: Let us move to individual consumers then. Are they being expected to take more or less responsibility for their own money?
Mark Zelmer: To a large extent, they certainly have been in my country. I assume that is true here, to the extent that, over time, you have had a migration away from things like defined benefit pension plans into various forms of retirement account, and from long-term saving products insurance companies would offer, which would cover risks of outliving your money and the like, to putting that decision back on consumers. Yes, the trend has been towards moving the management of risks away either from employers or financial institutions and delegating it more to the households themselves.
Q34 Stewart Hosie: Is that freedom or liberalisation always a good thing or could it be characterised as, “We nationalised all the old problems when the banks collapsed, and we are now privatising the future risk by giving it over to the consumer”?
Mark Zelmer: There are a couple of issues that one needs to be very mindful of. One is that people may have a perception of the role of the banker based on the distant past. The banker was your financial adviser, in some respects, who was perceived to be looking out for your interests. With the way things have migrated over time—Wells Fargo would be a good example from the United States—it is not clear to me, looking at it from afar, that people’s perceptions of the role of bankers have necessarily kept pace with reality.
That begs the question, first, about financial education for consumers and households to make sure they understand the kinds of transaction they are getting into and the kinds of relationship they should expect to have with banks, insurance companies and other financial intermediaries. It also begs a question about the kinds of obligation one should put on financial institutions when it comes to dealing with clientele who are far less sophisticated than they are.
One can go different ways along this path. One can, in a very overt sense, put detailed rules and regulations on banks, but I suspect you will end up with a lot of boilerplate templates. It will be a little bit like when you open a Google account and just click “accept”, because you have no idea and you do not have any choice anyway.
You can think of other paths. Under the Obama Administration, the Americans were toying with putting a higher standard of duty of care on institutions, from a fiduciary angle. There may be pros and cons to that as well, but it behoves policymakers to think about the core issues they see in this regard, and the pros and cons of different ways to address it.
Q35 Stewart Hosie: I do not want to digress too much, but what you have said is right. The danger is that people with more modest savings, assets and incomes will not get the help and advice they need at all, because it is not worth it for the banks to provide that.
Mark Zelmer: It is not just banks; insurance companies would do the same. They would have clear incentives to offer this commodity—and I mean it with a small “c”—to consumers, because they will make their money from high volumes. For high volumes they need standardised transactions.
Q36 Stewart Hosie: How easy is it for prudential regulators first to identify and then to assess the potential conduct risks being run by the firms they regulate?
Mark Zelmer: This is very underexplored territory at this stage. The focus of prudential regulators has been and was, up to the point I retired a few years ago, on the traditional balance sheet risks that manifested through the crisis. They are aware of conduct issues and that they can morph into prudential issues. As I was retiring, the focus was turning much more to the risk culture of the institution and seeing if they could affect the prudential issue by going after the culture of the organisation. These are early days in saying whether that is sufficient, but it is certainly underdeveloped terrain, not just in the UK but around the world.
Q37 Stewart Hosie: Let me ask that question in a slightly different way. Were an individual in a training room to be shouting across a room of 100 people, looking for a particular LIBOR, as has happened in the past—the evidence is all there—surely to goodness not a single bank in the world now would not pick that up, or would they?
Mark Zelmer: I do not know. Canada was very much a bystander in that kind of stuff, so I do not have first-hand knowledge to give you a proper opinion on it.
Q38 Stewart Hosie: On the work you specifically did, to what extent should the PRA, in stressing banks for conduct risk, make a call on how well the regulator functions and operates?
Mark Zelmer: That is an interesting question. Whether it does so explicitly or implicitly, it should certainly factor in. When it is assessing the institution, it is assessing its governance, controls and the like. It is increasingly looking to assess the culture of the organisation, which should get it into that territory to some degree. At this stage, it is still feeling its way. More rigour and structure need to be brought into that process, but these are early days. The UK has lots of company in that regard. It is not like the UK is lagging in any way.
Q39 Stewart Hosie: Over the period you reviewed, the FSA was preparing to become the PRA. To what extent did that influence or impact the work being undertaken to supervise the Co-op? Are you aware if, at any point, the regulator took its eye off the ball in relation to its supervision of the Co-op Bank?
Mark Zelmer: The combination of multiple institutions being on fire with the radical changes that were taking place at the FCA meant that resources were scarce. There was also a fair degree of staff turnover taking place. Understandably, you tend to put your best and brightest on the bigger institutions that pose the biggest risk to the system. The Co-op may have been a major bank, but it was not RBS or some of the others. Resourcing is going to be handled accordingly. The access of frontline supervisors of the Co-op Bank to technical expertise to support them in their work was commensurately less as well. They were in a difficult place and you had a lot of staff turnover taking place within that team as well.
I did not like how it handled the merger process, but I understand why it took the path it did. If you are going down that path, you are going to expose the creditors of the Co-op predecessor bank to greater risk by allowing a merger to take place—in fact actively facilitating a merger between the Co‑op and Britannia—when you know the financial condition of the bank is vulnerable. If you are not prepared to deal with it directly through the public purse but are going for a private solution, the first order of business after the merger should have been to deal with that capital position. Instead, it went chasing after 50 points of light on a range of issues that were more important, but more chronic in nature. It did not fundamentally deal with stabilising the financial condition of that bank as quickly as you can, because you have exposed some creditors to more risk than they realised they were taking on.
Q40 Stewart Hosie: So you need better staff, better retention and a better focus on the important things, and look out for the second-tier banks, rather than only the big ones?
Mark Zelmer: It is not instead of the big ones, but do not underestimate a potential systemic situation, especially because of a cluster of small or second-tier banks. That is one issue to be mindful of. Yes, very large institutions are systemically important in their own right, but the history of this country’s banking crises has been how small or smaller institutions, as a group, can also pose systemic heartburn for the country.
Q41 Catherine McKinnell: I want to ask about one of the comments in your report about frontline supervisors being generalists by nature and therefore quite often having to rely on specialist support when required. What are your views on how easy it is to ensure that supervisors are able to exercise their judgment, rather than a more checklist-based approach, when they rely on generalist staff? How can we get that balance right?
Mark Zelmer: You want to use generalist staff as your first point of contact with an institution, because there will be a wide range of issues, and somebody needs a sense of the bigger picture for the organisation and to decide which issues are more important than others. It is not surprising you would want somebody that can look at that broader picture as your point of contact with the institution. The key is that they have the resources to support them when they engage with the institution.
You made an interesting point about the judgment-based versus compliance approach. If you are going to run a principle or judgment-based supervisory approach, not only do you need to make sure that you have access to the technical experts who can back you up in making arguments to an institution; you also need to demonstrate that you are dealing with similar situations in a similar way, over time and across institutions. That does not mean you have to do things exactly the same, but you have to be able to demonstrate that you are not engaging in capricious behaviour. It is more likely the regulated institution will accept your demands if you can demonstrate that you are treating everybody equitably.
That requires certain basic things. Some of these issues I flagged in the report. The supervisors need to know what the actions have been in similar situations, in the past, hence you need decent information and record-keeping systems within the institution and regulatory agency that are readily accessible to the supervisors. Secondly, you need to make sure that there are appropriate amounts of guidance for the supervisors about how the agency expects them to assess different kinds of risk, and the controls and mitigants that exist within the bank or insurance company. I struggled to find evidence of written guidance within the PRA in this respect, and I tried hard to look for it. There is lots on process, but not so much on how you assess risks, transactions and the like. Fundamentally, you also need the powers vested in the supervisory agency to be able to carry out that supervision without being unduly subjected to the risk of the bank going around you and trying to appeal to the political process to intervene and protect it from the nasty supervisor.
Q42 Catherine McKinnell: It is interesting that you say you looked for evidence of guidance, because the big challenge comes when staff need to operate outside of standard processes. You said in your report, “A key lesson to be drawn from this period is that supervisors should focus more on the key urgent priorities for a bank, and not necessarily follow the standard process of supervision when the firm is at risk or nearing risk of entering a crisis situation”. Is guidance the key to making sure that firms are properly equipped in those situations?
Mark Zelmer: The guidance is useful in more normal circumstances to guide day-to-day supervisory activities and to have some criteria, so that the bank or insurance company understands the basis upon which it is being supervised. When you start to get into troubled waters, the supervisor needs the expertise and confidence to say to its management and peers within the agency, “Look, we are into troubled waters. These are the most critical issues that we need to deal with and we need to deal with them fairly quickly.” It needs to garner support from the agency to do just that, because it will probably need to start bringing in more senior people to help manage the case. To my mind, that means that you need a good peer-review process within the organisation, a willingness to share experiences and information within the organisation, which is a cultural issue, and support from senior folks in the organisation to go down that path.
Q43 Catherine McKinnell: From looking at your review into the Co-op, your report notes that the FSA relied on external support in its supervision. However, you have said, “Such use of third parties does not take away from the regulator the responsibility to critically weigh their inputs”. Do you think that the regulator used external reports during that review period in an appropriate way? Were there things that could or should have been done better?
Mark Zelmer: I do not think there was the culture of challenge back then that exists today. The expectation did not exist in the organisation to be as assertive as it should have been to provide challenge, both in the scope of the review and in the assumptions and findings of the reviews that were taking place. In a situation like this, for expediency purposes, it chose not to take on a section 166 review and go down the path of allowing the bank to engage a third party to do the review. If you are going to do that, you had better make sure you have a great deal of influence over the scope in the process. To leave it to the bank to engage the third party and to not be there to be aggressive in providing that challenge, as we saw, was a recipe for not getting a useful benefit from that exercise.
Q44 Catherine McKinnell: What is the risk-benefit analysis of that? What are the pros and cons of getting that external analysis against dealing with these things in-house?
Mark Zelmer: The experience I had from being at OSFI in Canada was that we often found it useful to have the bank or insurance company pay for a third-party review when we had a difference of opinion. A former colleague of mine, who headed up the supervisory side, called it the seven stages of grief you have to go through. Denial is the first one and it takes a bit of time to get through that one. Part of getting through those seven stages is bringing in a third party to provide an opinion. From OSFI’s perspective, it would have been to give support and weight to OSFI’s arguments to the bank to help it understand that perhaps its practices were not in line with its peers’.
In our case, the supervisors were always very nervous, as consulting firms generate a lot of income from servicing banks on many fronts. Even though the bank or insurance company might have to pay for it, you do not leave it to them to manage the process, because you know that the incentives are naturally in favour of the external party to be sympathetic to the management, because there is hope for future ongoing relations. As a result, OSFI was aggressive in wanting to have a say over scope and insisting on dealing directly with the third-party reviewer to make sure that it was satisfied with the quality of the analysis, rigour, et cetera. They have useful benefits to offer, both if you do not have expertise internally on a particular issue and in helping to facilitate the passage through the seven stages of grief, but use with caution and do not assume they are naturally going to give you what you want.
Q45 Catherine McKinnell: It sounds like it comes down to judgment again. Are there steps a regulator can take to make sure external reports are as useful and reliable as possible, rather than contributing to the challenge?
Mark Zelmer: The key thing is to make sure that you are satisfied with the third party the bank is engaging, that it is reputable and has quality people assigned to the task at hand. Make absolutely sure you are comfortable with the scope, and do not let the thing run on auto-pilot. Pay very close attention and engage regularly with the third party through the process to make sure it is not going off the rails and down a path that you do not think is appropriate. Make the institution pay for it, because it is a way of getting extra resources for the supervisor.
Q46 Chair: Before I bring in Simon, I want to ask about management actions at times of stress. You highlight in your report about the FCA having allowed for a relatively broad range of management actions to be taken into account. How far should the PRA allow potential management actions during a time of stress to be taken into account?
Mark Zelmer: I am a cynic when it comes to management actions because, if you are not careful, they are a wonderful way of assuming away the problem. In the early days of recovery planning, banks would have plans with a wonderful range of management actions, and you have to ask yourself if they are credible in that kind of stress circumstance. The onus should be on the bank to give comfort that the management action is indeed plausible under the circumstance. My inclination would be to say no, unless there is a good case otherwise.
Q47 Mr Clarke: Thank you, Mr Zelmer, for being with us today. Talk of management leads neatly on to the decisions surrounding Verde and the abortive bid. In your report, you note that the FSA’s role was that of a regulator: “It supervised firms, but was always cognisant of the need not to stray into the role of management”. Obviously the Co-op Bank was an organisation that had serious management shortcomings. In that scenario, where is the line between regulating and straying into management territory?
Mark Zelmer: In this case, the FSA, which was in the midst of becoming the PRA, was fairly adroit. The key is to make clear the hurdles any bank or insurance company has to overcome if it wants to pursue a transaction like that. You have to do that early on, then keep beating them over the head with it. You should not back down. In that way, you try to make it as clear as possible to the bank or insurance company what it will take for them to get over the line.
From looking at the Verde process, I took it that Co-op Bank management reacted to those hurdles. That they were willing to walk away at one stage, before the transaction was restructured to reduce many of the risks, was an example that they realised this was not going to be an easy task. Then the transaction gets restructured; it tempts them back into the process. They kept being reminded of what the hurdles were and nudged along to make some progress. Broadly speaking, the FSA was playing it appropriately in hammering away at them, “This is what you have to do,” and having regular touch points, occasionally with senior management, to review progress. We did not see it in this case because the transaction was ultimately aborted.
The outstanding question in my mind is that I am not convinced, from the records I saw from the Treasury, that the Treasury appreciated how big the hurdles were. I was concerned that the FSA/PRA was starting to put itself in a box because, the longer the process goes on and the more work that is done by the Government to facilitate the transaction, begs the question that, if the bank could not meet the hurdles, would the FSA/PRA have blinked or not. We do not know because it never got there, but the fact that the Treasury did not understand how big the hurdles were introduces the risk that there could have been an uncomfortable situation down the road. We will never know. The people involved will say they would have stood firm until the bitter end. They would say that, but we do not know.
That is why I made the point in the report that early engagement between the regulator and the official sector is critical, so that everybody on the public sector side understands what is going on. It then enables the hurdles to be not only clearly articulated but clearly credible.
Q48 Mr Clarke: One of the concerns about the Co-op and Verde was that the Treasury may have placed a degree of undue pressure on the FSA around the bid. You note in your report that “there was a reasonably clear line between HMT and FSA, in terms of HMT not impinging on the FSA’s remit in terms of financial stability or prudential supervision of the Co-op Bank”. Why is it only “reasonably clear”?
Mark Zelmer: It is for the reason I just mentioned. It was fine. I did not see any written evidence to suggest any inappropriate treading across borders, but I was worried about that risk building up. It could have spilled over at a later date, had the transaction persisted and had the FSA had to lower the boom, at some stage. No, there was no evidence of any inappropriate mixing of roles, but the risks that could have happened later in the process were there. That is why I felt a “reasonable” qualifier was appropriate.
Q49 Mr Clarke: That seems a reasonable decision. On management strategy more broadly, coming back to the role of the regulator in this somewhat fiddly area, is there merit in the regulator pointing out what it deems to be potential flaws in the management strategy, or does that blur the line? To what extent does that change in a crisis situation?
Mark Zelmer: The role of the regulator is probably to push to make sure that the management and board of an organisation fully understand the risks that they are dealing with, and to probe to find out how strong the measures being taken are to manage those risks. It is not necessarily to question the strategies per se, because banks and insurance companies are commercial entities. They should be allowed to fail and be responsible if they fail in their strategies, as businesses. But in my mind, the job of a prudential authority is to push them hard to make sure that they know what risks they are taking on and what risks they are exposing depositors and creditors in the system to. Be prepared, if push comes to shove down the road, to take control of the organisation before it causes a mess.
Q50 Mr Clarke: I appreciate there must be scope for failure, but there must also be robust supervision. With that in mind, how easy is it to objectively determine management weakness and was the regulator slow to do so here?
Mark Zelmer: The supervisor needs to form its own opinion about what it perceives the strategy and the risks of that strategy to be. One of the advantages of employing people from the private sector, who have worked on the other side, in their supervisory activities is that they have some familiarity with the commercial side of the business. They can test their own understanding and analysis by engaging with the bank and having those conversations. It is an iterative process. They will learn and refine; hopefully, through the dialogue, both parties will.
Q51 Mr Clarke: To press you on that second point, was the regulator slow to do so here, because the decisions that were taken were, at best, doubtful? Could the regulator have been firmer faster, in terms of Verde?
Mark Zelmer: If it had been the same transaction all the way through, I would have said yes but, because the transaction morphed over time and became increasingly less risky for the Co-op Bank, and because the regulator hoped it might be a source of capital for the Co-op Bank given the values being used for the transaction, I can understand why it allowed it to play out in the way that it did. We subsequently learned that there was no hope in hell of that transaction actually succeeding but, based on the information they had at the time and the way the transaction was evolving, I can understand why it tried to give the bank every opportunity to get on side.
Q52 Mr Clarke: This is a more big-picture question. Should the regulators pay more attention to the way in which firms comply with the Senior Managers and Certification Regime? It is just on how robust that is.
Mark Zelmer: To be honest, I come from a jurisdiction that does not subscribe to that kind of process, so I am probably not well placed to comment on how it works.
Q53 Mr Clarke: Very fleetingly, why does Canada not subscribe to that? What is the policy rationale for not having such a regime?
Mark Zelmer: It is a very different financial system. It is a highly concentrated system. It is inherently a highly profitable system, especially for the major players. As a result, there is what academics would call charter value for participants in the system. By that I mean that they naturally have incentives to think about the broader repercussions of their actions, because they do not want to kill the golden goose. That has led to a reasonably conservative banking and insurance culture in the country to protect its profitability. There were many factors, but that contributed to avoiding some of the worst angst that went on in other parts of the world. There is a fair degree of trust between the regulator and the regulatee, but it is fragile equilibrium and I would not dismiss the possibility that, if the system were to open up to more competition, our fragile equilibrium, which probably looks a lot like what the UK might have looked like in the past, would evolve.
Q54 Mr Clarke: I was going to talk of golden geese not being slain, because it would not be in anyone’s interest and we would have a lot of dead geese.
Mark Zelmer: I can see the system might evolve but, at this time, given our situation, we have not felt the need to go down that path.
Q55 Chair: Before I bring in Wes, I want to ask you about the PRC’s regard to competitiveness. There is often a suggestion that competitiveness is an objective suggested for regulators, particularly for the new FSA, in the sense of your review. Do you have a view on what impact that might have, if it was an objective for the regulator?
Mark Zelmer: I do not know all the ins and outs behind that objective, but it reminds me of an objective that exists in Canada. Yes, you are pursuing safety and soundness for depositors, policyholders and creditors, but you must be mindful that the institutions compete, have to be allowed to take risks and bear their own responsibility for failure. There are certain objectives you want to achieve as a regulator or supervisor, but you need to be mindful that this is a competitive marketplace. You should try to design your requirements in a way that facilitates competition, as opposed to turning the financial system into a graveyard.
Q56 Wes Streeting: Following on from that question, the final paragraph of your report notes, “If the UK experiences a protracted benign environment in future, there is a risk that prudential oversight could fade into the background at the BoE and receive commensurately less Executive attention and resources in an institution where the culture is heavily skewed in favour of macroeconomics”. Is that danger magnified by the One Bank ethos, in that it may distract from the visibility of the work of the PRA?
Mark Zelmer: It is a good question. I do not see an issue at the moment. I want to be clear about that. That there is a prudential regulatory committee that includes non-executive directors is a helpful safeguard. Having said that, if you look down years to come, assuming things are reasonably benign, there is a precedent here that is not a happy one. Prudential regulation moved from the Bank of England into the FSA when it was created. The FSA was focused on conduct, as I understand it, and prudential regulation became a bit of a backwater.
You can have an organisation with committees on top and with outside people from industry, as part of your non-executive directors, to pay attention to these kinds of thing, but my experience, having worked at a central bank and at a separate regulatory agency, is that the culture is different at the staff level. Over time, there is a natural temptation for people who are ambitious and want to pursue their careers to go where the action is. There may be good reasons for trying to foster a one-bank culture, but you need to be mindful that the business of prudential supervision and regulation is different from a macroeconomic-oriented business. You need to make sure that you do not find that one culture overly dominates the other to the detriment of an institution. It is not purely a case of resources shunted somewhere else. It could be that, but I would hope the non-executive directors would be jumping up and down to prevent it, and involving you folks if necessary.
The issue is more insidious than that. It could be perceived, if you have an extended period of many years, that that is not where the action is, so you do not necessarily get the best and the brightest doing it. If the business is fundamentally different, which it is, there are limits to how far you can apply a one-bank culture. Things like IT system needs will be different; the types of skill set that people need to carry out their activities will be different. You need to find a way to make sure the different businesses within an organisation are viewed as being important and able to attract the skill sets they need, so the people carrying out the activities feel they have career prospects.
Wes Streeting: That is helpful. Next, I would like to elicit some free consultancy from you, on behalf of the Committee. Thinking about those risks, how should we as a Committee be guarding against such an outcome? One of the difficulties we have when scrutinising the work of the PRA and the PRC is that much of the communication and the information they deal with it is commercially sensitive, and we understand that. How do you suggest we counter that problem, and what kinds of questions should we be asking to keep the Bank on its toes and make sure that this important function is not overlooked?
Mark Zelmer: You can push the non-executive directors to help you understand how the PRA, for example, is performing as a business, making sure that they know enough about the business, because they do not necessarily come from that background, unless you have non-executive directors who do come from that background.
The other thing you might want to think about is every five years the International Monetary Fund comes to town to do a financial stability sector assessment for the United Kingdom. When it does these exercises, it brings in experts from other countries to kick the tyres of prudential supervision, both from the banking and insurance fronts. You may want to ask yourself if it makes sense to have them visit you. You can pick their brains about what they think of the quality of the business, relative to their own experiences. That is one example.
There are other kinds of peer review that go on. You have to pay close attention to the background and quality of the people doing these reviews to see if you feel comfortable that they are able to give you the insights that you want. It varies, from cycle to cycle, as to what individuals come as part of that exercise. I would not suggest just asking the regular IMF staff to do this. You need to get to the experts who are part of that process, who come from other jurisdictions. That will be more enlightening for you. Any opportunity you can find, where there are people being brought in to give advice through exercises like that, could give you a window.
I appreciate it is commercially sensitive to discuss situations like this, a specific bank or insurance company, but I presume and hope that the PRA, from time to time, does what I would call a thematic review or cross-sector supervisory exercises. You might want to engage with them and discuss how they are going. You may consider engaging with the internal audit folks at the Bank, who do independent evaluation exercises, to see what you can learn from those reports as well. There are different ways that you can gain some perspective on how the business is running, over and beyond talking to the non-executive directors, without necessarily treading on commercially sensitive information.
Wes Streeting: That is all immensely helpful, and you have also volunteered a number of people to appear before the Committee. I am sure they will be grateful for that, but that is genuinely very helpful and constructive advice. Thank you.
Q57 Chair: Mr Zelmer, is there anything else that we have not covered this morning? Your report is extensive and there is lots in there. I hope we have covered the main points but, if there is anything else, now is the time to speak. You have the floor to raise our attention to any final issues before we conclude.
Mark Zelmer: No, you have covered the range of territory I would have expected you to cover. Between the report and what I have told you today, I have pretty much said everything I would have wanted.
Chair: Thank you very much. I hope you have been thanked by the Treasury for the report and review that you prepared as well. Thank you for your time this morning, particularly for making yourself available from Canada. We are very grateful.
Mark Zelmer: It has been my pleasure and honour. Thank you.