Evidence heard in Public

Questions 3766 - 3869



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


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


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


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

Oral Evidence

Taken before the Parliamentary Commission on Banking Standards

on Wednesday 6 February 2013

Members present:

Mr Andrew Tyrie (Chair)

Baroness Kramer

Lord Lawson of Blaby

Mr Andrew Love

Mr Pat McFadden

Lord McFall of Alcluith

John Thurso

Lord Turnbull

The Archbishop of Canterbury also attended as a Specialist Adviser with power to examine witnesses.

Examination of Witnesses

Witnesses: Stuart Gulliver, Group Chief Executive, HSBC Holdings, and Douglas Flint CBE, Group Chairman, HSBC Holdings, examined.

Q3766 Chair: Good morning. Thank you very much for coming to give evidence. May I begin by asking whether you support the proposal of this Commission that a reserve power be taken for full separation in order to make the ring fence more robust?

Douglas Flint: Yes, we do. In the event that it is judged that participants are circumventing or frustrating the purpose of the ring fence, it seems to me quite reasonable for there to be a sanction. Yes, we support that.

Q3767 Chair: Do you think there should be, as we also propose, a periodic-perhaps five-year, perhaps slightly longer, perhaps slightly shorter-review, independent of the regulator and the Government, of the design and effectiveness of the ring fence?

Douglas Flint: Again, yes. I think it would be remarkable if everybody who has been involved in this process is able to anticipate everything that might arise from the operationalising of the ring fence over the next three to five years. I actually think that the first independent review might be earlier than five years, because problems, if they arise on either side of this application, will probably arise quickly in its operation, rather than later. So I think maybe three years afterwards would be a good first time.

Q3768 Chair: I think we went for four.

Douglas Flint: Okay. Well, somewhere between three and five.

Q3769 Chair: Four seems roughly in the middle. Do you also support the Commission’s recommendation that the burden of proof for any exemption from the requirements on primary loss-absorbing capacity should rest with the bank?

Douglas Flint: On reflection, yes we do. It is clearly going to be a dialogue between the institution, the regulator and the Treasury. The ultimate risk lies with the Treasury, because in the event that the institution does bring risk from overseas to this country, it is the Treasury that ultimately will have to consider what proportion, if any, of that it will bear. As the contingent risk is with the Treasury, it seems to me that it is perfectly reasonable to ask the institution to demonstrate why there is no risk. I think what we would say is that because of the sensitivity and the considerable judgment involved, the test should be one not of absolute risk, because absolute certainty is never going to be possible as you look far into the future, but of reasonably foreseeable risk in that the deliberation should take place, in our view, at the institutional level at board level and at Government level in the highest levels of the Bank of England, the Prudential Regulatory Authority and, indeed, the Treasury. The judgments will be just as much about the people of whom assurances are being sought as about the detailed facts and figures that will be presented by us as an institution and by other institutions that are affected.

Q3770 Chair: May I reinterpret that as saying that if we are going to have a system based on judgment, the judgment must be exercised by very senior people with the self-confidence to vary that judgment, not box-tickers?

Douglas Flint: I would agree with that.

Q3771 Chair: Do you think there needs to be some sort of qualification based on a phrase in statute such as "reasonably foreseeable"?

Douglas Flint: I do, because I think if you make it an absolute test it is very easy for someone to say you can never be absolutely certain of anything.

Q3772 Chair: You, as a business, have a good opportunity to look at how it is conducted around the world. The US hands out, on the whole, larger regulatory fines-more have been announced today-has a tougher individual sanctions regime and, some say, has more politicisation of its regulation and, to some degree possibly, its judiciary. Is the US a tougher place for banks to do business? Does that toughness help engender a better culture?

Douglas Flint: It is certainly a very difficult place to do business when you get it wrong, which I think we know very well. I support the view that tough but fair regulation is a comparative advantage. I think that financial systems benefit from strong, tough but fair regulation. The US has a tough system and it has a particularly penetrating examination of failure. It is the way it is, but I think we should aspire to having a tough, fair system here. That is what you are attempting to achieve.

Q3773 Chair: Do you think the American system is fair? You are answering the question already, actually.

Douglas Flint: I am not the judge of fairness. I think that where you make a mistake you are punished, and you are punished in a more penal way than other jurisdictions choose to do. The US has a concept of corporate criminal liability, which is not shared by many other jurisdictions. Having said that, those who operate in that jurisdiction are aware that those are the rules under which they operate. So it is not for me to comment on someone else’s legal system. It is the legal system they have. We knew about that, and we know about that when we operate there.

Q3774 Lord McFall of Alcluith: Mr Flint, when you were here last time, I asked you about HSBC being too big to manage, and I think you said that it was a good question. I want to look at the issue of the fines that you had over money laundering in the United States.

The US Senate’s report has clarified that "overwhelming" amounts of Government information were available to HSBC regarding the specific high risk of money laundering in Mexican banks. This information was "inexplicably excluded" from the HSBC risk assessment matrix every year from 2002 to 2012 and from 2002 to 2009. HBUS gave Mexico its lowest risk rating for AML purposes. However, internal correspondence between members of the London compliance team refers to the "specific risks" of money laundering in Mexico-for example, a 2008 e-mail from Susan Wright, anti-money laundering compliance head for HSBC Group. "People on the ground", such as John Root and Paul Thurston, also repeatedly expressed concerns about Mexico being a high-risk money laundering jurisdiction.

Both of you had engagements in America and were directors of the Latin American holdings, etc. Given that HSBC paid almost $2 billion in fines over the money-laundering scandal, and the Senate reported that HSBC had "long-standing, severe, anti-money laundering deficiencies", what does that failure tell you about the culture and the organisational weakness in HSBC?

Douglas Flint: That is a very good question. One of the clear lessons that we have had as a result of this is that certain of the standards that we believed were being applied globally that were set from the centre were not being applied as they should have been. Certain of the matters that should have been shared and escalated were not being shared and escalated as well as they should have been, and certain of the external data sources-it is absolutely the case, and we have accepted that it is very much true, that there was a great deal of information regarding Mexico that was not taken into account in the risk assessment. That was an error, but I think we have also recognised that there are broader areas of knowledge and intelligence that we ought to now seek very hard to make ourselves knowledgeable of, so that we can ensure that such things do not happen in the future.

I would like to ask Mr Gulliver to comment on how we have dealt with the operational side of dealing with these issues that arose, but you are right to say there were things we were not aware of and standards that we believed were being applied that were not.

Q3775 Lord McFall of Alcluith: Given the almost cavalier approach of HSBC to the money-laundering issue, to the customer redress on PPI-for which you have put $2.7 billion aside so far-and to the acquisition of Household, which was an absolute disaster in the United States, relating to sub-prime investment quadrants, not one person has stood up and taken responsibility for this and said, "As a result of this shameful situation, I’m out the door." Why has that been the situation? Why is everybody sat cosy in their chairs?

Douglas Flint: I don’t think anyone is sitting cosy in their chair at all.

Q3776 Lord McFall of Alcluith: Nobody has taken responsibility.

Douglas Flint: On a number of occasions, my predecessor and I, since I became chairman, have said at almost every single AGM that the board takes collective responsibility for the acquisition of Household and for many of the other things you mentioned. We did take collective responsibility for that. Many of the issues that you have talked about in terms of detail are some time ago. The people who may have been closest to those issues are no longer with the group.

What the board did two years ago was ask Stuart Gulliver to set about repairing some of the issues that had been identified as where we were deficient. Stuart has been working tirelessly for two years, since he became chief executive, to repair that which is broken.

Q3777 Lord McFall of Alcluith: Mr Gulliver, after acquiring the Mexican bank, it was known right up to board level that the bank had few, if any, money laundering controls, and that the affiliate did not meet group standards. So why was it allowed to continue correspondent banking, when it was known that it didn’t meet group standards? Was it wilful ignorance or were the systems not in place for that?

Stuart Gulliver: I think the answer lies-I think the Commission has seen this in a number of instances-with culture. The culture failures were at two levels. We bought a bank in Mexico that we bought cheaply because it was in distress. That bank, as you can see from the documentation, clearly had inadequate anti-money laundering systems. We ourselves were too slow to put in place anti-money laundering systems that would be up to the standards we would all expect-

Q3778 Lord McFall of Alcluith: But you knew from day one of acquiring the bank that there were problems.

Stuart Gulliver: We knew from day one. As you will also see in the PSI report and the accompanying document, there was a considerable amount of effort made, but obviously it was insufficient to deal with it.

The other cultural thing-this goes to the root of the questions about Household and so on, and this is a significant change that I put in place in January 2011, having become group CEO on 1 January 2011-is that HSBC ran with this culture of "the country head is king". This is a firm that had a group of 400 international staff who moved around the world who carried the DNA of the firm. That served us incredibly well for a long period of time. Its historical roots sat around in Asia-Pacific, where the firm was established in 1865, and where you would have had to rely on the country head to operate without any great supervision because communications weren’t great.

We kept this going for far too long, so we were reliant, as a board, on the information coming from, if you like, a single pair of eyes on the country; the country head was omnipotent. The presumption of the board would be, because that country head was a DNA carrier for the firm, they would be doing the right thing, but without the check and balance-forgive me; this is quite long, but it is really important.

So in January 2011, I changed the organisational structure of the firm-this is hugely fundamental-from being run by 88 separate country heads who reported to the group CEO to being run with 10 global business heads: commercial banking, global banking, markets, retail banking, wealth management and private banking; 10 global functions: finance, legal, risk and compliance; and our systems people. What that now means is that you have 15 pairs of eyes going into every country. Therefore, your escalation process is much more direct, because it is now the case that the head of compliance and risk hires, appraises and pays all the compliance and risk people globally. Therefore, the information comes to the centre, because it is the person at the centre who is in control of everything. That information did not come to the centre previously.

It is very easy from outside to see this as a trivial change, but it is the biggest organisational change in this firm-I am not exaggerating-since 1865. We did it in order to deal with the weaknesses that you have just highlighted. Our structure was not fit for purpose for a modern world, where, to be honest, our geographical footprint became very attractive to transnational criminal organisations, whether they are terrorist or criminal in origin.

Q3779 Lord McFall of Alcluith: But you were chief executive of global banking markets at HSBC Global Asset Management from 2006 to 2011. You also had responsibility on the FSA register as a significant influence function of HSBC Bank USA and HSBC Securities (USA) from 2003 to 2009. Why didn’t you wake up and smell the coffee long before this?

Stuart Gulliver: First of all, my responsibilities were in relation to global banking and markets. My responsibility for the entire firm began on 1 January 2011. These activities did not take place in the global banking and markets business. Secondly, the information that came to those boards, which would include the regulatory reports that we were getting from the US regulators, gave the board no indication that there were these problems. That would also be true for the holdings board, with respect, both from the US regulators and, indeed, from the Mexican regulators.

Q3780 Lord McFall of Alcluith: But it gets harder to believe when you look at the written testimony from David Bagley, who was in group compliance, and he said: "As the Head of Group Compliance, my mandate was limited to advising, recommending, and reporting. My job was not-and I did not have the authority, resources, support or infrastructure-to ensure that all of these global affiliates followed the Group’s compliance standards. Rather, final authority and decision-making rested with local line management in each affiliate." In October 2002, a month before HSBC acquired the bank in Mexico, David Bagley said in an e-mail: "There is no recognisable compliance or money laundering function". Is that not amazing?

Stuart Gulliver: That is the key point as to why I made the organisational change. The quote that you have just set out is the reason for the change-

Q3781 Lord McFall of Alcluith: The thing is, Mr Gulliver, as Mr McFadden keeps saying, that people come along here and say, "That was in the past. We have all changed. Forget what is in the past." When scandals are emerging every day-they seem to be choreographed for the Parliamentary Commission on Banking Standards’ meetings-we just take it with a pinch of salt. It needs to be proved that the culture has changed. It cannot just be mere words. Is that right, Mr Flint?

Douglas Flint: I agree with you, but I also think that we were aware when we acquired Mexico in 2002 that, in a whole bunch of control functions, there was very little there, because the bank was effectively broken. As we had done in Brazil and Argentina, we set out effectively to start from scratch and to build them. Our weakness or deficiency, which we put our hands up and admitted to, is that we did not make sufficient progress fast enough. We perhaps underestimated the scale of the challenge. We knew that we were starting with a blank sheet of paper and with very little to work with. We set about recruiting a whole bunch of people and establishing systems and processes and controls. We inducted a whole bunch of people. It was not fast enough and, with hindsight, it was not penetrating enough. That is a deficiency that we have recognised. That is what led to the organisational changes that Stuart talked about having started in 2011.

Q3782 Chair: But John’s question does seem to be pertinent. We have had a string of people before us telling us, "Of course, it was all a terrible mess, but we have sorted it all out now." What we want to hear this morning is as much evidence as you can provide for us that something really has changed that is going to make this sort of thing happening again dramatically less likely.

Douglas Flint: If I may, and Stuart may want to expand on this, one of the things that is a simple but effective illustration of change is that the group management board, which now encompasses the four global businesses, the 10 functions and the technology division, effectively is jointly and severally-in other words, although someone is there with an HR hat, a technology hat, a risk hat or a retail business hat, they are collectively responsible for everything that happens. Someone cannot say, "That is not in my area of responsibility." They are collectively responsible. That was evidenced, and Stuart will help me here, by the fact that the top 60 or 100 or more-perhaps it was 300 people in the firm-were all given the PSI report, which is a shocking document. They were required to read it and to certify that they had read it and that they understood the lessons and the learnings in it. It was a very sobering read and everybody at the senior management level of the firm has had to read it. We put it on our website for all of our employees to read. We have done a huge amount to use the excruciating examples within that report and the DOJ statements to make people believe that, no matter where they are in the world geographically or in business, this is their issue.

Stuart Gulliver: We are not saying the job is done. This is actually a permanent journey, because the people that we are dealing with are highly sophisticated and will continue to change their ways of trying to penetrate the banking system. And we have reorganised HSBC. We have now moved to introducing the highest standards, which are actually US standards, everywhere in the world, where previously we tried to be the best in a country, which may not be the same standard as the United States applies. We have recruited a completely new team of legal and compliance people. And it is an absolute journey for us. We are in a five-year deferred prosecution agreement. So this is not something that in any way, shape or form we are saying is fixed.

However, during the last two years I have also sold 43 businesses to remove complexity from the firm. We have never, in the history of the firm, disposed of any businesses. We have done that to make it easier to control, to de-risk the firm, to make it more straightforward, so that we can assure you all that we are making the best efforts we can to make sure that this does not happen again. But it’s a journey that will not finish.

Q3783 Lord McFall: The key point is that David Bagley said in his e-mail that he knew the weakness there, but did not tell the US bank. Does not that tell us, one, that banks are too big to manage and Andy Haldane is right when he says that there is a regulatory case for smaller banks; two, that the corporate criminal liability measures in the US should be adopted here, Mr Flint; and three, that banks on their own will not be able to change, so it has to be done from the outside with banks?

Stuart Gulliver: May I take up at least one of those? I do not believe that this bank is too big to manage or too big to control. Having sold a large number of businesses, having reorganised the way that the firm is run, this is essentially four businesses and those four businesses each do about 10 or 12 things. So we are actually doing about 50 activities globally, and the disposal of the businesses is to remove things that were not logically linked. If you’re large and complex, yes, I can appreciate the challenge, but if you’re large and reasonably straightforward I believe you can manage these things and control these things.

Q3784 Chair: May I come back to the point you made earlier, Mr Flint, when you were saying that a good number of people are now responsible, and there is not just one pair of eyes? I think that is a point that Mr Gulliver made too. We have had quite a bit of evidence to suggest that there are advantages in making clear that one person is on point, that they are personally responsible and that it’s no good their saying, "Well, you know, we all collectively took a decision, so I suppose we’ll all take some small share of the blame." Isn’t it the case that if everyone is responsible, no one may be responsible?

Douglas Flint: I think you want both. I mean, there are clearly people who have specific responsibilities, which give them a heightened level of responsibility for their functional or business area. But what I think we have now enforced, or reinforced, is that you can’t sit in the room and say, "My gosh, so-and-so has got a difficult challenge, it’s his challenge not mine." So it’s a shared challenge, although there are clearly specific point responsibilities as well.

Stuart Gulliver: Also, it is made more complicated with HSBC because we are in 80 countries. So the point about having a single person was actually the weakness in Mexico. So, unless that single person communicates in to the head office-what you are describing, I could see that it would work well for a bank that is predominantly in the UK. But once you are into an international firm, which this is, I think that having joint and several responsibility-not "no one’s responsible"-is the way to manage the risk.

Q3785 Chair: That can be interpreted as an argument against your broader structure and against the structure of global banks of your type more generally.

Douglas Flint: It could be, unless you have embarked upon a process to simplify. I mean, doing lots of complicated things in lots of places is difficult. Having a simplified business model and repeating it in multiple jurisdictions, particularly we would say within a subsidiarised structure, gives you a considerable amount of checks and balances and control.

Q3786 Mr Love: But Mr Gulliver, did not those compliance failures-money laundering-go much further than just Mexico or south America? In the opening statement to the US Senate sub-committee, they cited an example of HSBC in London, the very heart of this organisation, coaching an Iranian bank on how to get payments through the US without inspection. Doesn’t this culture of failure go much deeper in your organisation?

Stuart Gulliver: But it is about the same response, with respect, of reorganising in the way I have just described, and that actually was done globally. It wasn’t just done to Mexico and the United States. We have reorganised the entire group to reflect exactly the weaknesses you are describing.

Q3787 Mr Love: Let me come on to culture in HSBC. Mr Gulliver, you have championed "courageous integrity". We have heard many fine words and many fine sentiments expressed to this Commission in recent weeks, by a variety of banks. What practical difference will courageous integrity make to the way individuals behave at HSBC?

Stuart Gulliver: Absolutely-it is critical that these things aren’t sloganeering, and are actually operationalised. So what we did was we introduced this in the first quarter of ’11, because I knew we had these cease and desist orders in respect of Mexico that hit us in the fourth quarter of 2010, just before I became the group CEO.

What we have done is we have effectively put a values appraisal or evaluation into the top people-group managing directors, group general mangers, so kind of the top 60 people in the firm. They have to get through a values appraisal, which is done every half year, before they even get to their balance scorecard. So if you don’t pass muster on values we don’t even get to the scorecard about what sort of money you made and what sort of return you’ve made.

So this really does affect people’s reward structure. They way values are judged is that you have got to select three peers, three subordinates and three people above you-it sounds complicated, but it isn’t-and they will actually opine on your values. You know about the "open, dependable, connected"-I think you have seen the full values matrix. So every half year, people are appraised on it. If they don’t hit it, we will actually exit people-and I have done so in the last two years.

So what I can honestly, genuinely say is that we have removed senior people from HSBC for values breaches on my watch. Therefore, people know that if they don’t get through the values gating they won’t get to the balance scorecard. The balance scorecard itself is split 60% financial, 40% non-financial. So this is not sloganeering, this actually hits people.

Q3788 Mr Love: I want to come on to the issue of the scorecard and some of the other issues you touched on, but the essence of this is, in a sense, the reverse of what you say. It is not about penalising people, it is about ensuring that the people who abide by courageous integrity are listened to and rewarded through the organisation.

Stuart Gulliver: It has got to be both. There has got to both a penalty and a reward. It has to be both, and it has to be straightforward in that regard. Effectively, what we are saying is that there is a way to behave at HSBC. Effectively, there is a behavioural standard that we are requiring. Again, one of the things that has changed-and this is an interesting statistic-is that our chief risk officer, whose job is effectively to stop things, is at the moment recommended to be the fifth highest paid person in the group in 2012, subject to board approval. The board hasn’t yet approved compensation. In 2006, they wouldn’t have been in the top 50. There is someone who is clearly being rewarded for stopping things. I think that is one of the biggest demonstrations of sincerity-where does the rank of your CRO fit in the overall pay structure of your company, as opposed to your revenue producers? I think that shows that we are really genuinely trying to operationalise this.

Q3789 Mr Love: Well, I understand that, but the cynics among us would say "Where was courageous integrity when the money laundering and the lack of compliance was going on? Where were the people standing up and saying this is all going wrong?" How much confidence do you have that people will exercise courageous integrity in your organisation?

Stuart Gulliver: Because of the fact that we have made a gating process for the whole appraisal system for the senior executives, I have confidence that for that group of people it will bite. Because they are then equally responsible, generally, for 20 people each, and there are only eight layers across the whole group, that will filter down through the firm.

Q3790 Mr Love: You have touched on it, but perhaps you could give the Commission a little more detail on this behavioural audit. How exactly will that work?

Stuart Gulliver: I think Barclays is doing the behavioural audit, not HSBC.

Mr Love: Let me move on, then, if that is not part of the process that you are going through.

Stuart Gulliver: Forgive me-clearly we have, over the last two years, in defining these values, done a ton of work. Remember, we are in 80 countries, so we have to get values that work across different cultures, languages, religions and ethnic groups. The things that can easily be criticised in a forum such as this are "Open, dependable, connected" and "Act with courageous integrity"-it all seems empty, but a lot of work went into defining those. We started filtering "Open, dependable and connected" into the organisation in about 2008 or 2009. I then made it mandatory in people’s appraisal process in 2011.

The behavioural stuff and what went on in Mexico is clearly absolutely shocking to us. I have been in the firm 33 years, and this gives me absolutely no pleasure whatever. You shouldn’t be under any illusion as to how seriously we take this and how upsetting the whole thing has been. We have crushed our reputation with the Mexican events.

Douglas Flint: One of the things that we have done recently that has been very successful for those of us who are somewhat prehistoric in our views is that we have an internal, online sort of YouTube television channel, and Stuart has done a number of extraordinarily crisp, no-holds-barred, three or four-minute segments as to what went wrong and how we should deal with it. What amazes me is the effectiveness of that channel. I think the last one was watched by 180,000 people. People really get this. Again, to make the point that Stuart has made in a different way, this organisation had, until recent events, an extraordinary sense of pride in what it had achieved, and 265,000 people did not recognise what they read. They desperately want to be part of making sure that it does not happen again.

Q3791 Mr Love: We will be watching with interest how this develops. Let me come on to the scorecard that you mentioned earlier. You indicated that there were financial and non-financial incentives within that. Is not it the case that "people and values", which was scored at 10% in 2011, has been reduced to 5% in 2012 while return on equity and progress dividend pay-outs are both still at 15% each? Isn’t that the wrong set of priorities on the scorecard?

Stuart Gulliver: The long-term scorecard has not changed at all; it is the short-term scorecard that will change each year. Remember, we publish the long-term scorecard. It is in the annual report and accounts, and we also publish how the board has evaluated me against both the short term and the long term. The key thing is that compliance and reputation have remained unchanged at 15% in the short term and 10% in the long term. You are right that "people and values" was adjusted, but in a way, some of the "people and values" stuff is double-counted by the behaviour values appraisal that I have just mentioned, which we run people through. If you have a values gate, you are effectively dealing with "people and values" through that. So I have to get through the values gate before I get to my scorecard, and that is why we felt that we could adjust down "people and values" because in a way we are double-counting.

Douglas Flint: Compliance and reputation stayed the same.

Stuart Gulliver: Compliance and reputation stayed the same, which is key from the control point of view. It was the introduction of the values gate that led us to adjust that a little bit, because it is dealing with the same issue.

Q3792 Mr Love: What proportion of your staff get through that values gate? This is critical. If it is everyone, then is it of any value? If very few people do not, is it an incentive?

Douglas Flint: It is not everyone, but very few people do not. There is a logical reason why it is very few people. I have been in the job two years. I have changed 18 of the top 21 people during that period, so in a way, there is a self-selection bias to a survivor group. You have selected what you think is a team fit for purpose in the post-2010 era, and you therefore hope that they will not immediately trip at the values piece. In a way, you have a survivorship bias in that initial period, but, no, it is absolutely not everyone.

Q3793 Mr Love: The people who pass will get a pat on the back and move on to the next stage and will hopefully be remunerated.

Douglas Flint: But it is a continuous process.

Q3794 Mr Love: What happens to those who do not pass? If there are a lot of people in a particular department, does that ring alarm bells? Do you evaluate those people? Do you assist them to ensure that in the next appraisal they will pass?

Douglas Flint: It will be a combination of probably two possible outcomes: one is remedial training, the other is that we may well dismiss those people.

Q3795 John Thurso: I would like to concentrate my questions around board-level corporate governance, particularly size and complexity. May I, however, quickly ask you a question relating to culture, Mr Flint? It goes to the original evidence that you gave to us, which I think you largely wrote, or certainly had editorial control over. There is a very interesting section in which you quote J. P. Morgan and make the point that it is as valid today as it was in 1933. You make two points. One is that where banks are successful is where they are "proximate to the communities they serve and where they are relevant to those communities." You also underline the need for professionalism-bankers as professionals.

My question is very straightforward. In a world where trading has come to dominate so much of banking, can bankers be the professionals that you clearly think they should be? Can that culture of professionalism and relationship overcome the culture of trading?

Douglas Flint: Yes, I think it can. It needs to be reinforced and in some areas re-established. Clearly, when the original J. P. Morgan wrote that statement, which is just as relevant today, there was not the capital market activity that we have today. But the capital market activity should in no way be regarded as a bad thing. Effectively, it is creating funding sources for larger companies and, hopefully, in due course in this country and in Europe, making more available for smaller companies as well, in that it allows them to raise money on their own credit without the intermediation of a bank. There is absolutely no reason why that cannot be done responsibly, ethically and very much in support of the community served by the banks, which are effectively accessing finance on behalf of customers in the markets, as opposed to lending directly.

Where I think you can raise more of a question-I have put this in evidence as well-is in the narrow area, and we believe that it is a very narrow area, of proprietary trading, where there is not, if you like, a customer; or, indeed, in some of the algorithmic and high-speed trading, where it is built into the systems and there is no personal relationship.

John Thurso: One of my colleagues may well ask you a question about that.

Douglas Flint: It is perfectly possible to have an ethical culture within markets businesses.

Q3796 John Thurso: Turning to corporate governance, I am fascinated by what Mr Gulliver has been saying, particularly about the move from the country kings, all of which makes a great deal of sense. I have asked a number of witnesses about the size and complexity of many institutions in the financial world. Everybody has said- many people have given us this evidence unprompted-that if you have something that is very big and simple, it is governable; if you have something that is complex and small, it is probably governable; but if you have something that is very big and very complex, it is probably ungovernable.

To what extent is what you have just said on money-laundering evidence that HSBC had tipped over into being ungovernable? Is your response, in simplifying by removing businesses and changing the structure, an answer to that?

Stuart Gulliver: Yes, it is. We put together six filters on which to evaluate every single business. The first two filters were about the size and the interconnectedness of the economy. The next three were about return on equity, advance deposit ratio-because we never want to be wholesale-funded-and expense ratio, and the sixth one was about financial crime risk. We ran every single country and every single business through it, and we very quickly ended up with a series of businesses that did not connect with other businesses-and there is your complexity.

At the end of 2010, HSBC was doing auto insurance in Argentina, sub-prime credit cards in the United States and corporate banking in Hong Kong. There is nothing in those activities that is remotely similar. There are no economies of scale from the systems that you can achieve, and there is no common risk platform that you can achieve. Therefore, we have sold the auto insurance business in Argentina and the sub-prime credit card business in the United States, so that is absolutely what we have been doing.

Q3797 John Thurso: That begs a question. Your reputation has always been as the guys who take culture and values seriously, probably more than many other banks, if not any other bank, in the world. Equally, you are clearly one of the top 10 in the scale and size of assets on the balance sheet. If you, who might well be put into the "trying to do it really well" group and who have that good reputation, could fall at that hurdle, what is the lesson for us as policy makers or regulators to put in place to help you and others like you not to get to that place in the future? We are all getting it right today, but it is our grandchildren that I am worried about when they come to run a bank -specifically mine.

Douglas Flint: The focus that is now being demanded is on purpose, and, whether it is social purpose or broader purpose, it is back to the proximate and relevant. Where our industry gets it more wrong than at any other time is when it loses sight of who the customer is. I genuinely believe that this generation of management gets it across the industry, because of what they have been through. If we have a regulatory system, and in particular a supervisory system, that is closer to institutions and has an understanding of where the risk the institution is taking is and where the relevance and proximity are being applied, we will end up with a better system. I genuinely believe that.

Q3798 John Thurso: Basically, you are the organisation that has the nearest to a complete subsidiary model. To what extent does your main board relate to, control or operate with the subsidiaries? Where is the overlap and where is the underlap? How does that fit in with the old country king model and the new model of "10 of that, 10 of that and a bit of the other"?

Douglas Flint: Let me start and maybe Stuart will come in. The board has spent a considerable amount of the last couple of years thinking very deeply about the governance challenges, because they say, "How do we govern this institution, given some of the issues that have arisen that surprised and shocked us?" There are two levels to that. Stuart has explained very clearly the move to functional management and business line, as opposed to geography. If you like, there is a horizontal to the matrix now, which is very much around business line and function.

At the same time, there is a vertical. We are subsidiarised. All our boards in all the top 20 or 30 or so countries have independent boards, so there is a legal entity balance sheet, a relationship on a bespoke basis with the local regulator, and all the boards have truly independent directors on them. We have gone further in the UK than we have gone anywhere else, other than the United States, in having a non-executive chairman of the board. That is something that is partly in response to the ring-fencing and getting ahead of it. We may well extend that.

We have something like 150 non-executive directors in the group. We bring them together and we share our experiences of what has gone well and what has gone not so well in governance. There are two lines of sight to governance: one is the management line of sight through businesses and functions, and the other is the legal entity oversight. I can absolutely tell you that boards today are very conscious of their responsibilities, because the issues that have arisen across our industry and within our own bank concentrate people’s minds on their accountability and responsibility. We spent a lot of time talking about it. We can talk about it now, or we can do it at another time.

We have augmented our board in the past two years, bringing on some additional expertise as the board was refreshed, but we have also, as you will have seen in the past couple of weeks, responded to the issues in Mexico and the United States. One of the things that we do not know enough about is the threats to the system that are arising in areas that we have simply not had line of sight to in relation to financial crimes. We have recruited advisers to the board, and we were delighted by the quality of the people prepared to work with us. They come from the intelligence world, organised crime detection and prevention, systems, terrorism financing and revenue and customs. They asked where we were worried the financial system could be abused, and how they could help us gain insight into the sort of thing we would hope a leading financial institution would do to prevent that access.

We are in the early stages-we have only just announced it-and we will work closely with those people to find out the areas that we are not even thinking about today where things might happen in future. One of the classics is cyber.

Q3799 John Thurso: If we are all agreed-and I think there is agreement-that if you are both big and complex, you have got a problem, then you are probably into the realm of ungovernables. The answer is either to be simple or be small, or possibly to be in subsidiaries. Therefore, I am testing-

Stuart Gulliver: Or probably big, simple and in subsidiaries, because then you have got multiple checkpoints.

Q3800 John Thurso: The danger then is presumably the need to ensure that you do not have underlap, because you have so many systems-

Stuart Gulliver: That things fall between the cracks.

John Thurso: And that becomes the function of the holding. The central becomes the internal audit for the governance for everywhere.

Q3801 The Archbisop of Canterbury: Mr Gulliver, one of the sub-panels of this group has been looking at high-frequency trading and algorithmic trading, which I think I am right in saying is an area that you have held back on very significantly. One of the questions that appeared virtually incomprehensible to those we asked was, "Where is the culture in the algorithm? Where are the ethics in the algorithm and in the high-frequency trading?" They are culture and ethics-free zones, apparently, once you have turned on the machine and pressed the button.

How can culture be embedded in institutions that are increasingly driven by automated technology? Presumably there are other areas of your activities which are not as potentially lethal but are significantly complex. You have just referred to threats to the system and cyber.

Stuart Gulliver: There are two levels. One deals with it at a practical level, and then at a more philosophical level. At the practical level, it will come to the extent to which boards and the risk and audit committees of boards get the people designing the systems in front of them and get them to explain what they have put into the system and how they have configured it. I would be surprised, perhaps positively, if the institutions at the leading edge of high-frequency and algo trading had boards that fully understood exactly what was going on in those activities, and fully understood the maths sitting behind it. That is a very common-sense check.

The second thing, though, as you say, is that once you have programmed it, there is no ethics or morality around it at all. It is whatever the programmer put in at the beginning, or the system spec. Therefore, if we are looking, as we should be, for the contribution of the banking industry to jobs and the economy, you have kind of cut it off at that point. So I think there is a philosophical issue about it. It may well therefore be an activity that should be conducted in hedge funds or outside the banking industry, rather than within large financial institutions.

Q3802 The Archbishop of Canterbury: That is a very interesting comment indeed, and it is obviously one that you have followed through. When we had algorithmic trading explained to us-both of us on the panel had considerable experience in trading and markets over many years-I am not exaggerating in saying that I was even more confused when they finished than when they started. I am sure your board members are a lot more intelligent about this than I am, but I simply fail to see how anyone could ever understand this. Does that not give us the danger that banks end up being IT companies run by banking experts? How are you dealing with the other areas in HSBC?

Stuart Gulliver: In terms of the complexity of particular approaches to trading, it is dealt with by a conventional risk limit structure. Any amount of high-frequency or algorithmic trading is going to lead to market risk, so you would capture it and the outputs at the market risk end. There is a philosophical question, as I said earlier, about whether-certainly on high-frequency-that kind of effective price arbitrage is something you want in a regulated bank.

On the complexity of the industry and whether banks become IT companies, the banking industry is, by definition, a huge customer of the big technology companies, and that is inevitable. We have changed the senior management of our IT area over the last two years, and we have on our board, in the form of one of our non-exec directors, Safra Catz, who is the chief executive of Oracle, someone who is specifically there to be able to ask those types of penetrating questions of our heads of IT. I think this has to be in terms of the selection of non-exec directors, so that that is how the board can be comfortable that the work of the IT area is up to speed.

Q3803 The Archbishop of Canterbury: Where you are running personal credit with the SME sector, even up to medium-sized companies, an awful lot of decisions on payments, vulnerability, risk, advancing credit, cutting overdrafts and advancing new loans are increasingly automated. How are you building into that- particularly in the SME sector, where volume, and, from your point of view, cost control, are really important-the human factor and the local knowledge? How does it happen that a decision is not simply taken by a machine far, far away that knows nothing of you or your area?

Stuart Gulliver: First of all, the automation of credit approvals for mass business comes from scorecards, and the scorecards will effectively filter for various criteria. It happens at two levels. One is that the construction of the scorecard itself can be sensitised to the specifics that you have just outlined. The second is an appeal process, and there is always an appeal process. There are people in branches-we still have 1,200 branches-and people will walk in and create an appeal. But, as you say, it is unlikely for mass business that you could have a relationship manager for every single client, because you are talking about millions of clients. So you have to automate it, but you absolutely have to have that check and balance in place.

Q3804 The Archbishop of Canterbury: Are you satisfied that there is a clear social-value filter and a clear moral-purpose filter in your credit scoring system?

Stuart Gulliver: Yes. We are absolutely aware of the fact that banks exist to finance businesses, which creates employment. Therefore, if the business is viable, we should be financing it.

Q3805 Baroness Kramer: I am rather interested in how culture and values get into the lower levels of your organisation-even more so, given some of your comments. If you are a bank whose particular vulnerability is going to be misuse by criminals and terrorists, and they cannot get at the senior level, it is obviously attractive for them to turn their attention to more junior levels, and there is no shortage of ingenuity in that respect.

When our panel on corporate governance below the board raised the issue of how values and culture, or your DNA, get down into lower levels-there may have been a passing mention of the TV channel, although I do not particularly remember it-your head of HR particularly relied on your group standards manual. With a bit more questioning, we got an estimate that it consisted of somewhere between 200 and 300 pages. I gather that every employee has to initial the bottom of every page to indicate that they have read it. There seemed to be great reliance on it. Some of us round the table have spent a lot of time looking at human nature, and the notion that culture and values are established through a 300-page manual struck us as fairly extraordinary. I wondered what your thoughts were on that.

Stuart Gulliver: The values are to be open, dependable and connected and to act with courageous integrity. The group standards manual is a policy document with policy statements about things like attitude to tax transactions, attitude to liquidity rules and attitude to capital rules; it is basically an operational manual, and the top 300 people in the firm have to read it every year, although they will not be initialling the pages, because it is now online. But that is not the values manual. The values are quite crisply annotated. The way they have been cascaded out is, as Douglas said, through the use of the internal TV channel. We have to deal with 80 countries, which have different cultures, different religions, different ethnic groups and different languages. We have therefore used modern technology to do a lot of it. I do a lot of stuff on it, as does Douglas. But the group standards manual says, "You will run with this capital ratio in this country. You will run with this liquidity in this country. Your AD ratio will be this. You cannot do the following things without approval from head office: you cannot go and lease premises, you cannot do this or that." It is not actually about the values; they are much crisper.

Q3806 Baroness Kramer: So you are basically saying that it is a compliance manual. In a sense, that almost makes me even more fraught, frankly.

Stuart Gulliver: It is not a compliance manual; it is an instruction as to how to run a bank.

Q3807 Baroness Kramer: Okay. Maybe you can help me to understand how compliance works, because this was represented to us as the way that compliance and values work down the system. Can you explain to me how compliance works down through the system? We have exposed that there is a real vulnerability in this area.

Stuart Gulliver: Within a particular country, you will have a compliance department with a head of compliance. That head of compliance will have a reporting line locally into the chief executive and globally up through to Marc Moses, who I think you met and gave evidence in front of you.

Q3808 Baroness Kramer: May I stop you there? As you know, everybody now has three lines of defence-it is standard terminology-and one of our primary concerns has been the blurring of the first line of defence, which is front-line staff who ought to do the right thing, and compliance. Over and over, in the various banks we have talked to, it is very clear that those two weave and interweave in and out of each other so that the second line is quite heavily compromised. If you are talking about a reporting line directly to the chief executive of your subsidiary, who is, by definition, front-line in that sense, have we not got a very serious blurring there?

Stuart Gulliver: No, I think you have to understand the structure of the firm. The subsidiary we have just described, as we were saying earlier, is a bank. It has an independent board, it is separately capitalised and it has separate liquidity. So that CEO has to have his compliance person. The regulator in Hong Kong, Malaysia or the UAE will insist that there is a reporting line there, because they equally want the escalation of that information and those compliance and rule breaches. It is compromised if the compliance person reports to the line revenue producer; it is not compromised if they report to the CEO of the bank in the country.

Douglas Flint: The thing that has changed is that the horizontal function split means that the UAE compliance guy is now assessed and remunerated by the head of compliance, not by the local guy in the UAE. He has double reporting-he obviously reports through the line that goes to the local regulator, but he also reports to his boss back in London. His boss back in London determines whether he gets promoted, how he gets graded and how he gets compensated.

Q3809 Baroness Kramer: May I follow up on the remuneration? You said earlier that your CRO is now one of the highest paid executives in the organisation, which is a significant change from the past. What about down the line? Historically, front-line people were very well rewarded, but compliance people were regarded as more boring, more standard, and were nothing like as well compensated. Has that changed all the way down the system?

Stuart Gulliver: Yes. We are doing exactly that all the way down the system for exactly the reasons you would expect us to, because we need to get the balance. Actually, we are also doing that because we need extremely talented people in those jobs.

Q3810 Baroness Kramer: I noticed that when you were talking about your senior staff, you have a very sophisticated way to reward them-gateways and balance plans and whatever else-in order to bring in some of the more value-type issues. How does that work at your more junior levels?

Stuart Gulliver: In the UK, for example, starting from the beginning of this year, we have removed all sales targets from individuals in the branches and from regions, and the plan for the retail banking business is actually set at the country level. The scorecard for people in the branches therefore contains criteria in relation to customer service and the values.

Q3811 Baroness Kramer: Sorry, are you saying that you are using a scorecard all the way down the system? When we talked to your folk, it seemed that the scorecard only applied at the senior level. After that there was none of that precision of criteria.

Stuart Gulliver: No, there are different scorecards at different levels. The one that is published in the annual reports and accounts is mine. It is the most detailed one, because arguably I have the biggest span of control. As you get down to a person sitting in a branch, it becomes much more specific. The values and the behavioural requirements are now set in their day-to-day requirements to conduct their job.

Q3812 Baroness Kramer: I will go back to the quote from Ann Almeida, who is head of your HR: "values are formally expected to be taken into account"-this is at all levels-"but there is not a distinct and separate process."

Stuart Gulliver: It is part of appraising someone in totality, in the round.

Q3813 Baroness Kramer: But on a much more subjective basis.

Stuart Gulliver: Absolutely, but it has to be. If you think about it, we are going from a situation where, in 2008 in the HSBC retail bank in the UK, we had a points system: if you sold X you got Y points; Y points equals £X. To remove that because of the mis-selling issues that we do not want to have repeated, inevitably you are going to have a discretionary scheme that has some subjective input. One is objective and hard-coded in points, the other is inevitably subjective.

Q3814 Baroness Kramer: It sort of does suggest that there is a vulnerability as this gets vaguer and washier. We asked your head of compliance, "Who knows what the philosophy of the firm is?" The reply was, "The senior compliance staff"-thank goodness; that is good-"and we hope that it permeates all the way down." I am trying to probe whether or not you have been effective. You have had a change in direction and a strengthening up at board level and senior level but there continues to be a vulnerability lower down the system, which those who wish to take advantage of the system will spot more quickly than I will. I raise that as an issue. It is not a very happy way to-

Stuart Gulliver: No, I believe that we are continuing to push this right down throughout the organisation.

Douglas Flint: Over 100,000 people went through values training last year.

Q3815 Baroness Kramer: But they were all senior, weren’t they?

Stuart Gulliver: The 100,000 were not senior.

Q3816 Baroness Kramer: I think we were told it was for senior people.

Douglas Flint: No, 100,000 staff.

Stuart Gulliver: Some 60,000 went through the year before, and the guys who did the two-day course-

Q3817 Baroness Kramer: We were told that 3,000 out of 3,600 people had gone on a two-day training course.

Stuart Gulliver: That is the two-day training course. In total, 100,000 people had some form of values training. The two-day course was run by Harvard business school for us.

Q3818 Baroness Kramer: Okay. So that is the really substantial course, and then there is a cut-down version, as it were.

Stuart Gulliver: Yes.

Douglas Flint: And then there is a huge amount of learning through what we call "HSBC academy", which is online learning. So everyone at a certain level or in a particular function-including me-will get an online module to complete. You have to go through it, do a test at the end and pass the test to stop getting e-mails saying, "You haven’t completed this to our satisfaction yet." A whole bunch of modules have come out on compliance, training, money laundering, and anti-bribery. They are distributed on a fairly regular basis to all relevant staff, who must do them online. They do the online learning, and then they do an online test. You can then monitor that everybody has done it, because they will have to have got the right answers to the questions.

Q3819 Baroness Kramer: One final question: I just wonder how you monitor the impact of all of this. When you look at this as an outsider, people are bombarded with vast manuals and endless things to fill in online. You will know that every bank has its own characteristics. Your bank’s is to write 1,000 pages rather than one. We looked at how many pages it took to open a bank account and you could hardly lift the number of pages. Is anybody looking to see whether all this stuff that is being bombarded out is actually registering in an effective, behavioural way, and providing the guard that you are obviously trying to see?

Stuart Gulliver: We do quarterly pulse surveys to get people’s feedback. The questions, in part, are to try to tease out exactly what you are describing. The pulse surveys should give you the feedback on whether the messages have sunk in. We are tracking through a quarterly process to see the extent to which our own messages are coming back to us or not.

Q3820 Mr McFadden: Mr Flint, I want to ask you a few questions about the relationship between pay and behaviour. You, as the chairman of the bank, are the guardian of that. I was listening to Mr Gulliver on all the measures the bank is using to try to inculcate the right structure from the top to the bottom. Over months of evidence, we have heard so much of this, full of the terms you use, the different value statements and mission statements that banks have, and e-mails to all the staff and so on; yet we sometimes find a contrast between all those rubrics and statements, and behaviour, which is influenced by pay. I think this is at the heart of the culture. I want to get your view on the relationship between pay and behaviour. Why is it that banking has this strange culture, with multiples of salary available as bonuses and incentives to senior staff? My understanding of Mr Gulliver’s potential package is that you have a salary; with that, the bonuses can be up to three times salary; then there is a long-term incentive plan that can go up to six times salary. Correct me if that is wrong.

Douglas Flint: That is correct.

Q3821 Mr McFadden: Apart from the amounts, which are obviously very large, for most people-for our constituents-that is a really strange way to pay people. What do you think that structure of pay does in terms of behaviour?

Douglas Flint: I think used properly it is a very powerful force for the good. Let me explain. One of the things that has evolved over a decade or so in the structure of compensation in the industry has been to seek to enforce control and, increasingly now, values by saying, "We expect to pay you 100, but we are only actually going to pay you somewhere between 10 and 20, guaranteed. The rest you are going to have to earn by a combination of meeting values-that is a gating procedure, so if you don’t meet the values there is no chance of getting anything else-and doing your job in a way that is consistent with the performance criteria that have been set for you: it is not just whether you have made money or not, it is the way you have made money, the clients you have dealt with, and the client satisfaction that has fed back from them."

Rather than paying someone what you expect to pay them and then just saying, "And try to do a good job," we pay them a good deal less on a guaranteed basis than we expect to pay; then, if they meet the values and their performance targets, they get paid some of the rest, though not all of it-it is most unlikely that anyone would ever meet 100% of their scorecard; then, for most, we defer the short-term elements of their pay for up to three years and the long-term element of their pay for five years, and for the more senior staff, even once that has vested, they can’t take it until they retire.

So, to the extent that you have rewarded somebody for a set of circumstances that turn out to be either overstated or different somehow from what you thought, there is the opportunity for you-and I think this is pretty much unique, still, to our industry-to have the ability to adjust their reward. I think that, in structural terms and control terms, the fact that people are leaving money in deferral, and ultimately until retirement, means they clearly have to behave, because otherwise they can see that pay reduced or cancelled. If they are leaving money in the stock at the front, should the firm suffer value destruction because of collective behaviour, their cumulative reward reduces to the same extent. Structurally it is a good system.

Q3822 Mr McFadden: But it’s completely weird, the way in which, with that money, you are saying, "We might pay you 100, but we will start off with 10 or 20." In any other walk of life, you would get paid 70 or 80 of that 100, and, even if there was a bonus system, it would be 20% or 10% of your salary. The way this is structured in banks is just weird. I am not convinced by your argument that this structure of pay produces exceptional performance and inculcates a good culture. I think it is a very strange culture-and that is apart from the amounts.

Douglas Flint: I should say that for the vast majority of staff who work for us it would be more likely to be 20: the people in the branches are not getting paid 20%; it is the other way around-they are paid a salary and they will get maybe 15% or 20% on top. That is for the vast majority of people by number. For the people who are taking significant responsibility and those who expose the organisation to risk because of their absorption of risk in activities that they manage or through their control of risk, through their controlling roles, there is, I believe, a very rational structure that says, "It is most unlikely that at the end of the year we will know exactly whether everything we think we have done well will have been done as well as we think."

For someone who comes into the branch every day and does the job, at the end of the year you are pretty much certain what they have done. For someone who is doing something complicated in the bank or who has a control function and may have overseen the expansion of credit in a particular area, you may in three years’ time say, "You know what? That wasn’t done very well. We would have adjusted your pay if we had known back then what you were doing." So to some extent what you are doing is giving yourself the ability to adjust for future facts and circumstances which would have been taken into account had you known them at the time you rewarded someone. So I think that mechanism-deferral and adjustment-is a protection both to shareholders and to the company, and to the system generally. You cannot create a circumstance where people get rewarded today because it all looks terrific but in three or four years’ time it does not.

Q3823 Mr McFadden: You talked about risk there. We were set up in the wake of the LIBOR scandal, which is still unfolding. Don’t you think the structure you are talking about had an impact on the behaviour of the traders in the various banks implicated in this? They knew that by trying to manipulate the rates they could significantly increase their own personal earnings. If they were more on an 80:20-type structure, surely the incentive to bend the rules, break the rules, do things that are dishonest would be far less? You have been talking between the two of you for over an hour about culture, values and all of that. I think this pay structure can conflict with those good intentions.

Douglas Flint: I will let Stuart say something because he has been much more involved in the trading area than I have ever been. Let me put it the other way to you. I don’t know what incentive people we have read about so far had to do what they did, but if you had the circumstance where someone did not have deferral-got paid 80:20-but got paid a lot of money, there would be no way of adjusting it thereafter. That person would have been paid the money. They might have left the organisation. There is no way you can get it back. The control aspect is that if someone is paid and is paid on a deferral basis-this is what the structure is designed to do-they know that if they have cut corners or done something that turns out to be mis-stated or overstated or has been dealing with the wrong kind of people, they will not be rewarded even though in the first instance they might have accrued that reward.

Stuart Gulliver: There are two answers to your concern. One is that there should be a whole series of controls within the finance and back office areas around dealing rooms to deal with some of the issues that have surfaced in other banks, separately from any compensation structure. But then on the compensation structures, as Douglas says, the problem with this is that a calendar year is a reasonably arbitrary period of time on which to evaluate someone’s performance. Part of what Mr Love was saying earlier about no one seeming to take responsibility is partly because all of this-in our case, Mexico-happened five or six years ago. Those people have gone. So you want a long period of deferrals so that people’s compensation remains at risk, for the simple reason it often takes a long time for these things to surface.

Part of the reason you want a significant amount of deferrals circling around a trader is you will not know within a particular calendar year, unless it is a foreign exchange trader where the position is closed out and realised, what the true value of that position is until the position is disposed, and if it is a banker till the loan has been repaid. So the structure of deferrals is important to contain risk. I can’t remember which bank it was, but there was some press coverage a couple of weeks ago about a trader from one of the European banks leaving behind £34 million, or something. That, to my way, is this type of structure working.

Q3824 Mr McFadden: Isn’t there an illusion in this of how talented bankers are? We have heard evidence of a sense of entitlement at the senior levels of the bank and this concept of the talent war. I look at sport and see someone like Lionel Messi or Cristiano Ronaldo who would be given the exceptional rewards. I also know they are exceptional sportsmen who would be extremely difficult to replace. In the case of Messi, maybe it would not be possible to replace him. But bankers aren’t Cristiano Ronaldo or Lionel Messi, are they? They are more replaceable than they think.

Stuart Gulliver: I would say that probably footballers are not a good analogy anyway, because footballers don’t buy sub-prime businesses in the United States. So I don’t think the footballer analogy works. What I would say for HSBC is that we have always been at the lowest end of payout ratios. I can only comment about HSBC, and from published data over the last couple of years, our code staff have been paid at a rate of about 50% of the code staff of RBS and Barclays. As you know, code staff numbers are now unpublished. All I can say is that in terms of ours, we pay the minimum that we believe we can pay to retain and attract top-quality people to drive the share price.

We had this 1:3:6 complicated structure approved at the AGM of 2011 by HSBC shareholders by a vote of 86.4%. I can understand why it appears complicated, but the owners of HSBC were reasonably comfortable with it. We pay what we require to keep people, but we are in no way inflating and if you talk to any other bank, we have always had a reputation of being very conservative in this regard.

Douglas Flint: May I make an additional point, which is contextual? For someone to aspire to earn an extraordinary amount of money, the organisation needs to give them the risk limits that would enable them to generate a volume of money benefit for the institution that would then enable them to be paid a certain amount themselves. We do not have the risk limits that would enable the profitability. The reason we pay people less is because we do not give people the kind of risk structures and risk limits that would enable them potentially to generate the revenue stream that would lead to a certain amount of reward. So I think we pay people very fairly and well for what they do. If someone wants to earn a lot more money than they can earn with us, it is almost certainly because they want to have a risk limit that we wouldn’t give them.

Q3825 Mr McFadden: Last question: we have heard a lot in submissions about how pay structures are changing and more is deferred. You talked about it just now. From where you are sitting, in terms of policy or good practice in banking, how far has this culture of deferral got to go, and what more will need to be done across the sector? I am not just talking about your bank. What more needs to be done to inculcate what we would want to see as the right behaviour, rather than reckless and risky behaviour? What else should be done?

Douglas Flint: I actually think the industry has gone a heck of a long way. You are seeing a number of institutions go up to the five-year deferral level for long-term reward when they used to be at three, and they are going for five-year cliff vesting as opposed to amortised. We are still on our own in terms of saying that even once it vests you stay with it until you retire.

Q3826 Mr McFadden: Could you expand on that point?

Douglas Flint: Well, for the most senior folk in the firm, after five years the money vests. Half the shares get sold to pay the tax; the retained portion is not sellable until the individual leaves the firm, retires. Essentially, why the shareholders were very supportive of that scheme is that it effectively creates the symmetry between shareholders and employees because the employees are genuinely shareholders who are still in the game. While it would not make a huge amount of difference to the current generation of very senior management, by the time the next generation comes through and they get to their mid-50s or whatever, they will have a significant portion of their wealth in the stock of the firm and they will care passionately about the value of the firm.

Q3827 Chair: I have two points on that. Given the very long lead time between the business decision and the discovery that it may have been flawed after all, as you have just begun to elaborate you need much more than five years for a high proportion of this remuneration, don’t you?

Douglas Flint: I think there is a balance. If you said to people that they do not get any reward for 10 years, I am not sure how you could employ people on that basis.

Q3828 Chair: You could pay them a higher base pay, as you have a different balance between the base pay and the bonus structure, as Pat McFadden was implying a moment ago.

Douglas Flint: The danger would then be that you do not have the ability to adjust if things turn out differently. There is a paradox. There is quite a lot of public support for clawback, which can only happen with deferral and deferral can only happen with bonus. You cannot defer someone’s salary. On the one side, we want the ability-and we, in the industry, want the ability-to adjust for circumstances that were not known at the time, which would have made the award less or non-existent. With that basis, there is a pressure to have more incentive pay because it can be deferred.

If you go the other way and say, "We don’t like bonuses", you don’t have bonuses and, if that is believed to be detrimental to behaviour, one would go that way. But then you don’t have anything deferred, and you don’t have anything to claw back if things turn out differently. There is a balance somewhere.

Stuart Gulliver: I think that there probably needs to be a different approach, depending on the age of the particular employee. At 25 years of age, deferring for 20 years may not be that appealing to them. At 40 years of age, deferring until retirement may be perfectly reasonable. On that earlier point, which is more likely to be the trader, a five-year deferral may be the more logical way to operate.

Chair: Maybe a mixture of deferral and clawback.

Stuart Gulliver: Yes.

Q3829 Chair: It also touches on the very last point you made in response to Mr McFadden, when you were saying that your staff have skin in the game because they have equity-the problem of interest in the equity value. Unless you have a very long lead time for that as well, you are at risk that equity encourages gearing. They will want to see a highly geared company.

Douglas Flint: I disagree. Gearing is terrific when it is rising, and pretty dreadful when it is falling.

Q3830 Chair: This is an exceptional time at the moment.

Douglas Flint: No, I disagree.

Q3831 Chair: For the past 15 years, we have had this long spell in which gearing looked good.

Stuart Gulliver: I don’t think that the regulatory environment will ever allow that gearing to come back. We never ran with it. If you recall, one of the things we got challenged on in 2006-07 by an act of a shareholder was that we were under-geared and needed to leverage our platform up.

Douglas Flint: I think that people have deferred compensation and equity. What they are looking for is for that stock to grow in line with the firm, which will be in line with GDP and, hopefully, plus a bit, and to pay a good dividend. They are not looking to leverage the equity. They have got their salary; they have got their compensation, and they have got their past compensation deferred all in one stock. You wouldn’t put all your eggs in one basket, and then lever it.

Q3832 Chair: There are a lot of detailed and difficult considerations to take account of here, but we all seem to be moving in the direction of saying that clawback and deferral are essential in your industry.

Douglas Flint: I believe so.

Q3833 Chair: In which case, do you think this is something that the PRA should be taking a much closer interest in? Should they regulate, and require deferral and clawback in a number of activities?

Douglas Flint: They do.

Q3834 Chair: Only very small at the moment.

Douglas Flint: For all code staff it is required.

Q3835 Chair: All risk takers are code staff?

Douglas Flint: All significant ones, yes.

Q3836 The Archbishop of Canterbury: I am increasingly baffled by some of the conversation that we are having at this stage of our questions. What is it essentially about bankers, code staff especially, that means that they need skin in the game? We do not give skin in the game to civil servants, surgeons or teachers; there is a whole range of people who do not have that. It seems to me that you are running what you quite rightly describe as, and are putting huge effort into, a values-based organisation, with a strong values-based culture. Yet, at the end of the day, particularly for your most senior staff who are most important as regards setting values and culture, you seem to be saying that the only way you can motivate them to any significant extent is with cash, deferred or otherwise.

It seems that they want both the bonuses deferred-yes, fine, with clawback-but surely if they are going to be there you should compensate by having much lower pay than equivalent professionals for your whole career and then you get a big balance at the end, or, alternatively, much higher pay and no balance at the end. But you seem to want the high pay and the balance at the end. I just do not understand the relationship between that and a values-based and culture-based organisation, which you are definitely trying to do; that is very clear. How do you deal with what seems to be an internal contradiction which we simply do not hear in so many other sectors of our society?

Douglas Flint: I think you will begin to hear it.

Q3837 The Archbishop of Canterbury: So you think surgeons will be saying, "I want skin in the game"? Teachers will be saying, "I get paid a bonus"?

Douglas Flint: No, no. I was talking in the private sector and corporately. One of the things that is not unique to banking but is very acute-our shareholders buy us because of an assessment of the long-term growth in the firm and the dividend stream that they will get for it. It is not the case that looking at an arbitrary period of a current year will necessarily give a very good indication as to whether the decisions that have been taken within that period of time are going to create the value that they expect in the future. Therefore, to create alignment, to say that for the decisions that have been made in this year, we are going to align the senior management with shareholders; that is, if those decisions turn out to be great decisions, then the pay that we have given them in relation to making those decisions will accrue to them, and if it doesn’t, we can take it back-you could take the same view in any long-term industry where you are making investment today but the reward will come in the future. That would not be the same as a surgeon, a teacher or anyone else; they do their job and they get paid. The input and the output are much proximate than they are in our industry. I think that is the difference.

Q3838 The Archbishop of Canterbury: But that never used to be the case in banking. This huge bonus culture is something that has grown up since the 1980s.

Stuart Gulliver: It probably was under the partnership structures pre-big bang, very similar actually. And post big bang.

Q3839 The Archbishop of Canterbury: But not at HSBC; I remember dealing with them in the 1980s.

Stuart Gulliver: But we remain at the conservative end.

Q3840 The Archbishop of Canterbury: Do you not accept the argument that you should have much more normal kinds of pay levels? Generous, huge pay levels, say £60,000, £70,000, £80,000 a year, and then get a bonus-a lot of stuff coming at the end, when the wonderful results of your long-term management come through.

Douglas Flint: The vast majority of our staff are actually paid-

Q3841 The Archbishop of Canterbury: But not your code staff.

Douglas Flint: No, the vast majority of our staff are probably at average or below average earnings. What we are talking about-and here is the difference from 20 years ago-is that the scale of the capital market’s activities globally in our industry is significantly greater, the durations are longer, the complexity is greater and the pay structures are too. I think partly reward people for being successful in those areas, but to protect the firm and shareholders from misrewarding people for things that they thought were terrific at the time, but turn out to be wrong in the future.

Stuart Gulliver: There is also a complexity that does not answer your question, your Grace, but it is a complexity. People invest in HSBC in part for the emerging markets, and the emerging markets have not had a financial crisis. In our operations in Brazil, Hong Kong, China, India and the middle east no banks have failed; there has not been a financial crisis. Therefore, there has not been the level of understandable introspection and thought about the things that you are raising. If we moved to what you have just suggested, we would not be able to retain a great many of our Chinese, Indian or Brazilian colleagues because the local market circumstances are completely different. We would therefore destroy the investment case as to why actually UK pension funds and insurance companies own HSBC-which is to get exposure to emerging markets. It is not an explanation, but it is a further complication if you are sitting where we are.

Q3842 Chair: While we are on the subject of the relationship between base pay and bonuses, do you think that, particularly for senior management, base pay should not be fixed but capable of going down as well as up?

Douglas Flint: There is no reason in theory why that shouldn’t be the case.

Q3843 Chair: You have used the words "in theory". In practice do you ever vary it?

Douglas Flint: I cannot think of circumstances-actually, when we aligned the group management, where we aligned everyone to the same fixed payout, there were-

Stuart Gulliver: There were a couple of adjustments down

Douglas Flint: There were a couple of adjustments down; but it is rare.

Stuart Gulliver: Logically you would take the whole variable piece out completely. If you are dealing with a situation where performance has been poor-and for example this is the case with both of us for financial year 2008; we did a rights issue in 2009, we went to our existing shareholders and raised an additional amount of capital: none of us took bonuses. So we zeroed the bonus. We did not adjust the base. So I think that if you have got rid of all of the variable and now you are attacking the base as well, aren’t you really just saying to the person, "Leave"? That is really the practical consequence of it.

Q3844 Chair: It depends what they are earning. If they are earning, as Bob Diamond was, £1.3 million in base pay, saying, "Well, we’re going to drop it to £1 million" might not necessarily be a "Leave" signal.

Douglas Flint: I think you have had examples in America of the pay being dropped to $1.

Q3845 Chair: So this isn’t theoretical; it is realistic.

Douglas Flint: No, but then there were some-

Stuart Gulliver: So, yes, in theory-but you would need to think of the circumstances.

Q3846 Lord Turnbull: You rightly stress the importance of behaviour, and I think we rightly have probed your efforts to entrench that in the firm. May I now turn to the other player in this, which is the regulator, the public authorities? Mr Flint, I read your excellent lecture of 29 May last year. You said, "I would encourage the regulatory and public policy bodies to think more deeply about how they can get to understand and if necessary shape the character and culture of the organisations" and that they should "care more about tone from the top, how individuals are screened for behavioural characteristics when recruited or promoted, how ethics and values are taught and reinforced" etc. In other words, you are saying they have got to come on this journey with you. Do you think they are able to do that, or are doing that at the moment?

Douglas Flint: I think the creation of the conduct authorities is certainly a step forward in that direction, but what I was trying to say is that we tend to have become very focused on metrics; so we love measuring capital and liquidity, and all sorts of things we can get our hands around, and look at. You know, "We ask you for 10.2 and you have got 10.1; that is not good enough" or "You have got 10.3; that’s fine"-but actually, as has been clearly evidenced through the last several years, it has been the wrong incentive structures, the wrong role models, the behaviour in senior management ranks as to how they appraised and promoted people, and so on.

So getting an understanding of how do you go about recruiting people-is your scorecard set for ethical values, or is it set for aggressive behaviour? You can do a tremendous amount of psychometric modelling and say: what kind of person are you trying to recruit to this firm? An aggressive person or a passive person? A collegiate person or a go-getter? How do you set your criteria? How do you train people? How do you assess them on performance? Is it purely on metrics or is it about softer issues, and so on? I think that, in a way, is an area of under-development in our regulatory framework, which of course is undergoing change at the moment, which was, for the last decade, very metrics-driven-and very metrics around hard things like capital and liquidity.

Also, I am unashamedly an accountant; I do think that there is an aspect to professionalism, which pervades medicine, law, accounting and so on, where you are taught ethics as part of your course, and you are taught that you have a duty to your profession that transcends whoever your employer is at a particular period of time. I think banking used to have that. HSBC and, curiously, Midland Bank, used to give a copy of George Rae’s "The Country Banker" to staff when they joined the firm. It was a book written in the 1930s about how to run a branch in a country village.

Q3847 Lord McFall: 1887.

Douglas Flint: It is a terrific book about a world that has long gone, but it encapsulated the behaviours and the prudence that were expected-it was kind of Micawber-like. It will be difficult to do and I know that you very focused on this, but trying to create a professionalism with a broader duty that transcends the profitability of the firm I think is really, really important-shared tasks between the public policy supervisory regimes and bank boards and management.

Q3848 Lord Turnbull: In the same speech, you warned about training a future generation of bankers to follow an immensely detailed rule book and training them, in effect, to work to the rules and not to the spirit. The new regulators say that they are aiming to concentrate on key issues and not follow a tick-box approach, but when you look at what is happening day by day in terms of the information they are asking you for, are you seeing sufficient signs that this change in philosophy on their part is taking place?

Douglas Flint: Gosh. At the top levels there is much more attention to the broader issues. Andrew Bailey in particular is extremely focused on the big issues-the issues that should impact the board. There is still a lower level of information-gathering to inform a regulatory supervisory system that is still commanding, partly for their own purposes and partly because the European Banking Authority requires it. I think we will give six times more information to our regulators under the EBA rules than we do even under FSA rules. There is still a tremendous amount of data capture, which is incredibly low-level detail. I hope that as we get into the new regime we can use a much more interactive relationship to get to the forest and away from the twigs on the branches of the trees.

Q3849 Lord Turnbull: One thing that was clear to us from examination of HBOS is that the supervisors did not spend enough time looking at the real things on the ground-the nature of the loan book. Too much of it was on process, particularly models and whether a model was good enough to qualify as an internal model. Is that approach of relying too much on models and the processes around them still there?

Douglas Flint: It is still there, but it has been challenged now. You will have seen the Basel Committee’s recent report, following quite a lot of market comment and analyst research that sought to understand, but failed to do so from published information, why risk-weighted assets were so different in many institutions against gross assets. There were explanations that made sense, but there was enough comment to ask whether we could be confident about whether the risk weighting was consistent around the world, and therefore whether we were measuring capital on a like-for-like basis.

There is beginning to be a much greater thoughtfulness around the use of models-whether there should be more of them and whether they should be more complex, or whether one should stand back and take broader measures such as the leverage ratio-Andy Haldane has written a lot on this-that are perhaps more indicative of the big issue rather than getting into the fourth derivation of the mathematics behind a particular element of a model. There is a lot to do in this area.

Q3850 Lord Turnbull: You have anticipated my next question, which is on the divergence between how people assess their own risk. The difference seems to be completely indefensible, other than that people will want to define it in ways that suit their own interests. So you would support more standardised but perhaps more rough-and-ready metrics, rather than allowing people in effect to write their own model as their own risk weightings.

Douglas Flint: It is not quite as bad as it sounds, because all the models used for regulating capital calculations are approved by the regulators. So what is being challenged in some respects is, first, did everyone use the same judgments when they were approving the models? What is also evident from the work that has been done is that in many jurisdictions, regulators would add buffers and add-ons because of model uncertainty. So part of it was the sophistication of the institution versus the supervisor in terms of how it constructed its model; part was supervisors adding buffers and add-ons for uncertainty. There is a bit of an unlevel playing field.

Maybe one answer is to have a centralised body approving models rather than doing it in multiple jurisdictions around the piece. Maybe part of it is moving to a simpler model basis. The problem with a simpler model basis is that you go back to Basel I, which had five risk weights, which is clearly deficient. Then we moved to risk-based modelling in Basel II. The problem with that is that you end up losing line of sight between the balance sheet.

One of the things that has been a tremendous step forward-we will see how far it is adopted-has been an enhanced disclosure task force co-chaired by a colleague of ours, our chief accountant Russell Picot. It was prompted by the Financial Stability Board to create a framework, alongside a lot of big fixed-income investors and regulators, to say: "Let’s take the board’s assessment of risk appetite, translate that to where the balance sheet is, translate that on a line-by-line basis or a subsection-by-subsection basis to where the risk-weighted assets are, then look at the return against those risk-weighted assets, then look at the return on capital by line of business." I think if we can make progress on that disclosure it will lift a great deal of the opaqueness of the business models of banks. This was produced only in the autumn and we and many other big banks have signed up to progressively implement it. I think it will help a lot to reduce some of the opaqueness.

One of the weaknesses of the crisis was an overreliance on models, whether it was bank managers, risk managers, regulators or rating agencies. The models did exactly what we were talking about in algorithmic and high-speed trading. They did what they were supposed to do, but if the inputs were flawed, if the mathematics was weak or if the experience period that was put into the model in terms of stress testing was different from what happened in the future, it didn’t work.

Q3851 Lord Turnbull: Can I ask you about people? If we are asking supervisors and regulators to exercise more judgment, they have to be people with pretty much the same intellectual talents as you are recruiting-if they are going to come into your business and talk about your business models, they have to do so with authority-but the regulatory authorities are not going to pay people on the basis of 1:3:7; they will be more like the 80:20 model. How will we ever get parity between the quality of people in the regulators and the people they are talking to, when their remuneration structures are so vastly different?

Douglas Flint: I would say two things. One is that there are hugely talented people in the Bank of England and Treasury, which have always been aspirational places to work. The people we talk to there have intellectual qualities that are at least as good as ours, so there is no problem there at all. The depth of talent is different than in financial institutions, and that is something regulators will have to work on. I read in the papers this week that if the ECB takes on supervision, it will require in the first instance to find 2,000 supervisors. That sounds like a big challenge.

The second question I put back to you as a thought. There are jurisdictions in the world, the most celebrated being Singapore, where they pay their supervisors and regulators equivalent to the people whom they are regulating. They would say that one of the reasons they have had fewer difficulties is that they have the same quality of people, and since the cost is passed to the industry, it is not a public cost and it is a much lower cost than having to clear up after a crisis.

Q3852 Lord Turnbull: Our report will clearly have to have a chapter on regulation and its future. Are there one or two messages that you would want us to include in it?

Douglas Flint: Focus on the big issues. If you are talking to the board, you should be talking about the issues that are at board level, which are the big structural issues: the risk appetite, culture, the quality of people, succession planning. Also, be much more interactive. We went through a period of regulation being very metric-driven and, if the metrics were okay, there was less point-to-workpoint contact.

The final thing that I will say-Stuart can add anything-is about the need for that more proximate interaction. We are dealing today with issues like PPI and interest rate swaps, where we are determining that it may not have been as it should have been for a very long time. We have to try to find a way, between the industry and the supervisor regulators, of being much quicker to identify that there is an issue and fixing it, rather than saying we have to compensate for something going back a significant number of years, because that makes it very difficult to think about what might happen in the future that you should have known about today. It also creates a climate of nervousness within the networks as to whether they should take on activities, because they were doing things that they thought were okay, but that are now judged to be wrong. There is a whole bunch of things that we all agree were wrong, but we have to try to find a way to get to these things when they happen in the future, because circumstances change much quicker.

Stuart Gulliver: I have nothing to add.

Q3853 Lord Lawson: May I follow up on some of the areas that Lord Turnbull has been asking you about and to which you have given very good answers? Clearly, what we have been through is not merely a series of banking disasters and scandals, but a serious regulatory failure both nationally and globally. I go back to what you said right at the beginning of this hearing: that you favour a strong regulatory system. I’d like also to divide it between the system and the people, and the regulators themselves and the system.

On the system, is it not the case that, while Basel III might not be quite as bad as Basel II, it is the same sort of approach and suffers from many, if not most of the same, defects in the complexity, box-ticking and all that? I do not know whether you are affected by Solvency II, but is not that also a pretty disastrous system? We have had evidence that it is. Indeed, this goes for the European system, as you have said. Do you agree with this? What shall we do about it?

Douglas Flint: Gosh. Andy Haldane did a really thoughtful piece about whether we have made it too complicated. I do think that complexity is a risk because you end up being guided by the output of that which is thought to have been constructed with huge intellectual rigour, whereas a common-sense approach might say that it doesn’t make sense. In particular, if you make the modelling very complicated, there is no line of sight to let you ask, "Does it feel right?" which is why I think that enhanced disclosure may help.

I think we have to get some simpler metrics. Capital has clearly been enhanced enormously and that helps; the liquidity improvements that have been done are immense, and again that will help. Many of the problems that have arisen in the last several years have come from very simple things like inadequate capital, funding models that were short-term funding illiquid long-term assets, an over-concentration in particular asset classes and as always, real estate, whether it is commercial or otherwise. None of these needed complicated models to say that the Spanish caixa or the Irish banks were massively exposed to property multiples that were historically very leveraged.

I believe that a much more interactive approach is needed, with the macro-prudential tools available to the Financial Policy Committee here-there are similar structures elsewhere-to look at the shape of the system and ask, "Is it doing what it is supposed to be doing? Are there strains and stresses that are arising from the concentrations that may not be apparent yet on a micro basis?" The problem with bubbles is that they look terrific when they are rising. We observe that some of the products, structures and issues that have caused immense pain in Europe and America have actually created enormous wealth in emerging markets, because in rising asset prices, leverage is a good thing; in falling asset prices, it is horrendous. In a sense it depends whether you are cheerleading in a rising market or commiserating in a falling market. That is why the macro-prudential oversight, which you know we believe very firmly has to be governed by the Treasury because it sets the risk appetite for the country’s system, is hugely important to get away from the micro of looking at an individual commercial property loan and so on.

Q3854 Lord Lawson of Blaby: It is also, as Lord Turnbull pointed out, the fact that some of the metrics-I am thinking particularly of risk-weighted assets-clearly can be gamed by the banks. I am sure you didn’t do that, but clearly they can be, and in many cases studies have shown that they were.

Douglas Flint: I’m sure that’s right, but I am sure it is also the case that if a risk-weighted asset is a product of probability of default and loss given default-when I talked about Basel II, I used to use the example of how between 1865 and 1998 the Hong Kong and Shanghai Corporation did not lose a single dollar lending on real estate in Hong Kong. Does that mean it is a risk-free asset? No, because in 1998 property prices fell by 60%. If you look back in this country over the last 20 years, property has been a risk-free asset-it has risen faster than inflation-so if you looked at the probability of default it is very low, and if you look at loss given default it is negligible. Therefore, your model gives you the wrong answer, because as the price keeps rising it looks like a better and higher quality asset and you need less and less capital against it-but that is where your macro-prudential comes in and says, "Hold on, we have got to a point that this is perhaps defying logic and maybe we should put in something to take the punchbowl away." Given the obvious sensitivity, particularly if it is around residential real estate, that is a political decision as much as financial system stability system, but models will give you the wrong answer in a rising market.

Q3855 Lord Lawson of Blaby: That being so, a lot more depends on the quality of the regulatory staff themselves.

Douglas Flint: Yes.

Q3856 Lord Lawson of Blaby: We have had a lot of evidence, including evidence from a very experienced banker who has also worked at the New York Fed and the Bank of England. Her view is that a lot of the problem was the poor quality of the regulatory staff, who were inadequate in quality and untrained. She suggested, for example, that one of the things that they should be trained in is the history of financial crashes, because there is a remarkable resemblance every time. I doubt whether any of the regulatory staff have any awareness of the history. You have pointed out how Singapore has had quite a good record by paying their regulatory staff what you might call competitive salaries. Do you think we would be well advised to go this sort of route ourselves-get better people, better trained, better paid?

Douglas Flint: Yes. I also think two other things. I think that to some extent where we are going is a product of history. If the role of the supervisor is much more to be high-level, understanding of the risks, and dealing at a senior level in organisations, that is a much more interesting career than the kind of much more minute data gathering, sticking it into a matrix and determining whether it comes out right. I think if the intellectual quality of the role is higher you will get better people.

I also think our industry should be challenged to support regulators by having an ongoing programme of secondment. When I was in accounting, the big four firms-actually, in my day it was the big eight firms, or even 10, I think-anyway, we used to routinely second people to the Bank of England to work in supervision. I suspect they still do it. I think it is incumbent on us as an industry to find a way-confidentiality respecting-to second good people into the regulatory world. The Fed, for example, in the States, has benefited from taking people from the banks into the Fed, and indeed, many people in the banks today have had their two or three years at the Fed. So I think we should support it; but yes, we do need a middle tier within our regulators that is capable of having strong, mature, experienced discussions with the people that report to Stuart.

Stuart Gulliver: I agree with what Douglas is saying. What I would say, though, is that within the British civil service you have got some incredibly talented people sitting in the Treasury, sitting in the Bank of England, sitting in the Foreign Office, so there is no shortage of people happy to do national service; but if there is a belief that to get some market expertise across we should go the Hong Kong, Thai or Singapore route, which is to pay a market-based salary, the way you could finance it is out of the bank levy, so actually it is a specific budget that is created-the industry pays for its own supervision, indirectly and at arm’s length.

Q3857 Chair: When you retire, Mr Flint, do you think you would be interested in joining a team of like-minded characters to assist the PRA in thinking through how they could better do their job?

Douglas Flint: I am sure I would give it favourable consideration if anyone ever thought I was up to it.

Q3858 Chair: Do you think that the idea that people who, frankly, enjoy the kind of remuneration that you have, might owe a societal duty to put something back in that way? Do you think it would be good if the industry developed that approach, where the people who are probably best placed of all to advise on how to handle risk discovery, towards the end of their careers, put it back?

Douglas Flint: Personally, yes I do.

Stuart Gulliver: Yes, I think it is totally reasonable to expect them.

Q3859 Chair: In these exchanges we have been discussing almost entirely-although just at the start you touched on other areas-prudential supervision, rather than the conduct side. You did refer to data collection as a general problem-misplaced data collection. That is one of the problems on the conduct side, is it, as well?

Douglas Flint: Yes. It transcends the whole supervisory-part of it is European; but, yes, I think on the data collection side one of the frustrations we have had in the past is short time scale for collection of data, without necessarily understanding the purpose for which it is to be used; because if we had understood the purpose we could have helped.

Also, one understands the driver is being asked to format information which we collect in a different format for our own purposes, to a standard format to enable some kind of aggregation to be done in this country, or indeed even broader, into Europe, or whatever. If one is given enough notice and given the purpose, one can work with people to say, "Well, there is a much better way of looking at this, so we can give you 90% of what you want very quickly, but the last 10% is tricky"; but in most cases 90%-in fact, in most cases 80% or 70%-is kind of what you need.

Q3860 Chair: What you are saying is that there is a lot of information collected that might not necessarily add much, if anything at all.

Douglas Flint: It is not clear to us because in many cases we don’t necessarily get the feedback from how that data was used.

Q3861 Chair: And you need more feedback on what use it is put to?

Douglas Flint: We would love to record the iterative process. We are trying to deal with this. Help us do that and we are very keen to support it.

Q3862 Chair: Looking at one specific area that the conduct regulators are becoming interested in, are you optimistic that product regulation will benefit consumers in the long run?

Douglas Flint: I have a personal view that some kind of product regulation that creates kitemark products is worthy of exploration. There needs to be some kind of safe harbour. If a product has been designed to a regulatory-approved specification and if the sales standards and practices designed for that product are followed, in terms of identifying to whom it should be sold and how it should be sold, there ought to be a confident belief that that works. The industry deals with situations where it has been challenged that it has failed in its sales practices over a long period of time. We have to try and find a way for our own, and society’s, protection to get to these issues earlier, and the public can be confident that we have a suite of products that do what they say on the tin.

Q3863 Chair: What about product regulation across the board? What that says is that there should be a set of kitemarked, safe-harbour products: a safe harbour both from the perspective of the firm and the customers.

Stuart Gulliver: From both. We have to think, "What in 2023 that we sold today in 2013 will be deemed to have been inappropriate?" That leads to all sorts of issues around financial exclusion, if you think about it, because it means firms back away. We hope that the FCA can provide kitemark-stamped products that, yes, protect the industry, because they protect the consumer and cannot be unpeeled in 10 years’ time.

Douglas Flint: We have a better job to do as an industry in terms of being much more transparent about risk. If you go into a supermarket and buy a tin of beans, you know exactly what you are getting. If you don’t like the beans you don’t go back there and buy them again. Apart from a simple deposit account or a simple loan, people buy things that are inherently risky. If you are trying to save for retirement, saving to buy a car or to pay for education, a lot of things can happen over that time which are simply environmental factors that may mean your expectations are unfulfilled. That does not necessarily mean there is a problem, just that life did not work out the way the base case was designed. There is an education issue.

It is fair to say that our industry has to be clearer in saying to people that the longer the duration of the product and the more the product is based on elements with uncertain cash flows, like equities, there is more of a risk. Having said that, there is an opportunity for a higher reward, but as long as you understand that, and it is incumbent upon us to demonstrate that they understand it, then that is okay. So many of the issues that we deal with are where people have had an outcome that is not the one they would have wished for, and they immediately say that there must be something wrong. Sometimes that is the case and sometimes it is not the case, it is just that the environment has changed.

Q3864 Lord Lawson of Blaby: I would like to go back to the culture dimension. Mr Gulliver said-I think I quote him correctly-"banks exist to finance business." I take it that this was not merely a kind of definition, but also a cultural statement. Mr Flint began by talking about the importance HSBC attaches to culture. I think we would agree with that, and it is what we were set up to look at. Given that, from a cultural view do you think there is a case for prohibiting regulated banks from engaging in proprietary trading, appropriately defined, and including high-frequency trading, which is also basically proprietary trading, in order to make it less likely that the culture of the bank as a whole will be subverted?

Douglas Flint: There is clearly a case, because it has been the way that America has gone with Volcker and the way that Europe has gone with the French and German proposals. We have gone a different way in this country with ring fencing, which says that the ring-fenced activities clearly cannot-

Q3865 Lord Lawson of Blaby: But these are for two different purposes. The ring fence is a very sensible mechanism to improve financial stability. We have suggested how it should be reinforced. What I am talking about now is the cultural issue.

Douglas Flint: There is a case for not having proprietary trading properly defined within regulated entities or groups.

Q3866 Lord Lawson of Blaby: Is that workable?

Stuart Gulliver: I would agree. In the purest sense, proprietary trading is almost asymmetric. The rewards effectively go to the staff members out of all proportion rather than necessarily even to shareholders. But it is the definition around it that is so critical. Paul Volcker has generally captured the issue, which is that you need to have market making and to be able to pre-empt what you think clients may do, so that you can hedge it. Let me describe it. If you are going to do a big push on 10-year mortgages, you may take some 10-year fixed money, but you are not going to know how much you are going to lend out to start with. That should not be prohibited, but it is technically proprietary risk. As long as we can distinguish proprietary risk, which is principle risk, from proprietary trading, then, yes, it should sit in a hedge fund.

Q3867 Chair: Do you think the Americans are going to solve these definitional problems?

Douglas Flint: Yes.

Stuart Gulliver: I think we can get 90% of the way.

Q3868 Chair: Do you think that we in the UK should have a Volcker rule in addition to the ring fence and bring it in at the same time?

Douglas Flint: I do not see any reason why one could not. I personally like the reverse Volcker-that almost sounds like a Tom Daley dive. What the French and Germans-

Q3869 Chair: You prefer Liikanen.

Douglas Flint: Yes. What the French and Germans have done is to define that which is not proprietary trading and then, by definition, everything else is. I think that, in many ways, is easier to apply.

Stuart Gulliver: That will capture some of the real interest rate liquidity risks that you have to do in a bank that may not have the final customer or the final tranche in sight. It is actually proprietary risk, but not proprietary trading. The French and German definition would get us closer.

In terms of your question about having these types of rules alongside the ring fence, I think that is fine provided that we get the definition right.

Douglas Flint: I also believe that the non-ring-fenced bank is systemically important. Therefore, if you believe that one of the risks to the non-ring-fenced bank is proprietary trading, you should not want it there, because the non-ring-fenced bank is going to be systemically important to the payments systems, to large corporates, to credit creation-

Stuart Gulliver: And the non-ring-fenced bank will need to borrow, will need the best credit rating it can get and so on, so it is going to need to convince bondholders, and indeed the overarching equity holders of the group, that it does not contain high parts of proprietary risk. The structure of the non-ring-fenced bank and its systemic importance would lead you this way.

Chair: We have ranged very widely and have ended up in a well-travelled place for this Committee with discussion of the Volcker rule. You have given us extremely interesting and thoughtful evidence. We are very grateful. We may come back to you with further requests for written material. Thank you very much indeed.

Prepared 12th February 2013